MACROECONOMICS

Jagdish Handa
McGill University, Canada

World Scientific
NEW JERSEY • LONDON • SINGAPORE • BEIJING • SHANGHAI • HONG KONG
• TA I P E I • CHENNAI

2
Published by
World Scientific Publishing Co. Pte. Ltd.
5 Toh Tuck Link, Singapore 596224
USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601
UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

MACROECONOMICS
Copyright © 2011 by World Scientific Publishing Co. Pte. Ltd.
All rights reserved. This book, or parts thereof, may not be reproduced in
any form or by any means, electronic or mechanical, including
photocopying, recording or any information storage and retrieval system
now known or to be invented, without written permission from the
Publisher.

For photocopying of material in this volume, please pay a copying fee

through the Copyright Clearance Center, Inc., 222 Rosewood
Drive, Danvers, MA 01923, USA. In this case permission to photocopy is
not required from the publisher.

ISBN-13 978-981-4289-44-3
ISBN-10 981-4289-44-2

3
Typeset by Stallion Press
Email: [email protected]

4
 To Sushma,
And our sons: Sunny and Rish,
And our delightful grandchildren: Riley, Aerin and Devyn

5
 Contents
Dedication

Preface

About the Author

Glossary of Symbols

Useful Mathematical Symbols and Formulae for Economics

Part I Introduction to Macroeconomics

1 Output, Unemployment, and the Basic Concepts
1.1 Introduction to Macroeconomics
1.1.1 The nature of macroeconomic analysis
The classification of goods in macroeconomics
Closed economy versus open economy analysis
Short-run macroeconomic theories, growth theories and business cycle
theories
1.2 The Classification of Economic Agents and Markets in Short-Run
Macroeconomic Models
Box 1.1: The Analytical Devices of the Short-Run Versus Long-Run
Analysis
Real Time Concepts: The Short Term and the Long Term
1.3 Introduction to AD-AS Analysis
1.3.1 The supply of commodities
The long-run (equilibrium) aggregate supply of commodities
The short-run aggregate supply of commodities
1.3.2 The components of aggregate demand
1.3.3 The diagrammatic AD-AS analysis
The possibility of disequilibrium in the economy
The actual aggregate supply of commodities
The impact of an increase in aggregate demand
The impact of an increase in the long-run productive capacity of the
economy
The stability of equilibrium
1.4 The Relationship between Output, Employment, and Unemployment in the
Economy
The labor force
Unemployment
1.5 Measures of National Output and Expenditures

6
 1.5.1 Gross domestic product (GDP)
Fact Sheet 1.1: Global Trends in GDP Per Capita, 1950–2003
1.5.2 Gross national product (GNP)
1.5.3 Net domestic product (NDP) and net national product (NNP)
Fact Sheet 1.2: Comparing GDP and GNP, 1965–2004
1.5.4 Measuring GDP
Box 1.2: GDP per capita as a Measure of the Standard of Living
GDP as a Measure of Welfare
1.6 Measuring the Price Level and the Rate of Inflation
1.6.1 Measures of the price level
1.6.2 The inflation rate
Fact Sheet 1.3: Inflation in Canada, 1915–2007
1.6.3 Core inflation
Mathematical Box 1.1: Calculation of the Price Index and Growth Rates
The construction of a price index: an illustration
Differences between Laspeyres and Paasche indices
Separating the rate of growth of nominal income into the real growth
rate and the inflation rate
1.6.4 Deriving the rate of inflation from a price index
Box 1.3: Mathematical Formulae to Learn
Fact Sheet 1.4: Measures of Inflation for the USA, 1960–2007
1.6.5 Disinflation versus deflation
1.7 Nominal Versus Real Output
1.8 The Economic Relationship between Real Output and Inflation
1.9 The Nature of Economic Relationships
Box 1.4: Definition of Equilibrium
Equilibrium conditions versus identities
Equilibrium versus disequilibrium
Stable versus unstable equilibrium
1.10 Exogenous and Endogenous Variables and the Concept of Shocks in
Macroeconomics
1.10.1 An illustration
1.10.2 Shocks
1.10.3 Multipliers
1.11 Growth Theory
1.11.1 Growth of the standard of living
1.12 Business Cycle Theories
Fact Sheet 1.5: Booms and Recessions in the USA Since 1960
Box 1.5: The Fundamental Role of Economics as a Science
Theories/models in economics
Economics as the ‘premier social science’
1.13 Paradigms in Macroeconomics

7
 1.14 Economic and Political Systems: Organization of the Macroeconomy
1.14.1 Capitalism
1.14.2 Marxism and communism
1.14.3 Socialism
1.15 Conclusions
Key Concepts
Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions
2 Money, Prices, Interest Rates, and Fiscal Deficits
2.1 What Is Money and What Does It Do?
2.1.1 The functions of money
2.1.2 The practical definitions of money
2.2 Money Supply and Money Stock
2.3 The Nominal Versus the Real Value of the Money Supply
2.4 Bonds and Stocks in Macroeconomics
2.5 The Definition of the Money Market in Macroeconomics
2.6 A Brief History of the Definition of Money
2.7 The Current Definitions of Money and Related Concepts
Extended Analysis Box 2.1: Current Meanings of the Symbols for the
Monetary Aggregates in Selected Countries
The monetary aggregates for USA
The monetary aggregates for the Canada
The monetary aggregates for the UK
Fact Sheet 2.1: Monetary Aggregates of the USA
2.8 The Monetary Base and Bank Reserves
2.8.1 The relationship between the monetary base and the money supply
2.9 The Quantity Equation
2.9.1 The quantity equation in growth rates
2.9.2 The implications of the quantity equation for a persistently high inflation
rate
Fact Sheet 2.2: Money Growth and Inflation in the USA, 1960–2008
2.10 The Quantity Theory
Extended Analysis Box 2.2: The Difference between the Quantity
Theory and the Quantity Equation
2.10.1 The transactions approach to the quantity theory
The adjustment period relevant to the quantity theory
Is velocity constant over time ?

Fact Sheet 2.3: Velocity of Money in the USA, 1960–2008

2.11 The Definitions of Monetary and Fiscal Policies
2.12 The Central Bank and Monetary Policy
2.13 The Economic Aspects of the Government and Fiscal Policy
2.13.1 The financing of fiscal deficits/surpluses and changes in the money
supply

8
 The implications of the independence of the central bank from the
government for financing deficits
2.13.2 The public debt
2.13.3 The selective nature of government expenditures, taxes, and subsidies
2.14 Interest Rates in the Economy
2.14.1 The Fisher equation on interest rates
Fact Sheet 2.4: Nominal and Real Interest Rates in the USA, 1982–2008
2.14.2 The concept of present discounted value (PDV) of a bond
2.14.3 Bubbles in asset prices
The importance of asset bubbles for output and business cycles
Box 2.1: The Determination of Stock Prices
Bubbles in house and land prices
A bubble in tulip bulb prices!
Conclusions
Key Concepts
Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions
3 Introduction to the Open Economy: Exchange Rates and the Balance of Payments
3.1 Exchange Rates
3.1.1 Three concepts of exchange rates
The (nominal) exchange rate
The real exchange rate
The effective exchange rate
3.2 Fixed, Flexible, and Managed Exchanged Rates
3.3 Purchasing Power Parity (PPP) as a Theory of the Exchange Rate
3.3.1 PPP at the level of a single commodity
Absolute PPP among countries
Extended Analysis Box 3.1: Does PPP Apply in the Real World? An
Illustration
3.4 Relative PPP and Shifts in the Relative Efficiency of Economies
3.4.1 Long-run changes in relative PPP
3.4.2 Short-run changes in relative PPP
3.4.3 Implications of short-run PPP for exchange rates and inflation rates
Box 3.1: International Comparisons of Standards of Living in Terms of
PPP
3.5 Interest Rate Parity (IRP) and the Determination of the Exchange Rate
3.5.1 The benchmark IRP theory
3.5.2 The benchmark IRP as a determinant of the domestic interest rate
3.5.3 The benchmark IRP as a theory of the exchange rate under flexible
exchange rates
3.5.4 The role of speculative returns to stocks in capital flows and IRP
3.5.5 Extending IRP to incorporate risk factors and risk aversion
Fact Sheet 3.1: Interest Rate Differentials Between Countries

9
 3.5.6 The relative importance of PPP and IRP in determining exchange rates
3.6 The Balance of Payments
3.6.1 The components of the balance of payments
Fact Sheet 3.2: United States Balance of Payments, 1976–2008
3.6.2 Equilibrium in the balance of payments
3.6.3 The change in foreign exchange reserves
Fact Sheet 3.3: US Foreign Exchange Reserves and Balance of
Payments, 2005–2008
Equilibrium in the balance of payments
Foreign exchange reserves and short-term bonds
3.7 The Balance of Payments in an Accounting Sense
3.8 The Market for Foreign Exchange and the Changes in Foreign Exchange
Reserves
3.8.1 The demand and supply of foreign exchange
3.8.2 Equilibrium in the foreign exchange market
Extended Analysis Box 3.2: The Demand and Supply of Foreign
Exchange Stated in Foreign Currencies
3.9 The Market Determination of the Nominal Exchange Rate
Fact Sheet 3.4: Exchange Rates against the US Dollar, 1980–2008
3.9.1 Hot money
Box 3.2: National Policies on the Balance of Payments and
Accumulation of Foreign Exchange Reserves
3.10 The Persistence of Balance of Payments Deficits and Surpluses
Conclusions
Key Concepts
Summary of Critical Conclusions
Extended Analysis Box 3.1.A: Comparison of the Actual and PPP Costs
of the Big Mac
Review and Discussion Questions
Advanced and Technical Questions

Part II Short-run Macroeconomics

4 Determinants of Aggregate Demand: The Commodity Market of the Closed Economy
4.1 Symbols Used
4.2 The Commodity Sector of the Closed Economy
4.2.1 Uses of national income
4.2.2 Sources of national expenditures
4.2.3 Equilibrium in the commodity market
4.2.4 National saving
4.2.5 The relationship between saving and investment
4.2.6 The (physical) capital stock
Box 4.1: An Unjustified Oversimplification for a Modern Economy
4.3 The Two Uses of Private Saving and the Drag of Deficits on Investment

10
4.4 Disequilibrium and the Role of Changes in Inventories in the Adjustment to
Equilibrium
4.4.1 The role of unintended changes in inventories
Extended Analysis Box 4.1: A Simplified Diagrammatic Analysis: The
45° Diagram
4.4.2 Limitations of the 45° diagram for macroeconomic analysis
4.5 For the Macroeconomic Analysis of the Closed Economy, Is There a National
Income Identity and One Between National Saving and Investment?
4.5.1 An accounting national income identity
Extended Analysis Box 4.2: The Distinction Between the Meanings of
Investment in Macroeconomics
The actual change in the capital stock and the definitions of investment
4.6 Demand Behavior in the Commodity Market
4.6.1 Consumption expenditures
Fact Sheet 4.1: Consumption and Disposable Income in the USA, 1980–
2008
Fact Sheet 4.2: Interest Rates and Saving in the USA, 1985–2008
4.6.2 The saving function
Extended Analysis Box 4.3: The Dependence of Consumption on
Wealth, Interest Rates
The concept of lifetime wealth as the present discounted value of
future incomes
4.6.3 Investment expenditures
4.6.4 Government expenditures and tax revenues
Fact Sheet 4.3: Interest Rates and Investment in the USA, 1960–2008
Extended Analysis Box 4.4: A more Realistic Investment Function
The impact of business confidence on investment
Fact Sheet 4.4: USA Fiscal Deficit, 1962–2008
4.6.5 The commodity market and the price level
4.7 The Commodity Market Model: The IS Equation/Curve
The impact of investment fluctuations on income: a partial investment
multiplier
Box 4.2: The Mechanism of the Investment Multiplier: An Illustration
The impact of fiscal policy on income: partial fiscal policy multipliers
Box 4.3: The Impact of a Balanced Budget: The Balanced Budget
Multiplier
A word of caution on IS multipliers
The IS curve
Shifts in the IS curve versus movements along it
Extended Analysis Box 4.5: The Slope of the IS Curve
4.8 Conclusions
Key Concepts

11
 Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions
5 Aggregate Demand in the Open Economy under an Interest Rate Target: Is-IRT
Analysis
5.1 The Number of Goods and Markets in the Open Economy
5.2 The Foreign Exchange Sector of the Open Economy and the Balance of
Payments, Review
5.2.1 Net interest payments and net transfer payments
5.2.2 Relationship between nominal and real interest rates
5.3 The Commodity Market of the Open Economy
Fact Sheet 5.1: Components of Aggregate Demand for USA, Canada,
and Thailand, 2007
5.3.1 The uses of private saving in the open economy
5.3.2 Three gaps: saving, fiscal, and external
Fact Sheet 5.2: The uses of private saving in the USA, 1980–2008
5.3.3 Commodity market equilibrium and capital flows
Extended Analysis Box 5.1: Financing high levels of domestic
investment during the development stages and foreign exchange crises
Foreign exchange crises and debt forgiveness
5.3.4 The open economy IS relationship
The complete model of the commodity sector for the open economy
Box 5.1: Derivation of the Open Economy is Equation
5.3.5 The open economy IS curve
5.3.6 Shifts in the open economy IS curve
Extended Analysis Box 5.2: The Mathematical Derivation of the Slope
of the IS Curve
Changes in the slope of the IS curve as the economy becomes more
open
Comparing the magnitudes of the fiscal multipliers for open economies
The stability of aggregate demand in more open economies
5.3.7 The impact of exchange rate changes on the IS curve
5.4 The Formulation of Monetary Policy
Box 5.2: Money Supply, the Stock of Money and Their Relationship
with the Interest Rate
5.4.1 Monetary policy through interest rate targeting
5.4.2 Rules versus discretion in setting the target interest rate
5.4.3 Diagrammatic depiction of the interest rate target
5.5 The Determination of Aggregate Demand under Interest Rate Targeting
5.5.1 Diagrammatic derivation of the AD curve
5.5.2 The downward slope of the AD curve in the open economy under an
exogenous interest rate target: a reiteration
5.6 The Policy Multipliers for the Open Economy Aggregate Demand

12
 5.6.1 The impact of investment fluctuations on aggregate demand: the
investment multiplier
5.6.2 The impact of fiscal policy on aggregate demand: the fiscal multipliers
5.6.3 The impact of monetary policy on aggregate demand: the target rate
multiplier
5.6.4 The distribution of incomes, consumption patterns, and the size of the
multiplier
5.6.5 The impact of exports on domestic aggregate demand
5.6.6 The lag in the impact of changes in the interest rate target
5.7 Diagrammatic Analysis of Fiscal and Monetary Policies
5.8 The IS, IRT Curves and the Determination of Output: A Caveat
5.9 But What about the Monetary Sector and Its Money Demand and Supply
Functions?
5.9.1 The demand for money
5.9.2 Ensuring equilibrium in the money market under an interest rate target
and the determination of the money supply
5.10 Managing Aggregate Demand in the Open Economy: Monetary and
Fiscal Policies under Flexible Exchange Rates
5.10.1 The effectiveness of fiscal policy for the open economy under interest
rate targeting
5.10.2 The effectiveness of monetary policy for the open economy under
interest rate targeting
5.10.3 Limitations on the effectiveness of a policy of interest rate targeting
5.11 Internal Versus External Balance
Extended Analysis Box 5.3: The Mundell-Fleming Model of the Open
Economy
5.12 Conclusions
Key Concepts
Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions
6 Aggregate Demand Under a Money Supply Operating Target: IS-LM Analysis
6.1 Monetary Policy
6.1.1 Reasons for choosing interest rate targeting over money supply targeting,
or vice versa
Lags in the impact of money supply changes versus those of interest rate
changes
Instability of money demand
The informal financial sector and black money in developing economies
6.1.2 Choosing the monetary aggregate as the target variable
6.2 The Demand for Money
6.3 The Motives for Holding Money
6.3.1 The volatility of the speculative demand for money
6.3.2 Other reasons for the volatility of the demand for money
6.4 The Standard Money Demand Function
Extended Analysis Box 6.1: Special Cases of Money Demand

13
 6.4.1 Equilibrium in the money market
Disequilibrium in the money market
6.5 The LM Equation
6.5.1 The LM curve
6.5.2 Shifts in the LM curve
6.5.3 Shifts in the LM curve versus movements along it
6.5.4 Final comments on the LM curve
Extended Analysis Box 6.2: The potential general shape of the LM curve
6.6 Deriving the Aggregate Demand for Commodities by Combining the IS and LM
Curves
6.6.1 The IS equation for the commodity market equilibrium
6.6.2 The relationship between the nominal and real interest rates
6.6.3 Diagrammatic determination of aggregate demand
The IS, LM curves and the determination of output: a caveat
6.7. The Impact of Expansionary Monetary and Fiscal Policies on Aggregate Demand
6.8 Bringing Aggregate Supply into the Open Economy Analysis, a Preview of
Chapters 7 to 9
6.8.1 The impact of an increase in aggregate demand on the quantity supplied
and the price level
6.8.2 The impact of monetary and fiscal policies on equilibrium output and
price level in the open economy
The short-run impact of aggregate demand changes on output and the
price level
The long-run impact of aggregate demand changes on output and the
price level
6.8.3 Disequilibrium in the domestic economy and stabilization through
monetary and fiscal policies
6.8.4 Summation on the roles of monetary and fiscal policies under flexible
exchange rates
Extended Analysis Box 6.3: The Mundell-Fleming Model of the Open
Economy
6.9 The Central Bank's Control Over the Money Supply
6.10 The Central Bank's Instruments for Changing the Monetary Base
6.10.1 Open market operations
6.10.2 Shifting government deposits between the central bank and the
commercial banks in Canada
6.10.3 A mechanism for commercial banks to change the monetary base
6.11 The Central Banks' Control Over the Monetary Base Multiplier through Reserve
Requirements
6.12 The Impact of Discount Rate Changes on the Economy
6.12.1 The central bank's discount rate and interest rate differentials in the
economy
6.13 The Determination of the Money Supply
Extended Analysis Box 6.4: The Creation of Demand Deposits
6.14 A Common Money Supply Formula for M1
6.14.1 The monetary base multiplier
6.14.2 Numerical examples

14
 6.15 Conclusions
Key Concepts
Critical Conclusions
Appendix
Determination of the LM curve
Derivation of aggregate demand for the closed economy from its IS and
LM equations
Intuition
Simplifying the AD equation for further analysis
Determination of the long-run price level under an exogenous money supply
Derivation of aggregate demand for the open economy from its IS and LM
equations
Review and Discussion Questions
Advanced and Technical Questions
7 Full-Employment Output and the Natural Rate of Unemployment
7.1 The Production Function
Fact Sheet 7.1: Diminishing Marginal Product of Labor
Mathematical Box 7.1: Examples of Production Functions and the
Derivation of the Marginal Product of Labor
The general procedure for deriving the MPL from the production function
7.2 The Labor Market
7.2.1 Demand for labor
Diagrammatic analysis
7.2.2 Supply of labor
Extended Analysis Box 7.1: The Intertemporal Analysis of Labor Supply
The Backward Bending Labor Supply Curve Over Long Periods
7.3 The Long-Run Equilibrium Levels of Employment and Output
7.3.1 The impact on long-run output of an increase in the price level or
inflation rate
7.3.2 The diagrammatic analysis of employment and output
The effect on output of an improvement in labor productivity
7.3.3 The impact of aggregate demand on long-run output
The pursuit of fiscal policies
The pursuit of expansionary monetary policies
7.3.4 The ineffectiveness of monetary and fiscal policies in LR equilibrium
Mathematical Box 7.2: The Derivation of the Demand for Labor,
Employment, and Output
The role of aggregate demand in determining output and the price
level, an illustration
The impact of expansionary monetary and fiscal policies
7.4 The Effects of an Increase in the MPL on LR Output and Employment
7.4.1 Diagrammatic analysis of the impact of an improvement in technology on
output and employment
7.5 The effects of an Increase in Labor Supply on LR Output, Employment and Price

15
 Level
Mathematical Box 7.3: The Effect of an Improvement in Technology
7.6 Conclusions on Changes in the Equilibrium Levels of Employment and Output
7.7 Full Employment and the Full-Employment Level of Output: Definitions and
Conditions
7.8 The Role of Supply-Side Policies in Changing Full-Employment Output
7.9 Stagflation
Fact Sheet 7.2: The Price of Oil in the USA, 1960–2008
Box 7.1: The Impact of Oil Price Shocks on Full Employment and Full-
Employment Output in Oil Importing Countries
The impact of oil price increases on oil producing countries
Can expansionary monetary and fiscal policies eliminate this
stagflation?
Extended Analysis Box 7.2: The Role of Demand-Side Policies in
Changing Full-Employment Output
An implicit assumption in the above conclusions
7.10 Crowding Out of Investment by Fiscal Deficits, Given the LR Supply of Output
in the Closed Economy
7.11 The Actual Level of Output in the Economy
7.11.1 Changes in the actual rate of output over time
7.12 The Rate of Unemployment
7.12.1 The LR equilibrium (natural) rate of unemployment
Shifts in the natural rate of unemployment
The lack of impact of monetary and fiscal policies on the natural rate of
unemployment
7.13 The Long-Run Equilibrium (Natural) Rate of Interest
The ineffectiveness of monetary policy and inflation in changing the long-run
real interest rate
7.14 Conclusions
Key Concepts
Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions
8 Output in the Short Run: The Role of Expectations and Adjustment Costs
8.1 The Role of Uncertainty and Errors in Expectations
8.1.1 The theory of rational expectations
8.1.2 Random errors and their predictability
8.1.3 Application of rational expectations to monetary and fiscal policies
8.2 Price Expectations and the Labor Market: Output and Employment in the Context
of Wage Contracts
8.2.1 Labor supply in wage negotiations during the first stage
8.2.2 Employment during the contract period (the production stage)
8.2.3 Diagrammatic analysis

The effect of a proportional increase in the expected and actual price

16
 levels
Conclusions from the contract wage analysis of production
The expectations augmented employment and output supply curves
Implications of the expectations augmented employment and output supply
equations for variations in employment and output over the business cycle
Box 8.1: Errors in Price Expectations, the Duration of the Wage Contract
and Cost-of-Living Clauses
8.3 Friedman's Expectations Augmented Employment and Output Rules
Extended Mathematical Analysis Box 8.1: Labor Demand when There
are Expectational Errors in the Context of Nominal Wage Contracts
The expectational equilibrium level of employment
The impact of expectational errors on employment
The Implications for Employment and Output of Wage Contracts with
Expectational Errors in Prices
8.4 Lucas Supply Rule with Errors in Price Expectations in Commodity Markets
8.4.1 Firms' responses to errors in expectations
Extended Analysis Box 8.2: Diagrammatic Analysis: Comparing the
FSR and LSR Curves (FSR and LSR) and the LRAS Curve
8.5 The Implications of the FSR and LSR for the Impact of Anticipated Demand
Increases
8.6 The Implications of the FSR and the LSR for the Impact of Unanticipated
Demand Increases
8.6.1 The impact of revisions in anticipations and real time
8.6.2 Practical aspects of the FSR and LSR analysis
Extended Analysis Box 8.3: The Implications of the FSR and LSR for
the Impact of Monetary and Fiscal Policies
The Implications of Rational Expectations for Systematic Monetary
and Fiscal Policies
The Scope for Monetary and Fiscal Policies for Stabilization when
Private Demand is Volatile
The Scope for Monetary Policies and the Political Economy of the
Government's Budget
8.7 FSR and LSR: The Impact of Anticipated Permanent Supply Changes on the
Economy
8.8 The Impact of Unanticipated but Permanent Supply Changes on the Economy
Box 8.2: Cost of Living Clauses
8.9 The Short- and Long-Term Relationships between Output and Inflation
8.10 Empirical Validity of the FSR and the LSR
8.11 Types of Adjustment Costs and Their Impact on Output
8.11.1 Firms' responses to increases in the demand for their products
8.12 Menu Costs and Price Stickiness as the Explanation of the SRAS Curve
8.12.1 Aggregate supply in the sticky-price hypothesis
8.12.2 Monetary policy and the sticky-price SRAS curve
8.13 Costs of Adjusting Employment: Implicit Contracts as an Explanation of the
SRAS Curve

17
 8.13.1 Variations in work effort over the short term
Box 8.3: Work Effort in Restaurants During the Day
Work Effort by Students over the Term and the Calendar Year
8.13.2 The Flexibility of the Employment-Output Nexus in the Short Term
Long-term labor contracts and labor hoarding
8.13.3 Okun's Rule: The Relationship between Unemployment and Output
Changes
8.14 The Implications of Adjustment Costs for Persistence of Output and
Employment Fluctuations Over the Business Cycle
8.15 Implications of Adjustment Costs for the Impact of Monetary and Fiscal Policies
8.16 Stagflation and the Recessions of 1973–1975 and 1980
Box 8.4: Empirical Evidence on Price Changes, Wage Contracts and
Output Response
8.17 Stylized Facts and Intuition on the Theories Explaining Variations in Output
8.18 Conclusions
Key Concepts
Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions
9 Actual Output, Disequilibrium, and the Interaction among Markets
Fact Sheet 9.1: Money Growth and Output Growth in USA, 1960–2008
9.1 The Relevance of Expectations on Variables Other Than Prices
9.2 Shocks to Aggregate Demand and Supply
Examples of shocks to consumption
Examples of shocks to investment and net exports
Shocks to money demand and supply
Shocks to the cost and availability of credit
9.2.1 Shocks to the aggregate supply of commodities

9.3 An Analysis of the Disequilibrium Following a Demand Shock

9.3.1 The assumptions for disequilibrium analysis
Effective aggregate demand and supply
Extended Analysis Box 9.1: The Dynamics Required for the Maintenance
of a Full-Employment Scenario
9.3.2 A plausible dynamic scenario
Initial effects of the emergence of a demand deficiency
Secondary effects of the demand deficiency
The transition from the old equilibrium to the new one
Output-price adjustment function [Optional]
Box 9.1: The Nature and Critical Role of Expectations in Dynamic
Adjustments
9.4 Diagrammatic Analysis Following a Fall in Aggregate Demand
Effective demand for labor
9.5 Disequilibrium with Flexible Prices and Wages
Extended Analysis Box 9.2: The Relative Rapidity of Commodity

18
 Markets in Adjusting Prices versus those by Firms and Consumers in
Adjusting Production, Employment, and Prices
The failure or sluggishness of the labor market in adjusting nominal
wages versus the faster responses of firms and workers in adjusting
employment and consumption
9.6 If the Real-World, Real-Time Economy is not on its LRAS or SRAS Curve,
Where will it be?
9.7 Can the Economy Get Stuck Below Full Employment?
Impact of the disequilibrium on the long run: hysteresis
9.8 Optimal Monetary and Fiscal Policies for the Demand-Deficient Economy
The use of expansionary monetary policies
Limits to the effectiveness of monetary policies in a demand-deficient recession
The use of fiscal policy: increases in government expenditures and/or cuts in
taxes
The policy dilemma
Fact Sheet 9.2: Economies in Disequilibrium: USA During the Great
Depression
Recapitulation of the roles of monetary and fiscal policies in a demand-
deficient economy
Extended Analysis Box 9.3: Disagreements Among Economists on the
Appropriate Policies in Real Time for a Real-World Economy
9.9 Excess Demand18 in the Economy and Appropriate Policies
9.10 Asymmetry in the Economy's Responses to Deficient and Excess Demand
9.11 An Analysis of Disequilibrium Following a Supply Shock
9.11.1 Supply shocks from labor productivity
The scope for monetary and fiscal policies
Box 9.2: Demand Shocks Emanating from Shifts in Long-Run Supply
9.11.2 The impact of a decline in credit supply on the effective supply of
commodities
9.12 ‘Crowding-Out’ or ‘Crowding-In' of Investment by Fiscal Expenditures in a
Demand-Deficient Economy?
9.13 Disequilibrium (Involuntary Unemployment) in the Labor Market due to a High
Real Wage
Expansionary monetary policy as a means of lowering the real wage and
restoring equilibrium in the labor market
Extended Analysis Box 9.4: Is the preceding demand-deficient analysis a
reflection of market failure or of markets' sluggishness in adjustment of
prices and nominal wages?
9.14 The Dual Implications of a High Saving Rate
The paradox of a high saving rate — also known as the paradox of thrift
9.15 Conclusions
Key Concepts
Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions
10 Employment, Unemployment, and Inflation

19
10.1 Definitions of the Labor Force and Labor Supply
10.1.1 Frictional unemployment and actual employment in labor market
equilibrium
The notions of matched labor supply and matched labor demand in the
presence offrictional unemployment
10.1.2 Discouraged workers
10.1.3 Other determinants of the equilibrium level of unemployment
Box 10.1: The Labor Force Participation Rate
The non-employment rate
Shifts in labor force participation rates in recent decades
Fact Sheet 10.1: Participation Rates in Canada and the United States,
Male and Female, 1980–2008
10.2 The Components of Unemployment
The diagrammatic depiction of the components of unemployment
10.3 Involuntary Unemployment
Involuntary unemployment due to a high real wage
Involuntary unemployment due to the emergence of a demand deficiency
Involuntary unemployment due to a credit crisis
The variation in the number of discouraged workers with the state of the
economy
10.4 The Actual Rate of Unemployment
Fact Sheet 10.2: Unemployment Rates in Selected Countries, 1985–2008
10.4.1 The division of unemployment into natural and cyclical unemployment
10.4.2 Changes in the actual rate of unemployment over time
Short-run and long-run variations in the natural rate of unemployment
Fact Sheet 10.4:
Estimating the natural rate of unemployment
Fact Sheet 10.5: Unemployment and Trend Unemployment in the United
States, 1980–2008
10.5 Policies to Reduce Structural Unemployment
Extended Analysis Box 10.1: Changes in Structural Unemployment
The impact of fiscal and monetary policies on structural unemployment
10.6 Policies to Reduce Frictional Unemployment
10.7 Measuring Involuntary Unemployment
Two ways of measuring involuntary unemployment
Monetary and fiscal policies to reduce involuntary unemployment
Box 10.2: Causes High Unemployment in Canada and Europe Relative
to the USA
10.8 The Costs of Unemployment
10.9 Relationship Between Unemployment and the Inflation Rate from the AD-AS
Analysis
10.10 Phillips Curve (PC)
10.10.1 Use of the Phillips curve as a policy trade-off
10.10.2 Instability of the Phillips' curve
10.11 Deviations of the SR Equilibrium Unemployment Rate from the Natural One

20
 Fact Sheet 10.6: Historical Phillips Curves in the United States, 1960–
1995
10.11.1 Friedman and the expectations augmented Phillips curve (EAPC)
10.11.2 Expectational equilibrium and the natural rate
Extended Analysis Box 10.2: The Friedman AND Lucas Supply Rules
Expectational errors and the commodity markets: The Lucas supply
rule
10.12 The Implications of the EAPC for Shifts in the Phillips Curve and for Policy
The relationship between the Phillips' curve and the EAPC
10.13 The EAPC and the Non-Accelerating Inflation Rate of Unemployment
10.14 The Implications of the EAPC for the Impact of Anticipated and Unanticipated
Demand Increases
10.15 The Determinants of Inflation
Fact Sheet 10.7:
Another way of stating the determinants of inflation
10.16 The Costs of Inflation
10.16.1 Inflation under perfect competition with fully anticipated inflation for
all future periods: the costs of inflation in the analytical long-run case
An intermediate scenario: anticipated inflation in the commodity and labor markets for
some quarters ahead

10.16.2 The costs of unanticipated inflation

A caveat
Box 10.3: The Impact of Hyperinflation on Long-Term Output Growth
and the Standard of Living: Practical Experience in Contrast to Theory
10.17 The Sacrifice Ratio
Fact Sheet 10.8: Sacrifice Ratio in the United States, 1950–2008
10.17.1 Gradualist versus Cold Turkey Policies of Disinflation
10.17.2 Indexation to the rate of inflation: should nominal wages and interest
rates be indexed?
10.18 Conclusions
Key Concepts
Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions
11 Paradigms in Macroeconomics
Stylized Facts on Money, Prices, and Output
11.1 An Analogy for the Two Main Paradigms in Macroeconomics
11.1.1 The approach of the classical paradigm to the pathology of the economy
11.1.2 The approach of the Keynesian paradigm to the pathology of the
economy
Extended Analysis Box 11.1: The Fundamental Assumptions of the
Classical Paradigm
11.2 Defining and Demarcating the Models of the Classical Paradigm

21
 11.2.1 The traditional classical approach
Extended Analysis Box 11.2: The Founders of the Classical Tradition in
Macroeconomics
David Hume (1711–1776)
Adam Smith (1723–1790)
David Ricardo (1772–1823)
11.2.2 The neoclassical model
Box 11.1: Some Major Misconceptions about Traditional Classical and
Neoclassical Approaches in the Classical Paradigm
11.2.3 The 1970s monetarism
11.2.4 The modern classical model
11.2.5 The new classical model
Extended Analysis Box 11.3: The Founders of the Classical Approach in
the Modern Period
Milton Friedman (1912–2006)
Robert Lucas (1937-)
Extended Analysis Box 11.4: The Economic Contributions of Milton
Friedman
11.2.6 Milton Friedman and the Keynesians
11.2.7 The relationship between the Monetarists and Friedman
11.2.8 Friedman and the modern classical school
11.3 The Keynesian Paradigm
Extended Analysis Box 11.5: The Founders of the Keynesian Paradigm
Knut Wicksell (1851–1926)
John Maynard Keynes (1883–1946)
11.3.1 Development of the Keynesian schools after Keynes
11.3.2 Frequent themes in the Keynesianparadigm
New Keynesian economics
11.3.3 The variety of Keynesian models
Monetary and fiscal policy effects in Keynesian models
11.3.4 The critical role of dynamic analysis when aggregate demand falls
Box 11.2: The Keynesian-Neoclassical Synthesis on Aggregate Demand
11.4 The Reformulation of Keynesian Approaches
11.4.1 NeoKeynesian building blocks for the reformulation of Keynesian
macroeconomics
11.4.2 Taylor rule and its incorporation into Keynesian macroeconomics
11.4.3 The new Keynesian model
11.5 The Credit and Economic Crisis of 2007–2010: Which Theory Can Explain It?
11.6 Which Macroeconomic Paradigm Should One Believe In and Use?
Box 11.3: The Anatomy of Two Quite Different Recessions: 1973–1975
and 2001–2002
11.7 Paradigms and Policies
11.7.1 Stabilization versus pro-active policies

22
 11.7.2 Rules versus discretion in the pursuit of policies
11.8 The Role of the Government in the Macroeconomy
11.8.1 Evolution of ideas about the role of the government
11.8.2 Classical and Keynesian approaches and the debates on the size of the
government
11.9 Conclusions
Keynesian versus classical economics in relation to the real world and real time
Key Concepts
Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions

Part III Topics in Open Economy Macroeconomics

12 The Foreign Exchange Market, IMF, and Globalization

12.1 Review of PPP
12.2 Review of Interest Rate Parity (IRP) Theory
12.2.1 IRP as a theory of the interest rate
The importance of stock market returns for capital flows: a problem for
the IRP theory
Fact Sheet 12.1: Interest Rate Differentials between Countries
Simplified form of the benchmark IRP for short-run analysis
Covered versus uncovered IRP
12.2.2 IRP as a theory of the exchange rate
The inconsistency between the IRP's implications and common
observations
12.2.3 The impact of expectations on exchange rates
12.3 PPP and IRP Combined
Long run analysis
Short run analysis
12.4 The Market for Foreign Exchange, a Review
Fact Sheet 12.2: Interest Rates and Net Capital Flows in the USA, 1985–
2008
12.4.1 The relationship between the balance of payments and (D$ — S$)
12.4.2 Diagrammatic analysis
12.5 Demand and Supply Elasticities
12.6 The J Curve for the Value of Net Exports in Real Time
12.6.1 Limitations of the commodity-based exchange market analysis
12.6.2 Policy implications of the J curve
Extended Analysis Box 12.1: The General Experience of Industrialized
Countries and LDCs on the J Curve
Implications of the above analysis for movements of exchange rates
under flexible exchange rates
12.7 Historical Experience with Flexible Versus Fixed Exchange Rates
12.8 Measures to Reduce Balance of Payments Deficits
12.9 The International Monetary Fund (IMF)

23
 12.9.1 The IMF and disequilibrium in the balance of payments of the member
countries
Extended Analysis Box 12.2: IMF Conditionality
For whose benefit are the IMF policies?
12.9.2 The IMF and capital flows
12.9.3 Special drawing rights (SDRs)
Extended Analysis Box 12.3: The World's Demand for Liquidity
Exchange and financial crises
Is the IMF a central banker for the world?
Does the world economy need a central bank?
The rising demand for the IMF to create a world currency
12.10 The International Transmission of Crises and Business Cycles
12.11 Economic Globalization
12.11.1 Causes of the push for global markets
12.11.2 Measures of the extent of globalization
12.11.3 Other aspects of globalization
12.11.4 Globalization and modernization
12.11.5 The persistence of nationalistic mercantilism — globalization limited
to truncated globalization
12.11.6 The possibility of net losses from globalization
12.11.7 Effects of globalization on commodity and capital flows
12.12 Conclusions
Key Concepts
Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions
13 The Open Economy Under a Fixed Exchange Rate Regime
13.1 The Balance of Payments (BP) Equilibrium Curve Under Fixed Exchange Rates
13.2 The IS-LM Model for the Open Economy Under Fixed Exchange Rates
13.2.1 The slope of the BP curve
13.2.2 The relative slopes of the LM and BP curves
Extended Analysis Box 13.1: Differences between the IS-LM Diagrams
for the Flexible and Fixed Exchange Rates
13.2.3 General equilibrium under fixed exchange rates, a diagrammatic
treatment
13.3 Managing Aggregate Demand in the Open Economy: Monetary and Fiscal
Policies Under Fixed Exchange Rates
13.3.1 The impact of an expansionary fiscal policy on aggregate demand
13.3.2 The impact of an expansionary monetary policy on aggregate demand
Extended Analysis Box 13.2: Monetary and Fiscal Policies in the Case of
Interest-Insensitive Capital Flows and Fixed Exchange Rates
13.4 Bringing Aggregate Supply into the IS-LM Analysis with a Fixed Exchange
Rate
13.5 Macroeconomic Equilibrium in the Small Open Economy with Full
Employment and IRP

24
 13.5.1 Fiscal policies under IRP and full employment
13.5.2 Monetary policy under IRP and full employment
13.6 PPP, the World Rate of Inflation and the Convergence in National Inflation
Rates
13.6.1 The world rate of inflation in the fixed exchange rate case

Box 13.1: The impact of a large economy on smaller economies

13.7 Economic Blocs, Fixed Exchange Rates and the Convergence in Inflation Rates
13.8 Disequilibrium in the Domestic Economy and Stabilization through Monetary
and Fiscal Policies: The Fixed Exchange Rate Case with Interest Sensitive
Capital Flows
A caveat
Extended Analysis Box 13.3: Disequilibrium in the Domestic Economy
and Stabilization through Monetary and Fiscal Policies: The Fixed
Exchange Rate Case with Interest Insensitive Capital Flows
13.9 The Need for Monetary and Fiscal Policies under Fixed Exchange Rates
13.9.1 Internal versus external balance
13.9.2 A note of caution on the use of monetary and fiscal policies
13.10 The Effect of Relatively Different Improvements in Productivity at Home and
Abroad under Fixed Exchange Rates
13.11 The Political Economy of Exchange Rates: The Choice between Fixed and
Flexible Exchange Rate Regimes
13.12 Other Tools for Handling Balance of Payments Deficits
Extended Analysis Box 13.4: The Gold Standard and the Gold Exchange
Standard
13.13 Summary of the Costs and Benefits of a Fixed Exchange Rate
13.14 Dollarization
13.14.1 Dollarization and different productivity growth rates
Extended Analysis Box 13.5: A Dual Currency System: Dollarization
Along with a Separate National Currency
13.15 Currency Boards
13.16 Conclusions
Key Concepts
Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions

Part IV Growth Economics

14 Classical Growth Theory
Fact Sheet 14.1: Long-Term Real GDP Growth in Selected Countries,
1960–2003
Fact Sheet 14.2: Real GDP Growth Rates in the USA
Extended Analysis Box 14.1: Setting the Boundaries of Macroeconomic

25
 Growth Theory
14.1 The Classical (Solow's) Growth Model's Assumptions
14.1.1 The technology of production
14.1.2 Saving, investment and the change in the capital stock
14.1.3 Labor force growth
14.2 The Analysis of the Solow Model
14.3 Diagrammatic Analysis of the Solow Model
14.3.1 The SS growth rate of output
Extended Analysis Box 14.2: Variability of the Saving Rate
14.3.2 The impact of shifts in the saving rate
Mathematical Box 14.1: Numerical example A
14.3.3 The impact of shifts in the labor force growth rate
Mathematical Box 14.2: Numerical example B
14.4 The Growth Implications of a More General Production Function
14.5 Convergence Versus Divergence in Output per capita among Countries
14.6 Assessing the Importance of Technical Change in the Solow Model16
14.6.1 Solow's estimates of the contribution of technical change to the
improvement in living standards
Mathematical Box 14.3: Numerical Examples on Technical Change
14.6.2 The residual as the unexplained component of growth
Mathematical Box 14.4: Detailed Estimation of the Contributions to
Economic Growth
Estimating the returns to different educational levels
14.7 Some Empirical Findings on the Contributions to Growth
14.8 The Solow Growth Model with Exogenous Changes in the Quality of Labor
14.8.1 Diagrammatic analysis
14.8.2 The role of human capital in changing the quality of labor
Extended Analysis Box 14.3: The Variability of the Labor Force Growth
Rate
14.8.3 Technical change with falling or negative labor force growth rates
14.8.4 Shifts in participation rates
Fact Sheet 14.3: Participation Rates in Canada and the USA, Male and
Female, 1980–2008
Box 14.2: The Implications of Increases in the Labor Force Participation
of Women
14.9 SS Growth versus Level Effects of Exogenous Shifts
14.10 The Implications of the Solow Analyses for Policy
Fact Sheet 14.4: Real GDP Per Capita, 2003 Population Growth Rates
Box 14.3: Historical Growth Patterns, Labor force Growth, and
Technology Shifts
14.10.1 Malthus's theory of economic growth and the Malthusian stage
14.10.2 The post-Malthusian stage

26
 14.10.3 The modern growth stage
14.11 Conclusions
Key Concepts
Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions
15 Advanced Topics in Growth Theory
15.1 Technology, Knowledge and Externalities
15.1.1 The distinction between endogenous and exogenous technical change
15.1.2 The definition of capital in endogenous growth theories
15.1.3 The externalities of new knowledge
Box 15.1: Inventions, Innovations, and Patents
Innovations versus inventions as forms of technical change
15.1.4 Patents and the protection of copyright, intellectual property and trade
secrets Recent changes in patent protection
15.2 The Production of New Knowledge
15.2.1 Diagrammatic analysis
15.3 The Dissemination of Knowledge Across Countries
15.4 Diagrammatic Analysis
15.4.1 Growth rates and capital flows among countries
15.4.2 Convergence and divergence in living standards between countries
15.4.3 Continuous growth in the advanced economies
Box 15.2: Foreign Aid and Investment — and Capital Drain
15.5 The Historical Experience of Growth
Box 15.3: Economic Globalization and Endogenous Technical Change
Technology Clusters
Globalization and Patents
15.6 Human Capital
Box 15.4: The Innovative Process and Its Creative Destruction
15.7 The Implications of Endogenous Growth Theories for Macroeconomic Policies
Extended Analysis Box 15.1: International Linkages and Growth
15.8 The Importance of the Inflation Rate and the Quantity of Money for Growth
Box 15.5: The Role of the Financial Sector in Growth: Some
Conclusions from Economic History
15.9 The Role of the Financial Sector in Growth: Recent Empirical Evidence
15.10 The Miracle of Economic Growth
15.10.1 The miracle of growth and financial crises in the economic tigers
15.10.2 The economic crisis of 2007–2010 in the USA
Extended Analysis Box 15.2: The Economic, Social, and Political
Environment for Growth: Markets, Competition, Capitalism, and
Entrepreneurship
15.11 Empirical Evidence on the Contributors to Growth
15.11.1 Empirical evidence on growth in recent decades

27
 15.12 Conclusions
Conclusions from endogenous growth theory
Conclusions on money, the financial sector, and growth
Key Concepts
Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions
16 Business Cycles, Crises, and The International Transmission of Economic Activity
16.1 Recessions and Booms in Economic Activity
16.1.1 The popular statistical designation of a recession
16.1.2 Monetary and fiscal policies to combat a recession
Fact Sheet 16.1: Output Gap in the United States (USA), 1980–2008
16.2 Business Cycles and the Growth Trend in Economic Activity
Box 16.1: Eliminating the Trend from the Data
Deriving the trend rate of output
The unemployment rate as the proxy for the output gap
16.3 Stylized Facts of Business Cycles
16.4 The Behavior of Variables Over the Business Cycle
16.4.1 Procyclical, acyclical, and countercyclical movements
16.4.2 Leading, lagging, and coincident variables
16.4.3 The volatility of variables over time
16.5 Business Cycles, the Full Employment Assumption and the Classical and
Keynesian Paradigms
Box 16.2: Business Cycles and The Assumption of Full Employment
16.6 The General AD-AS Theory of Business Cycles
Fact Sheet 16.2: The Structure of the AD-AS Model
16.6.1 The propagation mechanism
Box 16.3: An Illustration of Technical Change Creating Cycles through
Aggregate Demand Shifts
16.7 The Modern Classical Approach to Business Cycles
16.7.1 Real business cycle theory
16.7.2 Policy implications of the RBC theories
16.8 Keynesian Explanations of the Business Cycle
16.8.1 The multiplier-accelerator model of aggregate demand
16.8.2 Limitations of the validity of the multiplier-accelerator model
Extended Analysis Box 16.1: A Taxonomy of the General Reasons for
the Occurrence of Business Cycles
16.9 International Linkages Among National Business Cycles
16.9.1 A world business cycle
16.10 Crises in Economic Activity
Fact Sheet 16.3: The Asian Crisis of the Mid-1990s
16.11 Long-Term Effects of Recessions and Booms
16.11.1 The accumulation of human capital

28
 16.11.2 The accumulation of physical capital
16.11.3 Technical change
16.11.4 The longer-term effects of cyclical fluctuations through aggregate
demand
16.11.5 The longer-term impact of recessions and booms
16.11.6 The implications of hysteresis for policy
16.12 Money, Credit and Business Cycles
16.12.1 Money supply and business cycles
16.12.2 Credit and business cycles
16.12.3 Economic crises
16.12.4 Some instances of financial crises
16.13 The Relevance of Monetary and Fiscal Policies to Business Cycles
Box 16.4: Anatomy of the Financial and Economic Crisis of 2007–2010
in the World Economy
The policy responses
16.14 Inflation and Unemployment
16.15 Conclusions
Key Concepts
Summary of Critical Conclusions
Review and Discussion Questions
Advanced and Technical Questions

Index

Supplementary content is available for customers at

http://www.worldscientific.com/worldscibooks/10.1142/7507#t=suppl.
For any enquiries, please contact [email protected].

29
 Preface
The state of macroeconomics

“What went wrong with economics? And how the discipline should
change to avoid the mistakes of the past ” was on the cover page,
and
“Where it went wrong — and how the crisis (of 2007–2009) is
changing it ”
was the lead article of the venerable The Economist in its 18–24
July 2009 issue. Among the subheadings of the related article were
“ Rational fools” and “Blindsided and divided ”. [The Economist,
18 to 24 July 2009, front page and pages 11 to 12.]

On the state of research and teaching of macroeconomics

Some of the opinions from prominent economists quoted by The

Economist were:
“The past thirty years of macroeconomics training at American
and British Universities were a ‘costly waste of time’”.
“Mr. (Paul) Krugman feared that most macroeconomics of the past
thirty years was ‘spectacularly useless at best and positively
harmful at worst’…‘We are living through a ‘Dark Age of
macroeconomics’, in which the wisdom of the ancients has been
lost’”.
“Mr. (Willem) Buiter argues that a training in modern
macroeconomics was a ‘severe’ handicap at the onset of the
financial crisis…”.
“The mainstream macroeconomics embodied in DSGE (Dynamic
Stochastic General Equilibrium Models) was a poor guide to the
origins of the financial crisis”…“According to David Colander,…
macroeconomics is often the least popular class…. Mr. Colander
asked a group of Chicago students. “Did you do the dynamic
stochastic general equilibrium model?’ ‘We learnt a lot of junk
like that,’ one replied.” [Above quotes taken from The Economist ,
18 to 24 July 2009, pages 11–12 and 65–67; material in
parentheses added.]
“The unfortunate uselessness of most state of the art monetary

30
 economics…. (research) since the 1970s has been motivated by
the internal logic, intellectual sunk capital and aesthetic puzzles…
rather than by a powerful desire to understand how the economy
works; let alone how the economy works during times of stress
and financial instability” (Willem Buiter, quoted in The Hindu,
Business Line, 18 March 2009; material in parentheses added).1

The Economist has not been the only magazine to express such views. The
last two years have witnessed a cacophony of similar views in serious
journals, in newspapers, and magazines, on the radio, TV, and the Internet.
Among others have been:
‘The emperor has no clothes’.
And we might add:
‘Economists are no better than astrologers and soothsayers. But the
public and the country suffer far more from their views’.

But then, what justifies another macroeconomics textbook?

In view of all the doubts and dismal opinions of the dismal science2 held
by the public and economists about the validity and relevance of
macroeconomics and its teaching, it must be a wonder that one not only
admits to teaching macroeconomics but even dares to write another
textbook on the subject. We do so because of our strong belief that
macroeconomics and its policy prescriptions are both relevant and need to
be learnt to understand what has been happening in recent years, while
recognising that the subject and its presentation do need to be changed.
Further, there is a glaring lacuna in macroeconomics in that it does not
incorporate an adequate analysis of the role of credit in the economy. It is
also unfortunate that the teaching of macroeconomics in recent decades
has dropped the analysis of disequilibrium and the scope of policies
relevant to disequilibrium.

The reformulation of macroeconomics

Macroeconomics is not only one of the pillars of economics, it is probably
the most relevant and exciting part of economics for most people, since it
deals with the economy and its impact on their job prospects and personal
portfolios, as well as explaining the analysis behind the economic and
business headlines in newspapers and periodicals. Macroeconomics is rich

31
in theories, but not all are equally good at explaining the stylized facts.
This book teaches the students the differences between their assumptions
and implications and the use of stylized facts to discriminate among them.
What the quotes above on the sorry state of macroeconomics clearly
indicate is that the exposition of macroeconomics needs to be revised,
perhaps drastically. As a starting point, its theories have to be tested by
their concordance with facts and, if they prove to be invalid (i.e., useless
for explaining the economic reality that we live in), they need to be
rejected or revised. The revision of theories has to start with the empirical
validity of their assumptions, and move on to testing the validity of their
implications and predictions. This book undertakes these tasks, and makes
the relevance of the established theories to the related stylized facts of the
economy a central part of its exposition.
While the general equilibrium assumption can be useful in learning
about the healthy state and good times of the macroeconomy, its
dominance in macroeconomics to the exclusion of disequilibrium analysis
has been central to the recent failure of macroeconomics and its teaching.
It should be now beyond dispute that economies do sometimes go into
disequilibrium and stay in it. The causes of potential disequilibrium and its
analysis have been missing from most macroeconomics textbooks for
some decades, and need to be reinstated. Chapter 9 of this book does so.
What also seems to be indicated by the proclaimed sorry state of
macroeconomics is that divisions within its schools have been overdone to
the detriment of the validity of macroeconomics.3 The presentation of and
emphasis on these divisions needs to be reduced. If any observation can be
explained by several theories (from different schools), we should integrate
all of them for a fuller explanation of the observation. Why care from
which school the various components originated? This implies
downplaying the schools in macroeconomics, and looking directly at the
relevant theories. This book adopts this approach (see especially Chapter
8) and subsequently collects the main theories of macroeconomics into the
classical and Keynesian paradigms and confines the presentation of these
two paradigms into one chapter. Its approach is: learn the relevant theories
first (in Chapters 1 to 10), independent of their origins, and then learn the
paradigms (set out in Chapter 11) and, to satisfy one's intellectual
curiosity, find out in which paradigm they originated.
The emphasis in macroeconomic textbooks has for some decades drifted
toward the presentation of long-run equilibrium and growth theories, to the
detriment of short–run and disequilibrium analysis. Such an orientation is
justified for studying the changes in the standards of living over long
periods. It is less justified from the perspective of understanding the

32
economic world that we constantly experience. This book redresses this
trend, and brings the presentation of the short-term economics to the early
parts of the book.
Understanding the economic crisis of 2007–2010 requires revision of
macroeconomics to address the glaring omission of credit in current
macroeconomic modelling.4 This crisis also firmly establishes the non-
neutrality of both money and credit in the short run. This book
incorporates the relevant analysis of credit (see Chapter 16) and implies its
non-neutrality, as well as the non-neutrality of money and financial
institutions, in the modern economy (Chapters 8 and 9).5 Even in the
context of very long-term growth, this book distinguishes between the
neutrality of money and the non-neutrality of the economy’s financial
structure (Chapter 15).

More specifically on this book

This book presents the current state of knowledge of macroeconomics
relevant to elementary and intermediate undergraduate courses. The goal
of the book is to provide students with adequate knowledge to: (i)
understand the stylized macroeconomic facts, (ii) to learn the rich variety
of economic theories and the differences in their implications, (iii) to use
the theories to explain the stylized facts, and (iv) to understand the major
differences between the developed economies and the developing ones.
This book brings to its treatment of macroeconomics a fresh perspective
on the coverage of macroeconomic theories and their validity. Economics
is a social science, whose objective is to explain economic reality. This
perspective focuses on the ability or inability of the existing theories to
explain the established stylized macroeconomic facts, especially on the
relationship between monetary and fiscal policies and output and
employment. The relevant stylized facts have been established over several
decades by intuition and empirical studies, though few macroeconomics
textbooks pay adequate attention to them. The recent (2007 to 2010)
economic crisis and recession in many economies provide a rich topical
confirmation of many of these stylized facts.
Among the stylized facts of macroeconomics is that related to the
behavior of the central bank in its pursuit of monetary policy. The general
assumption of most macroeconomics books is that the central bank
controls and operates on the money supply as the primary monetary policy
tool, so that the appropriate assumption for the exposition of
macroeconomics is that the money supply is held exogenous. This
assumption leads to the IS-LM model for the determination of aggregate

33
demand. This assumption and the resulting IS-LM model are quite
unrealistic for many countries, especially developed ones, whose central
banks have claimed at least since the mid-1990s that they operate on
interest rates as their primary monetary policy tool. This claim changes the
exposition of short-run macroeconomics to one in which the interest rate is
held exogenous. Few textbooks set out this exposition. This book does so,
and calls the resulting short-run macroeconomic model of aggregate
demand the IS-IRT model. However, for completeness and to allow
instructors a choice between the IS-LM and IS-IRT expositions, this book
presents both these models.

Major features of this book

Among the highlights of this book are:
• Rigorous economic analysis and intuitive insights.
• Stylized macroeconomics facts, which are used to discuss the validity
and usefulness of theories.
• Treatment of open economy macroeconomics early in the book and
throughout the book. Open economy material is integrated into the
macroeconomic analysis from the very beginning (Chapters 3 and 5).
This treatment is supplemented by Chapters 12 and 13, which include
coverage of many topics such as the foreign exchange market, persistence
of deficits and surpluses in the balance of payment even when the
exchange rate is floating, the J curve, Globalization, Dollarization and
Currency Boards, the Gold Standard, and the IMF (including discussion
of whether the IMF should evolve into the world’s central bank).
• As an illustration of the use of stylized facts to select among theories
and/or modify them, this book (Chapters 3 and 13) questions the validity
of Purchasing Power Parity on the basis of the stylized facts on price
differences across economies, and does not assume PPP in setting up its
macroeconomic models. Similarly, Chapters 3 and 13 question some of
the unrealistic implications of the Interest Rate Parity theory and modify
the Interest Rate Parity condition to allow for deviations of interest rates
among countries.
• A fundamental stylized fact relates to how central banks seek to control
or influence output and employment through their monetary policies: do
they act in such a way that macroeconomic theory can build its model on
an exogenously given money supply (leading to the IS-LM model of
aggregate demand) or do they act in such a way that macroeconomic
theory can build its model on an exogenously given interest rate (leading
to the IS-IRT model, which is explained in Chapter 5 of this book (and

34
 almost never in other macroeconomic textbooks). This is a factual
question on which many central banks, especially in developed
economies, claim that they directly operate on interest rates, not the
money supply, in the economy. This monetary policy procedure renders
the traditional IS-LM model of aggregate demand irrelevant for many
economies, though it is still the mainstay of most macroeconomics
textbooks. This book also examines the conditions under which the
central bank would prefer to use interest rate targeting or monetary
targeting, and vice versa.
• The assumption that the central bank uses the interest rate, rather than the
money supply, as its primary monetary policy instrument for controlling
aggregate demand is central to the currently popular Taylor Rule and the
New Keynesian economics. However, the relevant analysis (called the
IS-IRT one in this book, with IRT standing for ‘interest rate targeting’ as
against ‘monetary targeting’ that leads to the IS-LM analysis) is not set
out in almost all other undergraduate and even most graduate, books on
macroeconomics. This book presents in Chapter 5 the relevant
macroeconomic analysis for the case in which the central bank primarily
sets the interest rate.
• For completeness, this book provides both the IS-IRT (Chapter 5) and IS-
LM (Chapter 6) models of aggregate demand for the open economy. The
instructor is thus provided with the choice of doing only the analysis
relevant to their particular country and its central bank’s monetary policy,
or doing both.
• This book separates the analysis of the long run (Chapter 7), the analysis
of the short-run based on errors in expectations, adjustment costs of
changing prices, employment, the capital stock and output (Chapter 8),
and disequilibrium (Chapter 9). A major argument differentiating this
book from others is that the economy can be out of both long-run and
short-run equilibrium — and therefore, in disequilibrium. Chapter 9
shows that the implications of disequilibrium for monetary and fiscal
policies can differ considerably from those of long run and short run
equilibrium.
• The treatment of short-run deviations of output from the long-run one is
explained in Chapter 8 as arising from many causes. These include errors
in expectations (due to the costs of acquiring adequate information and
deriving its implications for the revision in expectations), and the
adjustment costs of changing prices (as in the menu cost theory). But they
also include the adjustment costs of changing employment (leading to
labor hoarding and Okun’s rule), the physical capital stock and output.
All these occur in the short-run and collectively determine the short-run

35
 output and employment.
• The analysis of disequilibrium and the likely behaviour of firms and
households and the policies relevant to disequilibrium have often been
neglected in many macroeconomics textbooks published over the last
three decades. This book devotes a whole chapter (Chapter 9) to this
analysis. The relevance of this chapter to the economic crises and
fluctuations in output and employment, as during 2007–2010, should not
be ignored.
• On the basis of the stylized facts established from earlier studies — and
reinforced by the recession of 2007– 2010 — one has to reject the
relevance and validity of the long–run equilibrium and short-run
equilibrium theories to the explanation of many of the stylized facts. The
disequilibrium analysis, especially of demand-deficient conditions,
presented in Chapter 9 of this book is more appropriate.
• This book provides an explicit analysis of credit as distinct from bonds,
and the impact of a credit crisis on the other sectors of the economy. This
material is especially relevant to understanding the causes of many
downturns in the economy, and to the study of financial and economic
crises. Such analysis is missing from most other macroeconomics
textbooks but is vital to understanding the Asian Crises of the mid-1990s
and the worldwide one during 2007–2010.
• This book discusses (in Chapter 11) the heritage and evolution of
theories of the major schools of thought in macroeconomics. It organizes
them into the classical and Keynesian paradigms, and explains why it is
desirable to study both. The exposition of the theories is followed by a
comparison of the predictions of the theories with the stylized
macroeconomics facts to decide on the validity and relative usefulness of
the various theories.
• This book divides its presentation of the analysis of economic growth
into two chapters. Chapter 14 focuses on the Solow/neoclassical model of
growth, without technical change and with exogenous technical changes.
An important topic treated in this chapter is the historical experience of
growth epochs, starting with the Malthusian theory for agricultural
societies and ending with the modern and post-modern stages of growth
of already industrialized societies. Chapter 15 sets out the endogenous
growth theories. It also studies the relationship between the growth of
output on the one hand and, on the other hand, money supply growth, the
inflation rate and the development of the financial structure of the
economy.
• The ups and downs of business cycles are very important to the average
person and to firms. Contrary to a belief prevalent in the 1980s and

36
 1990s, cyclical fluctuations have not been eliminated or sufficiently
moderated. Chapter 16 explains their causes and the evolution of their
stages. It also explores the relationship between financial developments
and crises and economic crises. Further, it provides an explicit treatment
of how a credit crisis can cause adverse shifts in both aggregate supply
and aggregate demand in the economy. This exposition helps students to
explain why the credit crisis of 2007 in the USA developed into an
economic one in 2008 and how the stimulatory monetary and fiscal
policies were able to shorten its duration and decrease its intensity.
The book also innovates in its presentation in several ways. These include:
• Fact Sheets: Each chapter has Fact Sheets to provide graphs on the
major points relevant to the chapter. These Fact Sheets use data from: the
USA, the world’s largest economy; Canada, a developed small economy;
and several developing economies, which include Thailand, Malaysia,
China, and India.
• Boxes: There are three types of boxes (Boxes, Extended Analysis Boxes,
and Mathematical Boxes) in each chapter. These boxes dig deeper into
the analysis or the relevant facts, so that they extend the material beyond
that in the text. Extended Analysis boxes provide advanced material and
are meant to give the instructor the choice between whether to omit them
or cover them. Mathematical boxes provide the students with the
knowledge relevant to the economic theories presented in the chapter.
• Review of material: Besides bulleted Conclusions, each chapter has at
its end Key Concepts and Summary of Critical Conclusions, which
facilitates the student’s review their knowledge of the material of the
chapter.
• Questions: The book divides the questions at the end of each chapter
into Review and Discussion Questions, and Technical and Advanced
Questions. Less mathematically oriented courses can omit the Technical
Questions.
• Technical Questions: The technical questions at the end of each chapter
require knowledge of high school algebra and the ability to solve linear
equations, but do not require any university level mathematical courses
and specifically do not require knowledge of calculus. While the
technical questions and their solutions can be omitted from courses using
only verbal arguments, intuition, and diagrams, they do illustrate and
deepen the knowledge of the theories and their implications presented in
the chapter.
• Calculus: While the book, especially its Mathematical boxes and
Technical Questions, rely on knowledge of high school mathematics,

37
 they do not require a prior knowledge of calculus (or teach it).

Coverage of the Textbook Over Two Semesters

This book consists of 16 chapters, which are meant to be covered over two
semesters, so that each chapter would take on average about two weeks to
cover. Instructors can choose the pattern of coverage of the chapters
according to:
• Chapters 1 to 9 are designed to provide a very good one-semester
introductory course in macroeconomics, with the major emphasis on
short-run economics. If desired, these chapters can be supplemented by
Chapter 16 on business cycles and economic crises. If this beginning
course is followed by an intermediate macroeconomics course, the latter
would cover the more advanced Chapters 10 to 16.
• If the desired focus of the first course is more on growth, the appropriate
chapters for it would be Chapters 1 to 3 and 14 to 16, along with some
additional chapters selected by the instructor.

Study Guide
The book comes with a Study Guide. Each chapter of the Study Guide
provides an introduction to the main topics in the chapter, followed by
answers to the Review and Discussion questions, as well as the Technical
and Advanced questions, at the end of the chapter.

1http://www.blonnet.com/2009/03/18/stories/2009031850070900.html.
2‘The dismal science’ was the name given to economics by the social
reformer Thomas Carlyle in the mid-19th century. The name stuck, and
justly so in many ways.
3“The fault lies not with economics, but with economists. The problem is
that economists became overconfident in their preferred models…”. “They
forgot that there were many other models that led in radically different
directions. Hubris creates blind spots. If anything needs fixing, it is the
sociology of the profession”. (Quotes taken from an article by Dani
Rubrik, in The Hindu, New Delhi, 14 March 2009).
4For virtually all textbooks, financial institutions providing credit do not
exist, so that a credit collapse can have no impact on output and
employment in the economy. The extent to which banks are envisaged is
as creators of money, but not as providers of credit.
5In many macroeconomic models, finance is a veil, as is money, so that a

38
credit collapse has no impact on output.

39
Acknowledgments
I am indebted to several generations of students who used my
macroeconomics course notes. Their comments helped me refine the
material in this book. I am especially indebted to several teaching and
research assistants over the years for writing or revising the answers in the
Study Guide.
I especially wish to acknowledge the assistance of Oana Ciobanu with
the Fact Sheets in the chapters, and of Marc Jean Baraghid with the final
version of the Study Guide.

Prof. Jagdish Handa

McGill University, Montreal, Canada
[email protected]

40
 About the Author
The author is Professor of Economics at McGill University, Montreal,
Canada, and has taught Macroeconomics for more than 40 years.
Prof. Handa has published several books and numerous articles in
journals, many of them related to macroeconomics and monetary
economics. Among his related textbooks is Monetary Economics (first
edition, 2000; second edition, 2008), which provides a comprehensive
treatment of monetary theory and policy at an advanced level.

41
 Glossary of Symbols

B nominal balance of payments

B c nominal balance of payments on current account
B k nominal balance of payments on capital account
c consumption expenditures
c 0 autonomous consumption
D$ demand for our currency ($) in foreign exchange markets
e aggregate expenditures (in real terms)
FR foreign exchange reserves
g government expenditures on commodities
i (planned/desired/ex ante) investment expenditures
i u unintended investment
i 0 autonomous investment
ia actual investment (≡ i + iu)
K real capital stock
L labor force
LR long run
M money
M0 monetary base
M1 narrow definition of money (= currency + demand deposits)
M2 broad definition of money (= Ml + savings deposits)
n employment (labor employed)
net inflows of funds to the country for interest and dividend
NR
payments (= IR — OR)
net inflows of funds to the country from unilateral transfers (= IT
NT
— OT)
OR outflows (inflows) of funds to the country for interest and dividend
(IR) payments
OT outflows (inflows) of funds from the country due to unilateral

42
(IT) transfers
r real rate of interest
r T target real rate of interest set by the central bank
R nominal/market rate of interest
P price level (i.e., consumer price index or GDP deflator)
P F foreign price level
S $ supply of our currency ($) in foreign exchange markets
s (private) saving
s n national saving (≡ s + t — g )
SR short run
net taxes paid (government's tax revenues less government transfers
t
to the public)
t 0 autonomous net taxes paid
u actual rate of unemployment
U level of unemployment
u n natural (i.e., long-run equilibrium) rate of unemployment
u* short-run equilibrium rate of unemployment
u*
deviation of short-run equilibrium unemployment rate from the
—u
n
natural one
u — unemployment rate due to disequilibrium (i.e., deviation of actual
u* unemployment rate from the short-run equilibrium one)
x c real amount of exports of commodities in domestic commodity
units
X c nominal value of exports of commodities in domestic currency units
aggregate quantity of commodities (also often referred to as real
y
output/national income)
Y nominal value of output ( = nominal national income = Py )
y d aggregate real demand for commodities
ys aggregate real supply of commodities
yf full-employment (i.e., long-run equilibrium) real output/income
y* short-run equilibrium level of real output/income
zc/ρ real amount of imports of commodities in domestic commodity
r units

43
 Z c nominal value of imports of commodities in domestic currency
units
π [pi] rate of inflation (≡ P ″ )
π e expected inflation rate (≡ P //e)
[‘rho’] nominal exchange rate (= amount of a foreign currency
ρ required to purchase one unit of the domestic currency, denoted as
£/$)
real exchange rate (= amount (i.e., number of units) of foreign
r
ρ commodities required to purchase one unit of the domestic
commodity)

The small letter symbol indicates the real value of the variable while the
capital one indicates its nominal (dollar) value. One exception to this rule
is the use of K to designate the real value of the physical capital stock.
The superscript d on a variable indicates its demand and the superscript s
will indicate its supply. The variable symbol without either of these
superscripts indicates its actual value. If there exists equilibrium, the actual
value is the equilibrium one, i.e., one at which the demand and supply for
that variable are equal. Thus, nd is the demand for labor and ns is its
supply, while n is actual employment.

Symbols Used in Growth Theory Chapters 14 and 15

K real amount of capital
k real capital-labor ratio
L labor force (= employment)
n growth rate of labor force
η rate of growth in the efficiency of workers
S total real saving
s saving per capita (= S/L))
σ average propensity to save (APS) out of output (= S/Y )
Y real value of output ( = real income = Py)
y output per capita (= Y/L )
Useful Mathematical Symbols and Formulae for Economics
[di] change in the following symbol, holding all other things, i.e., other
exogenous variables, the same (e.g., ∂y/∂i means change in y for a very
∂ small change in i, ceteris paribus). ∂ is a mathematical symbol

44
 pronounced as ‘di’, so that ∂y/∂i is pronounced as ‘di y di i’).
[delta]change in the following symbol, holding all other things, i.e.,
other exogenous variables, the same. y is pronounced as ‘delta y .’
≡ identity (that is always true) or identical to
= equilibrium condition or equal to
' change in (the preceding variable)
″ rate of change (growth) in (the preceding variable)
Superscripts
d demand for (the preceding variable)
e expected value
F foreign (e.g., PF is foreign price level)
s supply of (the preceding variable)
Special use of subscripts
y d disposable real national income

Greek Letters Used

α alpha
β beta
γ gamma
ρ rho
σ sigma
η eta
θ theta
∂ di
delta

Designation of the change in a variable

Economics often focuses on the change in a variable caused by a change
in another variable, if all other variables were to be held constant. The
Latin phrase used for ‘holding all other things (variables) constant’ is
ceteris paribus. The most appropriate symbol (borrowed from calculus) for
a change (which need not be ‘small’) in y caused by a ‘very small’ change
in x, ceteris paribus’ , is ∂y/∂x. ∂ is a Greek symbol, which is pronounced

45
as ‘di’. The term ∂y/∂x is pronounced as ‘di y, di x ′ and is often spoken as
‘the derivative of y with respect of x ′ . Using this symbolism is much more
convenient than writing the longer form ‘change in y for a very small
change in x, all other things held constant’. Therefore, we will often use
this symbol, though without invoking the use of calculus for any derivation
connected with it. That is,

∂y/∂x = a change in y for a very small change in x, with other things being
held constant.

Designation of the rate of change (growth rate)

in a variable
Economics often examines the effect of a change in a variable over time.
The symbol for the change in a variable y over time is ∂y/∂t. We will
usually use the even shorter symbol y ′ for it. For period analysis, will
equal (yt+1 — yt), or yt (where is a Greek symbol pronounced as capital
delta), where the subscript t stand for the time Periodch li and stands for
‘change in’. However, for continuous time analysis, the appropriate
symbol is ∂y/∂t . That is,

where (= ∂y/∂t) means the change in y for a very small change in t .

The rate of growth (or change) of the variable y over time will be
designated as It is obtained by dividing y ′ by yt , so that:

In continuous time analysis, the rate of growth is also often designated by

a dot placed over the variable symbol. However, our usual usage for the
rate of growth (over time) will be .
If output y grows at the rate 5% per period, it is often written as 0.05.
Conversely, writing the growth rate 0.05 in percentage terms requires
multiplying it by 100, which gives 5%.

Useful mathematical formulae on growth rates to

remember
If

46
then,

That is, if a variable is a multiple of two other variables, its growth rate is
the sum of the growth rates of the two component variables.
if

then,

That is, if a variable is a ratio of two other variables, its growth rate equals
the difference between the growth rates of the numerator and the
denominator.

Splitting the growth rate of nominal (value of)

output into real output growth and inflation
An example of the preceding formulae occurs from the definition of
nominal output Y (which is the dollar value of the quantity produced) as
equal to the price level multiplied by real output y (which is the physical
amount of the quantity produced). That is, Y ≡ Py. The symbol ≡ indicates
an identity, while = indicates an
equilibrium. From Y ≡ Py , we derive Y ” ≡ P ” + y ”. Since P″ is the rate
of change of the price level, which is the definition of the rate of inflation,
for which our symbol will be π , we can write:

This equation asserts that the growth rate of the nominal (dollar) value of
the commodities produced in the economy equals the growth rate of their
real value (i.e., the quantity produced) plus the rate of inflation.
From the preceding discussion, we can derive the following formula:
If

then,

That is, if a variable is a ratio of two other variables, its growth rate is the
difference between the growth rates of the numerator and the growth rate
of the denominator. Therefore, since y ≡ Y/P , we have:

47
Hence, the real output growth rate is the growth rate of its nominal value
less the inflation rate.

48
 PART I
Introduction to Macroeconomics

49
 CHAPTER 1
Output, Unemployment, and the Basic
Concepts

Macroeconomics studies the functioning of the economy at the

aggregative level. It encompasses the study of the economy when it
is performing at its optimal level as well as when it has deviations
from this level. In both cases, macroeconomics examines whether
the economy’s performance can be improved through the use of
monetary, fiscal, and other governmental policies.
The fundamental questions of macroeconomics concern the
levels of output and its rate of growth, unemployment and inflation
in the economy, and changes in them. Their determination is
studied in the contexts of the short run and the long run .
Economics is a science. It builds theories to explain the real-
world, real-time observations on its variables of interest and
subjects the predictions of its theories to statistical tests through
the use of econometrics .

1.1 Introduction to Macroeconomics

Macroeconomics is the study of the functioning of the economy at the
macro or aggregative level. Its main variables of interest are:
• output and the standard of living,
• unemployment,
• the price level and inflation,
• interest rates and the money supply,
• balance of payments and exchange rates,
• the impact of government expenditures, taxes, and deficits on the
economy, and
• the central bank’s policies on money supply and interest rates and their
impact on the economy.
A proper study of these topics involves the study of numerous other
variables. Among these are wages, consumption, saving and investment,

50
exports and imports, capital flows between countries, labor demand and
supply, money demand and supply, etc.

1.1.1 The nature of macroeconomic analysis

A general study of the whole economy can be formulated in two distinct
ways:
i. As a general equilibrium microeconomic model.
This formulation includes a separate specification of the market for each
good, as in microeconomic analysis. Such a system is a very detailed one,
studying as it does the separate demand, supply, and price of each good
in the economy. It is, however, quite cumbersome for macroeconomic
purposes where the objective is to focus on a few macroeconomic
variables only.
ii. As a macroeconomic model.
Such a model aggregates the very large number of goods in the economy
into a small number of categories. This formulation reduces the number
of goods and their markets to be studied to a manageable level.
Obviously, the degree of aggregation used must depend upon the
intended use of the resulting model or framework.
The classification of goods in macroeconomics
Short-run macroeconomics normally classifies all goods in the economy
into five categories.
1. commodities (commonly known as ‘goods and services’),
2. money (i.e., currency and demand deposits),
3. bonds (i.e., non-monetary financial assets, so that the term ‘bonds’ in
macroeconomics includes equities),
4. labour,
5. foreign exchange (mainly foreign currencies and gold).
Closed economy versus open economy analysis
For heuristic reasons, macroeconomics is often first presented in closed
economy models that are later extended to the open economy. A closed
economy is one that does not have international trade in commodities or
capital flows with other countries. An open economy is one that has such
linkages. These linkages imply an additional degree of complication of
analysis, so that one needs to master the closed economy models to better
understand the open economy ones.
The closed economy macroeconomic models analyze the markets for

51
four goods: commodities, money, bonds, and labor, and assume that there
is no trade in these between the domestic economy and foreign ones. The
open economy models of macroeconomics assume that there is trade in the
above four goods between the domestic and foreign economies. They also
include the analysis of the market for foreign exchange.
Short-run macroeconomic theories, growth theories and business cycle
theories
The main focus of the short-run macroeconomic theories is on the impact
of shocks to the macro economy and the effects of fiscal, monetary, and
other policies on aggregate output, unemployment, and the other variables
of interest mentioned above. The main focus of growth theories is on the
long-run evolution of the total output of the economy and of the standard
of living. The main focus of business cycle theories is on the variations in
output and unemployment, inflation and interest rates over the business
cycle.

1.2 The Classification of Economic Agents and

Markets in Short-Run Macroeconomic Models
Macroeconomics treats labor as an input in the production of commodities,
whose purchase for consumption or investment is transacted with money
as a medium of exchange/payments. Labor is provided by
workers/households who receive wages as payment for the work.
Households receive not only wages but also interest and profits from their
ownership of physical capital (including land and housing). The sum of
these receipts constitutes national income. The demands for domestic and
imported commodities (including physical capital), bonds (non-monetary
financial assets), and money depend on national income (which is
generated by production) and national wealth.
The basic classification of economic units in the economy is into:
• Households, which supply labor as workers and are the owners of capital.
They receive wages and profits (or interest) from their ownership of
capital, and decide on consumption, saving, and money and bond
holdings.
• Firms, which engage in the production of output by hiring labor (and any
other inputs) to use with their capital. They incur investment to change
their capital stock, and issue bonds (including equities) to finance their
investment. Financial institutions, such as banks, are firms under this
classification. Firms are assumed to maximize profits.

52
• While non-government organizations (NGOs)1 usually do not attempt to
maximize profits, macroeconomic analysis, for simplification, does not
include their separate analysis and just lumps them in the category of
firms.
• The government (‘the fiscal authority’), which decides on the
government expenditures and taxes. It finances its fiscal deficits by
raising funds in the financial markets by selling bonds to the public. It
uses fiscal surpluses to buy back some of its (outstanding) bonds from the
public. In contrast with firms, it is not assumed to be a profit-maximizer.
• The central bank (‘the monetary authority’), which decides on the money
supply and interest rates.
• Foreign economies, which trade in commodities (as exports and imports)
and bonds (which result in financial capital flows between the foreign
and the domestic economies). These flows of commodities and financial
capital are captured in the balance of payments. The amount of foreign
currencies, foreign bonds, and gold held by the country are the major part
of its foreign exchange reserves and the exchange rate is the rate of
exchange between the domestic currency and foreign ones.
As mentioned earlier, there are five distinct, composite goods in
macroeconomic analysis: commodities (including physical capital), bonds,
money, labor, and foreign exchange. Their ‘markets’ are referred to as
markets or sectors, the two terms being used interchangeably. Therefore,
the macroeconomic model has five markets:
1. the commodity (or product) market,
2. the money market,
3. the financial (i.e., bonds, and equities) market,
4. the production-employment sector composed of the labor market and
the production function, and
5. the foreign exchange market, whose analysis captures payment flows
occurring due to the country’s economic exchanges with the rest of the
world.
The specification of each market requires (i) its demand function, (ii) its
supply function, and (iii) an equilibrium condition.
Only the first four markets are studied for the closed economy – that is,
an economy that does not have commodity or capital flows with other
countries. Virtually all countries now have an open economy – that is, an
economy with exchanges of goods with other economies. The common
procedure in macroeconomics is to first build models for the closed
economy and then to modify them to capture the open economy elements.

53
 Box 1.1: The Analytical Devices of the Short-Run Versus Long-Run
Analysis
Macroeconomics separates the forces that operate on economic variables
into long-run ones and short-run ones. The long run is an analytical (not
chronological) period during which all variables are free to change and
there are no adjustment costs. The short run is an analytical (again, not
chronological) period during which some variables are fixed in value
and/or there are costs to adjusting them. The main differences between
the short-run and the long-run analysis are: the long run allows changes
in the physical capital stock, population, and technology, while the short
run holds them constant (fixed).
The analytical distinction between the long run and short run is used to
distinguish between economic forces that mainly operate over longer
periods and those that operate mainly over shorter periods. The
analytical device of the long run is especially useful for the analysis of
the growth of output and standards of living over long periods. The
analytical device of the short run is especially useful for the analysis of
several variables of topical interest: differences in the actual rate of
output from its long-run potential, unemployment, interest rates,
inflation, balance of payments, exchange rates, etc. The short-run
analysis constitutes the greater part of the theories in macroeconomics.
Most of the chapters of this book present the short-run theories.
Real Time Concepts: The Short Term and the Long Term
The corresponding chronological concepts in real time are the long term
and the short term. The short term is sometimes used to cover a period
as short as the current quarter and sometimes a period as long as the
business cycle, which can run for about ten years. The long term is taken
to cover a period of several years.
The correspondence between the long term and the long run is vague
and ambiguous, as is that between the short run and the short term. In
general, the long run embodies economic forces that operate all the time,
even during the short term. Similarly, the short run incorporates
economic forces that operate not only in the short term but will also be
in evidence in the economy over long chronological periods since a long
chronological period incorporates a sequence of short runs.
Business cycles occur in real time. Their explanation incorporates both
short-run and long-run forces.

54
1.3 Introduction to AD-AS Analysis
The dominant concern of macroeconomics is with the determination of the
economy’s marketed output — defined so as to include that provided by
firms, government, and non-government organizations (NGOs)—as well
as the rate of inflation. Their analysis is encapsulated in the AD-AS
diagram, in which the demand for all the (marketed) commodities in the
economy is represented by the concept of aggregate demand (AD) and the
supply of all (marketed) commodities is represented by aggregate supply
(AS). In the AD-AS diagram with the price level P on the vertical axis and
the quantity of commodities y on the horizontal one, the AD curve is
downward sloping, just as in the case of the demand for a single
commodity.
The aggregate supply behavior of the economy is somewhat different
from that of the supply of a single commodity: macroeconomic analysis
implies that the AS curve for the economy is upward sloping in the short
run but vertical in the long run.

1.3.1 The supply of commodities

The long-run (equilibrium) aggregate supply of commodities
Long-run output is that level of the output that would be produced in the
economy if all its labor, capital, and other factors of production were ‘fully
employed’ in production with the technology available at the time in the
economy. Economics interprets the term 'fully employed in a special way.
It is to be interpreted as the level of output—and employment—that can be
sustained by the economy over the long run with its current supplies of the
factors of production and its current technology — given its current
economic, political, and social structures as well as the wishes of the
owners of the factors of production. To illustrate the last point, the full
employment of labor does not mean that the workers are working 24 hours
a day, 7 days a week, since workers will choose to work a smaller number
of hours. After all, they need to sleep and have time for leisure also.
Similarly, factories and offices are rarely used at night but can still be
‘fully employed’ for macroeconomics analysis if they are in use for the
time that their owners consider to be optimal. Therefore, the full
employment of the factors of production is to be interpreted as the levels
of employment that their owners wish to maintain over long periods. The
level of full-employment (i.e., the long-run equilibrium) output is also
loosely referred to as the ‘potential output’, meaning by this ‘the capability

55
to produce sustainable output over long periods’, of the economy.
As shown later in this book, macroeconomic theory establishes that the
full-employment output — determined by the existing level of technology
and the full employment of the factors of production—is independent of
the price level in the long run. Therefore, if we were to draw the ‘curve’
for full-employment output in a diagram with the price level on the vertical
axis and output on the horizontal one, this curve would be a vertical line at
the level of output equal to the full-employment one. The associated full-
employment curve is designated as the long-run aggregate supply of
commodities and is labeled as the LRAS curve. It is vertical, as shown in
Figure 1.1.
This book designates the full-employment output by the symbol y f,
which is to be read as ‘full-employment output’. Its determination is
specified more explicitly in Chapter 6.
The short-run aggregate supply of commodities
The short-run equilibrium level of output can differ from its long-run
equilibrium level yf because of mistaken expectations about future demand
and prices and/or because of the costs of adjusting prices, output and
employment to meet fluctuations in demand or supply. These reasons are
explored in Chapters 7 and 8. Basically, an increase — especially an
unanticipated increase — in demand leads profit-maximizing firms to
accommodate some part of it by increasing their output by:
• increasing the work effort of existing employees and using the existing
capital stock more intensively and
• increasing employment in the form of overtime and by hiring additional
workers, and buying types of equipment that are more easily variable.

Note that a firm’s profit-maximizing short-term response to an increase in

demand is likely to include both increases in prices and output and is likely
to depend on the relative costs of changing them. Hence, the short-run
aggregate supply (SRAS) curve is likely to indicate an increase in the

56
output of commodities at a higher price level — i.e., to have a positive
slope, as shown in Figure 1.1.

1.3.2 The components of aggregate demand

The aggregate demand (AD) for commodities results from expenditures on
commodities. It is composed of:
• consumption expenditures by households,
• investment expenditures by firms,
• expenditures on commodities by the government,
• the value of net exports (i.e., exports minus imports), which is foreigners’
expenditures on our commodities less our expenditures on theirs.
The sum of all these types of expenditures constitutes AD. Aggregate
demand decreases as prices rise and commodities become more expensive,
so that the AD curve in Figure 1.1 is downward sloping. The detailed
specification of aggregate demand is presented in Chapters 4 and 5.

1.3.3 The diagrammatic AD-AS analysis

Figure 1.1 shows the aggregate demand and supply curves. This figure has
the quantity of commodities, designated as y , on the horizontal axis, with
the price level, designated as P , being on the vertical one. Aggregate
demand is shown in these figures by the curves marked AD, which shows
the quantities demanded at various prices. This curve slopes downwards,
thereby showing that the aggregate demand for commodities falls as the
price level rises. The long-run aggregate supply of commodities is shown
by the curve marked LRAS (for long-run aggregate supply), and the short-
run supply of commodities is shown by the curve marked SRAS. The
LRAS curve is drawn at the full-employment output yf.
The economy is in short-run equilibrium at the intersection of the AD
and SRAS curves. It is in long-run (full-employment) equilibrium at the
intersection of the AD and LRAS curves. It is in disequilibrium if it is at
neither of these points. Figure 1.1 shows that full-employment equilibrium
exists in the economy at the point a, with y = yf. But if the three curves
intersect at different points, as in Figure 1.2a, the economy for the current
short run will be in equilibrium at the intersection of the AD and SRAS
curves. This is shown by the point b in Figure 1.2a. At this point, the
economy’s output will be y 1, higher than the economy’s long-run
productive capacity shown by y f. Since the economy at the point b is not
at its long-run equilibrium, economic forces will move the economy away

57
from the point b toward the long-run equilibrium at the point c .
Macroeconomics focuses on the determination of the equilibrium values
of the major macroeconomic variables, the shifts or shocks that change
them, and the forces that tend to return — or not to return — the economy
to equilibrium.
The possibility of disequilibrium in the economy
The commodity and labor markets do not always respond so fast to shifts
in aggregate demand or supply that we can take them to reach the
equilibrium position instantly. Following any such shift, the achievement
of the price level and output to its new equilibrium takes time and involves
lags, with changes in output and prices occurring in this period. During
this adjustment period, the economy is said to be in disequilibrium. The
analysis of disequilibrium levels of output and employment is presented in
Chapter 8.

The actual aggregate supply of commodities

Therefore, there are two potential sources of differences between the
actual output of the economy and its full-employment level. These are:
1. Short-run equilibrium deviations: these occur in response to shifts in
aggregate demand, if the economy moves along the SRAS curve and
away from the LRAS curve.
2. Disequilibrium deviations: these occur if shifts in aggregate demand or
supply move the economy away, for some time, from even the short-run
equilibrium position of the economy.
The long-run aggregate supply of output will be designated as LRAS (or y
f). The short-run aggregate supply will be designated as SRAS. Actual
output will be denoted by the symbol y .
The impact of an increase in aggregate demand
To see how a point such as b in Figure 1.2b can come about, assume that

58
the economy had the initial curves as AD0, SRAS0, and LRAS, so that it
was in full-employment equilibrium at the point a in Figure 1.2b. Now,
suppose that a sudden spurt of investment by firms increases their demand
for commodities for investment purposes. This shifts the AD curve from
AD0 to AD, causing the economy to move temporarily to the point b. At b,
since the level of output is higher than the economy’s long-run capacity at
y f, it cannot be sustained on a continuing basis. Workers will begin to
push for higher wages and profitmaximizing firms will also want to take
advantage of the high demand to increase their prices and profits. Hence,
the excess demand pressure (i.e., demand > yf) will cause a sufficient
increase in prices to force the economy to gradually adjust from the point b
to the point c . As the price level rises, the quantity demanded adjusts (i.e.,
falls) along AD1 to match the long-run supply. Further, as nominal wages
rise to match the increase in prices and other adjustments occur, the SRAS
curve will shift up to SRAS1. Eventually, through the adjustments in
prices and wages, full-employment equilibrium will be restored in the
economy at the point c.

Compare points a and c : both have the same level of output but prices
are higher at c than at a . Therefore, the effects of the increase in demand
were: output increased for some time but then fell back to its full-
employment level while the price level initially increased by less than its
long-run increase.
The impact of an increase in the long-run productive capacity of the
economy
A similar analysis to the above one can be performed for increases in
aggregate supply. Increases in the aggregate supply can come about
through increases in the supply of the factors of production and/or their
productivity through technical change. Suppose one of these occurs and
increases the quantity supplied at any given price level. This is shown by

59
the shift of the LRAS curve from LRAS0 to LRAS1 in Figure 1.3. Given
the aggregate demand curve as AD0, the new long-run equilibrium is at the
point b. Hence, the increase in supply has the long-run effect of increasing
full-employment output from y f0 to y f1 while lowering the price level
from P 0 to P 1.
The stability of equilibrium
An equilibrium is said to be stable if a deviation from it sets up forces that
cause the market to return to equilibrium. We illustrate this in the context
of Figure 1.3. The commodity market was initially in equilibrium at the
point a (P 0, y f0), with aggregate demand equal to aggregate supply y f0.
With the increase in supply to LRAS1, the equilibrium shifts to the point b.
What forces would move the economy from a to b? Suppose that supply
has become y f1 while demand is still at y f0. Since demand is less than
supply, the price level will fall. It will keep on falling as long as supply
exceeds demand, i.e., until the price level falls to P 1, so that the market
adjustment of the price level will take the economy to the new equilibrium
at b.
Since a similar argument can be used for any other divergence between
demand and supply in this diagram, any equilibrium position in this
diagram will be stable. The stability of the short-run equilibrium with the
AD and SRAS curves can be shown in a similar manner, but is left to the
students.

1.4 The Relationship between Output,

Employment, and Unemployment in the
Economy
The relationship between output and employment is specified by the
production function:

where
n = employment and
K = (physical) capital stock.
This production function assumes that for the modern economy, the main
factors of production are labor and capital. There are obviously other

60
factors of production such as land, but these are considered to be of minor
significance and ignored in the usual analysis. Capital is allowed to vary in
growth theory but kept constant in short-run macroeconomics. Therefore,
the production function used in the short-run analysis is simplified to:

This relationship is shown in Figure 1.4 by the ‘output curve’ marked y. In

this diagram, the horizontal axis shows employment n and the vertical axis
shows output y. Output and employment are positively related but output
increases less than proportionately with employment. This means that the
marginal product of labor — that is, the increase in output for a unit
increase in labor — is positive but diminishing as the number of workers
increases with the given capital stock.
The labor force
The labor force can be defined in one of two ways as:
1. The number of workers willing to work at a given market wage,
irrespective of their skills, abilities, and other characteristics and,
therefore, irrespective of their productivity.
2. The number of workers already employed plus those looking for a job at
a given market wage, irrespective of their skills, abilities, and other
characteristics and, therefore, irrespective of their productivity.
For practical, empirical purposes, the labor force is often simply defined as
the number of persons in a specific age group, such as between 15 and 65
years.
Unemployment
Unemployment is defined as the labor force less the number of workers
actually employed. Hence,

U = L - n,

where
U = the level of unemployment and
L = the labor force.

61
The unemployment rate u is:

An unemployment rate can be written as either 0.05 or 5%, with the latter
expressing it in percentage terms.
The ‘natural rate of unemployment’ u n is defined as the rate of
unemployment that exists when the economy is at full employment. It is
the long-run rate of unemployment. Hence,

where
u f = the ‘full-employment rate of unemployment’ (≡u n),
u n = the natural rate of unemployment (≡u f), and
n f = the full-employment (long-run equilibrium) level of employment.
Note that
• The natural rate of unemployment is not a constant. It can be altered by
changes in technology or in labor supply behavior.
• The actual unemployment rate often differs from the natural one. This
occurs whenever actual output is different from the full-employment one.
The determination of the natural rate of unemployment and the deviations
from it are among the most important topics studied in macroeconomics.

1.5 Measures of National Output and

Expenditures
National output is often measured by statistical measures such as GDP,
GNP, and related variables. Their definitions are as follows.

62
1.5.1 Gross domestic product (GDP)
GDP is the statistical measure of the amount of commodities — our name
for what are commonly called as ‘goods and services’ — produced in the
domestic economy during a specific period and (a) intended for sale in the
market or (b) provided by agencies, including the government and non-
government organizations (NGOs), through production involving the
payment of factors of production, even though a (full marketdetermined)
price is not necessarily charged to the buyers or users of the resulting
products.2 The nominal value of GDP is the sum, in current prices, of the
value of the commodities in (a) and the payments to the factors of
production — which constitute their cost of production — in (b).
GDP differs from all goods produced in the economy in the period in
question because:
• It does not include commodities that are excluded by (a) and (b) above.
These include the great deal of production that goes on in the home.
Therefore, GDP does not include most home and leisure activities.
• It does not include economic activities that result from unpaid volunteer
work and goods and services for which there is no charge.
• It does not include items that are not commodities (including services).
Among the excluded items are notes and coins, and financial instruments
such as bonds (remember, their definition in macroeconomics includes
equities shares corporations).
GDP is calculated for each period, such as a quarter or a year. The GDP
for a given period does not include items that were produced in earlier
periods. Therefore, old (meaning, produced in some earlier period)
paintings, old houses, and old refrigerators are not part of the current GDP,
though the services they render in the current period are.
Fact Sheet 1.1: Global Trends in GDP Per Capita, 1950-2003
This Fact Sheet shows that different groups of countries have seen quite
varied patterns of growth of GDP per capita over the past 50 years. In
1950, western Europe and its offshoots (the United States, Canada,
Australia, and New Zealand) already had much higher GDP per capita
than other countries. Since 1950, they have diverged further from the
rest of the world. Though aggregate production in Asia has surpassed
that of many European countries, their high population has limited
growth in their GDP per capita. In Africa, GDP per capita has seen very
little growth since 1950. Countries that were formerly part of the USSR
saw a substantial fall in GDP per capita after 1990 due to difficult

63
 transitions to the market-oriented economy, but are already on the road
to recovery.

1.5.2 Gross national product (GNP)

GNP measures the gross national income of the residents of the country.
This income accrues in the form of wages, profits, and rents. Most of it
results from the production of output by factors of production owned by
the residents and located within the country. However, the residents may
also derive income from other countries through their investments abroad.
Such foreign income is not part of the country’s GDP. Conversely, some
of the income from the production of goods included in the country’s GDP
may accrue to residents of other countries because of their ownership of
capital and labor used in the domestic economy. Therefore, GDP needs to
be adjusted for these flows of income to arrive at GNP.
For many countries, the net flow of income — that is, the inflow of
income to residents less the outflow to non-residents — is very small as a
proportion of GDP and is usually ignored in macroeconomic modelling.
GDP is, therefore, normally used as the appropriate proxy for both
domestic output and national income.3

1.5.3 Net domestic product (NDP) and net national product

(NNP)
NDP equals GDP less the depreciation of capital used in its production.
‘Net’ stands for ‘net of the depreciation of the economy’s capital during
the relevant period’.
Similarly, NNP is GNP less the allowance for depreciation.
Fact Sheet 1.2 : Comparing GDP and GNP, 1965-2004
The following graph plots the ratio of GDP to GNP for a selected
number of countries. For many countries the differences between GDP

64
 and GNP are insignificant. For the United States, the ratio of GDP to
GNP is consistently close to 1. In others, where foreign owned
companies produce a considerable share of aggregate product, GNP can
differ substantially from GDP. In the following graph, Canada and to a
larger extent Malaysia are examples of countries where GNP is less than
GDP, because foreigners have invested more in those economies than
residents of these countries have invested abroad, so that more of their
output is used to make dividends and interest payments to foreigners
than their residents receive from abroad. Notice the sudden jump in the
GDP to GNP ratio in China during the last half of the early 1990s, i.e.,
after the market economy reforms put in place led to the jump in China’s
inflows of foreign investments.

1.5.4 Measuring GDP

Nominal GDP is (roughly) the value of currently produced goods and
services at current prices. The common methods of compiling the statistics
on it lead to three concepts of GDP. They are:
1. ‘GDP at market prices’,
2. ‘GDP by value added’, and
3. ‘GDP at factor cost’.
GDP at market prices is the most direct method of measuring GDP. This
method sums:
• The expenditures on the various commodities included in GDP and
marketed by firms. These expenditures are at market prices and include
any indirect taxes, such as sales taxes, which are charged on the purchase
of commodities. The taxes collected by the firms are remitted to the
government. The remainder is used to pay wages, interest, and rents, with
the residual becoming profits.

65
• The cost of production of the commodities included in GDP and
provided by government organizations (GOs) and non-government
organizations (NGOs).
GDP by value added sums the value added by the firms, government and
NGOs in production. The value added to the economy by these production
units is the increase in the value of GDP by their activities. For a firm, its
value added equals its sales revenue minus (i) the cost of (raw materials
and intermediate) inputs, other than labor, purchased from other firms and
(ii) sales taxes paid by it. For the government and NGOs, which do not sell
their output at market prices and/or are not supposed to have profits, value
added is calculated as the sum of the payments to factors of production.
Note that value added by the various economic units does not include the
amount of indirect taxes forwarded by the firms to the government.
Compared with the calculation of GDP at market prices, this method
avoids the potential for miscalculating GDP because of ‘double counting’,
which is explained later.
GDP at factor cost is calculated on the basis of the factor incomes paid
in the production of commodities plus the profits of firms, which are
assumed to be a cost of production by private profit-maximizing firms.
Since profits are included as a cost, the total payment to the factors of
production is identical with the amount collected by firms from purchasers
— i.e., total expenditures on commodities — less the indirect taxes that are
collected from purchasers but directly remitted to the government and,
therefore, not paid to inputs. Therefore, indirect taxes are not part of GDP
at factor cost and represent the difference between GDP at market prices
and GDP at factor cost. Compared with the calculation of GDP at market
prices, this method avoids the potential for miscalculating GDP because of
double counting.
Note that the amount of indirect taxes collected by the government is
only included in GDP at market prices, but not in GDP at factor cost or by
value added.
The calculation of value added by an economic unit is illustrated later by
an example.
The calculation of GDP and double counting
In calculating GDP at market prices, care has to be taken not to double-
count (i.e., count twice) expenditures on raw materials and intermediate
goods that go into the production of other goods. Double counting arises
because the revenues of firms that produce raw materials and intermediate
goods occur through sales to other (finished-goods producing) firms and
become part of the costs and sales revenues of the latter. If the sales

66
revenues of both types of firms are added together in calculating GDP at
market prices, the sales revenue of the former types of firms will be
counted twice: once as the sales revenue of firms producing raw materials
and/or intermediate goods and once as part of the sales revenues of firms
producing finished goods.4 Such double counting has to be avoided in the
calculation of GDP at market prices.
The problem of double counting and how it can be avoided is illustrated
by the following example. Note that in this illustration, we simplify by
setting the production by the government and NGOs to zero.
An illustration of the three methods of calculating GDP
Assume that firm A produces 200 bales of fabric. It sells 150 bales for
$1,000 to firm B to be made into clothes and sells 50 bales directly to
consumers as fabric for $750. Most of the clothes made from the 150 bales
are sold for $5,000. The rest of the clothes (unsold clothes), worth $200 at
market prices, become an addition to inventories. The fabric-producing
firm A had bought raw cotton for $600. The wages paid by both firms total
$4,000 and rents are $500. There are no indirect taxes, such as sales taxes,
though there may be income or corporate taxes. Assume that these are the
only transactions related to GDP for the year.
(a) The calculation of GDP at market prices
Under this method, GDP consists of the commodities sold to the final
purchasers and unsold goods. These are the 50 bales sold for $750 to
consumers, the clothes sold for $5,000 and unsold production worth $200,
so that the nominal value of GDP would be $5,950.
The above calculation of GDP does not include the 150 bales sold for
$1,000 as an intermediate good to other firms. The reason is that these
bales are indirectly included in the clothes made from them, so that
including both would represent double counting of these bales, once as
fabric and then as part of the clothes made from them. To avoid double
counting, GDP at market prices only includes the values of the final goods
and unsold or unused intermediate goods — with the latter constituting the
value of the changes in their inventories.
(b) The calculation of GDP by value added
First, note that the fabric-producing firm A had bought raw cotton for
$600. The value-added calculation for GDP is as follows:

Value of basic raw materials $600

Value added by firm A (=$1,000 + $750 - $600) $1,150

67
 Value added by firm B (=$5,000 + $200 - $1000) $4,200
GDP at market prices by the value-added method $5,950

(c) Calculating GDP at factor prices

This method sums the labor incomes paid, rents paid, and interest and
profits paid by each of the firms, and also arrives at the same number
($5,950). The wages paid by both firms totaled $4,000 and rents were
$500. Since the firms’ revenue is $5,950,5 their profits equal $1,450 and
are treated as a factor cost of production. Note that the calculation of
profits is done in a manner that ensures that the GDP at factor cost always
equal the GDP at market prices.
Therefore, in the absence of indirect taxes, the nominal value of GDP at
market prices is the same under each of the three methods. If there are
indirect taxes, GDP at market prices is higher than GDP by value-added or
at factor prices by the amount of indirect tax revenues. However, the actual
calculations usually involve statistical errors in the collection and
computation of data, so that the estimate of these is used to bring about the
equality of the numbers derived for GDP under the three methods.
Cautions on the use of GDP data
Some types of production and transactions that should be part of GDP
under its economic definition are not included in the usual calculations of
GDP or are included more or less approximately since it is difficult to
collect accurate data on them. The most important of these are (i) illegal
(black market) activities and (ii) legal activities but with underground
transactions. Examples of the former are the growing and processing of
illegal drugs. Examples of the latter are home repairs by casual workers
who are paid informally without a receipt so as to evade the payment of
sales taxes by the homeowner on the transaction, as well as the payment of
income taxes by the recipient. Small shopkeepers, farmers, and restaurants
sometimes follow such practices to avoid or reduce their tax payments.
The payments in many of these transactions are in currency.6
Further, the actual calculation of GDP is elaborate and costly. It is done
better in some countries than in others. Even in countries that do it very
well, the data collected for GDP is not likely to accurately reflect some
items that fall within the definition of GDP. Therefore, the calculated GDP
data always suffers from errors and omissions. To adjust for these, the
reported GDP data often includes estimates for such errors and omissions.
As a result, the reported amounts are often revised for some years after the
initial data is released.

68
 Further, although the three methods of measuring GDP are supposed to
yield identical data, they never do. To address this problem, a ‘statistical
discrepancy’ item is added or subtracted to present consistent estimates for
the three methods.
Box 1.2: GDP per capita as a Measure of the Standard of Living
Simply put, GDP, or rather its approximation GNP, is the sum of
incomes in the economy. Therefore, one way of measuring the standard
of living in a country is by dividing its real GDP by its population —
that is, by GDP per capita — since this reflects the average income
levels in the economy. We can break GDP per capita into three
components in the following manner.

where n is employment, L is the labor force, and Pop is population. y/n

is labor productivity, n/L is the ‘employment rate’ — which is the
converse of the unemployment rate,7 since a lower value of n/L means
that a smaller proportion of the labor force is employed and, therefore,
indicates a higher unemployment rate — and L/Pop is the labor force
participation rate. Therefore, the standard of living rises with labor
productivity, the employment rate and the participation rate. It declines
as the unemployment rate increases. In the short and medium term,
changes in all three factors can be significant. Over long periods, the
most important determinant of the standard of living is the growth of
labor productivity. Its growth is studied in Chapters 14 and 15 on growth
theory.
GDP as a Measure of Welfare
Increases in GDP (and GDP per capita) often do not accurately reflect
increases in society’s (family’s) welfare. While this welfare normally
depends to a considerable extent on the availability of commodities
included in GDP (i.e., provided by the market or government), there are
also many other elements that also contribute to the quality of life.
Among these are ‘production activities’ in the home8 — whose value, if
quantified at market prices, would be very high as a percentage of the
market-based GDP — and leisure activities. GDP also does not
encompass the quality of the environment — e.g., air, water, forests, etc.
If the increase in GDP comes about without a decrease in home
production, leisure, and the environmental quality, the increase in GDP
per capita can be more readily accepted as an increase in society’s

69
 welfare. Examples of such beneficial improvements include the
information technology (IT) revolution, which increased productivity
(because of more capital per worker or technical change) and increased
GDP per capita—without necessarily higher labor force participation
rates, without a reduction in leisure and without the deterioration in the
environment.
However, in other cases, the increase in GDP per capita could be due to
a higher labor force participation rate, longer working hours, and
decreases in leisure—all of which tend to increase stress levels in the
home and decrease the availability of parental time for children — or
increases in pollution. In these cases, the change in social and the
family's welfare has to be adjusted for these additional costs. An
increase in the labor force participation rate increases GDP per capita
but reduces leisure time, so that the net increase in welfare is less than
that indicated by the increase in GDP per capita. Similarly, the increase
in production at the cost of more pollution increases welfare less than
the GDP increase.
Another side of the picture is also to be noted. As factor incomes rise,
the demand for leisure tends to increase, so that more holidays tend to be
taken. In this case, all other things remaining the same, the family's
welfare increases by the increase in its income and the increase in its
leisure, so that the increase in the family's income, as indicated by GDP
per capita, does not fully indicate the increase in the family's welfare.
Further, higher incomes usually increase the demand for a better
environment, and allow the society to ‘purchase’ cleaner air by
legislating anti-pollution requirements or by using home air and water
purifiers.

1.6 Measuring the Price Level and the Rate of

Inflation

1.6.1 Measures of the price level

The economy's price level can be defined as the average price of the
commodities sold in the economy. The data on the price level is usually in
the form of an index. These indices assign a base figure of 100 for the
price level in a designated (base) year.
Several measures are used to represent the price level. The most
commonly used measures are the GDP deflator, the consumer price index

70
(CPI), and the industrial product price index (IPPI).
The GDP deflator
This index provides the average price level for all the commodities
included in the economy's GDP. Note that the GDP deflator includes the
prices of both consumer goods and capital goods (i.e., those goods which
are sold to other firms for investment purposes). The GDP deflator is the
more appropriate measure of the price level relevant to the GDP. Note also
that the GDP deflator is a price index — which measures the change in the
economy's price level over time — while an index of GDP (or the GDP
index) is a quantity index — which measures the amount of the quantities
that are produced.
The consumer price index (CPI)
This index measures the average price of a ‘basket’ of commodities
purchased by the economy's consumers. The CPI includes only the prices
of consumer goods and is more relevant for deriving the impact of
inflation on the real value of consumer incomes.
Other indices similar to the CPI can be also constructed for specific
groups, such as for students and pensioners, which would use as quantity
weights the particular set of commodities bought by the specified group.
Note that the GDP deflator includes both capital and consumer goods,
while the CPI excludes capital goods.
The industrial product price index (IPPI)
This index measures the average price of industrial products produced in
the economy during the designated period. It includes the prices of many
commodities – such as raw materials and intermediate (i.e., semifinished)
goods used in further production — which are not directly purchased by
consumers. It excludes the prices of other commodities — such as services
— that are not industrial products, so that it differs from both the GDP
deflator and the CPI. In particular, as compared with the CPI and the GDP
deflator, it measures the prices of commodities at an earlier stage of
production and distribution than at the final retail stage to consumers. This
index includes the output of semi-finished commodities in its weights.
There are also many other types of price indices. Among these could be
a ‘students’ price index’, which would measure the changes in the average
prices of goods and services bought by students, and a ‘pensioners’ index’,
which would measure the changes in the average prices of goods and
services bought by pensioners. In general, these different price indices
would not show identical changes in any given year since they incorporate

71
different weights and different methods of computation. The selection
among them has to be relevant to their usage. Therefore, if the purpose is
to proxy the cost of living of consumers, the CPI is the most appropriate
one. But if the purpose is to find real GDP from nominal GDP, the GDP
deflator is the most appropriate one. Since an increase in the prices of
producer goods increases the costs of production of goods at the retail
level and leads to an increase in the retail prices, the IPPI is useful as a
leading (i.e., occurring before) indicator of changes in both the GDP
deflator and the CPI.
The GDP deflator is the more appropriate proxy for the theoretical
concept of the price level for the economy as a whole.

1.6.2 The inflation rate

The inflation rate is the rate of change in the price level. If the price level
increases by 5%, we can specify the inflation rate as being 5% or as 0.05.
In practice, barring unusual periods such as those of wars and severe
economic depressions, the price level rises in most years. Consequently, a
graph of the price level over a long period has an upward slope. However,
the inflation rate tends to fluctuate a great deal. The inflation rate tends to
be positive and especially high during wars, when government
expenditures are mostly financed by increasing the money supply. It can
be negative, i.e., prices fall, in the deflation following some wars and in
some recessions. The biggest declines in the price level in peacetime
occurred during the Great Depression of the 1930s. Fact Sheet 1.3
provides an illustration of the CPI and the inflation rate based on it for
Canada.
The central banks of many of the developed economies now follow
‘inflation targeting,’ which is a policy of keeping the inflation rate to a low
level. To illustrate, since the 1990s, the Bank of Canada has tried to attain
the average inflation rate of 2%, with fluctuations in inflation kept between
1% and 3%.
Fact Sheet 1.3: Inflation in Canada, 1915-2007
This Fact Sheet illustrates movements in the inflation rate that can occur
in a country. This illustration is taken from Canada's experience. Over
the last century, the price level in Canada has had an upward trend.
However, the inflation rate has followed a more cyclical pattern. Periods
including World War I, World War II, the Korean War, and the two oil
crises of 1973-1975 and 1980-1981 saw substantial price increases and
therefore sharp spikes in inflation. Deflation (i.e., negative inflation rate)

72
 occurred in post- WWI period as well as during the Great Depression.

For each of the preceding measures of the price level, there is a

corresponding measure of inflation.

1.6.3 Core inflation

The prices of some of the goods in the economy are very volatile and
fluctuate a great deal, so that increases are sooner or later offset by
declines, and vice versa. Among such goods are food9 and energy prices.10
A measure that excludes their prices from the price index and focuses on
the more stable prices of the other commodities is that of ‘core inflation’.
Core inflation is thus meant to be a measure of price changes that, having
occurred, are likely to persist for several years. Core inflation measures
can be prepared for the CPI, GDP deflator and other price indices.11
Mathematical Box 1.1: Calculation of the Price Index and Growth Rates

The construction of a price index: an illustration

Suppose that we are given information on the quantities bought and the
prices of the commodities A and B for 3 years 0, 1, 2 and we need to
construct a price index for all the three commodities. The information on
prices and quantities is given in Table 1.1, where qi,0 refers to the
commodity i bought in year t, t = 0, 1, 2, and pi,t specifies its price in
year t .

73
To construct a price index for the specific basket of commodities that
contains the three commodities in the amounts bought in year 0 and
specified in column 2, we first calculate the expenditures on the three
commodities for each of the three years. These are shown in the next
table.

Suppose we now want to set the price index for total expenditures on the
designated basket of commodities bought in year 0 at 100. Then, the
price index for year 1 would be calculated as (3900/3300)100 = 118.19
and the price index would for year 2 will be (2900/3300)100 = 87.88.
The price index and the rate of inflation calculated from the price index
is:

Differences between Laspeyres and Paasche indices

Note that the quantities bought are held fixed at the level bought in a
designated year. In the above example, this was done for the year 0. The
year selected for this purpose is called the base year. The index
calculated with a past base year for quantities is called a Laspeyres
index.
However, the quantities bought tend to change over time. Some price
indices, therefore, use the basket of commodities bought in the current
year. If the last year (year 2 in the above table) were to be the current
year, the quantities bought in that year would be used to calculate
expenditures for that and all previous years. These are likely to be
different from the ones in the base year. The index calculated with the
basket of commodities bought in the current year as the standard basket
is called a Paasche index.
There are, therefore, two types of price indices:
i The Laspeyres (base-period weights) index:
This type of index uses base period quantities as weights in constructing
the price index. The disadvantage of a Laspeyres index with base period

74
quantity weights is that as time passes, new commodities come into
existence and new versions and qualities of base period commodities
emerge. Since these were not available in the base period, their prices
are not incorporated in the index, so that the index becomes increasingly
inappropriate over time. Examples of a Paasche index are the CPI
(which uses the amounts of the commodities bought for consumption in
the economy during the base year) and the IPPI (which uses the amounts
of the industrial commodities produced in the economy during the base
year).
ii The Paasche (or current-period weights) index:
This type of index uses the current-period quantities as weights in
constructing the price index. In this case, the bundle of commodities
used in calculating the index consists of commodities consumed (or
produced) in the current (not in the base) period. Therefore, a Paasche
index uses current-period quantity weights. The advantage of a Paasche
index over the Laspeyres one is that it can take account of the new
products and the improved versions of older products, which did not
exist in the base period.
The GDP deflator is often constructed as a Paasche index and uses for
its weights the amounts of the commodities included in the current year's
GDP.
Also note that the price index for each year is relative to a given year,
which is often the base year (i.e., year 0 in the above example), with the
value of the index in that year set at 100.
Separating the rate of growth of nominal income into the real growth
rate and the inflation rate
Table 1.3 showed the calculation of total expenditures for our numerical
example. These expenditures can be treated as nominal GDP, which is Y
in our usual symbols. Suppose that we wish to find the growth rate of
real GDP, i.e., which is y” in our usual symbols. To do so, we first need
to derive the expenditures on each commodity for each year by
multiplying the actual (not base year ones) quantities bought/produced
in each year by the prices in that year and then obtaining total
expenditures for each year by summing over the expenditures on each
commodity. This is done in Table 1.4.

75
 The last row in Table 1.4 shows the value of Y for each year. We next
use this data and that on the price level P calculated in Table 1.2 to
derive the value of y, which equals Y/ P , for each year and then derive
its change y′ and its growth rate y”. These calculations are shown in
Table 1.5. Note that the relevant formulas are: y′t = yt +1 – yt and y ′t = y
′t /yt .

A comparison of the quantities in Table 1.4 shows that there was a 10%
increase in the quantity of each commodity from year 0 to year 1, but
there was no change in them from year 1 to year 2. This is confirmed by
the derived growth rate y″ of total output in Table 1.5. Note that we
cannot derive the growth rate of output in year 2 since we do not have
the data for year 3.
The growth rate of real output can also be calculated from the formula
y” = Y – P ”. This calculation is presented in Table 1.6.

The calculation of y″ in Table 1.6 yields 0.12 (12%) while its calculation
in Table 1.5 was 10%, which, in fact, is the correct value, as we can see
from the increase in the quantities of the individual commodities in
Table 1.4. The reason for this difference is that the formula y″ = Y ″ – P″
is only correct for ‘very small changes’ in magnitudes but introduces an
approximation error for large changes. In our illustration, the change in
magnitudes was 10%, which is not a very small change. For analytical
purposes, the formula is very convenient and is the one to be used.
However, when working with data that embodies large changes, the
calculations should be done as in Table 1.5.

1.6.4 Deriving the rate of inflation from a price index

76
The rate of inflation during a period is the rate of change in the price level
during the period. The rate of inflation during a period t is, therefore, given
by:

where P “ is our symbol for the rate of change in P , so that P″ and π are
synonyms for the rate of inflation. Pt is the price level at the beginning of
period t and P t+1 is the price level at the beginning of period t +
1.Suppose the former is 100 and the latter is 105, we have: 105 – 100

The rate of inflation is more familiarly expressed in percentage terms, so

that, for this example, πt will equal 5% (0.05 x 100).
The inflation rate is usually reported on an annualized basis. Suppose the
length of the period is one year and we want to calculate the inflation rate
for the year 2000. The calculation will be as follows:

But suppose the data was reported on a quarterly basis. To annualize the
rate, the procedure is:

where we have multiplied by 4 since there are 4 quarters in the year.12

Box 1.3: Mathematical Formulae to Learn
Although this box presents mathematical formulae, they are essential for
economic analysis and must be learnt. The following derivations are also
included in the section on Useful Mathematical Formulae for Economics
on page xliii of this book. Since they are being utilized for the first time
in this book, their derivation is repeated here.
For the nominal value of GDP, designated as Y , the definition of the
change in Y , designated as Y ‘ (≡ΔY) is:

where the subscript refers to the time period, so that Yt' is the change in
Y during period t. The definition of y″ — that is, the rate of change
(growth rate) of Y — is:

77
Nominal GDP equals the price level multiplied by real GDP. Omitting
the subscripts, Y = Py. Note that the change in Y is decomposed into the
changes in P and y in the following manner:

Dividing the left side by Y and the right side by Py, we get the growth
rate (rate of change) of Y as:

Note that P″ is the rate of change of the price level, i.e., the ‘inflation
rate’, and y″ is the rate of growth of real GDP. This is an example of a
very useful formula: if a variable is a multiple of two other variables, its
growth rate is the sum of the growth rates of the two variables. The
above derivation implies another useful formula. Since y = Y/P , we
have:

This is an example of another useful formula: if a variable is a ratio of

two other variables, its growth rate is the difference between the growth
rates of the variable in the numerator and the variable in the
denominator .

Fact Sheet 1.4 : Measures of Inflation for the USA, 1960—2007

As the indicator for the change in price levels, inflation measures vary
depending on the basket of goods used in its calculation. Inflation
measures derived from the CPI and GDP deflator tend to follow the
same general pattern but can also vary considerably. The following
graph illustrates the differences in inflation measures. During the 1973
and 1979 energy crises, consumption goods experienced a large increase
in prices as a result of the increased cost of crude oil. This is captured in
a spike in CPI inflation much larger than that depicted by the GDP
deflator, which is based on a much larger group of goods and only those
that are domestically produced.
Nowadays, the preferred measure of core inflation for the Federal
Reserve System of the USA is the PCE (Personal Consumption
Expenditure) Index. It provides a better index for long-term inflation as
it does not reflect prices of goods that can experience temporary price
shocks such as energy or food. Notice that the PCE Index is less volatile
than either inflation based on the GDP deflator or on the CPI.

78
1.6.5 Disinflation versus deflation
‘Inflation ’ is the rate of increase in the price level. ‘Disinflation ’ is a
decrease in the rate of inflation, so that the inflation rate falls, though still
remaining positive. By comparison, the term ‘deflation’ can be used to
refer to a deflation/decline in the level of economic activity or, in some
cases, a negative inflation rate, which would be a persistent decline in the
price level.
Economists and policymakers usually prefer a low rate of inflation. For
example, the central bank of Canada has set its desired/target rate of core
inflation at 2%. If the actual rate goes above, especially substantially
above, this level, the Bank of Canada pursues contractionary monetary
policies to cause a disinflation to lower the inflation rate to the target
inflation rate. Therefore, disinflation is often viewed as a desirable
squeezing out of inflationary pressures from the economy.
Economists and policymakers are usually wary of deflation that produces
a negative inflation rate, since it is associated with significant and
persistent declines in economic activity, so that output is falling and
unemployment is rising over several quarters and years.

1.7 Nominal Versus Real Output

The data on GDP and its related variables has to be collected in the form
of the dollar value of its various components. This dollar value of GDP is
known as the nominal value of GDP. The corresponding real values of
GDP is obtained by dividing the nominal GDP by its price index, which
measures the price changes of its component commodities and was
designated above as the GDP deflator.

79
 Our symbol for nominal national output is Y and that for its real value is
y. The relationship among Y , P , and y for a given period is:

In terms of the rate of change of the variables,

so that nominal income grows by the rate of inflation plus the growth of
real output.
Alternatively, we have:

This formula is useful for calculating the growth rate of real output.
The data for GDP is collected in nominal (dollar) terms, so that it needs
to be deflated by the price level to find the real GDP and its growth rate.
The procedure is as follows:

To calculate the growth rate of real GDP for a given year t , the procedure
is:
Real GDP growth rate in t = nominal GDP growth rate in t – rate of
inflation in t.
That is, for each year, the real GDP growth rate is calculated by
subtracting the rate of inflation from the nominal GDP growth rate.

1.8 The Economic Relationship between Real

Output and Inflation
The AD-AS analysis presented in this chapter implies both a short-run and
a long-run relationship between real output and the inflation rate. In the
long run, output is independent of the price level and, therefore, of the rate
of inflation. However, in the short run, increases in aggregate demand can
increase the price level and cause inflation while also increasing real
output. Figure 1.5 shows the short-run relationship by the curve marked (y,
π) SR and the long-run one by the curve marked (y, π) LR. There is a
separate short-run AD curve for each expected rate of inflation, so that the
curve marked (y,π) SRπe=0 is drawn for a zero expected inflation rate.

80
 Note that, in terms of output increases, there is no benefit in the long run
from having inflation but there can be some benefit in the short run.
Chapters 8 and 10 will examine the precise causes and limitations of this
benefit.

1.9 The Nature of Economic Relationships

There are three types of relationships among economic variables. These
are:
1. Identities. These arise because of definitional concordance between
economic variables or because of mathematical and linguistic
relationships. An identity always holds no matter what the nature of the
economy.
2. Behavioral relationships. These reflect economic behavior and are the
result of choices made by economic agents. Examples of these are the
aggregate demand and aggregate supply relationships or curves in the
preceding analysis.
3. Equilibrium conditions. The equilibrium condition for a particular
market is that the demand equals the supply in that market. In addition,
general equilibrium in the macroeconomy specifies the requirement that
demand equals supply in all the markets of the economy. There is no
presumption that equilibrium will always exist in all markets or that it
will always exist in any given market. If equilibrium does not exist—
that is, when demand does not equal supply — in a market, the market is
said to be in disequilibrium.
Box 1.4: Definition of Equilibrium
Equilibrium is that state in a market when the demand and supply in it
are equal, so that, at the equilibrium price, the buyers can buy as much
as they want and the sellers can sell as much as they want. Such a
market is said to ‘clear’ at the equilibrium price. For the macroeconomic

81
model as a whole, the clearance of all markets in the economy is referred
to as ‘general equilibrium’ .
Equilibrium conditions versus identities
Equilibrium conditions differ from identities. The latter are meant to
apply under all conditions, irrespective of the nature of the economy.
The former hold only sometimes — i.e., when the economy is in
equilibrium — but do not apply in disequilibrium. For example, the
equality of the demand and supply in the market for a product (apples)
represents an equilibrium condition. Since the demand for a product
(apples) does not sometimes equals it supply, the equality of demand
and supply is not an identity.
Equilibrium versus disequilibrium
When the demand for a good does not equal its supply, there is said to
be disequilibrium in the market for that good. For the macroeconomic
model, we can have equilibrium in some markets while there is
disequilibrium in other markets. For example, at a particular point in
time, the demand and supply of commodities may be equal while those
for labor may not be equal. In this case, the commodity market will be in
equilibrium while the labor market will not be in equilibrium, even
though it may be moving toward equilibrium. Therefore, in this case,
general equilibrium will not exist in the economy.
Stable versus unstable equilibrium
It is always important not only to know the condition for equilibrium in
any given market but also to know the forces that exist — or are needed
— in disequilibrium to bring about changes (in price, the quantity
demanded and/or the quantity supplied) in the market. Further, we need
to know if this movement will be toward equilibrium or away from it. In
the former case, the equilibrium is said to be stable. In the latter case, it
is unstable.
Even in the case of a stable equilibrium, movement from a
disequilibrium situation to the equilibrium one usually takes time, which
could mean a few days, a few quarters, or a few years. For example,
estimates show that the impact of a change in the interest rate by the
central bank continues to have effects on nominal GDP for more than
six quarters. In this case, we say that a lag exists in the impact of a
change in the interest rate on nominal GDP. Lags in the economy are
common and important if we want to explain the performance of
economies in real time. Economists try to measure the length of such

82
 lags.

1.10 Exogenous and Endogenous Variables and

the Concept of Shocks in Macroeconomics
The AD-AS model included the three basic elements of an economic
model, which are:
1. Endogenous variables
These are variables whose values are determined within the model.
2. Exogenous variables
These are variables whole values are not determined within the model
but are specified to it. As such, they are said to be ‘exogenous’ or
‘given’ to the model. However, it is expected that they will change over
time or be changed by the modeller. One purpose of the model is to
examine the impact of such changes on the endogenous variables.
Exogenous variables can be:
• Policy variables, whose values are determined by a policymaker. They
are not determined within the model itself. Examples of policy
variables are the money supply, which is at the discretion of the
central bank, and government deficits, which are at the discretion of
the government.
• Non-policy exogenous variables.
3. Parameters
These are the elements of the model that are specified as constants and
as such are not expected to change. Therefore, they are also known as
the constants of the model. They usually appear as coefficients attached
to variables. However, they are sometimes changed to investigate the
impact of such changes on the endogenous variables.

1.10.1 An illustration
Suppose our model solely explains consumption expenditures and
specifies that consumption expenditures c depend on disposable national
income y d which equals national income y less taxes paid t. Let us assume
that this relationship is of the form:

c = a + b(y – t).

83
Here, c is an endogenous variable (explained by the model), y is an
exogenous (not explained by the model) non-policy variable, t is an
exogenous policy variable determined by the government, and a and b are
parameters.

1.10.2 Shocks
A shock can be defined in a broad sense as a shift in the value of an
exogenous variable or parameter. Such a change may be anticipated or
unanticipated. However, in some ways, the notion of a shock is more
appropriate for an unanticipated shift than for an anticipated shift and
some economists use this term in this narrower sense.

1.10.3 Multipliers
A major purpose of macroeconomic modelling is to study the impact of
changes in the policy variables or the non-policy exogenous variables on
the endogenous variables. This impact is often captured through the notion
of ‘multipliers’. A multiplier indicates the change in an endogenous
variable for a unit change in an exogenous variable. For example, if the
endogenous variables are nominal income Y and real income y, while the
policy variable is the money supply M , the relevant multipliers are
designated as: 13 and . Here, the interpretation of is ‘the
change in Y for a very small change in M ’, with ‘all other variables being
held constant’. The symbol is a Greek letter pronounced as ‘di’ and
designates ‘a small change ’ in the following variable, when all other
variables are held constant. is pronounced as ‘di y di M ’.
Our earlier analysis of the long-run real output in the economy implied
that changes in aggregate demand do not affect real output in the long run,
so that we have for the long-run LR = 0. Our earlier analysis also
showed that in the short run, was not zero along the short-run
SRAS curve, so that SR > 0. Hence, the values of the economic
multipliers can differ between the short run and the long run. The long-run
and the short-run multipliers are analytical ones, without a precise
chronological correspondence in real time. These analytical multipliers
have to be distinguished from the chronological or real-time multipliers.
In terms of chronological time, with data on a quarterly basis, the impact
multiplier is the value of the multiplier during the first quarter after a
shock. If the data are on an annual basis, the impact multiplier is the value
of the multiplier during the first year after the shock. If there exist lags in

84
the effects of the shock, the impact multiplier would be less than the short-
run and the long-run multipliers. This is a common occurrence for the
policy and other multipliers in macroeconomics. Therefore, in assessing
the impact of any policy on the real-time economy, we not only need to
know the analytical multipliers but also have knowledge of the impact
multipliers and the likely lags — or, at least, possess some intuition on
them.

1.11 Growth Theory

Economics studies the very long-run growth of real output and standards
of living under the heading of growth theory. The focus of this theory is on
the analytical forces contributing to growth over long periods.
Our earlier discussion, in the AD-AS analysis, of the long-run AS curve
shows that the long-run growth of output will mainly depend on the
growth of the labor force and physical capital, and improvements in
technology. Growth theory analyses their role and importance, as well as
the contribution of other forces to growth. Growth theories are covered in
Chapters 14 and 15.

1.11.1 Growth of the standard of living

We argued earlier that a rough measure of the standard of living is output
per capita. Using this measure, the standard of living in the long run would
equal the full-employment output divided by the population. We can,
therefore, express the long-run standard of living as:

where nf is the full-employment level of employment, L is the labor force,

and Pop is population. Therefore, in the long run, the growth rate of the
standard of living is the sum of the growth rates of its three components. In
symbols, the above equation implies that the growth rate of the standard of
living is given by:

where”; indicates the growth rate of the ratio in the preceding bracket. Of
these ratios, nf/L cannot grow indefinitely and neither can L/Pop , though
both can increase (or decline) over some periods. Therefore, the long-run

85
growth of living standards can only result from the growth rate of yf/nf,
which is long-run output per worker or labor productivity. This increase in
labor productivity occurs because of improvements in the techniques of
production (technical change) and increases in the amount of capital per
worker. As Chapters 14 and 15 on growth theory indicate, empirical
estimates show that the major part of the long-term increases in living
standards has been due to technical change, i.e., changes in the methods of
production, such as those which occurred in the Industrial Revolution or
have been occurring in recent decades in what may be called the Computer
and Internet Revolution. The ‘carriers’ or embodiment of this technical
change have been human and physical capital.

1.12 Business Cycle Theories

Real output does not always increase at a constant growth rate. It grows
faster in some years than in other years; it may even decline in some years.
Overall, there is a cyclical pattern with some years of rising output,
followed by some years of falling output, which are followed by rising
output, and so on. Since output has a long-term rising trend — that is, it
increases over long periods — the deviation of output around this trend is
used to specify the cyclical behavior of output.
Since the unemployment rate is negatively related to real output, the
fluctuations in unemployment follow an opposite pattern to that in output
and are said to be counter-cyclical.
Business cycles are defined as short-term fluctuations in real output and
its related variables such as unemployment. These fluctuations are called
cycles since they are somewhat periodic, with a downturn following a peak
and an upturn following a trough. In Canada, the typical business cycle in
output has an upturn that lasts several years while the downturn is usually
milder and lasts at most a couple of years. Other popular terms associated
with business cycles are booms and recessions. The economy is said to be
in a boom if its output is above its long-run potential level or above its
average level over the business cycle. This happens toward the top of an
upturn and the early parts of the downturn. The economy is said to be in a
recession if its output is below its long-run potential level or below its
average level over the business cycle. This happens toward the bottom of a
downturn and the early part of the upturn.
The internal structure of the economy is a significant factor in producing
its cyclical activity. Some economies may have an internal mechanism that
produces fluctuations while others need periodic shocks to do so. In the

86
latter case, the economy must respond to shocks through fluctuations in its
performance. A real-world economy may possess both these
characteristics.
The preceding AD-AS analysis — with the SRAS curve rather than the
LRAS one — implies that the changes in output can come about because
of shifts in either aggregate demand or in aggregate supply. This leads to
two radically extreme types of business cycles theories. They are:
1. Business cycle fluctuations in output are mainly caused by shocks
originating in aggregate supply.
2. Business cycle fluctuations in output are mainly caused by shocks
originating in aggregate demand.
Intuitively, as our earlier basic treatment of the AD-AS model shows, both
demand and supply shocks can cause changes in output and
unemployment. In practice, both are likely to occur and lead to or
contribute to the actual business cycles experienced in the economy. This
is spelled out in Chapter 16 on business cycles.
Business cycle theories study the determinants of the real-time, real-
world cyclical fluctuations in output and other economic variables.
Therefore, the concern of business cycle theories is with chronological,
rather than analytical, periods during which both shifts in both AD and AS
routinely occur. Business cycle theories attempt to sift through the relative
importance of these AD and AS shifts. Some theories claim that the main
sources of real-world fluctuations lie in real-time shocks to aggregate
demand while others claim that the main sources of fluctuations lie in real-
time shocks to aggregate supply. An extreme version of the latter asserts
that only shocks originating from the supply side of the economy can
cause business cycles. The theories that explain business cycles in this
manner are called real business cycle models. They form a subset of
business cycle theories. However, as pointed out above, shifts in both
aggregate demand and supply do occur frequently and each can cause
fluctuations in economic activity, so that there are two broad sources of
business cycles. Their analysis is pursued in Chapter 16.
Fact Sheet 1.5: Booms and Recessions in the USA Since 1960
Economic fluctuations are a persistent pattern of output variations in
every industrialized economy. This Fact Sheet illustrates such
fluctuations for the USA. These economic fluctuations are composed of
booms, recessions, and the turning points of troughs and peaks.
Recessions are prolonged non-positive growth periods in GDP. In the
United States, the dates of recessions are officially announced by the

87
National Bureau of Economic Research (NBER), which is a think-tank
in economics. Over the past 50 years, eight recessions have been
recorded with an average length of 10 months. Some of the deepest have
been (1) 1973-1975, due to the rapid rise in world oil prices and the end
of the Vietnam War; (2) early 1980s, with the second oil crisis and tight
monetary policy; (3) early 1990s, following the bursting of the Internet
dot-com bubble, which led to a precipitous decrease in the prices of their
stocks and the consequent decrease in the wealth of their shareholders;
and (4) following the financial crisis of 2007-08. The following graph
shows the official dates of the recessions by shaded columns.

Box 1.5: The Fundamental Role of Economics as a Science

In studying macroeconomics, keep in mind that economics is a science.
A science is a discipline whose objective is to explain the real world.
This is done through its ‘theories’. Theories are attempts to understand
specific segments of the economic reality. The real world is too complex
and has too many dimensions to encapsulate within the limited construct
of theories. Therefore, theories are simplifications — such as caricatures
— of reality. As such, they may be valid or not, or better for explaining
some aspects of reality than others. Economics tries to judge their
success or failure in explaining the relevant observation of the real world
by intuition and statistical methods using data. Both are needed and
useful in judging the validity and relative usefulness of theories. The
science of using statistical data to test economic theories is known as
econometrics, which is a component of economics.14
The role of economics in explaining the real world is also referred to
as its ‘positivist role‘ . Another aspect of economics is its ‘normative
role’ — i.e., offering policy prescriptions to improve on the performance
of the economy, hopefully as a means of increasing the welfare of its
citizens, or of the functioning of one of its markets.

88
Theories/models in economics
A theory is a simplified system of relationships — in other words, a
‘model‘ — for understanding or explaining some aspect of the real
world. The AD-AS model discussed earlier is a theory. A theory may be
valid (i.e., applicable) or invalid (i.e., at odds with the observations).
Clearly, theories that are invalid should be discarded. All theories are
invalid to some extent or under some circumstances, so that faith in
them usually depends on experience about their degree of usefulness or
failure.
Macroeconomics often continues to maintain several theories for
explaining any given economic phenomenon — e.g., the current rate of
inflation or unemployment. Using the inflation rate as an example of a
variable whose determination is to be explained, economists have to
choose that theory, among the competing ones, which does the best job
in explaining the experienced rates of inflation. However, judgments on
which theory performs best in this respect often differ among
economists and lead to controversies. As mentioned above, a science
seeks to explain specific aspects of the real world. Economics is a
science in this respect. However, not all of the theories in any of the
sciences are valid under all conditions. Some sciences do a better job of
explaining their field of reality than others, and are held in greater
respect by the public than those that often do poorly. Historically, the
public's faith in the accuracy and usefulness of economics, especially as
a guide to policy, has tended to wax and wane.
Economics as the ‘premier social science’
Sciences are broadly separated into two categories: the natural and the
social sciences. The former deal with natural phenomena. Examples of
natural sciences are biology, chemistry, and physics. Social sciences
study aspects of society. Examples of social sciences are anthropology,
economics, political science, and sociology. Economics is sometimes
called ‘the queen of the social sciences’ because it is the most highly
developed one of the social sciences. In general, the natural sciences
have had historically a longer period of development than the social
sciences. Further, the subjects of the social sciences are human beings,
whose behavior seems harder to predict than natural phenomena. Part of
the reason for this is that their behavior can change in response to
policies based on the predictions of the theory. These provide some of
the reasons the public is more sceptical about the success of economic
policies in practice than of natural sciences.
Sciences differ from disciplines that do not try to explain the real world.

89
 Among the latter are languages, which are means of communication and
analyses, and are built on systems of definitions and identities.
Mathematics is a language in this sense. Economics needs to make
extensive use of languages, including mathematics, in the formulation
and development of its theories. Therefore, economic theories include
identities among their relationships. However, theories must also include
some relationships that are not identities but rather assertions about the
real world – thereby imparting to the theory its critical element (as a
component of a science) that it may not be valid for the phenomena that
it seeks to explain.

1.13 Paradigms in Macroeconomics

There are several schools of thought or paradigms in macroeconomic. The
currently dominant ones are the classical and Keynesian schools. The
classical paradigm originated in the 18th century and periodically becomes
the dominant doctrine in macroeconomics. It views the capitalist economy
as one that tends to perform at full employment and produce full-
employment output. It does allow for the possibility that the economy will
sometimes produce less or more than this level, but views any such
deviations from full-employment output as transitory and self-correcting,
so that the central bank and the government do not need to pursue
monetary and fiscal policies to speed the economy to its full employment
level. As a corollary, even if the economy enters into an economic crisis, it
can and should be left alone to recover. Further, the adherents of the
classical paradigm maintain that intervention by the central bank and the
government could worsen, or worsen as often or more often than improve,
the performance of the economy.
The Keynesian paradigm originated in (John Maynard) Keynes' writings,
especially with the publication in 1936 of his book, The General Theory of
Employment, Interest and Money. This paradigm takes the view that the
economy often does not perform at full employment; it may do so
sometimes but not at other times. Further, the deviations from full
employment are often not transitory and self-correcting. Hence, central
banks and governments need to pursue monetary and fiscal policies to
speed the economy to its full-employment level and to maintain it there.

1.14 Economic and Political Systems:

90
 Organization of the Macroeconomy

1.14.1 Capitalism
Capitalism is an economic system in which production, income
distribution, and other aspects of the economy are controlled by individual
economic units and allocations of products and inputs are based on prices
determined by freely operating markets. In particular, capital (physical or
financial) is owned by individuals and firms (which are owned by
individuals). In a pure form of capitalism, the state's role in the economy is
minimal and is confined to such public goods as providing law and order,
defence, etc. In this form, the government does not engage directly in
production and does not attempt to change the distribution of incomes in
the economy. A corollary of the latter is that the government does not
provide subsidies or make transfer payments either to firms or individuals.
The fundamental argument in favor of capitalism is based on the theory
of prefect competition, which implies that individual economic units acting
in their best interests (i.e., pursuing profit or utility maximization)
maximize efficiency in production and allocation of inputs among
different uses and firms under the existing technology. An additional
argument in favor of capitalism is that it provides incentives in the form of
higher profits and personal incomes for invention and innovation, which
produce changes in production techniques and raise the growth rate of the
economy over time.

1.14.2 Marxism and communism

There are also other approaches to the macroeconomy. One of these is that
of the Marxist school, which is based on the writings of Karl Marx around
the middle of the 19th century. In general, this school maintains that the
capitalist system is inherently flawed. From a social perspective, it
produces large (and unfair) income inequalities, which create a strong
distinction between the working and capitalist classes, resulting in conflict
and unrest. Further, the nature of the capitalist system is such that it
periodically breaks down and collapses (as in crises, recessions, and Great
Depressions). While the classical and Keynesian schools seek to
understand the workings of the capitalist economy and to improve on its
performance, Marxists see the replacement/overthrow of the capitalist
system as the essential way out of its flaws. Marxist economic analysis
provided the basis of communism, which has highly or fully centralized

91
economies with the government controlling all production and distribution
in the economy. While capitalism relies on markets to determine the prices
of commodities, production, and income distribution, these roles are
performed by bureaucrats in a communist system. While, in principle
under the rather unrealistic assumption of perfect knowledge, such
centralized/bureaucratic allocations can be made the same as under the
market system, in practice they induce major inefficiencies in production
and factor allocations. Further, compared with capitalism, the communist
system does not provide incentives to individuals and production units to
invent and innovate, so that the long-run growth rates of the economy tend
to be lower under a communist organization of production than under a
capitalist one.

1.14.3 Socialism
Compared with communism and capitalism, socialism (defined in such a
way as to exclude communism) provides a system and philosophy that
allows for a largely capitalist economy but wants to modify it in many
respects. Some of these departures from capitalism include government
ownership of production units, sometimes involving public ownership of
industries as a whole (thereby excluding the private sector completely
from the selected industries). The resulting macroeconomy with some
production units owned by the government and some by private economic
agents is often termed a ‘mixed economy’. Other departures from the pure
form of capitalism may not involve any ownership of production units or
exclusive ownership of industries by the state but try to achieve a better
income redistribution through progressive taxation, subsidies, and
government expenditures. Social programs in this category include income
support programs to individuals and households (often called welfare
programs, intended to ensure that everyone has a minimum income level),
child support, old age security, unemployment insurance, etc. The intent
behind such programs is to obtain the allocative and production efficiency
provided by a capitalist, market-determined, production system, while
generally ameliorating the income inequalities that arise under capitalism,
and reducing the uncertainty of incomes for the poorer segments of
society.
Note that, sometimes, a country, e.g., China at present, tries to combine a
communist political system (in China's case, inherited from the total
communist system in earlier years) with a capitalist economic mode of
production and factor allocation in specific economic sectors of the
economy. The latter is intended to capture the allocative efficiency and

92
long-run growth that private economic incentives in the form of higher
profits and incomes tend to provide, thereby avoiding the inefficiency in
production and lower long-run growth usually experienced under a
communist economic system with allocation of inputs and production
decisions by bureaucrats.

1.15 Conclusions
• The full, most productive use of the economy's factors — consistent with
the desires of the owners of the factors of production and the structure of
the economy — produces the full-employment level of output. This is
called the long-run equilibrium level of output. Changes in aggregate
demand do not change this output.
• The actual level of output in the economy at any given time can differ
from the full-employment level. Some of these differences occur in short-
run equilibrium, which allows increases (decreases) in output as
aggregate demand increases (decreases). The actual output level may
even differ from the short-run one if the economy is not in short-run
equilibrium. Over the business cycle, trend-adjusted output increases in
upturns and decreases in downturns.
• The rate of unemployment is negatively related to output. It increases in
downturns and decreases during upturns in the economy.
• The statistical measures of national output and income are GDP (or
NDP) and GNP (or NNP). These are calculated at market prices or at
factor cost or by value added.
• For the real value of output, these measures are deflated by a price index.
The most appropriate measure for deflating nominal GDP to arrive at real
GDP is the GDP deflator. The CPI is the more appropriate index for
studying changes in the standard of living of consumers. The CPI for
Canada uses base-period weights for the quantities of commodities while
the GDP deflator for Canada uses current-period weights.
• Growth, business cycles, crises, and booms are inherent features of free
market economies.
• The role of economics is to explain the real world. It formulates theories
and tests them for their ability to explain real-world and real-time
economic variations. However, the core analytical apparatus of
economics uses the analytical concepts of the long run and the short run,
whose correspondence with the chronological passage of time – as in the
real-time concepts of the short term and the long term – is at best very
rough. Intuition plays a major role both in the formulation of theories and
their application to real-time economies. Mathematics plays a major role

93
 in deriving rigorously the implications of theories.
• The two main paradigms in economics for the study of free market
economies are the classical and Keynesian ones.

KEY CONCEPTS

Aggregate demand
Aggregate supply
Price level
Short-run aggregate supply
Long-run aggregate supply
Full-employment output
The natural rate of unemployment
GDP and GNP
GDP in market prices
GDP at factor cost
GDP deflator
Consumer price index
The rate of inflation
Identities versus equilibrium conditions
Short run versus short term
Long run versus long term
Business cycles
Short-run macroeconomic model versus growth theory
Paradigms in economics
Capitalism, communism, and socialism, and
The definition of a science, social sciences, economics as the queen of the
social sciences .

CRITICAL CONCLUSIONS

• GDP per capita can be used as a proxy for living standards over time and
across countries.
• The AD-AS diagram is an important tool for studying the determination
of real output and the price level.
• The long-run level of output is called the full-employment level of
output. The long-run level of unemployment is called the natural rate or

94
 full-employment rate of unemployment.
• Inflation may cause an increase in output in the short run but does not do
so in the long run. It may reduce unemployment in the short run but does
not do so in the long run.
• In the long run, increases in aggregate demand only cause inflation, not
increases in output or decreases in unemployment.
• The long-run increase in living standards only comes from increases in
output per capita, which depends on increases in capital per worker and
technical change. These are studied in growth theory.
• Short run and long run are analytical concepts. These allow the study of
economic changes in analytical time. Short term and long term are
chronological concepts that are needed for the study of real-world, real-
time economies. In general, the short-term effects — in the year ahead —
of any shock include both short-run and long-run effects, but may not
encompass even all the short-run effects.
• Business cycles are fluctuations of output and unemployment over time.
They occur in real time. Business cycle fluctuations are explained by
taking into account the structure of the economy and both the short-run
and long-run effects of shocks originating in the supply and demand sides
of the economy.
• Growth, business cycles and crises are inherent and interrelated features
of free market economies.
• Identities always hold. They differ from equilibrium conditions, which
hold when there is equilibrium but not when there is disequilibrium.
• Economics is a science. It uses theories/models to explain economic
phenomena and to make predictions about the real world. It uses
econometrics to test its theories.
• The common form of organization of modern industrial economies is the
free market system arising out of capitalism, with socialism representing
modified forms of capitalism to cure some of the perceived social and
economic ill-effects of capitalism. Communism in its purest form
replaces the free market economy by a centralized, bureaucratic-run one.

REVIEW AND DISCUSSION QUESTIONS

1. Show diagrammatically the long-run and short-run levels of real output

and the price level with equilibrium at full employment. Now show the
effects of (a) a permanent increase in demand, (b) a temporary (i.e.,
short run but without being long run) decrease in supply, and (c) a

95
 permanent (i.e., both short-run and long-run) decrease in supply.
2. What are the main components of aggregate demand (AD) in the open
economy?
3. Why can output differ in the short-run from the full-employment one?
Can it be even higher? How do you explain this?
4. Specify the equation relating the unemployment rate to the employment
rate.
5. For the open economy, what are the two main methods of measuring
gross income and output of the economy? What is the difference
between their gross and net versions?
6. What are the three methods of calculating GDP and what are the
differences between them? What is double counting?
7. Developing countries typically have much smaller participation of
women in the labor force and produce a larger proportion of their
consumption in home-produced goods. Their agricultural sector also
produces a larger proportion of the food consumed by the farmers
themselves. Would these patterns produce lower official values of GDP
per capita than for the developed economies? Why? How might the
measures of GDP per capita be adjusted to eliminate this mis-
measurement of the standard of living in cross-country comparisons?
8. How does a closed economy differ from an open one? What are the
main markets studied for the closed economy macroeconomics?
9. When would there be disequilibrium between aggregate demand and
supply? Show diagrammatically? What do you expect to happen to
prices in your shown situation? Define a stable equilibrium. Does your
diagram indicate the stability of equilibrium? Give reasons for your
answer.
10. Specify the distinctions between long-run, short-run, and actual
output? Illustrate your answer with appropriate diagrams of the
relationship between long-run and short-run output and the price level?
Explain the reasons for the different relationships?
11. What are business cycles? What two types of explanations are possible
for their causes? What explanation do the real business cycle theories
favor?

ADVANCED AND TECHNICAL QUESTIONS

Basic model for Chapter 1

You are given the following information for the year 2000 on an economy.

96
All amounts are in current dollars.

Goods and services produced by factories 10,000

Revenue from sales during the year (goods sold) 8,000
Labor costs 7,000
Interest paid to domestic residents 1,600
Interest paid to foreign residents 400
Taxes paid by factories to the
100
government
Goods purchased by some factories from
900
others for inputs
Value of the factories as of January 1,
150,000
2000
Value of the factories as of December 31,
150,000
2000
Goods carried over in the economy on
200,000
January 1, 2000
Value of home activities, such as meals,
14,000
child-care, etc.
Interest and profits received by residents
100
on foreign investments
Paid volunteer activities by NGOs 200
Unpaid volunteer activities, such as
300
home visits to the elderly, by NGOs
Sales taxes paid by residents to the
150
government

T1. For the basic model,

a Calculate GDP for this economy by three different methods. Do these
calculations yield the same amount?
b Calculate GNP and the difference between GDP and GNP at market
prices.
c If the value of home production, such as care for the elderly, increases
by $90, recalculate GDP.
d If persons engaged in home production reorganize their production
arrangements into factories and start selling all of their output to others,
while buying back a corresponding amount for their own consumption,
re-calculate GDP.

97
 e If the NGOs begin to raise the funds needed by donations from
residents and pay all their (including formerly unpaid) volunteers,
recalculate GDP.
T2. Assuming that real GDP remains constant, suppose that at the
beginning of year 1, nominal GDP was $1000 and the price level was 2. If
the inflation rate in year 1 was 10%, what would be their values at the
beginning of year 2?
T3. Suppose that at the beginning of year 1, nominal GDP was $1000 and
the price level was 2. If the inflation rate in year 1 was 10%, what would
be their values at the beginning of year 2? If the inflation rate in year 2 was
25%, what would be their values at the beginning of year 3? What is the
amount of inflation over the two years? Does it differ from the amount
obtained by adding the inflation rates over the two years? If not, why not?
T4. Are productivity shifts endogenous or exogenous in the AD-AS
model? Why? Suppose productivity increases. Show its effects on the two
main endogenous variables of that model?

1NGOs are private organizations that produce goods and services but are
not profit maximizing. Examples of NGOs are the Salvation Army,
Oxfam, and other charitable organizations.
2Examples of such products are free education in government schools,
partly subsidized education in universities, and activities of service
organizations such as the YMCA.
3GNP can be significantly lower than GDP for countries that have
significant foreign ownership of its assets and foreign workers, without a
corresponding ownership by it of foreign assets and its workers working
abroad.
4This problem does not arise for the ‘in-house’ production of intermediate
goods within the same firm.
5Note that unsold products are evaluated at market prices and included as
part of this revenue.
6Therefore, some economists use an exceptionally high usage of currency
as an indicator of unreported transactions and production. They use the
excess usage of currency to estimate the levels of unreported production
and sales and add these amounts to the reported levels of production to
compute GDP.
7The unemployment rate equals (L - n)/L.
8Examples of this are: home-cooked meals, care of children, personal

98
interaction, etc.
9Food prices tend to be sensitive to weather and seasonal influences. An
exceptionally rainy season or a drought can cause such prices to rise very
significantly, and then be offset some weeks or months later as supplies
come in from other areas.
10Energy prices are sensitive to weather conditions such as exceptionally
cold weather, to manipulation by the international oil cartel, OPEC, and to
its ability to stabilize overall oil supply levels by enforcing production
quotas on its members.
11Core inflation measures are especially useful for the formulation of
monetary policy since the central bank needs to know which price
increases are likely to persist and which are likely to be reversed later.
12This formula is an approximation to the correct one, which requires
compounding of the growth rate.
13To reiterate, this expression is a short form for ‘the change in Y for a
small change in M, other things remaining the same’.
14The relationship between econometrics and economics is somewhat
similar to that between engineering and physics.

99
 CHAPTER 2
Money, Prices, Interest Rates, and
Fiscal Deficits

Some of the fundamental questions of macroeconomics concern

the proper role of monetary, fiscal, and other policies in managing
the economy.
Money is the medium of payments. The money supply and
interest rates have important effects on the economy. Changes in
the central bank’s monetary policy on these variables are major
determinants of aggregate demand in the economy and, therefore,
of changes in the economy’s output, unemployment, price level,
and the inflation rate.
Governments usually run fiscal deficits or surpluses, which also
impact on aggregate demand and interest rates, and then on
output and unemployment.

Among the most important questions in macroeconomic analysis are

whether, how and to what extent changes in the money supply, interest
rates, the price level, and inflation affect the economy’s real variables,
especially national output and employment. The control of the money
supply and interest rates rests with the central bank or ‘the monetary
authority’ . Its policies with respect to the money supply and interest rates
are known as monetary policy.
The government or ‘the fiscal authority’ determines the policies on
government spending, taxes, and the fiscal deficit/surplus. Its policies with
respect to these variables are known as fiscal policy.
Both monetary and fiscal policies impact on a considerable number of
macroeconomic variables. In particular, they affect aggregate expenditures
and interest rates in the economy.

2.1 What Is Money and What Does It Do?

100
2.1.1 The functions of money
Money is not itself the name of a particular asset and is best defined
independently of the particular assets that may exist in the economy at any
one time, since the assets which function as money tend to change over
time in any given country and among countries. At a theoretical level,
money is defined in terms of the functions that it performs. The traditional
specification of these functions is:
(i) medium of exchange/payments,
(ii) store of value, sometimes specified as a temporary store of value or
temporary abode of purchasing power
(iii) standard of deferred payments, and
(iv) unit of account.
Of these functions, the medium of payments is the most essential
function of money. Any asset that does not directly perform this function
— or cannot indirectly perform it through a quick and costless transfer into
a medium of exchange — cannot be designated as money. A developed
economy usually has many financial instruments that can perform such a
role, though some do so better than others. The particular instruments that
perform this role vary over time, with currency being the only or main
medium of exchange early in the evolution of monetary economies. It is
supplemented by demand deposits with the arrival of the banking system
and then by an increasing array of financial assets as other financial
intermediaries become established.

2.1.2 The practical definitions of money

Historically, the definitions of money have measured the quantity of
money in the economy as the sum of those items that serve as media of
exchange in the economy. However, at any time in a developed monetary
economy, there may be other items that do not directly serve as a medium
of exchange but are readily convertible into the medium of exchange at
little cost and trouble and can simultaneously be a store of value. Such
items are close substitutes for the medium of exchange itself.
Consequently, there is considerable controversy and disagreement about
whether to confine the definition of money to the narrow role of the
medium of exchange or to broaden it to include close substitutes for the
medium of exchange.
The theoretically narrow definition of money is that it is the good that
directly serves as a medium of payments. This narrow definition of money
is given the symbol M1. It is defined in practical terms as the sum of the

101
currency (i.e., notes and coins) in the hands of the public and the public’s
demand deposits in commercial banks. Payments for commodities are
usually made by the transfers of currency and demand deposits (through
checks and debit cards) from the buyer to the seller. Close substitutes to
M1 are referred to as near-monies.
A broader definition of money that has won the widest acceptance
among economists is known as the (Milton) Friedman’s (or broad)
definition of money and has the symbol M 2. It defines money as the sum
of currency in the hands of the public plus all of the public’s deposits in
commercial banks. The latter include demand deposits as well as savings
deposits in commercial banks.
A still broader definition of money is M 2 plus deposits in near-banks —
i.e., those financial institutions in which the deposits perform almost the
same role for depositors as similar deposits in commercial banks.
Examples of such institutions are savings and loan associations and mutual
savings banks in the United States (USA); credit unions, trust companies,
and mortgage loan companies in Canada; and building societies in the
United Kingdom (UK). The incorporation of such deposits into the
measurement of money is designated by the symbols M3, M4, etc., or by
M2A (or M2+), M2B (or M2++), etc. However, the definitions of these
symbols have not become standardised, so that their definitions remain
country specific.

Financial institutions in the economy

Financial institutions are firms involved in the process that determines the
money supply and interest rates. They also intermediate between the
borrowing and lending processes in the economy. In practical terms,
financial institutions include the central bank, commercial banks, near-
banks such as credit unions, trust companies, brokerage companies, postal
banks, pension funds, etc. They do not engage in the production or
consumption of commodities but receive funds from some sources and
channel them to others (i.e., invests them).

2.2 Money Supply and Money Stock

Money is a good, which, just like other goods, is demanded and supplied
by the various participants in the economy. There are a number of
determinants of the demand and supply of money. The most important of
the determinants of money demand are national income, the price level,
and interest rates, while that of the money supply is the behavior of the

102
central bank of the country which is given the power to control the money
supply and bring about changes in it.

The equilibrium amount in the market for money specifies the money
stock , as opposed to the money supply , which is a behavioral function.
These are depicted in Figure 2.1a with the nominal quantity of money M
on the horizontal axis and the market interest rate r on the vertical axis.
The money supply curve is designated as M s and the money demand
curve is designated as M d. The equilibrium quantity of money is . It
equals the quantity of money supplied at the equilibrium interest rate .
Note that the quantity is strictly speaking not the money supply, which
has a curve or a function rather than a single value. However, is the
money stock that would be observed in equilibrium.1
The money supply and the money stock are identical in the case where
the money supply is exogenously determined, usually by the policies of the
central bank. In such a case, it is independent of the interest rate and other
economic variables, though it may influence them. In this case, the money
supply ‘curve’ will be a vertical line, as shown by the line M s in Figure
2.1b, with the money supply as M 0. Much of the monetary and
macroeconomic reasoning of a theoretical nature assumes this case, so that
the terms ‘money stock’ and ‘money supply’ are used synonymously. One
has to judge from the context whether the two concepts are being used as
distinct or as identical ones.
Note that the symbol M in Figures 2.1a and 2.1b can represent any of the
relevant money supply variables, i.e., M 1, M 2, etc. These symbols were
defined earlier.
The control of the money supply rests with the monetary authorities.
Their policy with respect to changes in the money supply and interest rates
is known as monetary policy.

2.3 The Nominal Versus the Real Value of the Money

103
 Supply
It is important to distinguish between the nominal and the real value of the
money stock. The nominal value of money is in dollars — i.e., in term of
money itself as the measuring unit. The real value of money is in terms of
its purchasing power over goods and services. Thus, the nominal value of a
$1 note is 1 — and that of a $20 note is 20. The real value of money is the
amount of goods and services one unit of money can buy and is the
reciprocal of the price level of the commodities (goods and services)
traded in the economy. It equals 1/P where P is the price level in the
economy. The real value of money is what we usually mean when we use
the term ‘the value of money’. The real value of M 1 equals M1/P and the
real value of M 2 equals M2/P , where P is the economy’s price level.
The demand and supply functions of money are often stated in nominal
terms in a general analysis involving both of them. However, the demand
for money is mainly by the public, which is concerned not so much with
its nominal as with its real value since the latter represents its ability to buy
commodities. Therefore, the demand for money is usually investigated in
real terms. However, the supply of money is mainly determined by the
central bank in nominal terms.2

2.4 Bonds and Stocks in Macroeconomics

The common definition of a bond is that it is a financial instrument with a
fixed payment, known as the ‘coupon ’, at pre-specified intervals plus the
repayment of the principal amount at a designated time — known as the
redemption or maturity date. Bonds are issued by the government
(including those of the states/provinces and municipalities) and
corporations and are normally traded/re-traded in the financial markets.
The common definition of a share — shares, equities, or stocks in the
plural — of a corporation is that it is a financial instrument which entitles
the shareholder to the declared dividends of the corporation and
participation in the ownership rights of the corporation. Normally, shares
do not have a (fixed) coupon payment and there is no designated
redemption date. However, hybrid forms of shares and bonds also exist.
An example of such a form is preferred shares and convertible bonds (i.e.,
bonds that may be converted to shares).
Bonds and shares are normally not acceptable to sellers in payment for
their commodities. Since both bonds and stocks are financial instruments
but neither qualifies as a medium of payments, macroeconomic analysis
aggregates them into the single concept of ‘bonds’ in macroeconomics.

104
Therefore, the term ‘bonds’ in macroeconomics and in this book refers to
non-monetary financial assets and differs from its meaning in ordinary
English language usage.
There exist in the economy many financial instruments that lie at the
interface between money and bonds. Such instruments are not directly a
medium of payments but can be easily and almost without delay or cost be
converted into demand deposits or currency. Among these are savings
deposits in commercial banks and other financial institutions. There is no
hard and fast rule about their classification. For some types of analysis,
they are included in money and excluded from bonds, while for other types
of analysis they are included in bonds and excluded from money.

2.5 The Definition of the Money Market in

Macroeconomics
The money market in macroeconomics is defined as the market in which
the demand and supply of money (M1, M2, or a broader money measure)
interact, with equilibrium representing its clearance. However, the
common English language usage of the term ‘money market’ refers to the
market for short-term bonds, especially that of Treasury bills. To illustrate
this common usage, this definition is embodied in the term, Money Market
Mutual Funds, which are mutual funds that invest in short-term bonds.
It is important to note that our usage of the term ‘the money market’ in
this book will follow that of macroeconomics. To reiterate, the ‘money
market’ in macroeconomics means the market for money, not the market
for short-term bonds. The analysis of the money market is based on the
demand and supply of money and of equilibrium between them.

2.6 A Brief History of the Definition of Money

The evolution from a barter economy to a monetary economy usually
starts with a commodity money — i.e., a commodity used in consumption
also begins to be used as a medium of payment. One form of commodity
monies is currency in the form of coins made of a precious metal, with an
exchange value that is at least roughly equal to the value of the metal in
the coin. These coins were usually minted with the monarch’s authority
and were made legal tender, so that the seller or creditor could not refuse
to accept them in payment.
Legal tender was in certain circumstances supplemented as a means of
payment by the promissory notes of trustworthy persons or institutions and
in the 18th and 19th centuries by ‘bills of exchange’3 in Britain. While

105
these bills could be traded by the sellers to other buyers or banks, they did
not became a generally accepted medium of payment. The emergence of
private commercial banks in the 19th century in Britain led to note issues
by them and eventually also led to orders of withdrawal (i.e., checks)
drawn upon these banks by those holding demand deposits with them.
However, while the keeping of demand deposits with banks had become
common among firms and richer individuals by the beginning of the 20th
century, the popularity of such deposits among the ordinary persons came
about only in the 20th century. With this popularity, demand deposits
became a component of the medium of exchange in the economy, with
their amount eventually becoming larger than that of currency.
Until the mid-20th century, demand/checkable deposits did not pay
interest but savings deposits in commercial banks did do so, though subject
to legal or customary ceilings on their interest rates. During the 1950s,
changes in banking practices caused these savings deposits to increasingly
become closer substitutes for demand deposits so that the major dispute of
the 1950s on the definition of money was whether savings deposits should
or should not be included in the definition of money. However, by the
early 1960s, most economists had come to measure the supply of money
by M2 — that is, as M1 plus savings deposits in commercial banks —
which does not include any types of deposits in other financial institutions.
M2 is known as the Friedman definition (measure) of money, since Milton
Friedman had been one of its main proponents in the 1950s and 1960s.
Financial innovation — i.e., innovation in financial instruments,
institutions, and processes — has been extremely rapid in the last few
decades. It has included innovations in the payments mechanism, which
has included technical changes in the servicing of various kinds of
deposits, such as the introduction of automatic teller machines and Internet
banking (i.e., banking from home through the use of computers, etc.). It
has also included the creation of new assets such as Money Market Mutual
Funds. There has also been the spread first of credit cards, then of
debit/bank cards, followed still more recently by the attempts to circulate
smart cards (i.e., cards which embody a certain amount of money from
which deductions can be made when making payments through them).4
Further, competition among the different types of financial intermediaries
in the provision of liabilities that are close to demand deposits or readily
convertible into the latter, has increased considerably in recent decades.
Many of these innovations have served to further blur the distinction
between demand and savings deposits and also blurred the distinction
between banks and some of the other types of financial intermediaries as
providers of liquid assets. This process of innovation and the evolution of

106
financial institutions into an overlapping pattern in the provision of
financial services is still continuing.

2.7 The Current Definitions of Money and Related

Concepts
We have already referred to several definitions of money. These
definitions are not completely standardized across countries for M 1 and M
2. The approximate generic definitions of M1 and M2 that we will use
through this book are:
M 1 = currency in the hands of the public plus checkable deposits in
commercial banks and other financial institutions and
M 2 = M 1 plus (small or retail) savings deposits in deposit-taking
financial institutions.
Note that M 1, M 2, and other measures of money exclude amounts held by
the commercial banks, the government, foreign banks, and official
institutions, since these amounts are not relevant to financing the public’s
expenditures. There are also many symbols — often differing among
different countries — used for measures of money broader than M 2. Some
of these are illustrated in Box 2.1 from their usage in Canada, the USA and
the UK.
Extended Analysis Box 2.1: Current Meanings of the Symbols for the
Monetary Aggregates in Selected Countries
The monetary aggregates for Canada
M 1 = currency in the hands of the public plus demand deposits in
chartered banks5
M 1+ = M 1 plus personal checkable deposits and non-personal
checkable notice deposits at chartered
banks, trust and mortgage loan companies, and credit unions (including
caisses populaires6)7
M 2 = M 1 plus personal savings deposits and non-personal notice
deposits at chartered banks
M 2+ = M 2 plus deposits at trust and mortgage loan companies and
credit unions (including caisses
populaires)
‘Adjusted M 2+’ = M 2+ plus Canada savings bonds and mutual funds at
financial institutions
M 3 = M2 plus non-personal fixed term deposits at chartered banks and

107
 foreign currency deposits of residents booked in Canada
The monetary aggregates for the USA
M1 = currency in circulation among the public (i.e., excluding the Fed,
the US Treasury, and commercial banks) + demand deposits in
commercial banks 8 (excluding interbank and US government deposits
and those of foreign banks) + other checkable deposits including
negotiable orders of withdrawal (NOW) + credit union (such as savings
and loan associations) share draft accounts + demand deposits at thrift
institutions (such as mutual savings banks) — cash items in the process
of collection and federal reserve float
M 2 = M 1 + savings deposits, including money market deposit accounts
+ small time deposits under $100,000 + balances in retail money market
mutual funds
M 3 = M 2 + time deposits over $100,000 + Eurodollars held by US
residents at foreign branches of US banks and at all banks in the UK and
Canada + money market mutual funds held by institutions
The monetary aggregates for the UK
M 1 = currency plus current account (checking) sterling deposits in retail
banks and building societies, held by ‘UK residents’9
M 2 = currency plus sterling deposits in retail banks 10and building
societies, held by UK residents
M 4 = currency plus sterling deposits at the central bank, other banks
and building societies, held by UK residents

As listed in Box 2.1, the detailed descriptions of M 1 and M2 for Canada,

the UK and the USA are more complex than their standardized generic
definitions. However, these definitions are reasonable proxies. Note also
that the definitions of the monetary aggregates beyond M2, e.g., M 3 and
M4, differ more radically among countries than of M 1 and M2. For M 3
and M4, the only common denominator is that they are broader than M 2
and include, besides M 2, other highly liquid assets held at financial
institutions. The reliance on these specific wider aggregates usually
reflects the peculiarities of the country’s financial structure.
Note that currency holdings C (i.e., currency, which consists of notes and
coins in the hands of the public) and M 1 are becoming increasingly
smaller proportions of M 2 and wider aggregates. We illustrate the relevant
magnitudes and ratios by the following. For Canada in 1995, the currency

108
in the economy was $26.8b, M 1 was $62.7b while M2+ was $618.4b, so
that C/M 1 was 42% and C/M2 was only 4%. The ratio of M 1 to M2+ was
only 10%. In the USA, at the end of 1995, the amount of currency in the
economy was $379b,11 M 1 was $1,150b, M2 was $3,680b, and M3 was
$4,954b. The ratio C/M 1 was 33% and C/M 2 was 10%. The ratio of M 1
to M 2 was only 31% and to M 3 was 23%. For the UK in 1995, currency
holdings were £20.8b, M2 was £439.4b, and M4 was £682.5b.12
Fact Sheet 2.1: Monetary Aggregates of the USA
This Fact Sheet illustrates changes in the relative importance of currency
(held by the banks and the public), demand deposits in banks, M 1, M 2,
and M 3 by examining movements in these monetary aggregates for the
USA. Since the 1970s, with banks providing ever closer substitutes for
checking deposits and the increasing use of debit and credit cards,
broader monetary measures such as M2 and M3 have become ever
larger proportions of the money supply.

2.8 The Monetary Base and Bank Reserves

The money supply equals the monetary base — sometimes also called the
reserve base 13 — multiplied by the monetary base multiplier , which is
defined as the ratio of M/M 0, where the symbol M 0 (pronounced as ‘M
zero’)14 stands for the monetary base. It generic definition is:

M 0 = currency in the hands of the non-bank public

+ currency held by the commercial banks + deposits held by the
commercial banks at the central bank.

The reserves held by the commercial banks are called bank reserves and
are designated by the symbol R. Bank reserves for the banking system as a
whole are defined as:

109
 R = currency held by the commercial banks + deposits held by the
commercial banks at the central bank.
Therefore, we have:

M 0 = C + R.

Note the difference between the monetary base M 0 and the money supply
M 1. M 1 = C + D , where D is demand deposits of the public in
commercial banks, while M 0 = C + R .

2.8.1 The relationship between the monetary base and the

money supply
The central bank controls the monetary base through its purchases and
sales of bonds to the public. A purchase of bonds means that the central
bank pays for them by transferring funds from itself to the public, which
increases the monetary base. A sale of bonds means that the central bank
receives funds from the public, which decreases the monetary base. These
sales or purchases of bonds in the financial markets are called open market
operations. They are a major element in the central bank’s control of the
monetary base and the money supply.
The money supply is related to the monetary base by the equation:

where:
M 1 = narrow money supply,
M 0 = monetary base, and
α1 = monetary base multiplier (= ∂M 1/∂M 0) for M1.
The coefficient a1 represents the increase in the money supply M1 for a
unit change in the monetary base M 0. Its value is usually greater than one.
In fact, a value of three or four is not unusual for it for the modern
economies. This implies that a change in the monetary base engineered by
the central bank increases the money supply by the multiple α1 , so that it
is called the monetary base (to the money supply) multiplier.15 The
broader the definition of the monetary aggregate, the larger will be the
relevant multiplier. For M 2, we have:

M 2 = α2M 0,

110
where α2 = ∂M2/∂M 0 and is the monetary base multiplier for M 2. Since
M 2 is always larger (because it also includes savings deposits) than M 1,
α2 > < α1.
The use of monetary base and the money supply as instruments of
monetary policy and their relationship to interest rates will be studied in
Chapter 5.

2.9 The Quantity Equation

Any exchange of goods in the market between a buyer and a seller
involves an expenditure that can be specified in two different ways.
1. Expenditures by a buyer must always equal the amount of money
handed over to the sellers, and expenditures by the members of a group
which includes both buyers and sellers must always equal the amount of
money used by the group multiplied by the number of times it has been
used over and over again.16Designating the average number of times
money turns over in financing transactions during a given period as its
velocity of circulation V , expenditures as $F, and the money stock in
use as $M , we have $y ≡ $MV , where ≡ indicates an identity rather than
merely an equilibrium condition.
2. Expenditures on the goods bought can also be measured as the quantity
y of the commodities (goods and services) traded times the average
price of commodities.17 Expenditures Y then always equal the quantity y
of the commodities bought times their price level P , so that, $Y ≡ $Py.
Obviously, these two different ways of measuring expenditures must
yield the identical amount. These two measures are

Y ≡ MV

Y ≡ Py .

Hence

Equation (2) is an identity since it is derived solely from identities. It is

valid under any set of circumstances whatever since it can be reduced to
the statement: in a given period by a given group of people, expenditures
equal expenditures, with only a difference in the computational method
between them. Equation (2) is true for any person or group of persons.18 If

111
it is applied, as it usually is, to the aggregate level for a country, the two
sides of the identity and its four variables refer to all expenditures in the
country. But if it is applied to the world economy as a whole, its total
expenditures and the four variables will be for the world economy.

2.9.1 The quantity equation in growth rates

The quantity equation specified above was MV = Py. This can be restated
in growth rates (See Chapter 1, Box 1.3. Also consult Useful Mathematical
Formulae for Economics page xliii of this book.) as:

M″ + V″ ≡ P″ + y″

where ″ indicates the rate of change (also called the growth rate) of the
variable. This identity can be restated as:

π≡M″+V″—y″

where π is the rate of inflation and is the same as P ″. This identity asserts
that the rate of inflation is always equal to the rate of money growth plus
the growth rate of velocity less the growth rate of output. Ceteris paribus ,
the higher the money growth rate, the higher will be the inflation rate,
while the higher is the output growth rate, the lower will be the inflation
rate. Note that velocity also changes over time and can contribute to
inflation if it increases, or reduce inflation when it falls. The spread of
banks and automatic teller/banking machines (ATMs) has tended to
increase velocity in recent decades.

2.9.2 The implications of the quantity equation for a

persistently high inflation rate
In normal circumstances in the economy, velocity changes during a year
but not by more than a few percentage points. Similarly, real output
growth rate is usually only a few percentage points. For the quantity
equation, we need only to consider the difference (V ″ — y ″) between
them. In the normal case, both velocity and output increase over time and
the difference in their growth rates is likely to be in low single digits.
Adding this information to the quantity equation implies that high (high
single digits or higher numbers) and persistent (i.e., for several years) rates
of inflation can only stem from high and persistent money growth rates.
This is particularly true of hyperinflations in which the annual inflation

112
rate may be in double (10% or more) or triple (100% or more) digits or
even higher.
To reiterate, the source of high and persistent inflation is high and
persistent money growth rates. Therefore, if the monetary authorities wish
to drastically reduce inflation rates to low levels, they must pursue a policy
that achieves an appropriate reduction in the money supply growth.

Fact Sheet 2.2: Money Growth and Inflation in the USA, 1960-2008

The quantity theory asserts a close positive relationship from the

quantity of money and the rate of inflation. The following graph is a plot
of these for the USA during 1960-2008. If the quantity theory held
perfectly during every period, we would expect to see a perfect positive
trend along the 45-degree line between annual money supply growth and
inflation. In reality, as we see that the relationship between the two is
only slightly positive. However, because the effect of money supply on
inflation is usually lagged by some quarters, we might expect that its
effect on inflation will not be noticeable in the same period, but is likely
to be more visible over longer periods of time (and in hyperinflation). In
the interim, money growth impacts on output and velocity, which do not
remain constant.

2.10 The Quantity Theory

The quantity theory was first formulated in the 18 th century. It is the
proposition that: ceteris paribus, a change in the money supply in the
economy causes a proportionate change in the price level. This
proposition only applies in the long run, with output at its long-run (full-
employment) level. It does not apply if the economy is away from its full-
employment level, as usually happens when the economy is in a recession
with output below its full-employment level. It also does not apply while
the economy is undergoing adjustments following a change in the money

113
supply.
The quantity theory was dominant as a theory of the determination of the
price level through the 19th and early 20th century, though more so as an
approach than a rigorous theory. Its statement varied considerably among
writers. One version of the form that it had achieved by the beginning of
the 20th century is presented below from the works of Irving Fisher, an
American economist in the early 20th century.
Extended Analysis Box 2.2: The Difference between the Quantity Theory
and the Quantity Equation
The quantity theory is vitally different in spirit and purpose from the
quantity equation. The quantity theory is not an identity , as is the
quantity equation. The quantity equation is not a theory , as is the
quantity theory.
A relationship or statement that always holds under any circumstances
is said to be an identity or tautology . Identities generally arise by the
way the terms in the relationship are defined or measured. Thus, the
quantity equation defined (measured) expenditures in two different
ways, once as MV and then as Py , so that it is an identity. Note that an
identity is different from — and a much stronger statement than — an
equilibrium condition which holds only if there is equilibrium but not
otherwise — i.e., it does not hold if there is disequilibrium. Further, a
theory may or may not apply to any particular economy in the real world
or it may be valid for some states — e.g., equilibrium ones — but not
for others, while an identity is true (or false) by virtue of the definitions
of its variables and its logic so that its truth or falsity cannot be checked
by reference to real-world, real-time observations. A theory usually
includes some identities but must also include behavioral conditions —
which are statements about the behavior of the economy or its agents —
and often also equilibrium conditions on its markets. The quantity theory
does so,19 but not the quantity equation. Further, note that the quantity
theory includes (long-run) equilibrium conditions. Since the quantity
equation is an identity, it does not need to do so and does not incorporate
any equilibrium conditions.

2.10.1 The transactions approach to the quantity theory

This approach is associated with the writings of Irving Fisher in the first
quarter of the 19th century. He started with the quantity equation. He
recognized it as an identity and added two assumptions:

114
(i) Except during transition periods , the economy‘s output is independent
of the changes in the money supply.
(ii) Except during transition periods , the velocity of the circulation of
money is independent of the money supply changes.
On assumption (i), the dominant theory of the 19th and early 20th century
on output and employment in the economy implied that the labor market
would be in equilibrium at full employment and that output would tend to
stay at the full-employment level except in temporary disequilibria. This
full-employment output is the long-run output, explained in Chapter 1. It is
independent of the money supply and prices. Similarly, in the long-run
equilibrium, the velocity V is also independent of changes in the money
supply. We discuss this point further in a subsequent subsection.
Fisher’s model can be stated as:

where * and the superscript f indicate long-run values which do not change
when M and P change. Hence, in the long run, MV * = Py f. Therefore,
noting that the rate of growth of the price level (P″) is the same as the
inflation rate π,

n* = M″ + V*″ - y f″,

where π* is the long-run inflation rate. Since the changes in M ” and n do

not cause V* and yf to change, 3V*/dM = 0 (so that V*” = 0) and 9yf/dM
= 0 (so that yf// = 0). Hence, for long-run equilibrium, changes in the
money supply imply that:

π* = M ”.

That is, if we examine the effect of changes in the money supply in long-
run equilibrium only, the inflation rate will equal the money growth rate.
This is the quantity theory proposition.
The adjustment period relevant to the quantity theory
Note the qualification ‘except during transition periods’ to the quantity
theory proposition. Following a change in the money supply by the central
bank, the economy usually takes time to absorb it and to return to long-run
equilibrium. During this adjustment, both output and velocity tend to
change. The inflation rate is usually less than the money growth rate

115
during the adjustment phase.
The adjustment period is often longer than a year, so that the real-world
economies rarely have an inflation rate equal to the money-growth rate
during the short term. This is even less likely if the real-world economy
had started with a recession, so that it was already out of its long-run
equilibrium, and increases in the money supply increase output. That is,
for real-time economies, the assumption of continuous long-run
equilibrium with full employment and long-run velocity is doubtful, so
that the quantity theory does not apply over short periods.20 However,
most macroeconomic theories accept that the quantity theory holds under
the assumptions of long-run equilibrium. This result is established in
Chapter 7.
Is velocity constant over time?
Velocity (of the circulation of money in financing purchases) is the
average number of times money changes hands in the economy over a
given period. This can be formally derived from MV ≡ Y which implies
that V ≡ Y/M .
Velocity does not stay constant over time in real-world economies. It
increases as real incomes rise because the income elasticity of real money
balances is less than unity — usually about 0.7.21 Since velocity is the
ratio of real income to real money balances, velocity increases as income
increases. Therefore, as income changes over the business cycle and over
the long run of growth theory, velocity changes. There are also other
reasons, such as innovations in the payments technology — an example of
which is the spread of automatic teller machines — for changes in velocity
both over the short and the long term. Hence, from a realistic perspective,
velocity is continuously changing in the economy. Estimates of the annual
fluctuation in velocity for the USA are about 3% to 4%.
Fisher’s version of the quantity theory recognized, as does economic
theory generally, the real-time variations of velocity. Hence, if you see a
(fairly common) statement in some book that ‘the quantity theory assumes
velocity to be constant’, you must take this statement to mean not the
constancy of velocity in real time or in disequilibrium, because that would
be patently invalid for any real-world economy in real time, but merely
that velocity is independent of the money supply and the price level in
long-run equilibrium — which is an analytical concept rather than a real-
time one — with full employment.
Statistics on velocity are hard to collect directly. Therefore, the common
procedure for deriving an index of velocity is to use the quantity equation.
Under this procedure, V ≡ Py/M . We can use this identity and the easily

116
available data on P , y , and M to calculate an index for velocity. It would
normally show changes in velocity during the year as well as from one
year to the next.
Fact Sheet 2.3 : Velocity of Money in the USA, 1960-2008

This Fact Sheet illustrates movements in the velocity of M 1, M2, and M

3 from data for the USA. None of them is constant over the short term or
the long term. Financial innovations like the interest-bearing checking
accounts increased the demand for M 1, which decreased M 1 velocity
(which equals Y/M 1) in the 1980s. Interest rates on non-monetary assets
(i.e., on savings deposits and bonds) fell after 1994, which raised the
demand for M 1 and further decreased velocity. Since savings deposits
pay interest and are a component of M 2, M 2 velocity proved to be
relatively more stable than that of M 1 . Nevertheless, M 2 velocity did
not prove to be constant.

2.11 The Definitions of Monetary and Fiscal Policies

Monetary policies consist of changes in the money supply or in interest
rates (induced by the central bank of the country) to bring about changes
in the economy. The change in the money supply is usually brought about
by the central bank through a change in the monetary base or in interest
rates.
Fiscal policies are the use of government expenditures or taxes and the
resulting fiscal deficits (or surpluses) to bring about changes in the
economy . While government deficits can be financed through increases in
the money supply (and surpluses be accompanied by decreases in it),
macroeconomics defines fiscal policy as one in which the money supply is
held constant , so that the deficits must be financed by government
borrowing through increases in its bonds sold to the public. The reason for

117
this definition of fiscal policy is to keep distinct the effects of monetary
and fiscal actions on the economy.
In the real world, fiscal and monetary policies are intertwined, more so
in some countries than others. However, macroeconomics treats them as
conceptually independent policies and performs their analyses under the
assumption that fiscal deficits are bond-financed.

2.12 The Central Bank and Monetary Policy

The control of monetary policy is usually allocated to the central bank of
the country. This means that the central bank determines the changes it
wishes to bring about in the money supply and in the interest rates in the
economy. As we have shown earlier, the central bank usually induces
changes in the money supply through changes in the monetary base,
brought about through transactions (sales or purchases) in the bond
market. It induces changes in the market interest rates through its changes
in the monetary base and also by changing its own interest rate, known as
the discount or bank rate.
If the control over monetary policy is shared between the central bank
and the government — as is currently done in Britain — the term
‘monetary authority’ seems more appropriate.
The central bank of Canada is the Bank of Canada and that of the USA is
the Federal Reserve System, often shortened to the term ‘the Fed’. In both
Canada and the USA, the central bank is effectively independent of the
government and the legislative bodies (the Parliament in Canada and the
Congress in the USA) and has sole jurisdiction over the formulation of
monetary policy. However, there is normally extensive ongoing
consultation between these central banks and their governments on the
formulation of monetary policy.

2.13 The Economic Aspects of the Government

and Fiscal Policy
Fiscal policy is pursued through changes in either government spending
and/or its tax revenues, in such away as to create a surplus or deficit in the
government budget.
Government spending , defined as the sum of all payments made by the
government,22 can be classified into two broad categories:
(i) Expenditures on (currently produced) commodities (i.e., goods and

118
 services), with payments by the government in exchange for the goods
and services purchased by it.
Examples of expenditures on goods and services are those on the civil
service, the courts, and the military, as well as those on schools,
hospitals, etc.
(ii) Transfers to the private sector, without a corresponding flow of
commodities from the private sector to the government. This category
includes:
a. payments to households under social programs,
b. subsidies to businesses, and
c. payment of interest on the public debt — which is the debt owed by
the government to the public. This debt is accumulated through past
borrowing by the government from the public in the form of bonds
sold to the public (not the central bank).
Note the differences in terminology: ‘expenditures ’ for the purchases of
currently produced goods and services, ‘spending ’ for total payments, and
‘transfers ’ for transactions in which the government does not receive
currently produced goods and services in exchange. This usage will be
maintained throughout this book. In 1999, total federal government
spending in Canada was about $158 billion while the tax revenues were
about $162 billion, with an estimated surplus of $4 billion. Expenditures
on commodities were only about one-third of the total government
spending while transfers were more than two-thirds of total government
spending.
Tax revenues are collected from a variety of taxes. Among the most
obvious taxes are income and corporate taxes, sales taxes and tariffs.
The government’s spending and tax revenues represent the two sides of
its budget balance sheet. If the two sides are equal, the budget is said to be
‘balanced’. Otherwise, there is budget/fiscal23 surplus or deficit.
A fiscal surplus occurs if:
Fiscal surplus[= tax revenues (including transfers) — government
spending (including transfers)] > 0.
If we define net tax revenues as net of transfers from the government to the
private sector, we have:

Net tax revenues = total tax revenues — transfers to the private sector.

Therefore, we can redefine a fiscal surplus and deficit as:

fiscal surplus [= net tax revenues — government expenditures on goods

119
 and services] > 0 and

fiscal deficit [= government expenditures on goods and services — net tax

revenues] > 0.

Therefore, a fiscal deficit occurs if the government buys goods and

services worth more than its net (of transfers) tax revenues. These
definitions of the fiscal surplus and deficit are the ones most directly
related to the economic analysis of the commodity sector and we will
adopt this usage.

2.13.1 The financing of fiscal deficits/surpluses and

changes in the money supply
A fiscal deficit represents government spending in excess of its income.
Just as with households, the government has to pay for its deficit. There
are two ways in which the government can pay for it. They are:
(i) Bond-financing of the deficit (without a change in the money supply)
Under this method, the government raises funds through the sale of
newly issued bonds to the public, receives money in exchange and then
uses these funds to pay for the excess spending. The funds received by
the government in this process are returned to the public through its
extra spending (i.e., equal to the deficit) on commodities.
(ii) Money-financing of the deficit (without a change in the public’s bond
holdings)
Under this method, the government issues new bonds, which it sells to
the central bank for newly created currency (notes and coins), which is
then put into circulation in the economy by the government’s
expenditures on commodities. This creation of currency increases the
monetary base and, as shown earlier, increases the money supply by a
multiple.
A fiscal surplus represents government income in excess of its spending.
Since the government usually has a lot of outstanding bonds, it can use its
excess income to buy back some of the outstanding bonds and cancel
them. There are two ways of doing this. They are:
(i) Bond financing of the surplus, without a change in the money supply
A fiscal surplus implies that the government collects more in taxes than
it spends, so that it has surplus funds on its hands. In order not to change
the public’s money supply, it has to return these funds to the public. To
do so, it buys bonds from the public. Hence, a ‘bond-financed surplus’

120
 involves the purchase of an amount of bonds (equal to the fiscal surplus)
by the government24 from the public. The funds received by the
government through its budget surplus are returned to the public in
payment for the bonds. Therefore, the monetary base and the money
supply do not change in this process — while the public’s holdings of
bonds decrease by the amount of the surplus.
(ii) Money-financing of the surplus, without a change in the public’s bond
holdings
A ‘money-financed surplus’ means a decrease in the monetary base
equal to the amount of the fiscal surplus. To achieve this, the funds
received by the government in taxes from the public are used by it to
buy bonds from the central bank — which ‘retires/cancels’ the currency
it gets back, so that it no longer circulates in the economy. In this
process, the funds are not returned to the public, so that the monetary
base decreases by the amount of the surplus, thereby decreasing the
money supply.
Therefore, money-financed deficits increase the money supply while bond-
financed ones do not do so. Money- financed surpluses decrease the
money supply while bond-financed ones do not. Fiscal policy will be
defined as running a bond-financed fiscal deficit or surplus with the
intention of changing aggregate demand in the economy.25 It does not
change the money supply.

The implications of the independence of the central bank from the

government for financing deficits

In Canada and the USA, since the central bank is independent of the
government, the government cannot rely on the use of money creation,
which is in the control of the central bank, for financing its deficits.
Therefore, the usual method of financing deficits is bond financing, which
increases the government debt but does not change the money supply.
Similarly, fiscal surpluses also are ‘bond-financed’, which decreases the
government debt but does not change the money supply.
Therefore, when the central bank is truly (i.e., factually and not merely
legally) independent of the government, monetary policy changes the
money supply, while fiscal policy does not. If fiscal deficits in a real-world
economy are money-financed, macroeconomic theory considers them to be
a mix of both fiscal and monetary policies.

2.13.2 The public debt

121
The public debt is the debt owed by the government to others (excluding
the central bank). It consists of the bonds sold by the government in the
past and not yet redeemed but held by the public (not the central bank).
The part of this debt held by residents of foreign countries is called the
external debt. The rest is held by domestic residents; this part of the public
debt is called the internal debt.
A bond-financed deficit means that the value of the new bonds issued by
the government equals the nominal value of the deficit. Therefore, the
public debt increases by the nominal value of the fiscal deficit.
Conversely, the public debt falls by the nominal value of the fiscal surplus.
Most countries have large public debts. In 1999, in Canada, the
accumulated debt of the federal government was estimated to be about
$576 billion. The debt/GDP ratio was about 71% in 1995, dropped to
about 60% in 1999, and continued dropping thereafter. The reasons of this
drop were the rise in GDP and the decreases in the deficits, with surpluses
emerging in the years after 1998.
Interest has to be paid on the public debt. Payment of this interest is
treated as a transfer in the government payments to the public.

2.13.3 The selective nature of government expenditures,

taxes, and subsidies
Most aspects of government activity impact on some sectors of the
economy more than on others. To illustrate, government expenditures for
education support schools and universities, those for medicare support
health services, subsidies for R&D expenditures encourage innovation, etc.
Similarly, progressive tax rates are intended to reduce after-tax income
inequality and taxes on cigarettes are intended to discourage smoking.
Therefore, the government’s policies on government expenditures,
transfers, and taxes constitute more than merely a way of managing
aggregate demand in the economy. However, the macroeconomic study of
fiscal policy focuses only on its impact on aggregate demand.

2.14 Interest Rates in the Economy

The interest rate is the rate of return on a loan (bond). This rate of return
has two elements: any fixed (also known as the ‘coupon’) payment on the
bond — or the dividend in the case of a share in a corporation — each
period and any capital gains and losses due to changes in the market value
of the bond. Since some of these elements are not known at the time the

122
bond is bought, we distinguish between the expected rate of return, and the
actual rate of return — i.e., the rate that will eventually materialize.
The rates of interest on different bonds differ due to differences in the
risk, the time to maturity of the bond, and other differences in their
characteristics. In general, a riskier bond will yield a higher return, as will
a bond with a longer maturity. However, for all bonds in the economy,
macroeconomics envisages a common underlying element of the rate of
return to which premiums are added for riskiness, the time to maturity and
any other relevant characteristics.
Macroeconomics focuses on the common underlying component of the
variety of interest rates in the economy. It, therefore, deals with a single
rate of interest in the economy and studies its determination. In particular,
macroeconomics studies the short-run and the long-run effects of monetary
and fiscal policies, and of the rate of inflation, on this underlying interest
rate.
One of the proxies used for the macroeconomic interest rate is the
Treasury bill rate, also called the T-bill rate. Treasury bills are government
bonds, usually with a 90-day term to maturity. They are riskless and have a
very short maturity. Another proxy is a medium-term interest rate, which
is usually interpreted as the rate of return on 3-5-year government bonds.
A third proxy would be the long-term interest rate, which is usually
interpreted as the rate of return on bonds with more than ten years to
maturity.

2.14.1 The Fisher equation on interest rates

Assuming that the market for bonds — the ‘capital market’ — is perfectly
competitive, Irving Fisher, in the early 20th century, derived what has
come to be known as the Fisher equation (for the market/nominal rates of
interest). The Fisher equation separates the market rate of interest into two
components: the real rate of return and a premium or allowance for the
expected inflation rate. It assumes that (a) lenders have the option of
investing in riskless durable commodities, including capital goods, whose
prices rise by the inflation rate and which do not have storage or
transactions costs, so that the investor could risklessly receive the real
return rr by investing in such commodities, (b) the capital markets are
perfectly competitive and efficient, and (c) both borrowers and lenders
hold identical expectations.
Most bonds in the economy are nominal bonds (i.e., paying a nominal
interest rate), so that the interest rates observed in the economy are mainly
market (or nominal) rates. However, there are sometimes also real bonds

123
(i.e., paying a real interest rate).
The (proper form of the) Fisher equation relating real and nominal
interest rates is:

(1 + R) = (1 + r)(1 + πe),

where:
R = market (nominal) rate of interest per period from the loan,
r = real rate of interest per period from a real loan, and
πe = expected rate of inflation over the period of the loan.
Note that πe can differ from the actual inflation rate π, which is unknown
at the beginning of the period, i.e., when the investment is undertaken. The
explanation for the Fisher equation is as follows. An investor investing one
dollar in a ‘nominal bond’ (i.e., paying a nominal rate of interest R) would
receive $(1 + R) at the end of the period. If he/she was to buy a ‘real bond’
(i.e., paying a real interest rate r), he/she would receive (1 + r) in real
terms (commodities) at the end of the period. Given the expectations on
inflation held at the beginning of the current period, the expected nominal
value at the end of the period of this real amount equals $(1 + r)(1 + π e).
The investor will be indifferent between the nominal and the real bonds if
the nominal return from both bonds was equal, i.e., (1 + R) = (1 + r)(1 +
πe).
Expanding the preceding equation gives,

(1 + R ) = 1 + r + π e + rπ e

so that:

R = r + π e + rπ e.

For small values of r and πe, rπe would be approximately zero, so that the
commonly used (‘standard’) form of the Fisher equation is:

Although this form of the Fisher equation is an approximation, we will use

this form as the Fisher equation in further analysis. The intuitive
justification for this equation is that the lenders want to be compensated
for the expected loss of purchasing power due to the expected inflation,
while the borrowers — mainly firms who benefit from the expected

124
increases in the prices of their products — would be willing to so
compensate. The (simplified/standard) Fisher equation implies that:

re = R . πe.

In this form, re is the real rate of interest that lenders expect to end up with
at the end of the period: since they expect the inflation rate πe for the
period of the loan, the real return the lender expects to get from a nominal
loan will be only (R – ne).
If inflation occurs during the period of a loan, the loan repaid at the end
of the period is worth less by the rate of inflation than its purchasing power
at the time the loan was made. We can then derive the actual — also called
the ex-post — real rate of return from the nominal one by subtracting the
actual rate of inflation from the latter. That is, the actual real rate of return
is specified by:

where ra is the actual real rate of interest. The actual/realized real rate of
return on loans (ra) differs from the expected one (re) if actual inflation
differs from the expected one, which does happen frequently.26 Note that
if the rate of inflation is greater than the market interest rate, the actual (ex-
post) real return would be negative.
Fact Sheet 2.4 : Nominal and Real Interest Rates in the USA, 1982-2008
The Fisher equation implies that the real interest rate is the difference
between the nominal interest rate and the expected inflation rate. In
simple applications, the expected inflation rate is approximated by the
actual one. The following graph illustrates the relationship over time
between the nominal interest rate on Treasury bills and the actual
inflation rate using USA data. Their difference yields the actual real
interest rate on Treasury bills. Since the real interest rate fluctuates much
less than the inflation rate, very high nominal rates are usually due to
very high inflation rates, as during the late 1980s. Since 2000, nominal
interest rates have been quite low because the inflation rates have been
low.
The real interest rate fluctuates due to changes in the real sectors of the
economy and the monetary policy pursued by the central bank under
interest rate targeting. In the USA, since 2000, the central bank has
maintained the real interest rate at a fairly low level.
Note that while the nominal interest rate normally remains positive,

125
 the actual real interest rate can become negative if the expected inflation
rate is below the actual one, as happened around 2003 and 2008.

2.14.2 The concept of present discounted value (PDV) of a

bond
A bond in ordinary English-language usage is a financial instrument that
entitles the holder a fixed coupon payment at specified intervals and a
payment at the specified maturity date. The general procedure for finding
the present discounted value (PDV) of the amounts to be received (or to be
paid) in the future requires discounting every future amount to its present
value by dividing by the relevant discount factor and summing over the
discounted values.27 The nominal amount received at the end of the first
period; is divided by (1 + R 1), where R 1 is the interest rate in period 1. If
the amount received is at the end of the second period, it is first divided by
(1 + R 2) to find its value at the beginning of period 2 (which is also the
end of period 1) and then divided by (1 + R 1) to find its value at the
beginning of period 1. That is, the PDV of any amount received at the end
of period 2 is divided by (1 + R 1)(1 + R 2). Similarly, PDV of any amount
received at the end of period 3 is divided by (1 + R 1)(1 + R 2)(1 + R 3) and
so on. Therefore, the PDVb of a bond with a fixed coupon payment c and
redemption value an after (n — 1) periods28 is:

where:
= price of the bond in period 1,
coupon payment each period on the bond, assumed to be constant
c=
over time,

126
 R = market interest rate
the expected price (i.e., amount received by the bond holder on sale
an = or at maturity) after n periods (= ), and
n = sale or redemption date.
The above formula assumes that c, a , and n are known, and that these are
paid at the end of the relevant period. If there is no risk in holding the
bond, investors would be willing to pay PDVb to buy the bond, so that the
bond market will establish this value as the price of the bond. Since
government bonds are normally riskless, the above formula can be applied
to government bonds. Since bonds issued by private firms are risky, the
discount rate used has to be the one that takes account of this risk.
If we were to assume — as is usually done and as we did earlier — that
the interest rates are going to remain constant at R, the discount factors
become 1/(1 + R) for the first period, 1/(1 + R)2 for the second period, 1/(1
+ R)3 for the third period, and so on. In this case, the PDV formula
simplifies to:

The present discounted value of a stream of payments on the bond in the

future is the sum of the present discounted value of each of these
payments. In perfect capital markets, this is the amount that the investors
in the bond will be willing to pay for it. Hence, in perfect capital markets,
the price of a bond will be the present discounted value of the future
receipts from it. That is:

PDVb = pb.

Given the price of the bond, the rate of return on the bond can therefore be
calculated from the formula:

The above formula can be used in two different ways:

1. It can be used to calculate the rate of return R on the bond for a given
bond price . The rate thus calculated is called the internal rate of
return. The preceding formula shows that an increase in the bond’s
current price lowers its rate of return.
2. The second way is to derive the market price p b of the bond for a given
market interest rate R and given the expected resale value. An increase
in the market rate lowers the bond price. Hence, there is an inverse

127
 relationship between bond prices and interest rates.
Therefore,
(a) If the aggregate demand for bonds increases in the economy, it would
raise bond prices (of the existing bonds), and lower the economy’s
(macroeconomic) interest rate.
(b) If the central bank causes the market interest rate to rise, bond prices
(of the existing bonds) will fall. Conversely, if the central bank causes
the market interest rate to fall, bond prices will rise.
In the limiting case of a perpetual bond — also called a ‘consol ’ —
defined as a bond with a constant coupon payment per period and without
a redemption date , the above formula simplifies to:

which clearly shows an inverse relationship between bond prices and the
nominal interest rate.

2.14.3 Bubbles in asset prices

A bubble in the price of an asset is said to exist if this price differs from
the ‘fundamental value’ of the asset, as determined by the present
discounted value of the asset given its current and future returns and the
market interest rate. The ‘bubble’ is the difference between the market
price and the fundamental nominal value, as explained in Box 2.1. While
the concept of a bubble is analytically clear, it is often difficult to
determine this fundamental value for assets whose future returns are
unknown and have to be based on very vague and incomplete, even
possibly erroneous, information. This is usually the case in the valuations
of corporations. Therefore, there are often disputes on whether or not there
is a bubble in stock prices and the extent of the bubble.
A positive (negative) bubble is said to exist if the actual market price
becomes greater (less) than the fundamental value of the asset. A bubble is
said to burst when the stock price adjusts, as in a stock market crash, to its
fundamental value. The bursting of a positive bubble is often followed by
a negative bubble, i.e., the downward correction in the price gets overdone.
The importance of asset bubbles for output and business cycles
Bubbles in the prices of widely held assets, such as stocks, land, and
houses, can cause severe fluctuation in output and employment. Positive
bubbles in them cause household wealth to increase significantly and lead
to an increase in consumption expenditures. Positive bubbles in the stock

128
prices of corporations make it easier for them to obtain funds for
investment, which leads to increases in investment expenditures. These
increases in consumption and investment expenditures cause booms in
economic activity. The reverse process is set in motion when the bubbles
burst: the rapid fall in the prices of widely held assets sends the economy
into a crisis and a recession, as happened when house prices in the USA in
2007 started collapsing after a positive bubble developed during 2000–
2006. Chapter 16 on business cycle explains this process in greater detail.
Box 2.1: The Determination of Stock Prices
The present discounted value formula also applies to the prices of
equities/shares. Assume perfect capital markets, a share with an
expected dividend at the end of period t, and a constant market interest
rate R. The present discounted value formula implies that the price of
the share at the beginning of period 1 would be:

where the superscript S stands for the share and is uie expected
(designated by the superscript e) sale price of the share S at the end of
period n. As interest rates rise, future receipts are discounted at a higher
rate, resulting in a lower present value, so that there is an inverse
relationship between the current share price and the market interest rate.
There is a positive one between the share price and its expected
dividends, as well as between the current share price and its expected
future one (at the expected resale date n).
If there comes into being an expectation of higher (lower) future
dividends or resale prices, this shift in expectations will raise (lower) the
current price of the share. The average expectation in the financial
markets is called ‘market opinion’ (i.e., the average degree of market
optimism or pessimism on future share prices), so that a major
determinant of current share prices is ‘market opinion’. Since this
opinion is often more volatile than the profits of firms or sectors, their
share prices also tend to be volatile. Such volatility of stock prices can
be seen in the fluctuations on a daily basis of the average values of
stocks on stock exchanges, such as the New York Stock Exchange, the
Toronto Stock Exchange, and the London Stock Exchange.
The volatility of the prices of the Internet and communications stocks
at the end of the 1990s and the early years of this century is an example
of the shifts in the market opinion about future prices causing volatility
in current stock prices. The almost worldwide decline in stock markets

129
 in mid-2006 was at least partly due to the increases in interest rates
induced by the central banks of many countries, especially the US
Federal Reserve System, which raised their discount rates in an attempt
to combat rising inflation.
To emphasize this conclusion, our analysis implies that, besides
changes in stock prices due to changes in dividends or interest rates,
stock prices rise (fall) just because they are expected to rise (fall). The
ways this volatility impacts on aggregate demand, output and prices in
the economy is laid out in Chapter 9.
Bubbles in house and land prices
Just as stock prices can have bubbles, so can house and land prices.
These are assets whose future resale value is not known, so that
speculation about the future resale value can drive up (a positive bubble)
or down (a negative bubble) the current prices.
An example of large bubbles in house and land prices occurred just
prior to the East Asian crisis of the mid-1990s and the financial crisis in
the USA before 2007. The bursting of the bubbles in these cases
decimated the wealth of the public and sent the economies into severe
recessions.
A bubble in tulip bulb prices!
Bubbles can occur in the price of any asset. To illustrate, a massive
bubble in the price of tulip bulbs occurred in Holland in the 18th century
soon after it was introduced to Europe from Turkey. The bubble was
such that the prices of the bulbs of some desired varieties of tulips
became greater than the prices of many middle class houses.

Conclusions
• Money performs the two main functions of the medium of exchange and
the store of value, with the former being absolutely critical to the
transactions role of money in the economy. These functions are
performed by a variety of assets, with their liquidity characteristics and
substitutability among them changing over time. Innovations in the types
of assets and the changing characteristics of existing financial assets
mean that the financial assets which meet the role of money keep
changing over time.
• While currency was considered to be the only form of money at one time
(prior to the 20th century), currency and demand deposits were taken to
be the two components of money early in the 20th century, so that the
appropriate measure of money was considered to be M 1. By 1960, the

130
 measure of money had been expanded to include savings (including time)
deposits in commercial banks, and therefore had become M 2. In
subsequent decades, as the liabilities of near-banks became more and
more similar to the demand and savings deposits of banks, the measures
of money were broadened to include the deposits in near-bank financial
intermediaries.
• The recent incursion of electronics into banking in the form of automatic
tellers, banking from home through one’s computer or telephone, and the
use of smart cards for payments, etc. represent a very fast pace of
technical change in the banking industry. It is a safe bet that the
empirically appropriate measure of money is changing and will keep
changing in the future. Because of these changes, disputes about the
proper measure of money have expanded beyond the monetary
aggregates M 1 and M 2 to encompass broader and more complex forms.
• Given that the growth rates of output and velocity tend to be in low
single digits, the quantity equation implies that high, sustained rates of
inflation are due to high, sustained rates of increase in the money supply.
This is particularly true of hyperinflations.
• The quantity equation differs from the quantity theory. The quantity
theory states that, in long-run equilibrium analysis with full employment,
an increase in the money supply will produce a proportionate increase in
the price level.
• Monetary policy is controlled by the central bank. It consists of changes
in the monetary base or in the interest rates set by the central bank.
• The central bank controls the money supply through changes in the
monetary base. The money supply changes by a multiple of the change in
the monetary base.
• The government has its own budget. If its total spending exceeds its
revenues, it has a deficit, which has to be financed. An independent
central bank forces the government to finance any deficits through the
issue of new bonds rather than through the issue of newly created notes
and coins. A bond-financed fiscal deficit increases the public debt and a
surplus decreases it.
• Fiscal policy is controlled by the government. It consists of changes in
government spending or revenues. In particular, it is exercised through
changes in the amount of the fiscal surplus or deficit. Macroeconomics
defines and studies fiscal policy under the assumption that any deficits
are financed by the issue of new bonds while any surpluses are used to
redeem a corresponding amount of the existing government bonds.
• The Fisher equation on interest rates shows that the macroeconomic
market rate of interest rises with the expected rate of inflation. The

131
 expected and the actual real rates of return on bonds will differ if the
expected inflation rate is different from the actual one.
• Bond prices are inversely related to the market interest rates.

KEY CONCEPTS

The functions of money

Money supply versus money stock
M 1, M 2, and broader definitions of money
An identity versus a theory
The quantity equation
The quantity theory
Fiscal deficits and surpluses
Public debt
The nominal rate of interest
The real rate of interest, and
The Fisher equation .

SUMMARY OF CRITICAL CONCLUSIONS

• The appropriate definition of money keeps changing. There are currently

several definitions of money in common usage. These include M 1, M2,
and broader monetary aggregates.
• All definitions of money include currency in the hands of the public and
demand/checking deposits in commercial banks.
• Banks are one type of financial intermediaries but are different from
others in that their liabilities in the form of checking and savings deposits
are the most liquid of all assets in the economy.
• The quantity equation is an identity while the quantity theory is a long-
run equilibrium condition. The latter asserts that, in long-run equilibrium,
increases in the money supply will produce a proportionate increase in
the price level. This will not be so in the transition from an initial
equilibrium to the subsequent one. The quantity theory does not hold in
the short term since the commodities and the velocity of circulation of
money usually respond (at least during the short-term disequilibrium
phase) to changes the money supply or to changes in its growth rate.
• The velocity of circulation of money is not constant over time, but
changes from one month to the next one, and from one year to the next

132
 one.
• There are numerous interest rates in the economy. The differences among
them are mainly due to differences in the riskiness and the term to
maturity of the loans on which they are paid.
• Monetary policy is the use of changes in the monetary base or in interest
rates to influence the economy.
• Fiscal policy is the use of changes in government spending, taxes, and
the resulting fiscal deficits or surpluses to influence the economy.
• The public debt is the total amount of government bonds outstanding in
the economy. Bond-financed fiscal deficits increase the public debt and
fiscal surpluses decrease it.
• The present discounted value formula explains how the price of a
financial asset depends on its stream of expected returns (the coupon
payment on the bond or the expected dividends on a share) and its
expected resale price at the time the investor expects to sell it.

REVIEW AND DISCUSSION QUESTIONS

1. What does ‘the medium of payments’ mean? Are credit cards a medium
of payment? Are debit cards a medium of payment? Are telephone cards
a medium of payment? Discuss.
2. Define (a) currency, (b) demand deposits, and (c) savings deposits.
What are their relative proportions that you tend to hold on average over
a typical month? Explain their likely variation near Xmas?
3. What are the proportions of your holdings of currency, M 1 and M 2
relative to your own expenditures over a typical month and year? Does
this proportion fluctuate? Using your own experience as indicative of
the representative consumer’s behavior, what can you conclude about
the constancy of the velocity of circulation of money over time?
4. What is the definition of ‘bonds’ in macroeconomics and how does it
differ from that of ‘money’? Compare the definition of ‘bonds’ in
macroeconomics with those of bonds and stocks in ordinary usage.
5. Define (a) treasury bills, (b) money market mutual funds, (c) coupon
payment on a bond, and (d) a perpetual bond (consol).
6. Define the monetary base and specify its general relationship with the
money supply (M 1 and M 2) in the economy.
7. Specify the quantity theory.
8. Specify the quantity equation. What does the quantity equation imply
for the causes of persistently high rates of inflation in the real world?
Explain your answer.

133
9. Can the actual real rate of return on loans become negative? Illustrate
with an example.
10. What is the definition of monetary policy?
11. Specify the relationship between government spending and
government expenditures (on goods and services).
12. Define fiscal deficits and surpluses in two ways (i.e., in terms of (a)
government spending and (b) government expenditures on
commodities).
13. Define the public debt. How do fiscal deficits and surpluses change its
amount? What else changes the debt/GDP ratio?
14. Specify the present discounted value formula for the price of a bond.
Explain the relationship it implies between the bond price and the
market interest rate.

ADVANCED AND TECHNICAL QUESTIONS

T1. For a given value of M 0, if M 2 was five times the size of M 1, what
would be the relationship between their monetary base multipliers?
T2. [Optional] What were the magnitudes of M 1 and a broader money
measure for the last year (for which the data is available) for your
economy? What were the relevant monetary base multipliers? How have
these multipliers and the ratio of M 1 to M 2 changed in the past two
decades? [Use the Internet for your sources and cite your sources.]
T3. [Optional] What were the magnitudes of government spending,
government transfers revenues, and deficits/surpluses for the last year
(for which the data is available) for your economy? What were the
government expenditures on goods and services and government
transfers? What were their proportions relative to each other and to
government spending? How have these proportions changed in the past
two decades? [Use the Internet for your sources and cite your sources.]
T4. What is meant by a theory versus an identity? Define these two terms
and specify their differences. Is the Quantity Theory a theory or an
identity? Explain your answer.
T5. Why is the interest rate paid by banks on savings deposits so much
lower than those paid by them on money market mutual funds and those
charged by banks on loans to the public?
T6. Stock prices rise if investors expect them to rise and fall if investors
expect them to fall. Why? Why does this make stock markets volatile?

134
1 or M 0 (with 0 in the subscript) in these figures is different from the
symbol M 0 used later to refer to the ‘monetary base’ — which is different
from the money supply but related to it.
2Note that in the equilibrium equation setting money demand equal to
money supply, the dimensions — which can be either real or nominal —
of both the money supply and demand must be identical.
3Bills of exchange were promissory notes promising to pay a certain
amount of money at a specified future date. They were issued by buyers of
goods to the sellers in exchange for goods, and represented a method of
extending trade credit by the suppliers of goods to the firms buying the
goods.
4Examples of these include telephone cards, bus cards, etc.
5The chartered banks in Canada correspond to the commercial banks in
our discussions.
6These are essentially credit unions in Quebec, Canada.
7TMLis the abbreviation for trust and mortgage loan companies and
CUCP is the abbreviation for credit unions and caisses populaires .
8Our usage of the term ‘commercial banks’ refers to ‘depository
institutions’ in the USA.
9‘UK residents’ is meant to exclude the public sector and the financial
institutions.
1010Since 1993, these deposits include both non-interest-bearing and
interest-bearing deposits.
11b stands for billion defined as 1,000 million.
12These figures are taken from Statistics on Payment Systems in the Group
of Ten Countries , published by the Bank for International Settlements,
various years.
13The monetary base is also known as high-powered money .
14The meaning of this symbol differs from that of M 0, which is meant to
indicate a particular value of the money supply M and is pronounced as M
sub zero.
15The value of this multiplier depends upon the nature of the banking
system, the public’s demand for currency, the banks’ demand for reserves
and the monetary aggregate used.
16Thus, a person buying $100 of goods pays $100 to seller 1. Suppose the
latter in turn buys $100 worth of goods from another seller (seller 2) of
goods. The total expenditure was thus $200, the amount of money used
was only $100 and it was paid over twice in financing the expenditures.

135
Suppose now that the initial seller had bought only $50 worth of goods
from Seller 2. Total expenditures would now be $150, the amount of
money in use remains at $100 but it has been paid over only 1.5 times on
average.
17Since the goods traded are generally of different kinds, there are
obviously problems in thinking of an aggregate measure of goods in
physical terms and of the price level to be associated with a unit of such a
conglomerate or composite good. Both the ‘quantity’ or ‘output’ y of this
good and its average price P must then be thought of as indices.
18Identities are said to be true or false. By comparison, propositions or
relationships about the real world are said to be valid or invalid.
19This is done through its statement that in the long run, both output and
velocity are independent of the money supply and prices. This statement
will hold in some economies but need not do so in all economies.
20This deviation is studied in the short-run and disequilibrium
macroeconomic models covered in Chapters 7 and 8.
21That is, if real income rises by 100%, the demand for real money
balances rises by about 70%.
22Our focus is on the current account of the government. Therefore, the
receipts (or payments) of funds in transactions involving the purchases (or
sales) of bonds by the government are excluded from this definition of
government spending/payments.
23To distinguish between this deficit and that in the balance of payments,
the government’s budget deficit is called the fiscal deficit .
24This operation is usually conducted through its banker, which is the
central bank. The purchased bonds are almost always ones that the
government had issued in the past and sold to the public. The government
retires/destroys the bonds it buys.
25Fiscal deficits often occur because the government cannot collect
enough revenue to finance its desired expenditures. Such deficits also
change aggregate demand, but this is an inadvertent consequence of a
failure to balance the budget. Our focus in the analysis of the government
budget as a tool of fiscal policy is on those deficits and surpluses that are
created so as to manage aggregate demand in the economy.
26In special circumstances, the actual real interest rate can turn out to be
negative. However, the expected real rate on loans would always be
positive if lenders have the alternative of investing — with the same
transactions costs, including time and convenience — in physical capital
(e.g., land and houses), whose nominal return will at least equal the

136
expected inflation rate. If the expected real return on loans were negative
while that on physical capital was positive, no one would want to make
loans.
27For the constant interest rate R , the discount factor for funds received
after one period is 1/(1 + R ). For funds received after n periods, the
discount factor is 1/(1 + R )n .
28This formula assumes that the future interest rates are known.
Otherwise, the future interest rates would be replaced by their expected
values, which would introduce risk into the equation.

137
 CHAPTER 3
Introduction to the Open Economy:
Exchange Rates and the Balance of
Payments

This chapter provides an introduction to the concepts and

determinants of the foreign exchange rates and the balance of
payments. It lays the basis for the open economy macroeconomics
and the determination of the exchange rates in later chapters.
This chapter also presents the theories important for open
economy macroeconomics. These are the purchasing power parity
(PPP) and interest rate parity (IRP) theories.

This chapter introduces the economic concepts related to the openness of

the economy. Its focus is on the understanding of the exchange rate and the
components of the balance of payments.
Every economy nowadays has considerable commodity and financial
flows with other economies. These flows—mainly of commodities,
currencies, and financial capital, though sometimes also of people—
between economies affect their domestic national income, employment,
prices, and other endogenous variables, as well as the scope and
effectiveness of national monetary and fiscal policies.
The flows of factors of production other than physical capital1 are minor
and are generally ignored in macroeconomic analysis. Further, the short-
run analysis of an open economy assumes that the physical capital stock is
fixed so that any flows of such capital that occur are out of currently
produced output and do not affect the productive capacity of any economy.
In fact, there is a special use of the word ‘capital’ in the context of
international trade. This is in its financial usage. Therefore, the
international flows of ‘capital’ are to be understood as being only financial
flows, that is, a flow of the currency and other financial assets of one
country to another country in exchange for financial assets. These flows
have become quite considerable in the last few decades with the global

138
integration of financial markets, and on a daily basis, are very much larger
than the values of commodity flows for many countries.
The open economy has one more good than the closed economy. This
good is ‘foreign exchange ’, which is the designation for the medium of
payments between the domestic and foreign economies. This good consists
of foreign currencies, gold, and Special Drawing Rights (SDRs) at the
International Monetary Fund (IMF). SDRs act as a kind of demand
deposits of individual countries with the IMF. The price of the domestic
currency against a foreign one is the exchange rate between the currencies.
Since macroeconomic theory usually treats the rest of the world as a single
unit, our analysis will be set out in terms of the composite category labeled
‘foreign exchange’—meaning by it the foreign currencies of the rest of the
world—and ‘the exchange rate’ between this composite and the domestic
currency.
Flows of commodities and financial instruments occur between the
domestic economy and the rest of the world. The flows of commodities
among countries take the form of exports and imports of commodities,
with the difference between them designated as net exports (of
commodities), which is also called the balance of trade. The flows of
financial instruments—in the form of bonds and other ownership claims—
are financial capital flows and are captured in the balance of payments on
capital account. The flows of physical capital are counted in commodity
imports or exports, rather than in international capital flows. They do not
directly appear in the balance of payments on capital account.

3.1 Exchange Rates

3.1.1 Three concepts of exchange rates

Almost all countries have their own national currencies. The rate of
exchange between a domestic currency and a foreign one is called the
exchange rate between them. The exchange rates between currencies can
be nominal, real or effective ones.
The (nominal) exchange rate
The (nominal) exchange rate is the rate at which a currency can be
exchanged against another currency. This exchange rate between any two
given currencies can be defined in two alternate ways. It can be defined as:
1. The price of the domestic currency ($) in terms of a foreign currency

139
 (£s). An example of this occurs if we cite the exchange rate for the
British £ in Canadian dollars as £0.4 per C$. [News reports in Canada
often use this definition of the exchange rate for the Canadian dollar in
terms of the US dollar, as in ‘today, in foreign exchange markets, the
Canadian dollar traded at 85 US cents’.]
2. The price (per unit) of a foreign currency in terms of the domestic
currency. An example of this occurs if we cite the exchange rate for the
British £ in Canadian dollars as C$2.40 per £. This measure of the
exchange rate is clearly the reciprocal of the former measure, so that the
exchange rate as Canadian dollars per pound would be specified by £1 =
C$1/0.4 = C$2.5. [News reports in Canada often use this definition of
the exchange rate for the Canadian dollar in terms of the euro and the
yen].
The definition of exchange rate that would be appropriate to use
obviously depends upon convenience and preferences. There is no
consensus on which definition is more useful. Both have certain
advantages and disadvantages relative to the other. We prefer using the
definition above. That is, the (nominal) exchange rate will be defined as
the amount (i.e., number of units) of a foreign currency required to
purchase one unit of the domestic currency. Taking the dollar as the unit of
the domestic currency and the £ as the generic symbol for a unit of the
foreign currency, the nominal exchange rate under our definition will be
the amount of £s per $ in the foreign exchange markets — compactly
expressed as £/$. We will use the Greek letter ρ (pronounced ‘rho’) for the
nominal exchange rate. The dimension of ρ for our definition is £/$.
The real exchange rate
The real exchange rate, designated as ρr, is the amount of foreign
commodities that can be exchanged for one unit of domestic
commodities.2 Its relationship with the nominal exchange rate is specified
by:

where:
ρ = nominal exchange rate (£/$) (amount of foreign currencies per unit of
the domestic one),
ρr = real exchange rate (amount of foreign commodities per unit of the
domestic one),
P = domestic price level, and
P F = foreign price level.

140
Therefore, the real exchange rate is the nominal exchange rate adjusted for
the relative price ratio between the countries. As mentioned above, the real
exchange rate specifies the amount of foreign commodities required to
purchase a unit of the domestic commodity.
The best way of intuitively grasping the concept of the real exchange
rate is to focus on a single commodity, say X. Suppose we had one unit of
commodity X and were to sell this unit at home, convert the dollars thus
obtained into the British currency (£) and then attempt to buy units of the
same commodity in Britain. Let the amount of the commodity obtained in
Britain be z units. Then, assuming that there are no transport or other
transactions costs, one unit of the commodity in the domestic market
allows us to purchase z units of the same commodity in Britain. Therefore,
the real exchange rate is z (units of the foreign commodity per unit of the
domestic one). If the commodity is (a) a standardized one, (b) without
transport and other transactions costs, and (c) traded on competitive world
markets, we would expect z to equal unity but it is unlikely to do so if any
of these conditions are not met.
At the level of the economy as a whole with a very diverse set of
commodities, different economies have different mixes of commodities, so
that each economy’s mix is treated as if it was a ‘composite commodity’.
The proxy for the price of this composite commodity is the price level.
Hence, the relative price ratio (P/PF) is used to adjust the nominal
exchange rate ρ to derive ρr .3
Since the standardised composite commodity on a world-wide basis does
not exist, many economists intuitively estimate the real exchange rate on
their travels by using the price of a standard commodity such as a
Macdonald’s hamburger as they move from one country to another. To
illustrate, a price of $1.20 in Canada and of £1 per hamburger in the UK,
with a nominal exchange rate of £0.50 per $, gives a real exchange rate of
0.60. The calculation is as follows:

ρr = pP/PF = 0.5(1.20/1) = 0.6.

Intuitively, in this example, it takes only 60% of a British hamburger to

buy a Canadian one: the Canadian hamburgers are relatively cheaper (by
40%).
The effective exchange rate
There is obviously a different exchange rate for each foreign currency,
even though economic analysis is formulated in terms of a single exchange
rate for simplification purposes. Such a rate can be envisaged as an

141
average exchange rate for all currencies, just as the ‘price level’ is an
average of the prices of commodities in the economy. This average
exchange rate between the domestic currency and all foreign currencies is
called the effective exchange rate. As an average, the effective exchange
rate is obtained by weighting the exchange rate against each foreign
currency by the proportion of the domestic economy’s total trade (exports
and imports) conducted with the foreign country. Changes in this effective
exchange rate provide a measure of the average change in the value of a
currency in terms of all other currencies.
There can obviously be both a nominal and a real effective exchange
rate. The effective rate is preferable for converting GDP, exports, imports
etc. from one currency to another one for international comparisons.4

3.2 Fixed, Flexible, and Managed Exchanged

Rates
The market for foreign exchange may be allowed to determine the
exchange rate. Alternatively, the exchange rate may be set by the central
bank (or the government). The former is known as the flexible/floating
exchange rate case and the latter as the fixed/pegged5 exchange rate case.
In general, there are three exchange rate regimes: (i) fixed, (ii) flexible,
and (iii) managed floating. The managed floating exchange rate regime
covers the case where the central bank (and/or the government) buys or
sells foreign currencies (from its reserves) in the foreign exchange market
to limit the range of fluctuations in its foreign exchange rate. Most flexible
exchange rate regimes tend to be managed ones to a greater or lesser
extent.
Depreciation and appreciation of the domestic currency
If the exchange rate (£/$) decreases, say from £0.50 per dollar to £0.40 per
dollar, the domestic currency decreases in value against foreign currencies
and there is a depreciation (under floating exchange rates) or devaluation
(under fixed exchange rates) of the domestic currency against other
currencies. If the exchange rate depreciates, it costs more in the domestic
currency to buy foreign goods and assets, while domestic goods and assets
become cheaper in foreign currencies. Conversely, if the exchange rate
increases, the domestic currency increases in value against foreign
currencies and there is an appreciation (under floating exchange rates) or
revaluation (under fixed exchange rates) of the domestic currency against

142
other currencies. In this case, foreign goods and assets will become
cheaper in terms of the domestic currency, while the domestic goods and
assets will become more expensive in foreign currencies.

3.3 Purchasing Power Parity (PPP) as a Theory of

the Exchange Rate

3.3.1 PPP at the level of a single commodity

PPP for a single, standardized and internationally traded commodity is that
it must cost the same in different countries, once its prices in the different
countries are converted into the same currency by the nominal exchange
rate. The assumptions required for this condition are that:
• The commodity is identical in different countries.
• There is perfect competition.
• There are no ‘transactions costs’ (such as from transport, insurance,
tariffs, quotas, etc.) to its movement or information asymmetries
among the countries.
Under these conditions, the internationally traded commodity will cost the
same (after conversion into any one currency) in different countries. If its
price was higher in one country, there would be an inflow of imports from
other countries in sufficient amounts to reduce the domestic price to the
foreign one. If the price was lower, domestic firms will export enough of it
to other countries to raise the domestic price to the foreign one.6

Absolute PPP among countries

In the limiting (but unrealistic) case where all the goods and services
between the domestic and foreign economies are of the above type, we
have absolute PPP between the domestic economy and the world one. At
this aggregate level, PPP is the condition that:

PPP is often called the ‘law of one price’ (for commodities in the
international context). All of the commodities in the economy do not meet
the assumptions for PPP. Among the reasons for this are that many goods
— for example, buildings and land, most kinds of services, etc. — are not

143
internationally tradable. Further, some of the inputs of even internationally
tradable goods, such as the labor services in retailing or local
transportation or the usage of immovables such as buildings and land, etc.,
have to be local ones. Therefore, even for the internationally traded goods,
these transportation and local retailing costs drive a wedge between the
domestic prices of goods and the prices in other countries. Hence, the
conditions for PPP are not met for all or possibly most final goods (i.e., at
the retail level) in the economy, so that the above absolute version of PPP
does not apply in practice.
Extended Analysis Box 3.1: Does PPP Apply in the Real World? An
Illustration
On the question of whether PPP applies in the real world, we turn to the
prices of hamburgers across countries as an intuitively appealing test.
The Economist annually publishes the prices of McDonalds’ Big Mac
among countries.7 Its reports invariably show very significant deviations
from PPP. The data reported is on the price in the local currency, the
exchange rate between the local currency and the US dollar and the
implied dollar price. The article on 27 April 2000, showed the
following:

The more detailed calculations for this table are given in the Appendix
of this chapter. This evidence shows that, relative to the PPP implied
values for the Big Mac as of 27 April 2000, the Canadian currency was
undervalued by 23% and the Mexican one was undervalued by 11%,
relative to the US dollar, even though these countries are neighbors of
the USA. The British pound was overvalued by 20% relative to the
USA. As a corollary, the British pound was overvalued relative to the
Canadian dollar by about 50% (=1.23/0.80 -1). There was also
considerable under- or overvaluation even among the national currencies
(in 2000) of the EU.
Over several years, The Economist’s time series of Big Mac prices do
not usually show a strong tendency toward PPP; in fact, they sometimes
even show movements contrary to it. This can be seen by comparing the
Big Mac index over several years. The issue of The Economist on 18

144
 July 2009, showed the following for the countries in the preceding table.

Comparing the extent of undervaluation and overvaluation in the

preceding data for 2000 and 2009, the extent of undervaluation of the
currency decreased for Australia and Canada, whose nominal exchange
rates against the US dollar shot up in July and August 2009; and the
extent of overvaluation of the currency decreased for Britain, whose
nominal exchange rate against the US dollar fell in 2009. But, for
Mexico, the extent of undervaluation increased from 12% to 33%.
Therefore, PPP does not hold as a reasonably realistic assumption over
fairly long periods, and its assumption in macroeconomic theory is
unjustified.
Although the Big Mac is a fairly standardized product, its local cost of
production only partly depends upon internationally traded inputs, such
as meat and flour, which should adhere to PPP. Its cost of production
also depends on inputs such as local labor, land, and buildings, which
are not internationally traded and for which there is unlikely to be PPP.
The degree of local competition for hamburgers could also differ, so that
McDonalds may vary its profit margin among countries. These factors
cause differences in relative prices among countries.
At the economy’s level, most final products are similarly combinations
of internationally traded and non-traded goods, and there are different
degrees of competition for each country’s goods. Therefore, PPP rarely
applies to individual goods or to commodities as a whole.

3.4 Relative PPP and Shifts in the Relative

Efficiency of Economies
If we assume that the real exchange rate has a particular value k, we have:

For absolute PPP, k = 1, so that commodities would cost the same at home
and abroad. However, travellers from a country with k less than unity find
that that the same commodities cost more abroad than at home — and vice

145
versa. Appendix Box 3.1A derives the value of k for selected countries for
27 April 2000. k is less than 1 for Canada, Mexico, and Australia, so that
the residents of these countries find a Big Mac more expensive in the USA
than in their own countries. However, k exceeds 1 for Britain vis-à-vis the
USA, so that Americans find a Big Mac more expensive in Britain than at
home. One of the reasons for the divergence of k from unity for Big Macs
is that some of the inputs — such as local labor and land — into their
production and retailing are not internationally traded and the Big Mac is
itself not internationally traded — i.e., cannot be exported or imported.
Another reason arises from the impact of capital flows on exchange rates.
If a country’s k exceeds unity, its exchange rate is said to be overvalued
(relative to PPP). If k is less than unity, its exchange rate is said to be
undervalued. The reasons why k can differ from unity, and, therefore, from
absolute PPP, are:
• The price indices for P and PF (such as the CPI) are based on the prices
of a bundle of commodities, which include some internationally traded
goods and others which are not internationally traded, e.g., services,
buildings, etc.11 The distinction between these types of commodities is
relevant to the determination of the country’s nominal exchange rate ρ ,
which is affected by its relative ability to produce and export
internationally traded commodities more efficiently than other
countries. The nominal exchange rate is not affected by the efficiency
and costs of goods and services that are not internationally traded, even
though they are included in the measurement of domestic and foreign
price indices, which affect the derivation of the real exchange rate.
• The nominal exchange rate also reflects the ability to attract net inflows
of capital. As explained later in this chapter and in Chapter 12, these
flows are likely to be the dominant short-term determinant of exchange
rates for countries with extensive capital flows. Countries that are
attractive havens for capital attract large net capital inflows from
abroad. These inflows push up the demand for their currencies, which
increase their exchange rates, with the result that their value of k
becomes greater than unity. The reasons for this attractiveness can be
higher interest rates, better business opportunities, greater security of
capital, etc. Countries that are relatively more attractive for capital
inflows tend to have k greater than unity.
• Under fixed exchange rate regimes, another cause of the overvaluation
(undervaluation) of the real exchange rate is if the government or the
central bank sets the value of the country’s exchange rate higher
(lower) than PF/P .

146
 Empirical observation suggests that less developed countries often have
values of k less than one relative to the developed ones. Even among the
developed economies, countries that are attractive havens for capital have
values of k higher than one relative to other developed economies.

3.4.1 Long-run changes in relative PPP

Rewrite the preceding equation of the relative PPP as:

In the long run, k will not be a constant since transactions costs, the
relative competitiveness of the domestic economy in commodity
production, the relative sizes of its internationally tradable and non-
tradable sectors, and its relative attractiveness for capital flows can change
over time. With k potentially varying in the long run, the rate of change in
ρ would be given by:

Note that ″ stands for the rate of change in the accompanying variable and
π is the rate of inflation.
This is the long-run version of relative PPP. It asserts that the rate of
change in the nominal exchange rate reflects the difference in the inflation
rates and the rate of change in the real exchange rate. An increase in the
value of k for our currency — i.e., k > 0 — appreciates it. A decrease in
the value of k for our currency — i.e., k″ < 0 — depreciates it.

3.4.2 Short-run changes in relative PPP

For the short-run analysis, the factors that determine k are assumed to
remain unchanged so that k is treated as a constant (i.e., its rate of change k
″ equals zero), so that in the short run:

This version of PPP is known as the short-run version of relative PPP.

Economists usually prefer using the relative version (either the long run
or the short run one) of the PPP theory rather than its absolute version in
explaining changes over time in exchange rates. However, PPP indices are
often constructed and used for calculating the PPP-based GDP series for
cross-country comparisons of output and output per capita. An example of
these is given in Box 3.1.

147
3.4.3 Implications of short-run PPP for exchange rates and
inflation rates
The short-run version of relative PPP is that p″ = πF - πF. This short-run
version of relative PPP implies the following two theories, depending on
whether the country is following a fixed or flexible exchange rate regime.
(i) Fixed exchange rate regime case: In this case, ρ″ = 0, so that n = πe.
That is, in the short run, the domestic inflation rate for a small open
economy is determined by the foreign one. Therefore, the domestic
central bank through its monetary policies cannot achieve an inflation
rate different from the world one. Hence, under a fixed exchange rate
regime, PPP becomes a theory for the determination of the domestic
inflation rate for a small open economy.
(ii) Flexible exchange rate regime case: In this case, p″ can differ from
zero. This assumes that the central bank through its monetary policies can
bring about a domestic inflation rate different from the world one, so that
the rate of appreciation/depreciation of the exchange rate will be given
by: p″ = πF - π. If the domestic inflation rate is maintained lower (higher)
than the world one, the exchange rate will rise (fall) correspondingly.
Therefore, under a flexible exchange rate regime, PPP becomes a theory
for determining the rate of change of the exchange rate for a small open
economy.
Box 3.1: International Comparisons of Standards of Living in Terms of
PPP
The nominal GDP (or GNP) measure discussed in Chapter 1 was in
domestic prices. The real GDP measure discussed in Chapter 1 was its
nominal value divided by an index of the domestic price level (e.g., the
CPI or the GDP deflator). The standard cross-country comparisons of
GDP are made by converting its nominal value into a common currency,
often the US dollar, at the nominal exchange rate between the domestic
currencies and the US dollar. For example, Canadian GDP is converted
to US dollars as:

Canadian GDP in US$ = (Canadian GDP in C$)(US$ per C$)

or as,

This mode of conversion is quite useful for some purposes — e.g.

148
comparing the relative sizes of economies in nominal terms for
international trade purposes. However, the GDP per capita derived from
it is not an appropriate measure of the relative standards of living among
countries when the nominal exchange rate is not consistent with PPP.
Appropriate cross-country comparisons of real GDP and real GDP per
capita require that each given commodity be given the same value
among the different countries. For example, a cup of coffee, ceteris
paribus, in one country should be properly counted as being equal to a
cup of coffee in other countries, even though its domestic purchase price
converted into the US dollars at the nominal exchange rate may be quite
different from that in the USA. For this purpose, first, the exchange rates
based on PPP have to be constructed.12 These provide the values of the
‘PPP exchange rate’. Second, the PPP exchange rates are used to convert
GDP is the national currency into the PPP GDP. This is done for
selected countries in the following table.
Using PPP real exchange rates for cross-country conversions of GDP
raises the GDP figures for countries whose real exchange rates (derived
from the nominal ones) are less than unity and lowers those for countries
whose real exchange rates (derived from the nominal ones) are greater
than unity. The following table13provides for a few selected countries
the comparisons of GNP and GNP per capita based on the nominal
exchange rate and those on PPP.

The general pattern in this sample of countries is for the PPP

calculations to be different from the (‘standard’) calculations based on
the nominal exchange rates. Some of the differences are very large. In
particular, the less developed economies have much higher PPP values
than the standard ones. For India and Kenya, the PPP-based per capita
values are about five times those based on the market exchange rate. For
Mexico, they are twice as much. This reflects a commonly observed
pattern of the relative — i.e., relative to PPP — under-valuation of the
currencies of the developing economies. But, for the UK, the PPP
figures are lower, reflecting its relatively over-valued exchange rate.

149
 This over-valuation pattern also holds for France, Germany, and most of
the other West European countries.

3.5 Interest Rate Parity (IRP) and the

Determination of the Exchange Rate

3.5.1 The benchmark IRP theory

Corresponding to the PPP for commodity flows, the interest rate parity IRP
condition is derived with reference to capital flows. It is based on the
assumption that investors seek the highest return on their investments.
Therefore, if the risk of investing in different countries were identical, for
capital to be shared among countries, the rate of return must be the same
among countries — otherwise, the country with a higher rate of return will
get all the capital flowing into it. This argument leads to the IRP condition.
Its assumptions are: (a) perfect capital markets (including the absence of
controls on capital flows), (b) zero transactions costs, and (c) risk
indifference or zero risk. Under these assumptions, an investor will invest
his capital in the country that offers him the highest return.
The nominal rate of return from investing in domestic bonds is their
yield, designated by the interest rate R . Designating the nominal yield on
foreign bonds as R F, the net nominal return (in the domestic currency) on
foreign investments is R F less any expected appreciation p″e of the
domestic currency — or alternatively, plus any expected depreciation of
the exchange rate.
For this comparison, consider an investor who wants to invest $100 for
one year. If he/she invests it in domestic one-year bonds yielding 5% per
year, he/she will get back $105 at the end of the year. His/her alternative is
to invest in British bonds at a current exchange rate of £0.50 per dollar, so
that he/she will be investing £50. Let British bonds pay an interest rate of
6%, so that he/she will get back £53 after one year. Further, assume that
he/she expects the exchange rate to become £0.51 at the end of the year, so
that the £53 will give him/her $103.92 (= £53/0.51) in dollars. He/She
therefore benefits by investing in domestic bonds (which will pay $105) —
even though the domestic interest rate is only 5% against the 6% interest
rate in Britain. The reason the British bonds provided a lower expected
return in this calculation is that the exchange rate for the domestic
currency was expected to appreciate (from £0.50 per dollar to £0.51 per

150
dollar) and the effect of this appreciation had to be deducted from the
British interest rate.
In general terms, investment of $1 at home will yield $(1 + R) .
Investment of $1 abroad will be worth £p in the foreign currency at the
time of investment and therefore, would yield £(ρ + ρ R F) in the foreign
currency. The latter’s expected value in the domestic currency will be $(ρ
+ ρ R F)(1/ρ e). This equals $(1 + R F)(ρ /ρ e). Therefore, the return from
investments at home and abroad is the same if:

Note that the denominator on the right hand side is ρ e and not ρ . Also
note that R in the IRP equation stands for the return including the coupon
payment and any expected capital gains or losses. This is especially
important for capital flows into equities whose values fluctuate a great deal
and have a strong speculative component. Note that for relatively small
changes expected in the exchange rate, the error induced by the
approximation can be ignored for the sake of simplifying the presentation.
It cannot be ignored for large expected changes, as illustrated in Chapter 2.
In the preceding equation, setting [(ρ e−ρ )/ρ e] as ρ ″e and approximating
(ρ /ρ e) by unity for small differences between ρ and ρe, the above equation
reduces to the approximation:14

where p″e is the expected rate of appreciation of the domestic currency and
equals (ρ e − ρ )/ρ e. This simplified version of the IRP equation is called
the benchmark IRP equation. Interpreting intuitively this benchmark IRP
equation, to be used for low interest rates and ‘small’ changes in exchange
rates, investors compare the nominal yield R on domestic bonds with the
expected return on foreign bonds. The latter equals (R F − ρ ″e ). If R (R F
− ρ ″e), investors will get a higher expected return from investing abroad,
so that they will only invest abroad. But if R > (R F − ρ //e), investors
(including foreigners) will get a higher return from investing in our
country and will only invest in it. In equilibrium, under the assumption that
the investments in each country have the same risk level, if the bonds of
both countries are to be sold simultaneously, the return must be the same
from the two forms of investment, so that market equilibrium requires that
R=RF−ρ″.

151
 To reiterate, the benchmark IRP equation is an approximation and holds
only for very small changes in exchange rates. While it is in common
usage and we will continue to use it for further analysis, it is an
approximation to the actual relationship between these variables. However,
the approximation is quite good for very small changes in exchange rates
and for relatively low interest rates.

3.5.2 The benchmark IRP as a determinant of the domestic

interest rate
If the country follows a fixed exchange rate regime or the exchange rate is
expected to remain unchanged, ρ″e = 0, so that the benchmark IRP theory
implies that R = R F. That is, the domestic interest rate will equal the world
interest rate. Countries with a fixed exchange rate, which is expected by
the investors to remain unchanged, would clearly have this determination
of the interest rate. Countries with a flexible exchange rate that is not
expected to change will also experience the same restriction on their
interest rate.
If the investors expect the exchange rate to change, then the IRP
condition implies that the domestic interest rate will diverge from R F by ρ′
e.

3.5.3 The benchmark IRP as a theory of the exchange rate

under flexible exchange rates
If the exchange rate is expected to change during the investment period,
IRP turns into a theory for the determination of the exchange rate. For this
theory, note that ρ″e F = (pe − ρ )/ρ e, so that IRP can be written as:

Hence,

Given IRP, this equation determines the exchange rate for given values of
the domestic and foreign interest rates and the given value of the expected
future exchange rate. It implies that if the central bank raises the domestic
interest rate relative to the foreign one, the country’s current exchange rate
will rise relative to the expected future exchange rate. Further, if the

152
increases in the interest rate and exchange rate induce the financial
markets to expect a further appreciation of the exchange rate (i.e., ρe rises),
the exchange rate will appreciate even more.
A simple illustration of the above arguments is provided by the
following example. Suppose that initially ρ e = ρ = 1. The central bank
now raises the interest rate by 5%. Keeping ρ e unchanged at 1, IRP
implies that the current value of the exchange rate ρ would rise by 5%.
Therefore, the exchange rate is expected to depreciate to the lower future
value, with this value equal to ρ e, which remains at unity.

3.5.4 The role of speculative returns to stocks in capital

flows and IRP
While the IRP theory is usually stated as if the capital flows were in and
out of bonds, international capital flows arising from the purchase and
sales of stocks have become increasingly important in recent decades, as
evidenced by international mutual funds and stock index funds. These
flows are dominated not by the present discounted value of a known
stream of coupon payments on bonds, but by the unknown and highly
speculative future stock prices (see Chapter 2). Consequently, the expected
returns on stock prices tend to be speculative and very different from the
returns on bonds. Further, the movements in bond and stock yields can
follow very different patterns. In particular, bubbles in the stock prices
(see Chapter 2) build up, often to different extents in different countries.
These bubbles, rather than bond yields and movements in interest rates,
can significantly influence capitals flows among countries and limit the
straightforward application of the IRT theory as a determinant of capital
flows and interest rates among countries.

3.5.5 Extending IRP to incorporate risk factors and risk

aversion
The interest rate on an asset incorporates a premium to compensate
investors for its riskiness. The extra risk (‘exchange-rate risk’) in investing
in foreign countries includes the potential for future exchange rate
appreciations and depreciations, which depend upon the possibility of
different inflation rates in different countries and other factors. The latter
in turn depend on the general health of the economy and the ability and
willingness of governments and central banks to control inflation and
promote growth. Further, the possibility of financial and exchange rate

153
crises and that of the imposition of controls on the redemption of capital
increase the risk for foreign investors and increase the required risk
premium.
To compensate for the relevant risk premium α , the benchmark version
of the IRP equation has to be modified to:

R = (R F - ρ //e) + a.

Assuming a lower risk and/or a greater preference for investments at

home than abroad, capital importing countries have to pay a premium to
foreign investors as an inducement to send their capital, so that α would be
positive. That is, capital importing countries have to maintain higher
interest rates at home than capital exporting countries because of the risk
premium required by foreign investors to send capital to them. Conversely,
capital-exporting countries would have a negative value of α and relatively
lower interest rates at home than capital-importing countries. Hence, under
the preceding assumptions,

For capital importing countries: α > 0, so that R > R F.

For capital exporting countries: α < 0, so that R < R F.

In the general case, where countries are neither consistently capital

importing nor exporting ones relative to each other, the sign and value of α
will depend on many factors, which include the degree of risk aversion of
investors, investor’s perception of the degree of (country) risk involved in
investing in at home versus abroad, etc. The latter depend on the likely
future economic prospects of the economies. Therefore, as these factors
shift, the sign and value of α can shift over time.
Fact Sheet 3.1: Interest Rate Differentials Between Countries
This Fact Sheet shows the interest rate differential (plotted as the USA
Treasury bill rate minus the Canadian one) between USA and Canada,
which have closely integrated financial markets, so that there should be
perfect capital flows between these countries. The Interest Parity Theory
implies that the interest rates should be identical between them.
However, as the following graph shows, there is almost always a
differential between them. The Canada T-bill rate was higher in Canada
from the early 1970s to the mid-1990s; outside this period, the Canadian
rate was sometimes higher and sometimes lower than in the USA.
Therefore, the value of the interest rate premium/discount α is usually
not zero, even for closely integrated economies, and also not constant

154
 over time.

3.5.6 The relative importance of PPP and IRP in

determining exchange rates
International capital flows are nowadays potentially much larger than the
value of commodity flows for developed economies with open capital
markets, so that anticipated changes in interest rates can set up massive
capital flows. These flows — entering on either the demand or supply side
of the foreign exchange market of a country — dominate over other flows
and determine the exchange rate — unless the governments have fixed the
exchange rate or try to manage it through offsetting sales or purchases of
foreign exchange from their reserves.15
Further, since capital is extremely mobile and very large amounts can be
transferred among countries with well-developed financial markets at a
few minutes’ notice, IRP determines exchange rate changes on a
continuous basis in free exchange markets, while, as we have already
argued, PPP holds at best in a long-run context. Hence, in well-developed
financial markets, IRP provides the short-term determination of the
exchange rates while PPP or its relative version at best only provides a
long-term tendency, so that exchange rates determined through IRP can
deviate from PPP for considerable periods.

3.6 The Balance of Payments

155
The balance of payments can be defined or presented in an economic or
accounting form. Economic analysis focuses mainly on its economic form.
The (economic) balance of payments is a statement of the inflows and the
outflows of funds from the domestic country and the difference between
them. If the economic balance of payments were arranged in the form of a
table, it would specify each of the sources of the inflows and outflows of
funds from a country, and the difference between them. This is illustrated
in Table 3.1. The inflows of foreign exchange during a period are the sum
of the payments we receive for commodities and financial assets sold to
foreigners plus interest and dividends on our investments abroad and
unilateral transfers (such as gifts) from foreigners. The outflows of foreign
exchange are the sum of the payments we make for commodities and
financial assets bought from foreigners plus interest and dividends on
foreign investments in our country and unilateral transfers (such as gifts)
to foreigners. Defining the (economic) balance of payments by B, we
have:

where:

B = balance of payments,
Xc = value of exports of commodities (goods and services),

Xk = value of capital exports (against our purchases of stocks and bonds

from foreigners),
Zc = value of imports of commodities (goods and services),

Zk = value of capital imports (against foreigners’ purchases of stocks and

bonds from us),
IR = inflows of interest and dividend payments,
inflows of funds due to unilateral transfers (gifts and donations)
IT =
from abroad,
OR = outflows of interest and dividend payments, and
outflows of funds due to unilateral transfers (gifts and donations) to
OT =
foreigners.
On the right side of the above equation, (Xc + Zk + IR + IT) are the inflows
of foreign exchange, with Xc as the inflows against the exports of

156
commodities and Zk as the inflows against the outflows of bonds
(including stocks and shares and other claims to ownership). (Zc + Xk +
OR + OT) are the outflows of funds. Of these,
Table 3.1 Stylized Balance of Payments Accounts

Zc are the outflows of funds to pay for the imports of commodities and Xk
are the outflows of funds to pay for our purchases of foreign bonds
(including stocks and shares and other claims to ownership).
IR and IT represent inflows of funds without a corresponding reverse
flow of currently produced commodities or bonds. IR captures the
payments of interest and dividends on foreign bonds held by the domestic
residents, with such bond holdings representing past investments rather
than being a current cross-border flow of bonds. Similarly, OT captures the
payments of interest and dividends on domestic bonds held by foreigners.
IT and OT capture cross-border remittances or gifts, which do not involve
explicit future debt obligations.
The preceding equations define the balance of payments in an economic
sense. The balance of payments is thus a statement of the exports and

157
imports of commodities and financial assets. Note that all of the
magnitudes in this equation are in nominal form.
The balance of payments can be decomposed as:

Balance of payments = Balance of payments of current account

+Balance of payments on capital account.

3.6.1 The components of the balance of payments

The (economic) balance of payments has two components:
(i) The balance of payments on current account (Bc):

where (Xc — Zc) specifies the net exports of commodities and is also called
the balance of trade. NR (= IR — OR) is the net inflow of interest and
dividend payments and NT(= IT — OT) is the net inflow of transfer
payments.
The balance of payments on current account has three components: net
exports of goods and services, net interest and dividend income from
foreign investments, and net unilateral transfers. If the sum of the net
interest and dividend payments and net unilateral transfers were zero, the
balance of payments on current account becomes identical with net
exports, which is also known as the balance of trade. The common
assumption on (NR + NT) in the macroeconomic analysis for most
developed economies is that this sum is zero or that it can be taken as
exogenously given in short-run analysis.16
(ii) The balance of payments on capital account (Bk):

This balance specifies the net inflow/imports of funds resulting from

(financial) capital flows. Zk is the inflow of funds/capital resulting from
the purchase by foreigners of financial assets from us while Xk is the
outflow of funds/capital resulting from the purchases by domestic
residents of financial assets from foreigners. Zk is called capital imports
and X k is called capital exports.
Fact Sheet 3.2 : United States Balance of Payments, 1976−2008
This Fact Sheet illustrates movements in the balance of payments and its
components by examining these for the USA. As this sheet shows, none
of these is ever in equilibrium, i.e., equal to zero. The USA has had a

158
 large and increasing trade and current account deficit since abut 1984.
Its capital account balance has tended to be positive since the USA
attracts foreign capital flows because of its political stability and its
economic performance.

3.6.2 Equilibrium in the balance of payments

Clearly, the value of B need not equal zero. It could be positive, negative,
or zero. If B > 0, there is said to be a surplus in the balance of payments,
which is paid to us in foreign currencies so that our reserves of foreign
currencies increase. If B < 0, there is said to be a deficit in the balance of
payments, which we pay for through running down our country’s reserves
of foreign currencies. If B = 0, the balance of payments is said to be in
equilibrium, so that there is no change in our foreign currency reserves. By
comparison, if B > 0 or B < 0, the balance of payments is said to be in
disequilibrium.
We have the following two implications of equilibrium in the balance of
payments:
a. If the balance of payments is in equilibrium, B = 0 so that,

(Xk - Zk) = (Xc - Zc) + (NR + NT). (13)

That is, in the balance of payments equilibrium, net capital inflows

(outflows) equal the balance of payments surplus (deficit) on the current
account. Hence, if our country is exporting more than it imports of
commodities, it must also be a net exporter of capital, so that it increases

159
its investments abroad on a net basis.17
b. Also, for B = 0, we have:

(Zc - Xc) = (Zk - Xk) + (NR + NT),

so that for the balance of payments equilibrium to exist, the net capital
inflows (plus NR and NT) must be enough to cover the cost of net imports.
Hence, countries with net imports of commodities must be importing
capital or pay for them from their foreign exchange holdings.18

3.6.3 The change in foreign exchange reserves

The difference between the inflows and outflows of funds equals the
change in the nation’s foreign exchange reserves. Evaluated in dollars, this
amount is designated as ΔFR. Therefore,

where $FR stands for the foreign exchange reserves evaluated in the
domestic currency (Canadian dollars).
Hence, the net change in our foreign exchange reserves is zero (i.e., ΔFR
= 0) if the balance of payments is in equilibrium (i.e., B = 0). There is a
net inflow of foreign exchange to the home country (ΔFR> 0) if the
balance of payments is in surplus (i.e., B > 0) and a net outflow (ΔFR < 0)
if the balance of payments is in deficit (i.e., B < 0). B is thus a measure of
the change in the country’s foreign exchange reserves.
Fact Sheet 3.3 : US Foreign Exchange Reserves and Balance of Payments,
2005-2008
As explained in the text, balance of payments surpluses and deficits
produce corresponding changes in foreign exchange reserves, so that
movements in the balance of payment and foreign exchange reserves are
closely related. This Fact Sheet illustrates this relationship for the USA.
For the USA, both balance of payments and foreign exchange reserves
experienced a drop in 2005, followed by a steady increase up to the last
quarter of 2008.

160
Equilibrium in the balance of payments
For pedagogical purposes, macroeconomic analysis assumes that the
country in question desires equilibrium in its balance of payments.19 This
corresponds to the assumption that the desired level of the balance of
payments and of the change in foreign exchange reserves is zero. This is
clearly a simplification, though it is commonly used in open economy
macroeconomics. We will also do so. In this case, ΔFR = B = 0, so that:

(Zc - Xc) = (Zk - Xk) + (NR + NT).

That is, the net capital inflows (plus NR and NT ) must exactly cover the
payments for the net imports of commodities, with the result that the
change in foreign exchange reserves is zero.
Foreign exchange reserves and short-term bonds
Central banks often invest part of their inflows of foreign exchange in the
short-term bonds, especially Treasury bills, of foreign governments, so that
they can earn some income. These bonds are highly liquid and earn
interest, while holdings of foreign currencies do not.
The country’s official holdings of foreign exchange consist of its
reserves of foreign currencies (including gold and SDRs) plus its holdings
of short-term foreign bonds held by the government and the central bank.

3.7 The Balance of Payments in an Accounting

Sense
Surpluses and deficits in the balance of payments have to be settled in
foreign currencies. To capture the changes in the country’s foreign
exchange reserves, we define the official settlements balance Bs as the

161
amount that has to be paid to ‘settle’ (or ‘balance’ for accounting
purposes) the net indebtedness arising from the balance of payments
deficits and surpluses. It is defined by the condition B + Bs ≡ 0, which
makes Bs identical to -B.20 To reiterate,

Bs ≡ -B. (15)

That is, the official settlements balance equals the negative of the balance
of payments B. Note that, with a balance of payments surplus, a positive
value of B (but a negative value of Bs) represents an increase in our
foreign exchange reserves.
The accounting balance of payments is given by (B+Bs) , which is
identically equal to zero, since (B+Bs) ≡ 0. Hence,

[(Xc - Zc + NR + NT) - (Xk - Zk )] + Bs ≡ 0, (16)

where ≡ indicates an identity.21 To reiterate, Hence, the accounting

balance of payments is a statement of the receipts and payments to
foreigners in a given period of time, in a form such that the total receipts
always equal total payments.
The accounting balance of payments is related to the (economic) balance
of payments by the identity B + Bs ≡ 0. The accounting balance of
payments has three components:
1. The balance of payments on current account, Bc .
2. The balance of payments on capital account, Bk .
3. The official settlements balance, Bs , such that Bs = -B = (-Bc - Bk).
Note that the balance of payments (in the macroeconomic sense of this
concept) consists only of (1) and (2).
An illustrative form of the balance of payments accounts is shown in
Table 3.1... For this tabular form, the inflows of funds are listed in one
column and the outflows in another one, with the balancing item Bs being
placed with an appropriate sign in one column or the other one. This table
lists the flows by type of activity or function. While the items in this table
are self-explanatory, a few comments would be useful. The interest and
dividend payment flows are the return to past flows of capital and appear
in the current account (of the balance of payments), while the capital flows
— through the purchases and sales of bonds (including loans, equities, and
deposits in foreign banks) — during the same period are part of the capital

162
account.22 To reiterate, Although there is a net surplus of 130 on the
current account, the deficit on the capital account of 145 (because of net
investments abroad) results in a net deficit of 15 on the overall balance of
balance of payments. This deficit is settled by net sales of foreign
exchange worth 15 by the country, so that the official foreign exchange
reserves decrease by 15 — and the official settlements balance is +15.
To conclude, from the accounting perspective of Table 3.1…, the
inflows of funds from the exports of commodities, financial instruments,
and foreign exchange equal 425. The outflows of funds from the imports
of commodities, financial instruments, and foreign exchange reserves also
equal 425. This is the essential aspect of an accounting balance of
payments: both sides of the statement are always equal. The element in
this balance sheet that ensures the identity of the two sides is the flows
out/in of the country’s foreign exchange reserves to settle the (economic)
balance of payments deficit or surplus.

3.8 The Market for Foreign Exchange and the

Changes in Foreign Exchange Reserves

3.8.1 The demand and supply of foreign exchange

As explained earlier, foreign exchange consists of all those assets that can
act as media of exchange in international transactions. These include gold,
currencies of foreign countries, and Special Drawing Rights (SDRs) at the
International Monetary Fund (IMF). The SDRs are created by the IMF and
held by individual countries as a form of demand deposits with the IMF
itself, and are designated in a specified basket of the national currencies.
The supply of our currency ($) in foreign exchange markets arises during
the payments process when we buy from foreigners their commodities
(which are our imports of commodities) or financial assets (which lead to
our export of capital), pay interest and dividends or transfer funds as gifts
to them. Conversely, foreigners’ demand for our currency arises because
they buy our commodities (which constitute our exports of commodities)
or our assets (which constitute our imports of capital), pay interest and
dividends or send gifts of currency to us.23 To reiterate, Hence,

$S$ = $Zc + $Xk + $OR + $OT (17)

$D $ = $Xc + $Zk + $IR + $IT, (18)

163
where:
S$ = supply of our dollars in the foreign exchange markets,
D$ = demand for our dollars in the foreign exchange markets,
OR (IR) = outflows (inflows) of funds for interest and dividend
payments, and
OT (IT) = outflows (inflows) of funds occurring due to unilateral
transfers.
Note that (IR − OR) = NR and (IT − OT) = NT , where NR is net inflows
of interest and dividends and NT is net inflows of transfers.

3.8.2 Equilibrium in the foreign exchange market

Equilibrium requires the equality of demand and supply, so that, in
equilibrium,

$S$ = $D$.

Since $S$ corresponds to the inflow of foreign funds in the balance of

payments and $D$ corresponds to the outflow of funds, equilibrium in the
balance of payments requires that B = 0.
Extended Analysis Box 3.2: The Demand and Supply of Foreign
Exchange Stated in Foreign Currencies
In terms of foreign currencies, the demand for foreign exchange (£s) is
the demand for an international mode of payment (other than one’s own
currency). This demand arises because the residents of the domestic
economy wish to purchase, i.e., import either commodities or financial
assets from other countries, pay interest and dividends and make transfer
payments.24 To reiterate, This demand corresponds to the outflow of
funds in the balance of payments. The supply of foreign exchange (£s) to
our economy arises because foreigners wish to purchase our
commodities or financial assets, as well as pay interest and dividends to
us, and send us gifts. This supply of foreign exchange to our economy
corresponds to the inflow of funds in the balance of payments. The
difference between the supply of and demand for foreign currencies
becomes the inflow of foreign exchange funds into our economy. A net
inflow (outflow) in the form of foreign currencies, gold and SDRs
increases (decreases) the country’s foreign exchange reserves.
The demand for foreign currencies (£DF) by the domestic economy is

164
 the sum of our imports of commodities (£Zc) and our imports of
financial assets corresponding to our capital exports or outflows (£Xk).
Similarly, the supply of foreign currencies (£SF) to the domestic
economy is the sum of our exports of commodities (£Xc) and our
exports of financial assets (corresponding to our capital imports) (£Zk),
plus net transfers of interest and dividends (£NR) and net unilateral
transfers (£NT ). Therefore,

where:
DF = demand for foreign currencies,
SF = supply of foreign currencies,
outflows (inflows) of funds for interest and dividend
£OR(IR) =
payments, and
£OT(IT) = outflows (inflows) of funds for unilateral transfer payments.
The demand for foreign currencies by a country (£DF) has its converse
in the supply of dollars ($S$) by us to foreigners to act as an element of
their foreign exchange balances. The supply of foreign currencies by
foreigners (£SF) is correspondingly the converse of the demand for
dollars ($D$) by foreigners from us. The market for foreign exchange,
therefore, can be looked at from two different viewpoints: as the demand
and supply of dollars in the foreign exchange market or as the demand
and supply of foreign exchange/currencies. We will often find it
convenient to switch from one to the other in the intuitive explanations
of our analysis.

3.9 The Market Determination of the Nominal

Exchange Rate
For the flexible exchange rate regime with efficient exchange markets,
Figure 3.1 illustrates the market for foreign exchange and the market
determination of the equilibrium exchange rate. The horizontal axis
measures the quantity of dollars in the foreign exchange market, while the
vertical axis specifies the exchange rate (in £s per $). This figure assumes
that the curve for $D$ follows the usual shape of demand curves and is
downward sloping, while that for $S$ follows the usual shape of supply

165
curves and is upward sloping. Equilibrium between $D$ and $S$ is shown
at the nominal exchange rate ρ* . Exchange rates above this equilibrium
rate would have ($D$ - $SD$ ) < 0, so that there would be a deficit in the
balance of payments. Exchange rates below this equilibrium rate would
have ($DD$ - $SD$ ) > 0, so that there would be a surplus in the balance of
payments.
Since the exchange markets for financially developed economies tend to
be efficient (i.e., go to equilibrium rapidly), the market exchange rate for
our economies would tend to be the equilibrium one. However, if the
exchange rate has been fixed by the government or the central bank, it
might be at the equilibrium level, or above or below the equilibrium level.
If it was set above the equilibrium level, the economy would have a
balance of payments deficit; if it was set below the equilibrium level, the
economy would have a balance of payments surplus.
Since both commodity and capital flows are components of the demand
and supply of foreign exchange, changes in the net exports of commodities
and net capital flows both produce changes in the market exchange rate. In
the short term, capital flows are usually more volatile, so that short-term
fluctuations in the exchange rate are usually due more to fluctuations in
capital flows than in commodity flows.
Chapter 12 will further expand on this analysis of the foreign exchange
market.

Figure 3.1

Fact Sheet 3.4 : Exchange Rates against the US Dollar, 1980−2008

This Fact Sheet illustrates movements in the exchange rates in terms of
the US dollar of the currencies of several countries. Some of these
countries had pegged exchange rates for part of the selected period. For
example, the decision to peg the national currency to the US dollar was
taken by Argentina from 1980 to 1988 and from 1991 to 2001, by
Mexico in the early 1990s, by Malaysia after 1997, and by China in

166
 1993. When this peg became unsustainable, they switched to a floating
exchange rate. The depreciation experienced after such a switch is
evident for Argentina in 1988 and 2001 as well as for Mexico in 1991.

3.9.1 Hot money

Some of the capital flows among countries are extremely sensitive to
expected interest and exchange rate changes, as well as to the political and
economic insecurity in the country. Speculation about a possible
devaluation/depreciation of the domestic currency can cause sudden,
heavy outflows of short-term funds seeking protection from it, or trying to
make a profit out of it. Funds whose flows among countries are very
sensitive to expected exchange rate changes, interest rate fluctuations or
security and convertibility (i.e., unhindered exchanges among currencies)
considerations are known as hot money.
Investments in short-term bonds can move easily among countries, so
that changes in them are part of hot money movements. Such investments
can be a major part of financial capital flows among countries.
Box 3.2: National Policies on the Balance of Payments and Accumulation
of Foreign Exchange Reserves
Countries hold reserves of foreign exchange with their central banks to
meet the possibility of deficits in the balance of payments. Deficits draw
down the country’s foreign exchange reserves by the amount of the
deficit and surpluses increase these reserves by the size of the surplus.
The size of the country’s foreign exchange reserves, or changes in it, is
a matter of national policy, so that ΔR F may be positive or negative for
long periods. For example, by the end of World War II in 1945, the USA

167
 had accumulated massive foreign exchange reserves that were really not
needed for financing its international transactions and could have been
reduced. It ran very significant deficits in its balance of payments
virtually throughout the 1950s and the 1960s. Germany and Japan had
hardly any foreign exchange reserves until the early 1950s and needed to
build them up. As their economies recovered and exports increased, they
maintained surpluses in their balances of payments for several decades
as a way of building up their reserves. Neither the US nor Germany or
Japan was especially concerned about their deficit or surplus for much
of the 1950s and 1960s. This example illustrates that countries often
deliberately accept continuing deficits or surpluses in their balance of
payments for fairly long periods. In 2005 and 2006, China provided a
clear example of continuing surpluses and the USA provided a clear
example of continuing deficits.

3.10 The Persistence of Balance of Payments Deficits and

Surpluses
A country may continue to have a balance of payments surplus even with
flexible exchange rates if its central bank is willing to accumulate the
resulting inflow of foreign exchange. Alternatively, if there is a deficit, it
may be willing to sell the required foreign exchange from its reserves. As
explained earlier, such a system is known as a managed exchange rate
system. Under a managed exchange rate, the exchange rate is manipulated
by the central bank through sales or purchases of foreign exchange from its
reserves or changes in domestic interest rates. This provides one reason for
the persistence over time of balance of payments surpluses and deficits.
There are several additional reasons for the persistence over time of
balance of payments disequilibrium. These have to do with the elasticities
of the exports and imports of commodities and lags in the adjustment of
foreign exchange markets from a disequilibrium position to equilibrium.
These reasons will be discussed in Chapter 12 and Chapter 13 on the open
economy.

Conclusions
• While some countries in the world have floating/flexible exchange rates,
others have fixed/pegged ones. Many countries with flexible exchange
rates try to control/manage the movements in their exchange rates.
• PPP and IRP provide the main theories for explaining the differentials

168
 between inflation rates, exchange rates, and interest rates across
countries. These theories are especially likely to apply to countries with
high ratios of exports and imports to GDP and with large inflows and
outflows of capital.
• The weight of empirical evidence is that PPP does not strictly hold
among countries even over several years. The empirical failure of PPP
occurs mainly because of the existence of goods that cannot be traded
across borders and because of capital flows.
• The weight of empirical evidence is that most currencies do not strictly
possess interest rate parity with others. The empirical failure of IRP
occurs because of imperfect capital markets and differences in the
riskiness of domestic versus foreign assets, especially the ‘exchange-rate
risk’ in investing abroad.
• Foreign exchange consists of foreign currencies, gold, and SDRs issued
by the IMF. These serve as the media of exchange in international
payments. A decrease in the demand for the domestic currency in the
foreign exchange markets induces a depreciation of its exchange rate,
while a decrease in the supply of the domestic currency in the foreign
exchange markets induces an appreciation of its exchange rate.

KEY CONCEPTS

Exchange rates: nominal real, and effective

real, and effective
Purchasing power parity
Interest rate parity
Balance of trade
Balance of payments on current account
Balance of payments on capital account
The balance of payments
The accounting balance of payments
Official settlements balance
Foreign exchange reserves
The demand and supply of foreign
exchange, and
Fixed, flexible, and managed
exchange rates .

169
 SUMMARY OF CRITICAL CONCLUSIONS

•The nominal exchange rate defines the value of a currency in terms of

another currency.
•The real exchange rate defines the conversion rate between the
commodities in one country and the commodities in other countries.
•The real exchange rate rather than the nominal one indicates the relative
prices of commodities in different countries.
•Relative PPP rather than absolute PPP is more likely to apply among
countries.
•IRP is an important determinant of interest rates across countries with
perfect capital flows.
•While the accounting concept of the balance of payments is that the
balance of payments will always be in balance, this is not so for the
(economic) balance of payments, which can have a positive or negative
value. The latter is said to be in equilibrium when its value is zero.
•Countries can have fixed, flexible, or managed exchange rates.
•Most countries with floating exchange rates usually try to manage the
movements in the exchange rate.

Appendix
Extended Analysis Box 3.1A: Comparison of the Actual and PPP Costs
of the Big Mac in Different Countries, 27 April 2000

Notes:
Nominal exchange rate = US$ per unit of local currency.
Implied PPP exchange rate = foreign (US$) price of the Big Mac
divided by the local price of the Big Mac.
PPP check is done by multiplying PPP-implied exchange rate by local
price and dividing by foreign price.
This should equal 1.
US$ per unit of local currency is the actual nominal exchange rate.

170
 Big Mac price in US$ = Big Mac price in local currency times the
nominal exchange rate.
% Overvaluation (+) against US$ = (actual exchange rate-PPP exchange
rate)/PPP exchange rate.
k equals the nominal exchange rate times the domestic price of the Big
Mac divided by its US price.

Findings.
With k less than 1 for Australia, Canada, and Mexico, the Big Mac is
relatively cheaper in these countries than in the USA.
With k more than 1 for Britain, the Big Mac is relatively more expensive
in Britain than in the USA. Source of original data:
http://economist.com/markets/bigmac/displayStory.cfm.

REVIEW AND DISCUSSION QUESTIONS

1. What are the two ways of defining the nominal exchange rate? Which
one is used in this book?
2. What is purchasing power parity? Explain its relationship to the real
exchange rate.
3. What does relative purchasing power parity imply for changes in the
nominal exchange rate?
4. Does PPP apply to different shops in the same city? Same country?
Across countries? Illustrate with some examples. Give reasons for the
deviations from PPP.
5. What is the relative importance over time of IRP and PPP in
determining movements of the exchange rates?
6. Define equilibrium, surplus, and deficit in the balance of payments?
What does each imply for changes in the country’s foreign exchange
reserves?
7. Is the accounting balance of payments always in balance? Give reasons
for your answer.
8. Is the (economic) balance of payments always in balance? Give reasons
for your answer.
9. Distinguish between fixed, flexible (floating), and managed exchange
rates? Which does your country have at present?
10. Show diagrammatically the demand and supply of the domestic
currency in the foreign exchange markets. For the figure as you have
drawn it, is the equilibrium exchange rate stable or unstable? Explain

171
 your answer.
11. How is the payment of interest to foreigners treated in the country’s
balance of payments?
12. Why do countries with flexible exchange rates continue to have
deficits or surpluses on their (a) current account, (b) capital account, and
(c) balance of payments?

ADVANCED AND TECHNICAL QUESTIONS

T1. Suppose that a given bundle of commodities costs C$50 in Canada and
£20 in Britain, and the exchange rate (£ per C$) is 0.5. What is the real
exchange rate between these countries? Is this different from the real
exchange rate that PPP requires?
T2. Why do underdeveloped economies tend to have real exchange rates
less than unity? What difference does it make in comparing standards of
living across countries?
T3. A simplified form of Canada’s balance of payments on current account
is presented below. Calculate the values of (a) net exports of
commodities, (b) net interest and dividend flows, (c) net transfers, and
(d) balance of payments on current account.

T4. A simplified form of Canada’s net inflows on capital account is

presented below. Calculate the balance of payments on capital account
(including the statistical discrepancy)?

172
T5. Given the information in the preceding two questions, what was
Canada’s balance of payments in the three years? Was it in equilibrium,
deficit or surplus?

1The remaining flows are of immigrants and transient migrants. While

these are not very significant for most economies, there are a few
countries, especially in the Middle East, for which such workers can be a
significant proportion of the labor force.
2To illustrate this point, assume that we are comparing the price of an
identical car that costs £10,000 in Britain and $24,000 in Canada, and the
nominal exchange rate is £0.5 per $.Hence, for this car, ρr = ρP/PF = 0.5 X
24,000/10,000 = 1.2 The units of measurement of ρr, from ρP/PF are: (£
per $)($;per Canadian car)/(£ per British car) = British cars per Canadian
car.
3Since the dimensions of ρ are £/$ and of P/PF are $/£, the dimension of ρr
is [(£/$)·($/£)], where the £ and $ signs cancel out, so that the dimension
of the real exchange rate is foreign commodities per domestic ones.
4The PPP index of nominal exchange rates — which is an index over time
of the nominal exchange rate which makes the real exchange rate equal to
unity — would be an even better measure for certain purposes such as
comparing standards of living across countries.
5The term ‘pegged’ is more appropriately used where the exchange rate by
the policymaker is changed periodically.
6This activity is known as arbitrage, which consists of buying in cheaper
countries and selling in more expensive ones so as to make a profit from
the price differences.
7http://www.economist.com/markets/bigmac/displayStory.cfm.
8This is the domestic price of the Big Mac divided by its price in USA.
9Local currency per US dollar.
10This is the domestic price of the Big Mac divided by its price in the

173
USA.
11Note that the price indices do not take account of capital flows.
12The mode of construction of this index first requires finding for a
common, pre-selected bundle of commodities the domestic price and the
foreign price. Suppose this bundle is labeled as A and its domestic price is
P PPPwhile its foreign price is P *PPPW. Under PPP, ρ PPP . (P PPP/P *PPP)
will equal unity, where ρ PPP is the PPP exchange rate. Hence,ρ PPP = (P
*PPP/P PPP).
13Source: John L. Allen, Student Atlas of Economic Development.
Dushkin/McGraw-Hill, 1997, pp. 69–72.
14This is an approximate formula that does fairly well at low values of the
expected rate of change in the exchange rate but not very well at high
values. It is only approximate since it does not, for foreign investments,
take into account the conversion of the earned interest rate into the
domestic currency.
15The ability to do so depends upon the country’s foreign exchange
reserves (and other support it can arrange from other countries or the IMF)
relative to the private capital flows. Since the latter can be potentially
much larger than the former for most countries, countries now possess a
very limited ability to manage exchange rates contrary to market forces.
16Such as assumption would be unrealistic for countries with large
positive or negative values of either NR or NT. Among these are countries
with a large externally held debt, as in the case of many developing
economies, or large net remittances as in the case of some oil-rich
countries on the Arabian Peninsula with a large number of foreign
workers.
17A clear example is provided by the case of China in 2004 and 2006. It
was both a net exporter of commodities and a net exporter of capital.
18A clear example is provided by the case of USA in 2004 and 2006. It
was both a net importer of commodities and a net importer of capital.
19However, note that in practice, central banks do often want to build up
their foreign exchange reserves throughB > 0. An example of this is
provided by China in 2005 and 2010.
20Note that Bs > 0 indicates an outflow of foreign exchange from the
central bank’s official reserves to pay for a balance of payments deficit (B
< 0). Conversely, if there is a balance of payments surplus (B > 0), Bs < 0
and there is a corresponding inflow of foreign exchange reserves, which
become an addition to the official reserves.
21This is an identity since the left side will always equal zero.
22High levels of net inflows (outflows) of capital over some years will

174
imply increased outflows (inflows) for interest and dividends in
subsequent years, creating a dynamic relationship between the present
capital account and the future current account. The donations to foreign
countries are unilateral flows and enter in the current accoun
23Note that our exports (imports) of capital Xk would be equal to our
imports (exports) of bonds (including equities) Zb .
24Note that capital equipment and services such as shipping and tourism
are already included in the definition of commodities (goods and services)
and are part of imports and exports.
25A minus sign indicates an outflow of capital due to an increase in our
claims on nonresidents or a decrease in liabilities to them.
26This item includes bonds and stocks.
27This item consists of loans, deposits, official international reserves, etc.
28This item consists of loans, deposits, official international reserves, etc.

175
 PART II
Short-run Macroeconomics

176
 CHAPTER 4
Determinants of Aggregate Demand:
The Commodity Market of the Closed
Economy

This chapter presents the analysis of the commodity market of the

macroeconomy. For this analysis, it encapsulates the relationships
of the commodity market into the IS equation and curve.

Chapter 1 listed four types of goods in the closed economy. These are:
1. Commodities, usually referred to as goods and services, and including
both consumer goods and physical capital
2. Money
3. Bonds (i.e., all non-monetary financial assets)
4. Labor
This chapter lays out the assumptions and their implications for the
commodities market of the closed economy. It encapsulates this analysis
into the compact IS relationship and curve.

4.1 Symbols Used

Many of these symbols used in this chapter have already appeared in the
preceding chapters. However, for convenience, the meanings of the
symbols used in this chapter are specified in Table 4.1.
In most cases, the small letter symbol will indicate the real value of the
variable while the capital one will indicate its nominal (dollar) value.
Thus, y is the real value of income/output while Y is its dollar value. One
exception to this rule is the use of K to designate the real value of the
physical capital stock.
The superscript d on a variable will indicate its demand and the
superscript s will indicate its supply. The variable symbol without either of
these superscripts will indicate its actual value. If there exists equilibrium,

177
the actual value will be the equilibrium one, i.e., one at which the demand
and supply for that variable are equal. Thus, n d is the demand for labor
and n s is its supply, while n is actual employment.
The symbol ≡ indicates an identity while the symbol = indicates an
equilibrium condition.

4.2 The Commodity Sector of the Closed

Economy
As specified in Chapter 1, this book defines commodities as what are
usually referred to in everyday English language usage as goods and
services. This category includes both consumer goods and physical capital.

4.2.1 Uses of national income

Commodities are produced by firms with the use of inputs (factors of
production). Firms incur expenses in the form of wages, profits, and rents,
the sum of which is aggregate income. Therefore, real national income (y)
represents both the aggregate cost of production of firms1 as well as the
aggregate income of all economic agents in the economy. In the closed
economy, the recipients of national income y can spend it on consumption
c , pay it in net taxes (i.e., total taxes paid to the government by the public
less transfers from the government to the public) t or put it away for future
periods as saving s .2 That is,
Table 4.1 Symbols used in this chapter.
Symbols Meaning
aggregate quantity of commodities (also often referred to as
y
output/national income)
yd aggregate demand for commodities
ys aggregate supply of commodities
yf full-employment output/income
e aggregate expenditures (in real terms)
c consumption expenditures
co autonomous consumption
s (private) saving
sn national saving
i (planned) investment expenditures

178
iu unintended investment
io autonomous investment
net taxes paid (government's tax revenues less government
t
transfers to the public)
to autonomous net taxes paid
g government expenditures on commodities
r real rate of interest
R nominal/market rate of interest
P price level
π rate of inflation
πe expected inflation rate
Nominal national income is written as Y and equals Py.

y ≡ c + t + s. (1)

Note that all the variables are in real terms, so that the data collected on
their nominal values would have been deflated by the appropriate price
index (see Chapter 1).

4.2.2 Sources of national expenditures

For the closed economy, the demand for commodities — which constitutes
the sales revenue of firms — originates as consumption expenditures c , as
government expenditures g on goods and services, or as (desired )
investment expenditures i which are deliberately undertaken by firms.
Such investment expenditures deliberately incurred by firms are for the
purpose of intentionally increasing their capital stock used for production
in future periods and are also known as ‘desired’, ‘intended’, ‘planned’, or
‘ex ante’ investment. We will just refer to them as ‘investment’, leaving its
adjectives as implicit. Designating the real aggregate expenditures on
commodities as e, where e is, by definition for the closed economy,
specified by:

e ≡ c + g + i, (2)

where g is real government expenditures on commodities produced in the

current period. Note that neither g nor t includes transfers from the
government to the public (see Chapter 1).

179
4.2.3 Equilibrium in the commodity market
National income y is the sum of the payments to the inputs in the
production process and is thus the aggregate cost of production in the
economy. Aggregate expenditures e is the aggregate revenue from the sale
of the output produced. Ongoing equilibrium in the production process
(i.e., according to firms’ production plans) requires that the firms’
aggregate costs and revenues be equal. That is, the commodity market
requires y = e for equilibrium.
To check that e = y represents the equilibrium position, examine what
happens in the economy if e < y or if e < y. If e < y , firms’ revenues
exceed their costs, so that they find it profitable to increase production —
and also possibly raise their prices. To increase production, they will hire
more workers and other inputs. This would increase income y , so that y
would tend to increase toward e. If e > y , firms find that their input costs
are higher than their sales revenue, so that they will cut back on production
and employment of inputs. This would decrease employment and the
incomes paid to the factors of production, so that y will fall toward e.
Therefore, equilibrium will only exist in the commodities market if,

y = e. (3)

When e = y, firms’ expenses exactly equal their revenues, so that they do

not have an incentive or compulsion to change their employment of factors
of production. Therefore, y = e represents equilibrium between income and
expenditures in the commodity market for the closed economy.
Since y = c + t + s and e = c + i + g, equilibrium (y = e) in the closed
economy commodity market requires that:

c + t + s = c + g + i,

which is the same as:

t + s = g + i. (4)

The equality of (t+s) and (g+i) is another way of stating the equilibrium
condition for the commodity market for a closed economy. As Chapter 5
will show, it differs from the equilibrium condition for an open economy.

4.2.4 National saving

National saving s n is defined as the sum of private saving s and

180
government/public saving (= t – g), so that:

s n ≡ s + (t – g). (5)

This equation is a definitional identity — i.e., a way of defining a variable.

Therefore, another way of stating the equilibrium condition for the
commodity market of the closed economy is:

s n = i, (6)

which asserts the equality of national saving and investment for

equilibrium in the closed economy, and is another form of the equilibrium
condition for a closed economy. Note that this equality of national saving
and investment for the closed economy does not hold in disequilibrium in
the commodity market and is not an identity.

4.2.5 The relationship between saving and investment

We specified i above as the firms' desired real investment expenditures
undertaken for the purpose of increasing their capital stock. These are
incurred because the firms want to change their capital stock. For
achieving this, at the level of a single firm, the firm can buy the
commodities from other firms or use some of its own output. The critical
element in this definition, determination, and measurement of the
macroeconomic concept of investment is not whether the firm buys
commodities needed to add to its capital stock from other firms or uses
some of its own output.3 The critical element is whether the firm wants to
change its capital stock and how much investment it wants to undertake for
that purpose. For these reasons, this concept of investment is also known
as desired, intended, or ex ante investment. Macroeconomics uses the
single word ‘investment’ for this concept, thereby leaving the adjectives
‘desired, intended, planned, or ex ante ’ as implicit. Note that this
investment can be partly in terms of buildings, machines, etc., which will
be used to produce more commodities in the future and partly in the form
of desired changes in the firms’ inventories of their goods to meet their
future expected sales.
Real private saving is the ‘residual’ of current output left over after
consumption and payment in taxes. This residual may be directly
committed by the savers themselves for future production in the form of
investment in physical capital (such as in a newly built house), but most of
it tends to be lent to others in exchange for bonds (which includes the

181
purchase of shares) or exchanged for money balances. Therefore, the
overall investment expenditures on commodities may not equal the supply
of saving in an economy in which other goods—such as bonds and money
— exist. Further, saving is done by households on the basis of their utility
maximization, subject to a budget constraint, while investment is
undertaken by firms based on profit maximization. There is no reason to
assume that the saving intentions of the myriad households in the economy
will in the aggregate just and always produce the aggregate level of saving
that will equal the aggregate investment expenditures undertaken by the
myriad firms.
National saving is based on the plans of households and governments.
There is again no basis for assuming that national saving planned by these
will always be an amount equal to the investment intended by firms. What
the above analysis shows is that the equilibrium of the commodity market,
when and if it comes about, will show the equality of national saving and
investment for a closed economy. But, this equality will not apply if there
is disequilibrium.

4.2.6 The (physical) capital stock

For a firm, the actual capital stock of a given period is defined as
consisting of all the commodities carried over from the preceding period.
The actual capital stock can differ from the desired capital stock.
Investment i is the desired change in the existing capital stock, and can
differ from the actual change in the existing capital stock. Further, one of
the simplifying assumptions of macroeconomics is that, in the short run,
this investment does not become operational as capital in production use
due to a gestation lag. Hence, the capital stock is held constant in short-run
macroeconomic theory, even though this theory envisages investment as
occurring.
Box 4.1: An Unjustified Oversimplification for a Modern Economy
If there is no government at all in the economy—which is quite an
unrealistic assumption for the modern world—or if the government
expenditures on commodities (goods and services) and the net taxes
received by the government are equal, the budget is balanced, so that g =
t. In this special case, the equilibrium condition simplifies to:

s = i. (4´)

The equilibrium condition for the commodity market is sometimes

182
 expressed as (4´), which requires the equality of saving and investment
for equilibrium in the closed economy. However, since this condition
requires a balanced budget (which rarely occurs) or the absence of the
government (which is now patently unrealistic), it is preferable to
specify the equilibrium condition for the closed economy as t + s = g +
i , rather than as s = i.

A cautionary note : (s + t) = (i + g) and s = i are not identities.

Since s + t = i + g (or its simplified form s = i ) is an equilibrium

condition, it would not hold in disequilibrium. Since it does not hold
under all circumstances — e.g., in disequilibrium — it is not an identity.
Further, saving and tax payments are done by households, while
investment is undertaken by firms and government expenditures are
expenses incurred by the government. There is no reason to assume that
in the real world in real time, the actions of these very different
economic agents will produce the value of (s + t that would be the same
as that of (i + g) over any given quarter or year. Whether they are equal
for any given quarter or year would depend on whether the commodity
market comes into equilibrium in that quarter or year. It may or may not.
However, for analytical convenience, the analysis is pursued under the
assumption of (short-run, though not short-term) equilibrium in the
commodity market.

4.3 The Two Uses of Private Saving and the Drag

of Deficits on Investment
We have so far derived the following alternative forms of the equilibrium
condition for the closed economy.

s+t=i+g,

s n = i.

We can restate the above conditions as:

s = i + (g – t ).

This equation states that, in the closed economy, private saving can be
used for two purposes: investment and a fiscal deficit (= g–t) . To the

183
extent that the government absorbs private saving to finance its deficit
(through borrowing by the issue of its bonds), it reduces the amount of
commodities that can be used for investment. The less left for the latter,
the slower the growth of the physical capital stock and the smaller will be
the future production capacity of the economy. Current fiscal deficits,
therefore, hurt future increases in output and the standard of living.
The above equilibrium condition can also be stated as:

s /y = i /y + (g – f )/y.

Hence, for a given level of fiscal deficits as a proportion of GDP, the

higher is the domestic saving rate (i.e.,s/y), the greater will be the
investment/GDP ratio and the greater will be the growth rate of the future
productive capacity of the economy. A high saving rate is, therefore,
desirable for the economy.

4.4 Disequilibrium and the Role of Changes in

Inventories in the Adjustment to Equilibrium
Note that there is no basis for assuming that a particular real-world
economy will always or mostly have commodity market equilibrium. In
fact, because the behavior of households, firms, and the government is
constantly changing and causes variations in private saving, investment,
and fiscal deficits, and because there are usually lags in commodity market
adjustments, the more common state in the real world is likely to be one of
disequilibrium, rather than of equilibrium. In this case, the relevant
assumption is that when the economy is in disequilibrium, it does make
adjustments to move toward equilibrium.
When the commodity market is not in equilibrium, we need to examine
the mechanism that might take this market toward equilibrium. This
mechanism operates through the responses of economic agents and
markets. To illustrate this mechanism, start from an initial equilibrium and
assume that a positive shock4 increases expenditures such that national
expenditures e become greater than national income y , i.e., e > y. Since e
> y , firms face a higher demand for commodities than their production,
since y represents not only national income but also the amount of
commodities produced in the economy. To meet this demand, since an
increase in production takes some time to implement, firms would have to
deplete their inventories below the planned levels or raise their product
prices. Firms’ revenues are also higher than their costs. Hence, they have

184
an incentive to increase production — as well as raise prices. To increase
production, firms have to increase their employment of labor and other
inputs, thereby increasing national income y in the economy.
Similarly, a negative shock to expenditures would produce for some time
e < y. In this case, firms’ revenues are less than their costs and they sell
less than their output, leading to losses and the unintended accumulation of
inventories of commodities. This will cause firms to decrease their
production and employment of inputs, thereby decreasing factor incomes
y. Such adjustments will continue until y equals e and the economy enters
its equilibrium state. In this state, firms’ demand and supply, as well as
their revenues and costs, are equal and there is no tendency for them to
further adjust their employment or output.
To recapitulate, national income y paid out by firms to inputs and
national expenditures e received by firms from the sale of their products
are not always equal. Over any given period, firms both individually and in
the aggregate may end up with more or less sales revenue than their costs.
If this happens, firms change their production plans and employment — as
well as their product prices and payments to inputs. If and when these
alterations bring about the equality of income and expenditures, we say
that the commodity market has reached equilibrium. An efficient economy
will continue with changes in the firms’ production plans until the
commodity market equilibrium is attained. But, since this process involves
the production and consumption plans of very large numbers of firms and
households in the economy, any disequilibrium is unlikely to be eliminated
instantaneously. Therefore, after a shock, the economy could be in
disequilibrium for several quarters, if not years.

4.4.1 The role of unintended changes in inventories

Firms hold inventories of both unfinished and finished goods because
production takes time and firms want to be able to meet orders for their
products within a desired time. The total value of these inventories is part
of the firm's capital stock. Changes in inventories can be desired or
undesired ones. The major reason for the former is an increase in the firm's
expected sales. Such desired changes in inventories are part of the firm's
(desired) investment, as this term is defined in macroeconomics.
Unintended changes in inventories are not part of the firm's (desired)
investment. They reflect an unintended difference between the firm's
output and its sales. Since inventories are costly to hold, the firm would
want to reverse any unintended accumulation of inventories by cutting its
production. It may also try to increase its sales through a reduction in the

185
price of its product.
At the macroeconomic level, unintended changes in the economy's
inventories are a reflection of the difference between the economy's output
y and its sales e . Such changes are a symptom of disequilibrium in the
commodity sector of the economy. Therefore, economic analysts pay
considerable attention to the movements in inventories since they serve as
an indicator of potential disequilibrium.
If y > e , firms will be producing more output than selling, so that there
will be an unintended increase in inventories. Firms would cut back on
production (a) to unwind any unintended accumulation of inventories and
(b) to meet the lower demand level. They may also decrease their prices
somewhat to increase sales. Conversely, if e > y, firms would have sold
more than they had planned for and there would be an unintended decrease
in inventories. Firms would increase production (a) to replenish
inventories to reach their desired levels and (b) to meet the higher demand
level. They may also try to reduce the quantity demanded by raising their
prices.
Extended Analysis Box 4.1: A Simplified Diagrammatic Analysis: The
45° Diagram
For the simplest diagrammatic analysis at this stage, it is better to
convert the variables into their nominal values. Figure 4.1a plots the
equations:

Equilibrium condition: E = Y
Composition of expenditures: E ≡ C + I + G,

where E = Pe , Y = Py, C = Pc, I = Pi, and G = Pg. Figure 4.1a shows E

(nominal expenditures) on the vertical axis and Y (nominal income) on
the horizontal axis. The 45° line represents E = Y and is, therefore, the
line for the first equation above, which is the specification of
equilibrium in the commodity market. By the second equation above,
national expenditures E are shown by the curve marked C+I +G. Figure
4.1a assumes that consumption C increases proportionately with income
Y , while investment I and government expenditures G are independent
of income. Equilibrium occurs at the point where the curve (C+I+G )0
intersects the 45° line. This equilibrium is shown by the point a at which
E* = Y *.5 Points on the (C +I +G) line, other than the point a, are ones
of disequilibrium. On this line, points to the left of the 45° line have E >
Y ,6 so that firms will have an unintended decrease in inventories and/or
not be able to meet the demand for their products. To restore inventories

186
to their intended levels and to meet the higher demand, as well as
increase profits, firms will increase production and/or prices. These will
increase firms’ revenues, which will raise national income. Therefore, Y
will increase toward equilibrium. On the (C+I+G) line, points to the
right of the 45° line have E < Y ,7 so that firms will be paying more in
factor incomes than they collect from sales revenues, which entails
losses. They will also be left with unintended increases in inventories.
They can restore inventories to the desired, lower levels, as well as
increase profits by cutting back on production and employment, so that
incomes will fall. Therefore, Y will decrease, moving the commodity
market toward equilibrium, at which Y is lower than initially but now
equals E . This type of the response pattern of firms to disequilibrium
between Y and E ensures the stability of the commodity market
equilibrium, which is the condition that E* = Y* .
Diagrams of the type used in Figures 4.1a and 4.1b are known as the 45°
diagrams. The meaning of the symbols Y and E on the axes of these
figures needs to be noted. In these figures, E is nominal national
expenditures and Y is nominal national income. Suppose that firms
become more optimistic about the future of the economy and increase
investment. Let this increase in investment be such as to shift the (C + I
+ G) curve from (C + I + G )0 to (C + I + G )1 in Figure 4.1b. The
equilibrium of the economy moves from Y*0 to Y*1.

187
 Figure 4.1b

4.4.2 Limitations of the 45° diagram for

macroeconomic analysis
The use of this 45° diagram for the analysis of the commodity market
suffers from two deficiencies:
• Since the price level is not determined in this diagram, the increase in
real expenditures and income cannot be determined from this diagram.
In reality, prices are likely to be pushed up by the higher expenditures
and the consequent higher demand for commodities, so that we need a
more elaborate analysis and diagram, which simultaneously determines
real expenditures/income, prices, and real output. Such a diagram was
the AD-AS one presented in Chapter 1. We will revert to the AD-AS
diagrammatic analysis in subsequent chapters.
• The 45° diagram only captures the dependence of consumption on
income but not that of investment on the interest rate. Because of the
latter, there is likely to be interaction between aggregate expenditures
and the interest rate, which is not captured in the preceding analysis
and diagrams.
To avoid these deficiencies, we need to use the more elaborate IS-IRT
and AD-AS analyses developed in Chapter 5.

4.5 For the Macroeconomic Analysis of the

Closed Economy, Is There a National Income
Identity and One Between National Saving and
Investment?
It is sometimes asserted that there is an identity between real national
saving and real investment8 — that is, i ≡ sn —so that what is saved
automatically and instantaneously becomes national investment. This is
not true for the modern economy, whether or not there is a balanced
budget. Intuitively, households and governments save while firms invest,
and the motives of these agents are very different. It would be unrealistic
to expect that, in the aggregate, the plans to save of the very large number
of very diverse households and of the government will identically coincide
with the plans to invest of the very large number of very diverse firms.

188
Further, in the modern economy, most households do not even directly
lend their savings to firms, but can hold them in currency or, more usually,
lend it to financial intermediaries, who may choose to lend out only part of
their own borrowing. Therefore, there cannot be a saving-investment
identity in short-run macroeconomic analysis.
Correspondingly, there is no ‘national income identity’ (i.e., y ≡ e )
between national expenditures and income in economic terms. These
variables can differ from each other, with a difference indicating
disequilibrium in the commodity market and leading to adjustments in
employment and output in the economy.

4.5.1 An accounting national income identity

What we have studied so far are economic relationships between the
various variables, especially between national expenditures and income
and between national saving and investment. We have to distinguish
between these relationships — from the seemingly similar though quite
different in economic theory — and accounting ones between the same
variables. To do so, assume that there is disequilibrium in economic terms
between expenditures and income. If expenditures are less than
income/output, firms are forced to hold goods that they had intended to
sell but could not sell. Hence, their inventories of goods increase to an
undesired extent. Conversely, if expenditures exceed income, firms have to
sell some commodities (out of inventories) that they had not intended to
sell. Hence, their inventories of goods decrease to an undesired extent.
Such unintended changes in the inventories of goods constitute
‘unintended investment’ and are identically equal to (y – e). That is:

iu≡y–e, (7)

where i u is unintended investment equal to the unintended increase in

inventories. If i u > 0, inventories increase — by an undesired amount —
and if i u < 0, inventories fall. Therefore:

y ≡ e + i u. (8)

To see why the preceding equation is an identity, substitute (y – e ) for i u.

This gives y = e + y – e = y, which is clearly an identity. Equation (8) is
known as the national income identity. Expanding both sides of this
identity,

189
 c + s + t ≡ c + g + i + i u,

so that:

i + i u ≡ s + (t – g) .

Setting i a = i + i u and calling i a the accounting measure of investment,

we have the identities:

i a ≡ s + (t – g) ,

i a ≡ s n, (9)

so that, for a closed economy, there is a national income identity between

the accounting measures of investment and national saving. i a is also
sometimes called actual or ex post investment, so that the national income
identity for a closed economy can also be stated as the equality of actual
investment and national saving. Data is easier to collect on i a than on i , so
that the statistics on investment are usually based on i a. This makes it
difficult to judge from the national accounts data whether there is
equilibrium or disequilibrium in the commodity market. Therefore, data on
changes in inventories is usually used as an indicator of unintended
investment and disequilibrium.
Extended Analysis Box 4.2: The Distinction Between the Meanings of
Investment in Macroeconomics and the Accounting Definition of
Investment
For the economic analysis, we included the investment expenditures of
firms in national expenditures. At the level of the firm, investment is the
amount of expenditures the firm (intentionally/deliberately) incurs in an
attempt to maintain or add to its capital stock (i.e., the stock of
commodities they want to possess for future production). As pointed out
earlier, such investment is also known as planned, intended, or ex ante
investment.
Accounting/actual investment equals these investment expenditures
and the unintended accumulation of inventories, labeled above as
unintended investment. At the level of the single firm, the latter results
from the firm's unintended change in its inventories because of a failure
to match production to the sum of actual sales and desired changes in
inventories. It often results from a failure to correctly anticipate the

190
 demand for the firm's products. At the macroeconomic level, unintended
investment results from the collective failure of firms to match
production less the intended additions to inventories to aggregate
demand. It often results from the failure of firms in the aggregate to
correctly anticipate aggregate demand in the economy and is a symptom
of disequilibrium in the commodity market—which will induce changes
in national production and income.
The actual change in the capital stock and the definitions of investment
The capital stock is defined as the amount of commodities carried over
one period — say the current one — to the next one. Most of these
commodities consist of buildings, machines, roads, dams, etc., which
contribute to the production of new commodities. But it also includes
inventories of commodities, produced in the current or past periods and
intended for sale in subsequent periods. In commodity market
equilibrium, these inventories will equal their desired level. However, if
expenditures are less than output in the current period, some of the
commodities that the firm had produced with the intention of selling
them will remain unsold. This amount will be added to inventories and
will constitute an unintended change in inventories. The measurement of
the capital stock at the end of the period will include them. Therefore,
the change in the capital stock during a period is identical to the
accounting definition of investment. That is:

ΔK ≡ i a ≡ i + i u,

where ΔK is the actual change in the capital stock, which includes any
unintended investment. As a corollary, the change in the capital stock
differs from the investment i unless the commodity market is in
equilibrium—which requires i u = 0. But, since i a is identically equal to
s n, ΔK is identically equal to s n — but not to (the macroeconomic
concept of) i !

4.6 Demand Behavior in the Commodity Market

In the closed economy, the only economic agents that create the demand
for commodities are households, firms, and the government. We now
study their demands for commodities.

191
4.6.1 Consumption expenditures
Consumption expenditures are expenditures by households on consumer
(durable or non-durable) goods. Their most important determinant is
current disposable income. The remainder (disposable income less
consumption expenditures) is saving, which is meant to provide for known
future consumption needs, such as for retirement, or because of the
uncertainty of future incomes or consumption needs such as medical
ones.9
Therefore, our model assumes that real consumption expenditures c
depend linearly upon real disposable income y d, which equals income y
less net taxes t paid to the government.10 Hence, disposable income y d
equals (y – t ). Therefore, our consumption function is:

c = c (y d) = c 0 + c yy d,

= c 0 + c y(y – t) 0 ≤ c y ≤ 111, (10)

where the symbol c in c(y) stands as a functional symbol and is indicative

of the dependent variable while c 0 and c y are parametric symbols,
indicating that they are exogenously determined (held constant unless a
change in them is hypothesized) rather than variables. The names given to
these two parameters are:
c 0 autonomous consumption — which is that part of consumption that
does not depend upon variables in the model
c y the marginal propensity to consume (MPC), since it specifies the
change in consumption for a unit change in disposable income.12
On average, a one-dollar increase in income increases consumption
expenditures by a positive amount but by less than one dollar since some
of the income increase is put toward saving. Hence, c y lies between zero
and one, and is usually between 0.8 and 0.95 for developed economies.
Note that the above consumption function implies that real (private)
saving s is given by:

s=yd–c (11)

= y d – c 0 – c yy d 0 ≤ c y ≤ 1,

= –c 0 + (1 – c y)y d.

192
The linear form of the saving function can also be written as:

s = s 0 + s yy d. (12)

Comparing the above two equations, we have:

s 0 = –c 0

sy=1–cy

so that the marginal propensity to save (MPS) equals (1–MPC). For c y

between 0.8 and 0.95,the MPS would be between 0.2 and 0.05.
Note that the average propensity to consume (APC) is measured by c/y
and is different from the MPC, which is given by and equals cy.
Similarly, the average propensity to save (APS) is measured by s/y and is
different from the marginal propensity to save (MPS), which is given by
and equals s y(= 1 – c y).

Fact Sheet 4.1: Consumption and Disposable Income in the USA, 1980-
2008
This Fact Sheet uses US data to illustrate the relationship between
consumption and disposable income. Plotting disposable income per
capita against total consumption expenditures per capita for various
years yields a virtually linear relationship, indicated by the line trend
with an intercept (which corresponds to what the text has called as
autonomous consumption). This suggests that the linear consumption
function used in the text is a good approximation to actual consumption
behavior.

193
 Figure 4.2a shows real consumption expenditures c and saving s as
functions of real disposable income y d. Autonomous consumption c 0 is
shown by the intercept of the curve c . The slope of the consumption curve
equals the MPC. The saving curve is marked s and represents the
difference between the 45° line, which in this diagram represents e = y d,
and thec curve. Its slope is (1–c y) and its intercept with the vertical axis is
at –c 0. At y d0 c = y d, thereby implying that at this point s = 0. The saving
curve marked s shows the behavior of saving. Autonomous saving equals
–c 0, which provides the intercept of the s curve with the vertical axis. The
slope of the s curve is (1 – c y). For c y = 0.9, the slope of the s curve is
0.1, while that of the c curve is 0.9. Therefore, the s curve will be very
much flatter than the c curve.
Figure 4.2b shows the effect of an increase in autonomous consumption
from c0 to c’0. This shifts the consumption curve up in a parallel fashion
from c to c’. Note that a change in the MPC, cy , will change the slope of
the consumption curve.

4.6.2 The saving function

Since consumption is a function of disposable income, so is saving, which
is defined as the part of disposable income that is not spent for
consumption during the period in which income is received. Since both
disposable income y and consumption c are in real terms, i.e., in
commodity units, so is saving. Saving occurs every period and differs from
the everyday usage of the term ‘savings’ , which is usually a synonym for
wealth. Saving is also different from savings deposits in banks, which are
in nominal terms and part of money holdings.
Some economists believe that saving depends not only on disposable
income but also on the real interest rate, which are the return to saving if it
were to be invested and not merely held in money (i.e., currency and bank
deposits). Fact Sheet 4.2 shows for USA the relationship between saving
and the nominal interest rate on Treasury bills (the T-bill rate). While this
graph establishes a presumption that saving depends on the interest rate, it
does not strictly establish that saving depends on the real interest rate.

194
 Fact Sheet 4.2: Interest Rates and Saving in the USA, 1985-2008
This Fact Sheet illustrates the relationship between saving and the
interest rate using US data. Note that the appropriate relationship is
between the saving rate (i.e., saving divided by GDP) and the real
interest rate, not the nominal one.
The graph below shows that saving and the real interest rate have
followed similar fluctuations over the past 20 years. Each time interest
rates rose, as in the late 1980s, between 1995 and 2000, and the mid-
2000s, the saving rate also rose.

195
Extended Analysis Box 4.3: The Dependence of Consumption on Wealth,
Interest Rates and Consumer Confidence

The concept of lifetime wealth as the present discounted value of future

incomes
Assume that the individual consumer expects to receive labor income of
y tL in period t, t = 1,2,…,R , until retirement at the end of period R .
Further, he/she has a current amount of wealth a 1 in physical and
financial assets. He/she expects the interest rate to remain constant at r .
Then, according to the present discounted formula presented in Chapter
2, his/her expected lifetime wealth w e1 at the beginning of the current
period 1 equals his/her current assets plus the present discounted values
of expected future labor incomes. That is:
Wealth ≡ Current assets plus the present discounted value (PDV) labor
incomes over the lifetime.

where we 1 is the present (i.e., in period 1) real value of wealth, a1 is the

real value of physical and financial assets, yLe t is the expected real
labour income in period t, and R is the period at the end of which the
person retires. Note that the real value of physical and financial wealth
in terms of consumer goods is not definite, since their nominal values
tend to fluctuate. Future labour incomes are also not known with
certainty. Therefore, we 1 is more or less uncertain.
The dependence of consumption on wealth and the interest rate
If we consider the individual's consumption over his/her lifetime, we

196
would have to take account of the possibilities of substitution between
present and future consumption and also replace current disposable
income by his/her disposable lifetime wealth.13
The cost of one unit of current consumption is unity. The present
discounted cost of one unit of consumption next period is 1/(1 + r). The
higher the interest rate, the cheaper a unit of future consumption
becomes, so that it becomes tempting for the consumer to substitute
more of future consumption for current consumption.14 Doing so for a
given level of budgeted expenditures on current and future consumption
requires cutting down on current consumption and increasing current
saving to invest it at the higher interest rate and increase future
consumption.
Clearly, changes in the level of wealth also influence consumption: an
increase in the consumer's lifetime wealth increases his/her consumption
expenditures. Therefore, stock market booms tend to increase
consumption, by increasing wealth, while stock market crises decrease
consumption. Similarly, expected increases in future labor incomes
increase the expected lifetime wealth and increase consumption.
Therefore, the analytical determinants of current consumption are often
specified as current disposable income, the PDV of disposable wealth
(i.e., net of taxes) and the interest rate r, so that the consumption
function becomes:

c = c (PDV of disposable wealth, r),

where an increase in disposable wealth or a fall in the interest rate

increases consumption. Note that current disposable income is a part,
though usually a small part, of the PDV of disposable wealth. This
consumption function can be restated as:

c = c (y d, PDV of future disposable wealth, r),

where y d is disposable income in the current period. This function adds

to the standard consumption function we have used for building our
model by explicitly adding future disposable wealth and r as two
additional determinants of consumption.
Empirical evidence on the dependence of consumption expenditures and
saving on interest rates
While expenditures on highly durable consumer goods that are financed
by borrowing, such as the purchase of cars and houses financed by
mortgages, are sensitive to changes in the interest rate, empirical

197
evidence indicates that expenditures on non-durable consumer goods are
not significantly sensitive to interest rate increases and decreases. Since
the latter are the dominant part of total consumption expenditures,
empirical estimates of the direct effect of interest rate changes on
aggregate consumption expenditures (and, therefore, also on aggregate
saving) do not show this effect to be very significant. We will therefore
ignore it and drop the interest rate from the consumption function.
However, note that a rise (fall) in the interest rate usually lowers (raises)
bond and equity prices, and is therefore associated with a decrease
(increase) in wealth. A decrease (increase) in wealth lowers (raises)
consumption expenditures. However, this indirect effect of changes in
the interest rate, through wealth on consumption, is usually left to the
analysis of the impact of wealth on consumption rather than directly of
the interest rate on consumption. This effect is encompassed under the
next heading.
The impact of changes in wealth on consumption
While changes in lifetime wealth clearly have an impact on current
consumption expenditures, the standard macroeconomic models
continue to specify the consumption function as:

c = c(yd)

For this simplified/standard consumption function, changes in

consumption induced by changes in wealth are treated as exogenous
shifts in autonomous consumption and/or the marginal propensity to
consume. To illustrate, a stock market boom that increases equity prices
and raises shareholders’ wealth (as happened in the late 1990s in the
western economies) will increase their consumption. This increase will
be captured through an exogenous shift of the consumption function and
an upward shift in the consumption curve from c to c’ in Figure 4.2b.
Conversely, a crash in the prices of shares in the stock market (as
happened in 2000 and 2001) will shift the consumption curve
downwards.
To reiterate, the standard macroeconomic model operates with the
consumption function c = c (yd ), and treats the impact of changes in
wealth as exogenous shifts in this function. The latter is a significant
effect and will be studied again in Chapter 9.
The impact of changes in consumer confidence on consumption
Our broad statement of the consumption function is:

c = c(yd , PDV of future labour incomes, present value of financial and

198
 physical assets, r)

Since the PDV of future labour incomes and even the present real value
of financial and physical assets are usually quite uncertain, expectations
of their value have a very significant impact on consumption
expenditures. These expectations are represented by economists by the
concept of ‘consumer confidence’ in the current and future state of the
economy – especially that on jobs and incomes. Over the short term, the
influence of changes in consumer confidence on consumption tends to
be even more significant than of changes in interest rates. Therefore, for
the explanation of changes in consumption expenditures, economic
analysts often have in mind a simple consumption function of the form:

c = c(yd, consumer confidence, stock market performance)

4.6.3 Investment expenditures

Investment was defined earlier as the amount of expenditures on currently
produced commodities, which firms intentionally incur for the purpose of
maintaining or adding to the capital stock. The real value of investment
expenditures i depends on the real (rather than the nominal15) rate of
interest, which is the real cost of borrowing funds to finance investment
projects. Note that by the Fisher equation on interest rates (see Chapter 2),
the real interest rate r equals the nominal/market interest rate R less the
expected inflation rate

(π e). Hence, we have for our model:

i = i(r),

199
which is simplified to the linear relationship:

i = i0 – ir r. (13)

Since the real interest rate r is the real cost of borrowing funds for
investment, an increase in the real interest rate will reduce investment, so
that ir ≥ 0. The names given to the parameters of the investment equation
are:

i 0 = autonomous investment

i r =the interest sensitivity of investment16.

The component (–irr) of investment demand is the decrease in

investment induced by rises in the interest rate and is often referred to as
‘induced investment’.
Figure 4.3 plots the above investment equation and shows a downward
sloping investment curve i. The intercept of this curve on the horizontal
axis is at i0 and represents the level of investment if r = 0. The slope of the
curve equals (–i r): as the interest rate rises, the level of investment falls.
Fact Sheet 4.3 plots the relationship between investment and the interest
rate on Treasury bills. Two curves are shown, one between investment and
the nominal interest rate and other between investment and the real interest
rate. Economic theory posits a relationship between investment and the
real interest rate.
Fact Sheet 4.3 : Interest Rates and Investment in the USA, 1960-2008
This Fact Sheet shows the relationship between investment and the
real interest rate using US data.
The interest rate, being the cost of borrowing, plays an important role,
along with real GDP as an indicator of expected sales, in investment
decisions. The graph below shows that sharp drops in the real interest
rate are often followed by periods of increased investment. Similarly,
increases in the real interest rate, as in the late 1970s and late 1990s,
sooner or later led to decreases in investment.

200
 Extended Analysis Box 4.4: A more Realistic Investment Function
Investment expenditures by firms also depend on their desired capital
stock, which depends on the projected demand for commodities, which
equals y e in equilibrium. Hence, the main determinants of investment
are the interest rate r and expected future sales y e. Therefore, the more
realistic investment function is:

i = i(r, y e)

For simplification, basic macroeconomic analysis simplifies this

function by dropping expected future demand, so that investment
becomes a function only of the rate of interest. That is, the standard
macroeconomic analysis is based on the simplified investment function i
= i(r), so that shifts in the expectations of future demand are represented
as exogenous shifts in the standard investment function.
The impact of business confidence on investment
In the short term, changes in the expectations of future sales do cause
quite significant shifts in investment expenditures. Economic analysts
try to capture these expectations in the concept of ‘business confidence’ ,
so that, for practical analysis, the investment function that is often used
is:

i = i(r, business confidence).

Business confidence shows a great deal of variation over the business

cycle and depends on political as well as economic events. These
variations make investment the most volatile component of aggregate
demand. Their role is examined under the concept of business
confidence in Chapters 8 and 16.

201
4.6.4 Government expenditures and tax revenues
Government revenues are collected from a variety of taxes, of which two
of the most significant ones are income taxes and corporate taxes. For
given tax rates, tax revenues rise with the level of national income, so that
the main determinant of government revenues is national income y (not
disposable income yd). Further, our concern is with the net tax payments,
which exclude government transfers to the public in the form of business
subsidies, income support (or welfare) payments,
unemployment/employment insurance benefits, etc. The need for such
payments decreases as incomes and employment rise. Further, as incomes
rise, more of the taxpayers end up in higher tax brackets under a
progressive tax system. Hence, net tax payments tend to rise with the level
of national income, so that the net tax revenue function can be written as:

t = t(y), (14)

where t is real net government revenues. Note that tax revenues have been
made a function of national income y, rather than merely of consumers’
disposable income yd. Assuming for simplification that the tax function is
a straight line, we have:

t = t 0 + tyy 0 ≤ ty ≤ 1, (15)

where:
t 0 = autonomous net tax revenues and
ty = marginal tax rate.
In the present context, autonomous net tax revenues are the amount
collected when income equals zero. These represent that component of
total tax revenues (or transfers) which does not change with changes in
income. Major examples of these are poll taxes (taxes per head) or
transfers to persons independent of their level of income, taxes on wealth
rather than income, taxes on house values, etc. Note that t 0 can be
negative, zero, or positive. It would be negative if the government has a
net transfer (i.e., transfers exceed total tax revenues) to the public when
national income falls to zero. This is a likely possibility in the short term,
especially under a progressive tax system and income support programs in
which the government ensures that each person has at least a minimum
amount (private income plus income transfers from the government) each

202
year.
The government expenditures on goods and services also depend upon a
variety of elements. Here, the simplest assumption is that their amount is a
matter of political choice by the government, rather than of the variables
such as income, unemployment, and interest rates in the model. Their
essentially political determination is reflected in the assumption that real
government expenditures g are exogenously determined at g 0, so that,

g = g0. (16)

Note that g 0 are real government expenditures on goods and services

(which would include civil service salaries) but do not include transfers
from the government to the public. The latter are also not included in our
corresponding variable t, which is tax revenues net of transfers. Hence, t
represents government net ‘withdrawals’ of real income/output from the
private economy, while g represents government ‘additions’ to the
expenditures on commodities.
Figure 4.4 shows the net revenue function by the line marked t. Tax
revenues increase proportionately with income, so that this line is upward
sloping. This line assumes that the tax rate is held constant at the rate ty.
An increase (decrease) in autonomous taxes t0 will shift the t curve up
(down) in a parallel manner, while an increase (decrease) in the tax rate ty
will make this curve steeper (flatter). Government expenditures on goods
and services do not increase with income, so that their line, marked g, is
horizontal. The budget is balanced at the level of income yBB (balanced
budget income). If the actual level of income is below yBB, there will be a
fiscal deficit. If the actual level of income is above yBB, there will be a
fiscal surplus.
Some economists argue that the tax rates should be set at a level such as
to yield a balanced budget at the full-employment level of income. If this
is done, a fall in income below the full-employment level would
automatically generate a fiscal deficit and an increase in income above the
full-employment level would automatically generate a fiscal surplus. The
former (latter) will increase (decrease) expenditures in the economy, so
that the government budget provides, to some extent, an automatic
stabilizing effect on national expenditures and aggregate demand in the
economy. Such a stabilizing effect on aggregate demand occurs whenever
the tax rates and government expenditures are held constant – while
national income fluctuates. Therefore, the budget is one of the automatic

203
stabilizers in the economy.

Fact Sheet 4.4 : USA Fiscal Deficit, 1962-2008

This Fact Sheet illustrates movements in fiscal outlays, revenues and

surpluses/deficits by using data for the USA. Before the 1970s, US
governments generally followed the philosphy that, in peactime,
governments should aim at a balanced budget. However, since about
1975, the USA has usually run a fiscal deficit, observable as periods
when total government outlays surpass total revenues. Notable
exceptions include the late 1990s when the economic boom of the period
raised incomes and the resulting rise in tax revenues created a fiscal
surplus. The recession that followed caused a sharp drop in incomes and
tax revenues, with the result that the government budget fell into a
deficit.
After the financial crisis starting in 2007 turned into an economic
recession, tax revenues fell while the US government pursued a fiscal
stimulus policy, which created unprecedented peacetime budget deficits.

4.6.5 The commodity market and the price level

204
Note that all the variables specified for the commodity market have been
assumed to be in real terms and independent of the aggregate price level.
In one-period analysis, this assumption means that consumers and firms
are free of price illusion in their consumption and investment decisions. It
is a convenient one in macroeconomic models and is often made.17

4.7 The Commodity Market Model: The IS

Equation/Curve
The above arguments collectively specify our model for the commodity
sector of the closed economy as:

y=e equilibrium condition

This model can be solved for y in terms of r, the policy variables and the
various parameters in either of two ways:
1. One method is to use the equilibrium condition s + t = i + g, and
substitute the functions for each of the variables in this equation.
2. The alternative method is to use the equilibrium condition y = e and
substitute for e its various components. We follow this procedure here.
The steps are:

To rearrange this equation to keep only y on the left side of the equation,
take all the terms in y to the left side. This gives:

y – cyy +cytyy = c 0 – cyt 0 + i 0 – i rr + g

which implies that:

205
so that,

This equation is derived from the equilibrium condition y = e, so that it is

another form of the equilibrium relationship. It specifies the locus of
values of y and r at which equilibrium exists in the commodity market and
is known as the IS (investment-saving) equation.18
The only two variables in this IS equation are y and r. It expresses
income y as a function of the interest rate r, exogenous parameters and the
policy variables/parameters (g, t0, and ty) of the model. It implies that the
equilibrium level of real income is a function of the rate of interest and not
of the price level.19 Note that y in this equation represents the demand for
commodities as a function of the interest rate, not of the price level, so that
this equation is not the aggregate demand equation. Further, in the above
equation, y does not represent the output – that is, the quantity produced –
of commodities in the economy; it only represents the demand for
commodities as a function of the interest rate.
The writing of the IS equation can be simplified to:

where:

As explained below, the name given to a is the ‘autonomous investment

multiplier ’. Note that all the terms in it are parameters, so that the value of
a is treated as given (unless one of its parameters shifts), and does not
change as the variables in the model vary.
The impact of investment fluctuations on income: a partial investment
multiplier
We can use the IS equation to derive the impact of changes in the
exogenous parameters and policy variables on income. Of all these, a
common concern is with the impact of shifts in investment on income
since investment is one of the most volatile components of expenditures
and government expenditures are an important policy variable. The impact
on income of a change in autonomous investment is captured by the
following multiplier, which is derived from the IS equation above. For this

206
derivation, note that all the parameters and variables on the right-hand side
of the IS equation are to be held constant except for i 0. Let the change in i
0 be Δi 0. Then, Δy = αΔi 0, which implies that Δy /Δi 0 = α. We use the
conventional symbolic term to replace Δy/Δi 0 when the change is
very small, ‘with all other things being held constant’. Therefore,

This multiplier is the autonomous investment multiplier , which is defined

as the increase in national income for a (very small) unit increase in
autonomous investment. For 0 < cy < 1, 0 < ty < 1, this multiplier is
positive and greater one.
For cy = 0.8, ty = 0.2, Box 4.3 derives the value of this multiplier as 2.78.
But suppose that the MPC was higher at 0.9, so that the parameter values
were cy = 0.9 and ty = 0.2. For these, the multiplier equals 3.57. Hence,
increases in the MPC increase the investment multiplier. Increases in the
tax rate will lower it.
Box 4.2: The Mechanism of the Investment Multiplier: An Illustration
Assume that the MPC (cy ) equals 0.8 and the marginal tax rate (ty) is
0.2. Now suppose that firms decide to increase their investment
expenditures by $1. They spend it on commodities, so that expenditures
rise by $1.
The firm that receives this dollar has higher sales revenue by $1, so
that it pays out an extra dollar in incomes (wage and profits). With a tax
rate of 0.20, disposable income only rises by 80 cents, so that the
recipients of the additional dollar spend 64 cents (80 cents times MPC of
0.8) out of it and add 16 cents to their savings.
The 64 cents spent on commodities are received by firms who
eventually pay them in wages and profits, thereby raising incomes by 64
cents. The recipients of this amount pay taxes of 12.8 cents (= 64 cents
times 0.2), so that their disposable income rises by 51.2 cents. They
spend 40.96 cents (= 51.2 cents times 0.8) as an increase in consumption
and save the rest. These 40.96 cents further increase the sales revenue of
firms, which will induce an increase in incomes of 40.96 cents and an
increase in disposable incomes of 32.77 cents (= 40.96 time 0.8). This
amount will cause a further increase in expenditures of 26.22 cents (=
32.77 times 0.8) – and so on.
This process continues until the last increase in expenditures becomes
zero. Eventually, the increase in expenditures becomes:

207
 $1 + $0.64 + $0.41 + $0.26 +…

The total increase is given by the formula:

Two points about this multiplier are worth noting. One is that it depends
on the MPC and the marginal tax rate. For most plausible values of these
parameters, the multiplier will be greater than unity. The second point is
that the process of expenditure increases and income payments is going
to take quite some time. It would be several quarters at least, so that the
increase in expenditures during the first quarter — called the impact
effect — is substantially less than the full increase — called the full or
long-run effect. The IS analysis, which does not incorporate real time
and is based on analytical time, only uses the full/long-run increase.
The impact of fiscal policy on income: partial fiscal policy multipliers
The impact of an increase in government expenditures on income is:

so that an increase in government expenditures increases income. Note that

the government expenditures multiplier equals the autonomous investment
one.
Correspondingly, the autonomous net tax multiplier —which is the effect
of a very small change in t0, with all other parameters and variables held
constant — is specified by:

Hence, an increase in autonomous net taxes decreases income. Further,

since cy < 1, this reduction in income is less than that induced by a
correspondence decrease in government expenditures.
Box 4.3: The Impact of a Balanced Budget: The Balanced Budget
Multiplier
Suppose that the government always balances its budget — that is, g = t
— so that any increase in its expenditures is matched by an equal
increase in its tax collection. The latter could be done through an
increase in autonomous tax revenues or in the tax rate. If this is done,
the increase in the government expenditures tends to increase income

208
 while the increase in tax revenues tends to decrease it.
We illustrate the balanced budget multiplier by examining the case
when the tax rate (ty) is zero, so that all net tax revenues are autonomous
ones. In this limiting case (with ty = 0), the balanced budget multiplier is
unity, as shown by:

To illustrate this value of the balanced budget multiplier, we use our

earlier numerical example with cy = 0.8, in combination with the
additional assumption that t y = 0. Then, the government expenditures
multiplier will be 5 and the net tax multiplier will be 4. The increase in
expenditures would be unity.
If the economy has a positive tax rate, which is always the case in
modern economies, the balanced budget multiplier remains at unity
whether the increase in tax revenues is due to a rise in autonomous taxes
or in the tax rate, or some combination of the two. Note that to achieve
this effect, the increase in tax revenues must equal the increase in
expenditures, so that the budget remains balanced.20

A word of caution on IS multipliers

Note that while we have derived the investment and fiscal effects on
income through the use of the partial multipliers based on the IS equation,
they are deceptive and even erroneous for a monetary economy. The
reason for presenting them is partly by way of an exercise in the
illustration of multipliers and partly out of an attempt at uniformity of
treatment with most textbooks in macroeconomics. The above multipliers
are based on the commodity market, which is a very limited part of the
economy, and ignore the monetary sector and the role of monetary policy,
which are also needed for determining aggregate demand in the economy.
Further, they ignore the supply side of the economy. The appropriate
multipliers are those derived after a general analysis of all the sectors of
the whole economy.21 Therefore, the usefulness of the above multipliers
for the commodity sector alone is only in studying the determinants of the
position and shifts of the IS curve.
The IS curve
The IS equation specifies the income and interest rate combinations that
ensure equilibrium in the commodity market. It is plotted in Figure 4.5 as

209
the IS curve. The IS curve represents the set of points in the (y, r) diagram
at which equilibrium exists in the commodity market. It has a negative
slope since r has a negative coefficient in the investment function. The
intuitive explanation for the negative slope of the IS curve is obtained by
comparing points a and b on the curve marked IS0. Starting from the point
a, an increase in income y, along the horizontal axis, increases both
consumption and saving, as well as increasing tax revenues. Equilibrium
requires that investment must also increase by the combined increases in
saving and tax revenues. This increase in investment has to be induced by
a decline in the interest rate. Hence, if equilibrium is to be maintained in
the expenditure sector, an increase in income must be accompanied by the
appropriate decline in the interest rate.
Shifts in the IS curve versus movements along it
Movements along the IS0 curve (e.g., from the point a to the point b in
Figure 4.5) occur if there is an exogenous change in the interest rate r,
while shifts in the curve occur if any of the parameters (e.g., c0, cy, i0, ir ,
etc.) or the policy variables (g, t0, ty) change. Some of these shifts will be
parallel ones while others even change the slope of the IS curve.22 As
shown in Figure 4.5, an increase in autonomous investment or government
expenditures will shift the IS curve in a parallel manner to the right from
IS0 to IS1, while decreases in them will shift it, again in a parallel manner,
to the left to IS2.
Intuitively, at the given interest rate r 0 and the IS curve IS0, induced
investment will be constant. An increase in autonomous investment will
raise the level ofinvestment, which through the multiplier will increase
expenditures from y d0 to y d2. Hence, at the given interest rate r0, the
economy will go from the point a on IS0 to the point c on IS1.

The IS curve shifts to the right from IS0 to IS1 due to:

210
• An increase in private expenditures, either from an increase in
autonomous consumption c 0 or in autonomous investment i 0 and/or
• An expansionary fiscal policy, which increases government expenditures
g or reduces autonomous taxes t 0.
Conversely, the IS curve shifts to the left from IS1 to IS2 due to decreases
in private expenditures or a contractionary fiscal policy.
Extended Analysis Box 4.5: The Slope of the IS Curve
The IS equation was derived above as:

y = α(c 0 + 10 + g 0 – c yt – i rr )

where α is the autonomous investment multiplier. It specifies the

increase in y for a unit change in autonomous investment. Note that the
increase in y caused by a unit increase in government expenditures also
equals α.
First, note that a mathematician plotting this equation would put the
variable on the left side, which is y, on the vertical axis and the variable
on the right side, which is r, on the horizontal axis. However,
economists have traditionally put y on the horizontal axis and r on the
vertical one: y is the more important variable of interest and the IS
equation reflects this by placing it on the left side. Second, note that the
slope of the IS curve is , which is the inverse of Third, we have
to use the IS equation to first derive and then take its inverse to
derive
From the IS equation:

Therefore, the slope of the IS curve is specified by:

Since both α and i r are positive, is negative, so that the IS curve

slopes downwards.
Changes in cy, ty, and ir change the multiplier value and the slope of
the IS curve. An increase in the marginal propensity to consume c y or a
decrease in the tax rate t y will decrease the numerator in the preceding
equation, thereby reducing the slope of the IS curve and making it

211
 flatter. In terms of the multiplier a, which equals 1/(1 – c y(1 – ty)), the
increase in the marginal propensity to consume cy and/or the decrease in
the tax rate t y will increase a. Since an increase in α reduces the slope
[1/(αi r)] of the IS curve, this would flatten the IS curve. Students can
check on this effect by plotting the IS curve for the alternative sets of
values: (i) c y = 0.8, t y = 0.2 and i r = 0.2, (ii) c y = 0.9, t y = 0.2 and i r
= 0.2, (c) c y = 0.8, t y = 0.1 and i r = 0.2.

4.8 Conclusions
• In the commodity market of the closed economy, the main components
of expenditures are consumption, investment, and government
expenditures. The main uses of income are for consumption, tax
payments, and saving. Their main determinants are income or disposable
income, wealth, and tax rates. The main determinant of investment is the
interest rate.
• The equilibrium condition for the commodity market is the equality of
national saving and investment. This condition is not an identity. This
equilibrium condition yields the IS equation and curve.
• Increases in government expenditures or decreases in tax rates raise real
aggregate demand (at the preexisting price level).
• The division of an increase in aggregate demand between increases in the
price level and the quantity produced of commodities also requires the
specification of the aggregate supply curve.
• The balanced budget multiplier for equal increases in government
expenditures and tax revenues is unity.
The nature and properties of the aggregate demand curve are analyzed in
the next chapter.

KEY CONCEPTS

Consumption function
Investment function
Government expenditures, transfers, and spending
Investment multiplier
Fiscal multipliers
The IS relationship/curve, and

212
Aggregate demand.

SUMMARY OF CRITICAL CONCLUSIONS

•The IS curve shows the set of combinations of y and r at which the

commodity market is in equilibrium.
• A change in investment or fiscal deficit changes the equilibrium value of
y by a multiple. In practice, this process takes several quarters. The value
of the multiplier depends on the consumption, investment, and tax
revenue functions.
• An increase in investment or government expenditures increases
expenditures on commodities at the existing interest rate and shifts the IS
curve to the right.
• An increases in taxes, for government held constant, shifts the IS curve to
the right.

REVIEW AND DISCUSSION QUESTIONS

1.Define (private) saving. What is the equilibrium condition for the closed
economy involving private saving and investment?
2. What is unintended investment? Suppose this investment is positive
Discuss firms’ responses to this and the impact on national income.
3. Is unintended nvestment positive, negative, or zero in equilibrium in the
commodity market? Explain.
4. Define national saving. What is the equilibrium condition for the closed
economy between national saving and investment?
5. Suppose the stock market booms and causes consumers to increase their
marginal propensity to consume. Show its effect on the IS curve. Is
there a change in the slope of the IS curve?
6. Suppose the stock market collapses and causes firms to substantially
decrease their autonomous investment. Show its effect on the IS curve.
Is there a change in the slope of the IS curve?
7. Suppose that a financial crisis reduces the credit that households can
borrow to finance their purchases of consumer durables and that firms
can borrow to finance their inventories and meet short-term obligations.
Analyze their effects on consumption, investment, and the IS curve. Can
fiscal policy reduce or eliminate this impact of the crisis and how?

213
8. Explain the impact of an increase in autonomous taxes on national
income and consumption. Does it shift the IS curve? If so, how (parallel
or not) does it shift the IS curve?
9. Suppose the government increases the income support (in the form of
transfers per head) provided to seniors in the economy while holding its
autonomous (poll) taxes and the tax rate constant. What effects would it
have on: government expenditures, the net revenue function,
consumption, and the IS curve?
10. What is meant by the government expenditures multiplier. Discuss the
process by which income/expenditures would increase by a multiple
following an increase of $1 in these expenditures.
11. If there is a balanced budget increase in government expenditures and
autonomous taxes, does the IS curve shift and, if so, in which direction?
Explain.
12. What is the ‘national income identity’? Does it imply an identity
between investment and saving? If so, in what sense of these terms?
Explain.
13. Does the IS curve incorporate the national income identity and/or the
identity between investment and national saving? Explain.
14. Suppose that the government wants to increase its expenditures g and
has the options of financing it by higher taxes, bond issues or increases
in the monetary base. Explain the impact of each of these on the IS
curve and show diagrammatically the effect on the IS curve.
15. What is meant by the ‘balanced budget multiplier’? Explain why it
takes this value.

ADVANCED AND TECHNICAL QUESTIONS

National income accounts data

You are given the following information for the year 2000 on an
economy. All amounts are in current dollars.
Goods and services produced by factories 10,000
Of which, goods and services sold during the year 8,000
Labor costs 7,000
Interest paid 1,600
Taxes paid by factories to the government 100
Goods purchased by some factories from others for inputs 900
Value of the factories as of January 1, 2000 150,000

214
Value of the factories as of December 31, 2000 150,000
Goods carried over in the economy on January 1, 2000 200,000
T1. Using the above national income account data, calculate GDP,
consumption, disposable income, government deficit/surplus, private
saving, and the average propensities to consume and to save.
Basic (commodity sector) Model for the closed economy in Chapter
4:

T2. Answer the following for the Basic (commodity sector) Model of this
chapter.
(a) What is the equation that describes the IS curve? [Do not use the IS
equation formula for this answer. Use step-by-step calculations. Start by
calculating disposable income. Then, use the equilibrium condition y =
e for your further steps.] Draw this curve in the IS diagram.
(b) What is the IS curve if g becomes 600? Draw this curve in the IS
diagram.
(c) Are the IS curves in (a) and (b) parallel?
(d) What is the IS curve if ir becomes 60? Draw this curve in the IS
diagram.
(e) Are the IS curves in (a) and (d) parallel?
Revised commodity sector model for the closed economy:

T3. Answer the following for the revised commodity sector model:
i. What is the equation that describes the IS curve? [Do not use the IS
equation formula for this answer. Use step-by-step calculations. Start by
calculating disposable income. Then, use the equilibrium condition y =
e for your further steps.]
ii. If the interest rate were to be set at 0.02 by the central bank, what
would be the equilibrium amounts of consumption, investment, and net
tax revenues?
iii. If the interest rate were to be set at 0.02 by the central bank, what
happens to the level of equilibrium investment if g increases to 6000?

215
 What is the reduction in private investment because of the increasing?

1 Total output is assumed to be fully distributed to the factors of

production within the same period. Some of this is done through
contractual income in the form of wages, rents and interest payments. The
remainder is considered the profits of the firms that produced the output.
In equilibrium, the amount of profits would be ‘normal profits’, which is
the amount that firms need to earn to produce the existing output and pay
their owners for the capital investment in the firm.
2Note that the word ‘saving’ is in its singular form and represents the
amount of currently produced commodities that are not paid in taxes and
are not consumed (i.e., destroyed) in the current period. This saving is in
real terms. It differs from the word ‘savings’ which is often used to refer
to‘wealth’ or to ‘savings deposits’ in banks. Wealth can be in a physical
form, such as houses, or a financial form, such as bonds and stocks.
Savings deposits are a form of money and are included in M2 (see Chapter
2) and are in nominal terms.
3One way of thinking about it would be to envisage each firm selling its
output and then deliberately buying some of it back for its investment
purposes.
4Examples of such positive shocks are those that increase consumption,
investment, government expenditure, or reductions in tax rate.
5At this point, nominal national saving Sn equals nominal investment I.
6At these points, S n < I .
7At these points, S n > I .
8The alternative assertion in nominal terms is I ≡ S n.
9In some societies, objectives such as sumptuous children’s marriages or
building a house can also be significant determinants of saving.
10‘Net taxes paid’ equals total government expenditures less transfer
payments by the government to the public. Since transfer payments are not
being deducted from income in calculating disposable income, disposable
income includes transfer payments.
11 The normal upper limit on both MPC (= c ) and APC (= c/y ) is
y
specified as unity (i.e., one). It does apply over long periods. The reason
for this is that one cannot indefinitely consume more than one’s income.
However, over short periods, APC does sometimes exceed unity, with the
excess of consumption over income financed out of wealth.
12Some economists assume that consumption depends negatively on the

216
rate of interest, so that the consumption equation should have included
another term (–c rr ) on the right side. The empirical dependence of
consumption on the rate of interest, for the usual range of interest rates in
the developed economies, is doubtful — and definitely much more
doubtful than the dependence of investment on the rate of interest. Further,
the impact of such a term within the macroeconomic model is identical
with that of the dependence of investment on the rate of interest.
Consequently, we have not included the rate of interest in the consumption
function; by implication, the rate of interest will also not be in the saving
function.
13Disposable lifetime wealth is defined as the present discounted value of
disposable income over the consumer's lifetime.
14The intuition behind this argument is as follows. Suppose a commodity
costs $1 this year and $1 next year. If the commodity is not bought this
year, the dollar saved will be invested at the interest rate r and yield (1 + r
) at the beginning of next year. As r rises, more of the commodity can be
bought next year for each dollar's worth of its consumption during this
year. That is, its future consumption has become cheaper relative to its
current consumption, so that current consumption will be reduced. The
amount thus saved will be invested at the higher interest rate, so as to buy
a larger amount of next year's consumption. Hence, as the interest rate
rises, future consumption is substituted for (is increased) at the expense of
current consumption. This is an application of the substitution effect
studied in microeconomics.
15Suppose a firm borrows funds in the loan market at the nominal interest
rate R (10%) when the expected inflation rate is π e (8%)and uses these
funds to purchase commodities for investment. The value of the
commodities purchased is expected to rise by the end of the year by π e, so
that, to the firm undertaking the investment, the real cost of obtaining the
funds is only R – π e (2%). It is this real cost, not the nominal one, which
the firm compares with the (real) productivity of investment.
16If saving depends positively on the rate of interest, this can be easily
incorporated into the model by redefining (–i r) to measure the decrease in
investment and the decrease in consumption (or less the increase in saving)
induced by the marginal increase in the rate of interest.
17It has to be modified if the analysis is over more than one period or if the
economy is an open one.
18It was given this name since it was originally derived for a model
without a government sector. Hence, it represented the equilibrium

217
condition i = s .
19This general statement remains valid even if consumption depends upon
income and the rate of interest, and even if investment depends upon the
interest rate and income. For this variation of the IS model, use the general
linear forms of the consumption and investment functions as: c = c(y, r) =
c 0 + cyy – c rr and i = i(r, y) = i 0 – i rr + iyy , with the assumption that c
y(1 – ty ) + iy < 1.
20For c = 0.8, t = 0.2, the government expenditures multiplier equals
y y
2.78, while the autonomous tax multiplier has the value –2.22, so that if
the increase in government expenditures of $1 was financed by an
autonomous tax increase of the same amount, expenditures/incomes would
rise by 0.56. In this example, if both government expenditures and
autonomous taxes had increased by $Δg, expenditures would rise by 0.56
times Δg. Therefore, the increase in income is only 0.56, which is less than
unity. While this result seems to contradict the balanced budget multiplier
proposition, it does not really do so. In this calculation, tax revenues rise
for two reasons: the initial increase of $1 in autonomous taxes—equal to
the increase in government expenditures—and the increase in tax revenues
because of the constant tax rates on a higher income. The total increase in
tax revenues becomes greater than the increase in government
expenditures, so that the budget changes are not balanced, and the
balanced budget multiplier does not apply.
21For the open economy, the appropriate multipliers are those from the
open economy model presented in Chapter 5.
22The slope of the IS curve depends on i , c , and t , so that the IS curve
r y y
shifts in a parallel manner if these do not change but one of the other
parameters or policy variables changes. Changes in one of these three
parameters will change the slope of the IS curve.

218
 CHAPTER 5
Aggregate Demand in the Open
Economy Under an Interest Rate
Target: IS-IRT Analysis

This chapter extends the commodity market analysis of the closed

economy in the last chapter to a small open economy with a
flexible exchange rate .
This chapter further uses the assumption that the monetary
policy being pursued sets an interest rate target, and the money
supply is adjusted to support this target. Under these assumptions,
it shows that the open economy aggregate demand curve has a
negative/downward slope.
Under interest rate targeting, expansionary (contractionary)
monetary and fiscal policies are both effective in increasing
(decreasing) aggregate demand.
This chapter also discusses the demand for money and derives
the money supply appropriate to the real interest rate target set by
the central bank.

The preceding chapter covered the analysis of the commodity market

through the development of the IS equation and curve for the closed
economy. This chapter revises the closed economy treatment of the
commodity market to accommodate trade in commodities and assets with
other countries. It will be assumed that the home economy, also referred to
as‘our economy‘, is small relative to‘the rest of the world so that changes
in it do not induce changes in the values of the economic variables in other
countries. Hence, the variables of the rest of the world can be taken as
exogenous to the domestic economy in question. The resulting model is
known as the small open economy macroeconomic model.
The IS curve analysis for both the closed and the open economy shows
that the critical variable determining the aggregate demand for
commodities is the interest rate. The determination of the interest rate
depends on the policy pursued by the central bank. This policy can be

219
classified broadly under the headings of monetary targeting or interest rate
targeting. The former leads to the exogenous specification of the money
supply, and leads to the IS-LM analysis presented in the next chapter. The
latter leads to the exogenous specification of the interest rate, and leads to
what we label as the IS-IRT analysis. The approach used in this chapter is
based on the assumption of interest rate targeting by the central bank.
This chapter combines the IS curve analysis for the commodity market
with a real interest rate target set by the central bank to derive aggregate
demand in economy. It also examines the influence of the rest of the world
on the domestic aggregate demand for commodities, as well as the impact
of monetary and fiscal policies in the open economy.
The definitions of the symbols used in this chapter are as given in the
preceding chapters.

5.1 The Number of Goods and Markets in the

Open Economy
The open economy macroeconomic model has five goods: commodities,
money, bonds, labor, and foreign exchange, and the corresponding five
markets. Of these, in equilibrium, the results obtained from the analysis of
four markets also provide the equilibrium solution for the fifth market, so
that the explicit analysis of one market can be omitted. We will follow the
conventional macroeconomic analysis to omit the bond market from
explicit analysis, so that the analysis will proceed with the specification of
the remaining four markets for commodities, money, labor, and foreign
exchange. We will start with the last one in the next section and then
proceed to analysis of the commodity market for the open economy. This
will be followed by the analysis of monetary policy as pursued by most
central banks and the determination of the aggregate demand for
commodities.

5.2 The Foreign Exchange Sector of the Open

Economy and the Balance of Payments,
Review
Chapter 3 defined the balance of payments (B) as the net inflows of
foreign exchange — that is, as the difference between the country's
inflows of funds (from commodity exports, capital imports/inflows and the

220
net inflows of interest, and dividend payments and unilateral transfers) and
its outflows of funds for commodity imports and capital exports. These net
inflows result in corresponding changes in the foreign exchange held in the
domestic country. The usual assumption for macroeconomic analysis is
that the foreign exchange authorities — usually the central bank — desire
equilibrium in the balance of payments and achieve it, so that B = FR = 0.
As Chapter 3 showed, the balance of payments B is specified as:
B = (inflows of foreign exchange from net exports of commodities)
+(inflows of foreign exchange due to net imports of capital)
+(inflows of foreign exchange from net interest payments to us)
+ (inflows of foreign exchange from net transfers to us).
Hence, in equilibrium,

where:
FR = foreign exchange reserves,
B = balance of payments, in nominal terms,
Xc = value of exports of commodities (goods and services, including
physical capital goods),
Xk = value of ( financial, not physical) capital exports,

Zc = value of imports of commodities (goods and services, including

physical capital goods),
Zk = value of ( financial, not physical) capital imports,
NR = net interest and dividend inflows,
net unilateral transfers (gifts and donations) to the domestic
NT =
economy from abroad.
Chapter 3 had also defined the real exchange rate as:

where:
ρ = nominal exchange rate, defined as the amount of foreign currency
(say, £s) required to purchase a unit, i.e., a dollar, of the domestic currency
($), so that the dimensions of ρ are £ per $ [not $ per £],
ρ r = real exchange rate, defined as the amount of foreign commodities
required to purchase a unit of the domestic commodity,
P = domestic price level, and PF = foreign price level.
We will henceforth simplify by assuming, unless otherwise explicitly
specified, that both exports and imports have price elasticities greater than

221
unity (see Chapters 12 and 13 for the reasons for this assumption). In this
case, our volume of exports (xc) and their value Xc decrease, as the real
exchange rate ρ r (= ρ P/PF) rises (which will occur if the domestic price
level P rises, ρ rises1 and/or the foreign price level PF falls) since our
commodities would become relatively more expensive compared with
foreign goods. However, the nominal value of our exports also increases
with an increase in foreign income yF. Similarly, under the assumption that
our imports have an elasticity greater than one, both the volume and the
nominal value of our imports Z c increase if ρ r rises (since a unit of our
commodities will buy more of foreign ones so that foreign commodities
become relatively less expensive) or if domestic income y rises.
Capital flows depend upon a range offactors, ofwhich the rates ofreturn
on domestic and foreign assets are likely to be the most important ones. It
will be assumed that capital exports Xk decrease, and capital imports Zk
increase, as the yield on domestic bonds rises or as that on foreign bonds
falls.2 The domestic yield equals the domestic interest rate R, while the
yield on foreign bonds to domestic investors is the sum of the foreign
interest rate RF and the expected appreciation of the domestic currency ρ ′
′e.
Putting these ideas formally into Equation (1), the equilibrium condition
for the foreign exchange market becomes:

where:
y = domestic national income,
yF = foreign national income,
R = domestic nominal interest rate,
RF = foreign nominal interest rate, and
(if positive) expected appreciation of the exchange
ρ ′′e = rate; (if negative) expected depreciation of

NR (net interest inflows) and NT (net inflows of transfers) tend to be

largely exogenous. To the extent that they are not exogenous, their
dependence on incomes and interest rates can be captured in either
commodity or capital flows.
We have assumed in the introduction to this chapter that the domestic
economy is small relative to the rest of the world: that is, the variables PF,
yF, and RF would not be affected by changes in our exports and imports of

222
commodities and capital. Therefore, they are exogenous variables, whose
values are given for our economy. Omitting the variables yF and RF from
Equation (2) since they are exogenous and omitting ρ ′′e for simplification,
we are left with the equation:

By substituting ρ P/PF for ρ r in Equation (3), we get:

This equation — designated as the BP (equilibrium) equation — now

becomes an element of the open economy macroeconomic model. Solving
it for the equilibrium value ρ* of the nominal exchange rate ρ yields ρ * =
f(P/Pv, y, R; other exogenous variables and parameters in the BP
equation).
For a country with a flexible exchange rate, the usual assumption for
convenience in macroeconomic analysis is that the exchange rate will
always be the equilibrium one, thereby ensuring continuous balance of
payments equilibrium for the given values of R and y. Therefore, under
flexible exchange rates, ρ * from Equation (3) is substituted for ρ in the
open economy IS equation to derive the final form of the IS equation and
curve. We will return to this point later in the next section.

5.2.1 Net interest payments and net transfer payments

NR is the net inflow of interest and dividend payments from ownership of
bonds, stocks, and physical capital in other countries. NT is the net inflow
of transfer payments from gifts and remittances. It is commonly assumed
in international economics that NR and NT are relatively insignificant
compared with net exports and net capital flows for most countries, so that
they are usually omitted from further analysis. However, this is not true of
certain countries. For example, Saudi Arabia has a lot of foreign workers
who send remittances back to their families, often in Pakistan and India. In
these cases, the balance of payments analysis should explicitly take into
account NR and NT and not omit them from balance of payments analysis.

5.2.2 Relationship between nominal and real interest rates

The nominal interest rate R is the appropriate interest rate for the analysis
of the balance of payments since foreign investors are interested in the
nominal, not the real, interest rate. However, the real interest rate r is
appropriate for the analysis of investment and the commodity market. As

223
Chapter 2 discussed, in prefect capital markets, the relationship between
the nominal and the real interest rate is given by the Fisher equation on
interest rates, which is:

where πe is the expected inflation rate. The approximate form of the Fisher
equation is:

This chapter assumes that πe = 0, so that R = r.

5.3 The Commodity Market of the Open

Economy
Now consider the specification of the commodity market of the open
economy. Let y stand for domestic output and national income, while e
stands for real expenditures on domestically produced commodities. The
symbol for consumption expenditures in Chapters 4 was c. The open
economy has two types of consumer goods, domestically produced and
imported ones. We now define c as the real value of all consumption
expenditures. Therefore, the consumption expenditures on domestic
commodities will equal total consumption expenditures c minus the
expenditures in real terms on imported commodities. For simplification,
we assume that investment is wholly out of domestic commodities. In
addition, foreigners buy our commodities through our exports, so that
foreigners‘ real expenditures on our exports, which equal xc, have to be
added to derive the expenditures e on domestic output.
Hence, the real expenditures e on the commodities produced in the
domestic economy, which constitute the aggregate sales revenue of all the
firms in the economy, are given by:

where all the variables are in real terms. Our exports are already in
domestic commodities and are in real terms. zc/ρ r( = (PF/ρ P)zc) are
expenditures in real terms on imported commodities. Why divide zc by ρ r?
Note that zc is the quantity of imported goods bought at foreign prices PF,
so that £PFzc is our expenditure in the foreign currency on imported goods.
This amount has to be converted into the domestic currency by dividing by

224
the nominal exchange rate ρ , so that $(PF/ρ )zc is the domestic/dollar price
of the imported goods. This nominal amount in dollars has to be deflated
into real terms at the domestic price level P to find the expenditures in
(domestic) real terms on the imported commodities. Therefore, the real
value of imported commodities in the above equation is not merely zc but
(PF/ρ P)zc, which equals zc/ρ r.3
In the aggregate over all domestic firms, firms‘ output equals their total
payments to the factors of production, which together constitute national
income. As in the closed economy analysis, national income can be spent
by domestic residents on consumption c (which includes the consumption
of both domestic and imported goods), saving 5, and payments of net taxes
t . Hence:

where:
c = real consumption expenditures,
s = real private saving, and
t = real net taxes paid (net of transfers).
For equilibrium in the commodity market, we need (as in the closed
economy models) the equality of national income (which constitutes the
domestic firms‘ cost of production of commodities) and national
expenditures (which equal their revenues from sales of commodities). That
is, the generic equilibrium condition for the open, as for the closed,
economy is:

In this equilibrium, firms‘ revenues and costs are equal, so that they would
maintain production at an unchanged level. As shown in Chapter 4 for the
closed economy, if e > y, firms‘ revenues would exceed their costs and
they would be tempted to expand production. If e < y, firms‘ revenues
would be less than their costs and they would attempt to reduce losses by
contracting production.
Hence, we equate e to y for equilibrium in the commodity market. Doing
so yields:

By cancelling out c, we get:

This equation represents one of the forms of the commodity market

equilibrium condition.

225
Fact Sheet 5.1: Components of Aggregate Demand for USA, Canada, and
Thailand, 2007
The following pie charts illustrate the breakdown of 2007 aggregate
demand (gross national product) at market prices for three countries
with quite different economies: USA is a large continental country with
a large economy, both Canada and Thailand are small economies but
Canada has a developed economy, and Thailand has a developing one.
Consumption expenditures (including those on imported goods) are the
largest item (59% or higher) for each country. Gross capital formation
(which is another name for gross (i.e., including replacement for
depreciation) investment) is the next largest item. While exports and
imports are large for each country, the pie charts show only net exports
(i.e., exports minus imports), which are a very smaller percentage of
GNP. In 2007, both USA and Canada had negative net exports, meaning
that they were net importers of goods and services. With a fast-growing,
export-led economy, Thailand allocates a larger proportion of its annual
production to net exports and investments than USA and Canada.
Government expenditures are relatively bigger for Canada than for USA
or Thailand because of Canada's greater provision of social services for
its residents.

5.3.1 The uses of private saving in the open economy

The commodity market equilibrium condition above can also be
rearranged to show the uses that can be made of private saving. For this
purpose, rewrite this condition as:

226
That is, private saving out of the currently produced output of commodities
can be used to pay for domestic investment, the government deficit or be
exported through positive net exports.
Uses of national saving
National saving sn is the sum of private saving s and government saving
(t – g), so that:

Hence, national saving can be used for domestic investment or net exports.
A fiscal surplus increases national saving, so that it increases domestic
investment and/or net exports, or both. A fiscal deficit decreases national
saving, so that it reduces the levels of domestic investment and/or net
exports, or both. The fiscal deficit and the foreign trade deficit are thus
closely connected.

5.3.2 Three gaps: saving, fiscal, and external

The preceding commodity market equilibrium condition can be rearranged
as:

where (i - s) is the (private) investment-saving gap, (g - t) is the fiscal gap,

and the last term [xc - zc/ρ r ] is the net export gap, also called the external
trade gap. Note that this equation expresses a relationship among the three
gaps, but without specifying a direction of causality — that is, the initial
change in which variable causes the other variables to change. However,
additional information on the exogeneity of any one of the gaps will insert
causality from an exogenous gap to the endogenous ones. For example, if
it is assumed that the government sets its deficit exogenously of the
economy, it would be better to rearrange the above equation as:

With (g - t) being given as exogenous, the exogenously created fiscal

deficit will induce either an investment- saving gap and/or an external gap.
To illustrate the implications of this relationship in the very unusual case
when i = 5, so that (i - 5) = 0,4 the fiscal deficit (i.e., g > t ) would imply a
balance of trade deficit of the same amount. Intuitively, a fiscal deficit
means that the government is buying ‘too much’ of commodities. This is
made possible through a net inflow of commodities from abroad.
Conversely, a fiscal surplus would induce positive net exports. In the

227
general case, in which saving does not necessarily equal investment, fiscal
deficits partly crowd out investment by making it less than saving and
partly reduce net exports (e.g., by inducing imports).
Therefore, the above equation raises the presumption that countries with
large fiscal deficits would also have balance of trade deficits while those
with large fiscal surpluses would have trade surpluses. Hence, the impact
of fiscal deficits and surpluses is not confined only to domestic aggregate
demand. They also have implications for the balance of payments and
therefore, for the foreign exchange markets and the exchange rate .
Fact Sheet 5.2: The uses of private saving in the USA, 1980-2008
This Fact Sheet uses US data to illustrate the allocation of private saving
to investment, the fiscal deficit and net exports. In an open economy,
equilibrium in the commodity market requires that private saving equal
the sum of investment, the budget deficit, and current account balance.
The following graph draws these variables as percentages of GDP. Both
private savings and investments were positive, though with a declining
trend, through the period, but investment was larger than saving.
Further, the US government ran budget deficits for most of the period.
The difference between investment and saving, plus the budget deficit
had to be financed by a net inflow of foreign commodities, so that the
current account had to be in deficit, as is clearly shown in the graph. To
cover this external account deficit, the USA was a net importer of capital
from other countries throughout the period.

National saving and the two-gaps equilibrium condition

In Chapter 4, national saving was designated as sn and was defined as:

Therefore, the three-gaps version of the commodity market equilibrium

condition for the open economy can be modified to a two-gaps one of the

228
form:

That is, if we save more as a nation than we invest at home, then we must
be exporting the excess national saving through positive net exports of
commodities. But if we save less as a nation than we invest at home, we
must have positive net imports of commodities.

5.3.3 Commodity market equilibrium and capital flows

We can combine the above condition for commodity market equilibrium
with the condition for equilibrium in the balance of payments, derived
earlier in this chapter. To do so, first convert the commodity market
equilibrium condition into nominal terms by multiplying by the price level
P. This gives the condition:

The nominal value of net exports is specified by [Xc(pr ) - Zc(pr, y)].

Under the balance of payments equilibrium condition, the nominal value of
net exports of commodities is related to nominal capital flows by the
equation:

so that combining the commodity market equilibrium condition and the

balance of payments equilibrium condition gives the uses of nominal
private saving. In nominal terms, S = I + nominal fiscal deficit + net
nominal exports of capital - nominal NR - nominal NT.
For countries for which we can justifiably assume that NR = NT = 0, this
equation says that nominal private saving can be used either for domestic
investment, the government deficit, or net capital exports .
Ownership of foreign assets and net external indebtedness
For nominal capital flows, the above condition for simultaneous
equilibrium in both the balance of payments and the commodity market
can be restated as:

Note that net nominal exports of capital = change in foreign exchange

assets (NFA) in the country.
For countries for which we can justifiably assume that NR = NT = 0, this
equation says that nominal national saving can be used either for domestic
investment or for net capital exports. The latter increase the country's

229
claims on the rest of the world. High national saving is, therefore, a way of
building the country's own physical capital and/or its ownership of foreign
assets at a faster rate. Conversely, low national saving hinders the growth
of the economy's own physical capital and/or increases foreigners‘
investments in our economy — thereby increasing the national
indebtedness to foreigners.
The ownership of foreign assets, which occurs through the accumulation
of net exports of capital, can take the following forms:
• ownership of foreign currencies,
• ownership of foreign bonds,
• ownership of foreign stocks, and
• ownership of foreign physical capital through foreign direct investment
(FDI).
Extended Analysis Box 5.1: Financing high levels of domestic investment
during the development stages and foreign exchange crises
Many countries, especially in their development stages, need to build up
their physical capital at a faster rate than their own saving permits. For
them, it is useful to turn the above equation to:
I - (Sn - net nominal outflow of interest payments5) = net nominal
imports of capital + nominal NT.
Note that past net inflows of capital have two effects:
1. If the past net inflows of capital increased investment in physical
capital at home, domestic output, current saving, and capital would be
higher. But if they were‘wasted‘ or used to finance consumption, they
would not have increased current output and saving.
2. The past net inflows of capital increase the current net outflows of
interest payments (on past capital imports). If net capital imports
continue over time, this outflow for interest payments becomes a
rising charge against both national saving and net exports.
If national saving increases because of higher output due to the
investment of past capital inflows, the increased outflow for interest
payments can be more easily accommodated and may not become a
problem. If national saving did not rise sufficiently because of the past
capital inflows, the increased outflow of interest payments reduces the
domestic funds available for increasing the domestic capital stock. This
scenario usually requires continuing net capital inflows and/or net
unilateral transfers. If these do not occur to a significant extent, domestic
investment will have to decrease.

230
Foreign exchange crises and debt forgiveness
Countries undergoing financial and exchange rate crises often have large
net capital outflows. In such cases, there could come about negative
investment — and a gradual reduction of the economy's physical stock
— which reduces their future income and saving capacity. One way out
of this trap is for the lenders to agree to some form of'debt forgiveness‘
— which is essentially the canceling of all or part of some foreign debts.
Lenders are reluctant to do so6 but it has been done to a limited extent
by some foreign lenders for the poorest countries with heavy foreign
indebtedness. From 1989 to 1997, US$33 billion of debt owed by 41
countries designated as the Highly Indebted Poor Countries (HIPC) was
forgiven.7
To see the potential of past capital inflows for causing financial crises,
rewrite the above equation as:
Net nominal outflow of interest payments + net nominal outflows of
capital
= Sn – I + nominal NT.
Countries sometimes build up a substantial external debt and, therefore,
become committed to high outflows of interest payments. When these
outflows as a percentage of domestic GDP become large, foreign
investors tend to become reluctant to invest further in the country and
domestic residents tend to choose foreign currencies or other assets, so
that there would occur a massive net capital outflow. In such
circumstances, the combination of the outflows (the left side of the
above equation) is likely to exceed the right side of the above equation.8
That is:
[Net nominal outflows of interest payments + net nominal outflows of
capital]
> [Sn - I + nominal NT].Extended Analysis Box 5.1: (Continued
We would also have:
[Net nominal outflows of interest payments + net nominal outflows of
capital]
> [net nominal exports + net nominal outflows of transfers].
A country faced with such a scenario would have to default on its
payments to foreigners, unless it follows policies to:
• Increase national saving. Since it is likely to be difficult to raise private

231
 saving rates, the government would have to cut its deficit drastically,
by cutting its expenditures, possibly on transfers to the public (e.g.,
subsidies to the poor and corporations) and on programs such as health
and education, and also by raising taxes.
• Reduce investment in domestic physical capital, which will hinder its
productive capacity in the future.
• Increase exports and decrease imports. This can require massive
depreciations/devaluations of its exchange rate, as well a decrease in
domestic aggregate demand through monetary and fiscal policies.
Each of these possibilities comes at some cost, which can be economic,
political, and social. Usually, all of the above policies have to be
followed. Often, their cost on the economy and the standard of living of
its citizens is quite heavy. Other policies — which are less painful in the
short term and are therefore often followed — are:
• Increasing capital inflows, e.g., by borrowing from international
organizations such as the International Monetary Fund, other
countries‘ governments or from private foreign sources.
• Increasing net inflows in the form of transfers (e.g., grants, which do
not require repayment or remittances by expatriates), or the waiver of
past debts (‘debt forgiveness’).
• Arranging rescheduling of payments (interest and capital) on the
existing debt, etc.

5.3.4 The open economy IS relationship

As specified above, the commodity market equilibrium condition for the
open economy is:

Our assumptions on the main variables of the commodity market (other

than exports and imports) remain unchanged from those on the closed
economy in Chapter 4. Therefore, the following equations from Chapter 4
are now taken to apply to the open economy.

where the symbols have the standardized definitions in Chapter 4. Note

that the above investment function assumes that the interest rate is a

232
function of the real interest rate r rather than the nominal/market one R.
The distinction between these and their relationship through the Fisher
equation was specified in Chapter 3. The reason for the assumption that
investment, which is in real terms in commodities, depends on the real
rather than the nominal interest rate is that the firm is interested in the real
cost, net of expected inflation, of the funds that it needs to borrow for its
investments.
Further, for the small open economy, assume that the export and import
functions are linear and are specified as in the following.
Export function

where xc is the real amount of exported commodities and xc0 is

autonomous exports — which are autonomous in the sense that their
amount is independent of ρ r and other endogenous variables, but can
depend on exogenous variables, such as yF. Exports decrease if the real
exchange rate ρ r rises, since this would make our exports more expensive
in foreign markets. This is reflected in the last term (—xcρ ρ r), where xcρ
is the amount of the fall in exports for a unit increase in the real exchange
rate.
Import function

where zc is the real amount of imported commodities and zc0 is

autonomous imports (i.e., the amount which is independent of disposable
income yd and the real exchange rate ρ r, but can depend on exogenous
variables). The marginal propensity to import out of yd is zcy, which is
positive. Imports increase if the real exchange rate ρ r rises/appreciates,
since this would make imported commodities less expensive. This is
reflected in the last term (zcρ ρ r), where zcρ is the amount of the rise in
imports for a unit increase in the real exchange rate.
The complete model of the commodity sector for the open economy
Gathering together the preceding equations, the open economy's
commodity sector has the equations:

233
 Substituting the equations for each of the variables on the right-hand side
of the equilibrium condition gives us a long equation that has y in some
terms on both side of the equation. If we rearrange this long equation to
bring y to the left-hand side and all other terms to the right-hand side, we
will get the following IS equation.
Box 5.1 presents its derivation.

This equation is the IS equation for the open economy. It makes y a

function of r and ρ r, where ρ r = ρ P/PF, so that replacing ρ r by ρ P/PF
makes y a function of r, ρ , P , and PF, in addition to policy and exogenous
variables and parameters. Under flexible exchange rates, the equilibrium
value ρ * of the nominal exchange rate is determined by the balance of
payments equilibrium condition (Equation (3) earlier in this chapter).
Replacing ρ by ρ * gives the final form of the IS equation. In this form of
the IS equation, y will be a function of P , r, the fiscal policy variables and
the exogenous foreign variables PF, RF, and yF, but ρ will no longer
appear as a separate determinant of y. Since this form is considerably more
cumbersome than Equation (15), our further analysis of the commodity
sector will be based on Equation (15), with the understanding that ρ r is
replaced by ρ P/PF and ρ is replaced by its equilibrium value from the
balance of payments equation equilibrium condition.
Equation (15) adds for the open economy the export and import
functions to the closed economy IS equation. In particular, [.] has two
terms involving the real exchange rate ρ r, with both preceded by a
negative sign, so that an appreciation of the exchange rate decreases
national income/expenditures y. The first term is xcρ ρ T . Its negative sign
occurs because as ρ r rises, exports fall, which decreases expenditures on
domestic commodities.9 The second term is Zcρ ρ r. Its negative sign

234
occurs because as ρ r rises, imports increase, so that the leakages from
domestic income (similar to those due to taxes paid) rise, thereby
decreasing the share of disposable income that can be spent on domestic
commodities.10
Replacing (.) by α in the above IS equation, the IS equation can be
written as:

where the symbol a has the meaning:

Box 5.1: Derivation of the Open Economy is Equation

The equilibrium condition specified for the commodity market is that:

Substituting the equations for each of the variables on the right-hand

side, we get:

where y d and t are given by:

Substituting for y d and t in the equilibrium equation, we get:

Expanding and collecting the y terms on the left-hand side gives:

Rewrite this equation as:

This equation can be restated as:

This form of the IS equation for the open economy shows the
commodity market's determination of y by the parameters and variables
on the right-hand side of the equation. Since r and ρ r are the variables
on the right side of the equation, and ρ r = ρ P/PF, this equation implies
that y depends on the variables r, ρ , and P/PF, in addition to the policy

235
 variables and parameters.
The equilibrium value of ρ* from the balance of payments is next
substituted in the above equation to get the final form of the IS equation.
The resulting equation would be quite complex and is omitted. The
general form of the final IS equation would be:

If P rises, our net exports fall, which will decrease expenditures on

domestic commodities. Hence, domestic income y will fall. Therefore,
for commodity market equilibrium, ∂y /∂P > 0.

The open economy fiscal and investment multipliers

The government expenditures and autonomous investment multipliers for
the preceding open economy IS equation are derived by the same
procedure as used in Chapter 4. Briefly, this procedure was to expand
Equation (5) to:

Now suppose that government expenditures g rise, with all other variables
held constant, so that the change in them is zero. Therefore:

which, for small changes in g, gives:

By similar reasoning when we only consider the effect of a small increase

in autonomous investment i0, we get:

Hence, α specifies the autonomous investment and government

expenditures multiplier for the open economy. As shown in Chapter 4 for
plausible values of the parameters, this multiplier has a positive value,
usually greater than one.
The multipliers for changes in exports and imports
Now suppose that autonomous exports x0 rise, with all other variables held
constant, so that the change in them is zero. Therefore:

which, for small changes in x 0, gives:

236
Therefore, the autonomous exports multiplier equals the fiscal
expenditures and autonomous investment multipliers above. If a has a
value greater than one, which is the usual case, an increase in exports
increases aggregate demand in the economy by more than the increase in
exports.11 Similar reasoning implies that the multiplier specifying changes
in aggregate demand due to changes in autonomous imports is:

Hence, an autonomous increase in exports increases y while an

autonomous increase in imports decreases y.

5.3.5 The open economy IS curve

Plotting the open economy IS equation in the IS diagram, with r on the
vertical axis and y on the horizontal one, gives the open economy IS curve,
which has a negative slope. Figure 5.1a shows such a downward sloping IS
curve. The reason for this negative slope is: an increase in income (from y0
to y1) increases saving and tax revenues, so that equilibrium requires an
increase in investment, which requires a decrease in the real interest rate
(from r0 to r1). Therefore, for equilibrium to be maintained in the
commodity market, an increase in y must be accompanied by a decrease in
r. Hence, commodity market equilibrium implies a negative relationship
between y and r.

5.3.6 Shifts in the open economy IS curve

The closed economy analysis of the IS curve implied that the IS curve
would shift to the right if there were increases in autonomous
consumption, autonomous investment, government expenditures, etc. The
reasons for this shift should be reviewed at this stage. The same arguments

237
also apply to the open economy IS curve. Such a shift to the right from IS0
to IS1 is shown in Figure 5.1b.
The open economy also has additional causes of shifts of the IS curve.
Among these, at any given interest rate, an increase in autonomous exports
increases expenditures on domestic commodities, so that the IS curve
shifts to the right. Conversely, for any given interest rate, an increase in
autonomous imports, for a given overall consumption, shifts expenditures
from domestic commodities to foreign ones, so that expenditures y on
domestic commodities fall. This shifts the IS curve to the left, as shown in
Figure 5.1b by the shift of the IS curve from IS0 to IS2.

Extended Analysis Box 5.2: The Mathematical Derivation of the Slope of

the IS Curve
Just as for the closed economy analysis of Chapter 4, the slope of the IS
curve is given by ∂r/∂y. To derive this, we note that ∂r/∂y is the inverse
of ∂y/∂r, whose value can be easily calculated from the above IS
equation12 as:

Hence, by using the inverse of this equation, the slope of the IS curve
for the open economy is given by:

Since α and ir are both positive, this slope is negative.

Changes in the slope of the IS curve as the economy becomes more open
A more open economy is one with higher imports and exports as a
percentage of its real GDP. The multiplier a becomes smaller as the
imports/GDP ratio increases. This implies that the slope of the IS curve
— which is given by ∂r/∂y = –1 /(αir ) — becomes greater in absolute
terms. Therefore, the IS curve will be steeper for a more open economy.
Hence, as the Canadian and American economies, as well as most
others, have become more open in the last several decades, their IS
curves have become steeper.
Comparing the smaller, more open Canadian economy with the larger,
less open American economy, the IS curve for Canada will be steeper
than for the USA. That is, a 10% decrease in the interest rate will have a
much smaller impact on aggregate demand in Canada than in the USA.

238
 Comparing the magnitudes of the fiscal multipliers for open economies
The fiscal multiplier depends on the propensity to import zcy. Since zcy
is in the denominator of the multiplier, the autonomous investment
multiplier a for a more open economy with a higher propensity to import
zcy is smaller in value than its counterpart for a less open economy.
Hence, a larger, more diversified economy — such as that
of'continental‘ countries such as the USA, European Union, China, and
India — is likely to have a larger multiplier than smaller, less diversified
economies such as that of Canada or Britain. As a result, an increase of
$1 in government expenditures will have a smaller impact on aggregate
demand in Canada than in the USA.
Further, as economies become relatively more open and increase the
ratio of their imports to GDP, their fiscal and autonomous investment
multipliers will decrease. This occurs since a larger part of the impetus
to aggregate demand from the fiscal expansion leaks away in the form of
higher imports and less is spent on domestic commodities.
Conversely, if foreign income yF rises, exports and total expenditures
on domestic commodities rise, which increases domestic income y.
Therefore, while the economy has smaller fiscal and investment
multipliers, its income rises because of higher exports.
The stability of aggregate demand in more open economies
With a smaller multiplier for a more open economy, national income
will be more stable for a given change in domestic investment in the
more open than in the less open economy. However, it will also be less
responsive to a given change in government expenditures. Therefore,
both the need for and the strength of stabilization policies will be less in
the more open economy. The larger the marginal propensity to import
zcy, the stronger will be this effect. As against this greater stability in
response to the domestic sources of disturbances, the fluctuations in
imports and exports due to shifts in foreign economies will have greater
impact on the domestic national income of a more open economy and
may require offsetting stabilization policies.

5.3.7 The impact of exchange rate changes on the IS curve

The movement along the open economy IS curve when the interest rate or
income changes
The relationship between the IS curve and the changes in the exchange rate

239
is quite complicated and needs to be carefully specified. The nominal
exchange rateρ is determined by the equilibrium in the foreign exchange
market. It rises if the domestic interest rate increases (because that induces
capital inflows), but falls if income y increases (because that induces an
increase in imports). Since the IS diagram has r and y on the axes, the
impact of any change in ρ because of prior changes in r or y will produce a
movement along (but not shift) the IS curve.13 The reasoning behind this
statement is illustrated by the following: starting from a given y0, an
increase in r and R will induce capital inflows, thereby creating a balance
of payments surplus, so that, in the flexible exchange rate case, ρ will
appreciate to restore equilibrium in the balance of payments. But this
appreciation will reduce net exports, which, through the multiplier, will
reduce income y in the economy.14 This reduction in y (due to the increase
in ρ caused by the increase in r and R) is captured through a sliding
movement upward to the new, higher, level of the interest rate on the
existing IS curve. Conversely, a decrease in r and R will induce capital
outflows, which will cause ρ to fall. This will increase exports, which will
cause y to increase. Hence, the decrease in r will mean a movement
downwards along the IS curve to a higher level of y. Similar analysis
applies to an initial change in y. Therefore, for the flexible exchange rate
case, if the change ρ is caused by an initial shift in r or y, the IS curve does
not shift, but there is a sliding movement along the IS curve to the new
equilibrium levels of r and y.
The shift in the open economy IS curve when prices change
Since ρ r =ρ P/PF, an increase in the price level P increases ρ r, which
makes domestic goods more expensive to foreigners and causes net
exports to fall. This will lead, because of the ensuing balance of payments
deficit, to a decline in p. This depreciation will moderate the increase in pr,
which will moderate the reduction in net exports in the IS equation but net
exports will remain below their initial level. The fall in net exports will
reduce national expenditures and income below their original levels.15
Since P is not on the vertical axis of the IS-IRT diagram, the reduction in y
due to the increase in P is captured through a leftward shift in the IS curve,
rather than through a sliding movement along it. That is, under flexible
exchange rates, an increase in P causes a leftward shift of the IS curve.
Conversely, a decrease in P causes a rightward shift of the IS curve.
Similar reasoning can be used to show that, under flexible exchange
rates, increases in net exports induced by exogenous increases in PF or in
yF will increase y for a given value of r and produce rightward shifts in the

240
IS curve.
In the more complicated case where net exports change because of
nominal exchange rate changes induced by changes in both the domestic
prices and domestic interest rates, there will be a shift in the IS curve, as
well as a movement along the curve.
The importance of the interest rate in determining national
income/expenditures
The preceding IS equation and curve show that one of the critical variables
determining the level of expenditures on domestic commodities is the
interest rate. The determination of the interest rate depends on the financial
sector of the economy and the policies followed by the central bank. On
the latter, the central bank's policies can be classified broadly under the
headings of money supply targeting or of interest rate targeting. The
former leads to the exogenous specification of the money supply. The
latter leads to the exogenous specification of the interest rate. The
approach used in the remainder of this chapter and this book, except for
Chapter 6, is based on the assumption of interest rate targeting by the
central bank, and differs from that in most other macroeconomics
textbooks, which assume money supply targeting. For completeness, the
latter is presented in Chapter 6.

5.4 The Formulation of Monetary Policy

In most modern and financially developed economies, the central bank has
been allocated the role of determining monetary policy. The two main
options of the central bank in the formulation of monetary policy are to:
1. determine the underlying nominal or real macroeconomic interest rate in
the economy; and
2. determine the supply of money (or of another monetary aggregate).
Box 5.2: Money Supply, the Stock of Money and Their Relationship to the
Interest Rate
Figure 5.2 shows the case of a money stock that is exogenously
determined by the central bank at a given level M0. In this case, the
money supply curve is a vertical line. The exogenously determined
money stock M0 determines the equilibrium interest rate r*0 along the
money demand curve. Suppose the following can be validly assumed.
• Stable money demand and supply functions (i.e., these functions do not

241
 shift).
• The existence only of the indirect transmission mechanism of money
supply changes on national income (and the absence of the direct
transmission mechanism).16
• An efficient money market — so that there are no lags in the impact of
changes in the monetary base on the interest rate.17
Under these assumptions, it does not matter whether the central bank
sets the interest rate or the monetary base supplied to the economy. If
these conditions are not met, then it does matter which policy is adopted.
In the real world, the economy's overall response occurs through three
sequential lags:
1. from the change in the monetary base to the change in the money
supply and thence to the change in the interest rate,
2. from the change in the interest rate to the change in investment, and
3. from the change in investment to the change in nominal national
expenditures and income.

Each of these lags is of significant duration. Our concern here is with the
central bank's choice between changing the monetary base versus
changing the interest rate. In general, there is a significant lag between
monetary base changes and its consequent interest rate changes.
Therefore, the impact of a change in monetary policy on the economy
would be faster if the central bank was to directly alter the interest rate
than if it was to initiate the change by altering the monetary base.
Consequently, most central banks tend to set interest rates and support
them through monetary base changes, rather than set the monetary base
and let the market determine the interest rate after some lag. An
additional reason for using the interest rate as the monetary policy
instrument becomes relevant if the money demand function is volatile
and unpredictable, which it has proved to be in many developed
economies for the past several decades. Therefore, the reasons for the
central bank's preference for setting the interest rate rather than
manipulating the money supply are:

242
 • lags in the impact of changes in the money supply on the interest rate,
• instability of the money demand curve, and
• instability of the multiplier from the monetary base to the money
supply, which makes the control of the money supply through the
monetary base unpredictable.
The reasons for the central bank to manipulate the money supply rather
than the interest rate are:
• Fragmented financial markets, several or all of which may not follow
the lead of the central bank's change in the interest rate.
• Inability of changes in the interest rate to affect investment and
consumer durable expenditures, because these are mainly financed out
of the investor's own saving or their black money hoards.
In financially developed economies without fragmented financial
markets and without significant amounts of black money, the central
bank is better able to change aggregate demand, and to do so faster, by
targeting the interest rate rather than the money supply. Therefore, the
central banks of the USA, the UK, and Canada, as well as of many other
countries, follow interest rate targeting. However, the standard
macroeconomic analysis of aggregate demand in many textbooks is
based on an exogenous money supply, so that it is based on the IS-LM
analysis presented in Chapter 6.

5.4.1 Monetary policy through interest rate targeting

As Chapter 2 pointed out, the economy has a variety of nominal interest
rates. The interest rate on a financial asset depends on many factors,
including the risk of capital loss in holding it and, if it is a bond, its term to
maturity. These interest rates tend to be related to each other, so that
economists focus on an underlying interest rate, changes in which trigger
corresponding changes in all other interest rates. As an approximation, this
underlying interest rate is often taken to be the return on riskless bonds
with short maturity, such as the‘T-bill rate‘ on three-month Treasury bills
issued by the government. Financially developed economies normally also
possess an interest rate on overnight loans (i.e., loans from late one day to
the next day) among its major financial institutions, usually its commercial
banks. In Canada and Britain, this rate is called the overnight lending rate,
but is called the federal funds rate in the USA. Both the T-bill and the
overnight lending rates are market/nominal rates.
The central bank can bring about changes in the interest rates by trading

243
(i.e., open market operations) in the relevant financial market, which
induces changes in the demand or supply in these markets. For example,
its purchase of T-bills (i.e., Treasury bills, which are short-term
government bonds) will push up the demand for them, increase their price
and lower the interest yield on them. Conversely, its sale of T-bills will
reduce the demand for them, reduce their price and raise the interest yield
on them. Similarly, if the central bank wants to manipulate the overnight
loan rate, it can change the demand for or supply of loanable funds in the
market for overnight loans, and thereby alter the overnight loan rate.
In financially developed economies, financial markets are closely related
and commercial banks are a major player in them, so that a change in any
one interest rate — either in the cost of obtaining funds from the central
bank (i.e., in the bank/discount rate) or in the overnight loans market (i.e.,
in the overnight loan rate) or in the yield on their assets (e.g., in the T-bill
rate) — tends to trigger corresponding changes in the interest rates in all
other markets. The central bank can therefore choose the interest rate that
it wants to directly manipulate, while leaving it to the markets to adjust
other interest rates in a more or less synchronous manner.
In some countries, the central bank is willing to lend to domestic
commercial banks for short periods. The interest rate on such loans is
called the bank rate (in Canada and the UK) or the discount rate (in the
USA). Since the central bank is the lender in this market, it sets the interest
rate that it charges.
In Canada, the Bank of Canada sets a range for the overnight loan rate,
as well as specifying that the bank rate would be the maximum of the
range it sets for the overnight loan rate. In the USA, the Federal Reserve
System ensures its desired federal funds rate by trading in the market for
overnight loans.
The value of the interest rate desired by the central bank is called its
target rate. While central banks actually set nominal interest rates, their
objective in doing so is to set the real interest rate. Our analysis will
therefore assume that the target interest rate is the real one.

5.4.2 Rules versus discretion in setting the target interest

rate
The central bank can determine its choice of the target rate by:
• Its evaluation of the current and emerging economic conditions or on
some other basis without a pre-set rule. Such a policy is called a
discretionary policy.

244
• A real interest rate rule. Examples of interest rate rules are:
i. Pursue a simple rule for determining the target real interest rate. The
simplest form of such a rule occurs if the central bank maintains the
target real interest rate at a constant level r0 (while retaining the right to
vary it when it so wishes).18 According to this rule:

Note that a simple policy rule does not have a feedback from the
performance of the economy to automatically change the interest rate
target rT.
ii. Pursue a feedback rule for determining the target interest rate rT. A
feedback rule is one in which the feedback from the state of the economy
(e.g., its inflation rate or unemployment, etc.) is allowed to change the
target interest rate. An example of a feedback rule is provided by the
popular Taylor rule. According to this rule, the central bank sets the real
interest rate according to:

where rT is the real interest rate target, y is real output, yf is full-

employment output, π is the actual inflation rate, nT is the inflation rate
desired by the central bank and the subscript t refers to period t . πT is
called the target inflation rate. Similarly, yf is the target output level. (yt
— yf) is (minus of) the output gap. Under this rule, the central bank would
increase its target interest rate if actual output (or the demand for it) was
too high or if inflation was too high, relative to their long run or desired
levels. The Taylor rule is a feedback rule according to which changes in
the actual performance of the economy change the interest rate target:
under this rule, the target interest rate is automatically decreased if output
is below its full-employment level and raised if inflation becomes greater
than the target rate.
This chapter restricts the analysis of interest rate targeting and its impact
on the economy to the case of the simple policy rule specified in Equation
(18), which is that the central bank exogenously determines the real
interest rate that it sets for the economy. That is, our assumption will be
that r = rT0 , where the superscript T stands for the‘central bank target’.

5.4.3 Diagrammatic depiction of the interest rate target

The assumption made above on monetary policy is that the central bank

245
successfully targets/sets the economy's real interest rate r at r 0. That is,
under this simple interest rate targeting:

Plotting this interest rate in the (r, y) space of the IS diagram, we have a
horizontal line at the target real interest rate. This is shown in Figure 5.3a
by the ‘interest rate target’ curve labeled IRT.19

5.5 The Determination of Aggregate Demand

under Interest Rate Targeting
The equation for the aggregate demand for commodities — known as the
aggregate demand (AD) equation — is obtained jointly from the IS
equation for the commodity market and the monetary policy equation
determining the interest rate. Our model now becomes:
The IS equation :

where the symbol a stands for the (autonomous investment) multiplier and
has the meaning:

The monetary policy equation:

Substitution of Equation (22) into (21) yields the AD equation, where y

has been replaced by yd to emphasize that this equation specifies the
aggregate demand for commodities. This AD equation is:

246
In this equation, the real exchange rate ρ r equals ρ P/PF, so that aggregate
demand is a function of the price level, just as in a demand equation in
microeconomic analysis. Further, aggregate demand yd is likely to be
negatively related to the price level P. This AD equation differs from the
IS equation since the interest rate has been replaced by the exogenous
target interest rate set by the central bank, so that in the AD equation, yd is
not a function of the interest rate.

5.5.1 Diagrammatic derivation of the AD curve

Aggregate demand in the economy is given by the intersection of the IS
and IRT curves. This determination is shown in Figure 5.3a. The
intersection of the IS curve and the given interest rate determines the level
of the aggregate demand for domestic commodities. This level is shown as
yd. Further, at the given interest rate, a rightward shift of the IS curve
because of increases in investment, government expenditures, exports, and
the other reasons mentioned above will increase aggregate demand. A cut
by the central bank in the real interest rate will also increase aggregate
demand.
We concluded above that, under flexible exchange rates, a decrease in P
causes a rightward shift of the IS curve. This shift in the IS curve, for a
given target interest rate, increases aggregate demand, as shown in Figure
5.3a by the shift of the IS curve from IS0 to IS1, with a consequent
increase in demand from yd to yd. Conversely, an increase in P causes a
leftward shift of the IS curve, which decreases aggregate demand.
Therefore, the aggregate demand curve AD in Figure 5.3b is downward
sloping.
The reason for the downward slope of the AD curve under interest rate
targeting in the open economy is that an increase in the domestic price
level makes domestic commodities more expensive relative to foreign
ones, so that foreign commodities are substituted for domestic ones in
consumption. Consequently, the demand for domestic commodities falls.
Therefore, as P rises, yd decreases. Conversely, if P falls, yd increases.
This is shown by the downward sloping AD curve in Figure 5.3b.
Note that the open economy IS and AD curves incorporate within them
the effects of changes in y and r on the nominal exchange rate ρ and
through it on exports and imports. Note also that an increase in P (or
decrease in PF) shifts the IS curve to the left while a decrease in P (or
increase in PF) shifts it to the right. However, since P is on the vertical axis

247
of the AD diagram, the changes in P do not shift the AD curve.

5.5.2 The downward slope of the AD curve in the open

economy under an exogenous interest rate target: a
reiteration
The open economy demand curve is downward sloping. To establish this
for the open economy, suppose that the domestic price level rises, while
the foreign price level and the nominal exchange rate held constant under
the ceteris paribus clause. The price level increase raises the real exchange
rate, which equals ρ P/Pf. This increase in the real exchange rate makes
domestic goods more expensive relative to foreign ones. The substitution
effect increases the demand for imported goods and decreases net
exports,20 which shifts the IS curve to the left, thereby decreasing
aggregate demand at the given IRT. Hence, a rise in P reduces yd.
Conversely, a fall in P would increase yd. Hence, the AD curve, which
plots yd against P, has a negative slope in the (P, y) Figure 5.3b.

5.6 The Policy Multipliers for the Open Economy

Aggregate Demand
The policy multipliers measure the impact of changes in the exogenous
policy variables on aggregate demand in the economy. The exogenous
policy variables in our model are the fiscal ones of g, t0, and ty, while the
monetary policy one is r0T. The more significant multipliers are the
investment ones, the fiscal ones, and the monetary policy (target interest
rate) ones. To derive these multipliers, we start with the simplified form of
the AD equation, which was:

where the symbol α stands for the (autonomous investment) multiplier and
has the meaning:

All of the elements of a are parameters, so that a can be treated as a

constant whenever one of the variables of the model changes. This AD
equation is almost the same as the open economy IS equation, so that its

248
investment and fiscal multipliers are identical to the ones derived from the
IS equation alone. The following subsections reiterate these briefly.

5.6.1 The impact of investment fluctuations on aggregate

demand: the investment multiplier
The most volatile component of expenditures is investment, so that we
want to capture the impact of shifts in it on aggregate demand. To do so,
note that if only autonomous investment changes:

Therefore, α also gives the value of the autonomous investment multiplier

for aggregate demand. This multiplier is identical with that derived for the
commodity market alone. Note that the usual value of a is greater than
unity.

5.6.2 The impact of fiscal policy on aggregate demand: the

fiscal multipliers
The impact ofchanges in government expenditure on aggregate demand is
specified by its multiplier. To derive this multiplier, assume that the
government increases its autonomous expenditures by g, while keeping
its other fiscal policy variables t 0 and t 1 constant. In this case:

Therefore, α is the value of the government expenditures multiplier for

aggregate demand. This multiplier is identical with that derived for the
open economy IS analysis alone. As the usual value of α is greater than
unity, an increase in government expenditures increases aggregate demand
by more than the increase in government expenditures.
Similarly, as shown in Chapter 4, the multiplier for the impact of an
increase in autonomous taxes is given by:

That is, an increase in taxes reduces aggregate demand and a decrease in

taxes increases aggregate demand.

5.6.3 The impact of monetary policy on aggregate demand:

the target rate multiplier

249
Under an interest rate target, the monetary policy multiplier for aggregate
demand is given by ∂yd/∂r, which is the change in aggregate demand for a
small increase in the interest rate. The value of this multiplier is given by:

Therefore, an increase in the target interest rate reduces aggregate demand

in the economy. This effect depends on the impact (—ir) of the increase in
the interest rate on investment and then on aggregate demand, so that the
target rate multiplier depends on the value of ir, which is the interest
sensitivity of investment. Conversely, a reduction in the target interest rate
increases aggregate demand in the economy.

5.6.4 The distribution of incomes, consumption patterns,

and the size of the multiplier
The distribution of income in the population affects the consumption
patterns, the average propensity to consume and the size of the multiplier.
In particular, the marginal propensity to consume imported goods reduces
the size of the multiplier. In developing countries, the rich usually have a
much higher marginal propensity to import than the lower income groups.
Therefore, if the initial increase in incomes accrues mainly to the rich, the
multiplier will be smaller than if it enters the economy as an addition to the
incomes of the poor. This point is relevant to the impact of government
expenditures and transfers on the economy. Assuming all other things
(including the saving propensity) being the same, income transfers to
lower income groups, whose consumption is usually more of domestic
goods and services than that of the rich, will have a higher multiplier and,
therefore, a larger impact on aggregate demand than if the transfers were
made to higher income groups. Further, the poor tend to have a higher
marginal propensity to consume than the rich, which would further
increase the size of the multiplier when the initial income goes mainly to
the poor.

5.6.5 The impact of exports on domestic aggregate demand

The preceding model of aggregate demand shows that an increase in
exports increases aggregate demand by a multiple. Conversely, a decrease
in exports decreases aggregate demand. Such changes in aggregate
demand tend to produce changes in the economy's output, at least for some
time, so that fluctuations in exports become a major source of the

250
transmission of business cycles across countries. This point is discussed
further in Chapter 16 on business cycles.

5.6.6 The lag in the impact of changes in the interest rate

target
Since investment depends negatively on the real interest rate, a reduction
in the target real interest rate increases investment, which increases
aggregate demand through the multiplier. In practice, this process takes
quite some time. First, following the reduction by the central bank in the
target rate, the financial markets act to reduce the various interest rates.
Second, these reductions in the different interest rates induce increases in
the related types of investment. Third, the multipliers related to the
increases in the different types of investment have to work. As shown in
Chapter 4, the multipliers take time to work through their various stages.
Therefore, it is likely that the full expansionary impact of an interest rate
reduction would take several quarters. Some estimates of this impact place
it at six or more quarters, though some part of the impact may start to
occur much earlier.

5.7 Diagrammatic Analysis of Fiscal and

Monetary Policies
For both the closed and the open economy, an expansionary fiscal policy
shifts the IS curve to the right. This is shown in Figure 5.4a by the shift of
the IS curve from ISo to ISi. At the given interest rate target rT, aggregate
demand increases from yd to yd. Conversely, a contractionary fiscal policy
shifts the IS curve to the left and, at the given interest target rate roT,
reduces aggregate demand.
As shown in Figure 5.4b, an expansionary monetary policy in the form
of a reduction in the interest rate target from rT0 to r T1, shifts the IRT line
downwards and increases aggregate demand from yd0 to yd1. Conversely,
a contractionary monetary policy in the form of an increase in the target
interest rate target from rT0 to rT1, shifts the IRT line upwards and
decreases aggregate demand from yd0 to yd1.

251
5.8 The IS, IRT Curves and the Determination of
Output: A Caveat
Note that the intersection of the IS and IRT curves determines the
aggregate demand for commodities and determines the AD function/curve
used in Chapter 1. Their intersection does not determine the aggregate
quantity of commodities that will be produced in the economy. This
requires knowledge of both the demand and the supply function/scurves of
commodities. Chapter 1 had shown this by incorporating both aggregate
demand and supply in the AD-AS diagram.
Because the intersection of the IS and IRT curves only determines the
demand for commodities but not output (i.e., the quantity produced), it is
strictly inappropriate to interpret y on the horizontal axis of the IS-IRT
diagram — as well as the AD–AS one — as the output (i.e., the quantity
produced) of commodities. The symbol y on the horizontal section of both
the IS–IRT and the AD–AS diagrams is to be interpreted as the‘aggregate
quantity of commodities , not as output (i.e., the quantity produced). This
usage means that the intersection of the IS and IRT specifies the quantity
demanded of commodities — not the output of the economy.

5.9 But What about the Monetary Sector and Its

Money Demand and Supply Functions?
We have so far analyzed the determination of the interest rate, inflation,
and output without any reference to the money market, which
encompasses the demand and supply of money and the equilibrium
between them. These are peripheral to the determination ofaggregate
demand under a successful interest rate targeting policy, since, as the
preceding analysis has shown, knowledge of their values is not needed for
determining aggregate demand.

252
 However, for the successful achievement of the interest rate target, the
central bank does need to know the money demand function and, thereby,
to know the amount of money supply that it has to ensure for the economy.
While the proper treatment of this topic is left to Chapter 6 on the money
and other financial markets, we present below a limited part of the relevant
analysis.

5.9.1 The demand for money

Theoretical and empirical studies show that the demand for real balances
can be taken to be specified as:
md = transactions demand for real balances + portfolio demand for real
balances. The transactions demand is the demand for money to finance
individuals‘ purchases/expenditures. Since it is used to finance
expenditures, which in equilibrium, equals national income, we assume
that the transactions demand equals m yy.
The portfolio demand for money is a component of the individual's
portfolio. If we designate the nominal value of this portfolio by FWand
assume that the portfolio includes only money and bonds (including
stocks), the portfolio demand for money is given by:
portfolio demand for real balances
= real value of financial wealth - real value of the demand for bonds.
The real demand for bonds bd will increase with their nominal return,
which is the nominal interest rate R. Therefore, bd = f(R). We simplify by
writing this demand function as bd = mRR, where mR > 0. Hence, the
portfolio demand for real (money) balances will equal (FW/P — mRR).
Therefore, the demand for real balances is given by:

where R is the nominal interest rate, which is related to the real one by the
Fisher equation,21 presented in Chapter 2. The preceding equation is
usually simplified in short-run macroeconomic models to:

where FW/P has been dropped from the equation. Although this
simplification is traditional and almost universal in macroeconomic
models, it does have some consequences for macroeconomic analysis
when the rate of change of the nominal value of financial wealth is
different from that of the commodity price level, as it usually is. This

253
situation often occurs since bond and stock markets usually go up by more
or less than commodity prices. Ignoring this issue so as to adhere to the
traditional treatment of the money demand function, the demand function
will be taken to be as specified by Eq. (26).
For the money demand function (26), the component of money demand
represented by myy is the transactions demand for money and the
component of money demand represented by (—mR R) or, more
accurately, by (FW/P — mRR), 22 is the speculative or portfolio demand
for money. Note that these designations are just meant to be simplified
ones for facilitating discussion of the reasons for money demand. The
component (—mR R) or (FW/P — mR R) can also be referred to as the
interest-sensitive component of money demand.
Figure 5.5 shows the demand curve for money by md. Note that the
nominal interest rate R is on the vertical axis and the demand for real
balances md is on the horizontal one. Since R is on one of the axes,
changes in R imply movements along the curve. An increase in R reduces
the amount of money demanded, so that there is a movement up along the
existing md curve: a rise in R from R0 to Ri means going from the point a
to the point b along this curve. However, income is not on one of the axes,
so that for a given rate of interest R0, an increase in income increases
transactions demand and shifts the md curve toward the right from m^J to
mi. Correspondingly, a decrease in income reduces transactions demand
and shifts the money demand curve left from md0 to md2.

5.9.2 Ensuring equilibrium in the money market under an

interest rate target and the determination of the money
supply
Equilibrium in the money market requires that nominal money demand

254
(Pmd) equal the nominal money supply MS. That is, under interest rate
targeting, equilibrium requires that the money supply be given by:

We had assumed earlier in this chapter that monetary policy sets the real
interest rate r. By the Fisher equation, R = r + πe. With r set by monetary
policy and ne specified by the public's expectations, the value of R can be
calculated. Plugging this value, as well as the value of yd from our earlier
IS analysis, into the above equation yields the nominal value of the money
demand and the required amount of the money supply. For successful
targeting of the interest rate, the central bank has to ensure the money
supply given by the preceding equation.

5.10 Managing Aggregate Demand in the Open

Economy: Monetary and Fiscal Policies under
Flexible Exchange Rates

5.10.1 The effectiveness of fiscal policy for the open

economy under interest rate targeting
Assume that the country in question has flexible exchange rates, with the
initial general equilibrium shown in Figure 5.6a by the point a. Now
suppose that an expansionary fiscal policy (i.e., an increase in government
expenditures or decrease in tax revenues, resulting in a fiscal deficit) is
pursued. Such a policy shifts the IS curve to the right from IS0 to ISi. This
raises income from y0 to y,.
The intuitive sequence of effects is as follows. The expansionary fiscal
policy increases income, which increases commodity imports, so that the
supply of domestic dollars increases in the foreign exchange market. To
restore equilibrium in the foreign exchange market, the exchange rate
depreciates so that domestic products become relatively cheaper than
foreign ones. This causes exports to rise and imports to fall, which further
increases aggregate demand. Hence, there are two effects of the fiscal
deficit:
1. A direct expansionary effect on aggregate demand and national income
through the fiscal multiplier.

255
2. An indirect expansionary effect on aggregate demand through the
increase in net exports, which results from the fall in the exchange rate
(which is itself due to the increase in imports resulting from the increase
in income in the first point).
Both of these effects are incorporated in the slope and shift of the IS curve.
Note that the fiscal expansion by shifting the IS curve to the right increases
aggregate demand. This is shown by the intersection of the new IS curve
IS1 with the IRT curve in Figure 5.6a. Note that under interest rate
targeting, the interest rate does not change since it has been exogenously
set by the central bank. With domestic and foreign interest rates
unchanged, there is no change in capital flows,23 so that the impact of
changes in capital flows on the exchange rate does not have to be taken
into account. Therefore, even under a high degree of capital mobility, the
final effect of the fiscal expansion is to increase aggregate demand. Hence,
in the flexible exchange rate case with IRT, fiscal policy is a powerful tool
for changing aggregate demand.

5.10.2 The effectiveness of monetary policy for the open

economy under interest rate targeting
Now consider the impact under flexible exchange rates of an expansionary
monetary policy, which in our context is a reduction in the target interest
rate. This is shown in Figure 5.6b by the downwards shift of the IRT curve
from IRT0 to IRT1. This policy pushes aggregate demand along the IS
curve from y0 to y1.
The intuitive sequence of effects is as follow. As the interest rate falls,
investment rises. The increase in domestic incomes (due to the rise in
investment) increases imports and decreases net exports, which reduces the
inflows of foreign exchange, so that the exchange rate falls. The lower
interest rate also reduces capital inflows, thereby contributing to the
depreciation of the exchange rate.24 The overall depreciation makes
domestic goods relatively cheaper compared with foreign ones, so that

256
under the assumption of elastic net exports, imports fall and exports rise,
which further increases domestic expenditures and income. All of these
effects are incorporated in the final shape of the IS curve.
Therefore, under flexible exchange rates and elastic net exports, the
effect of an expansionary monetary policy has two components. These are:
1. A direct expansionary effect on aggregate demand that occurs through
an increase in investment at the lower interest rate and increases
aggregate demand and income in the economy.
2. An indirect expansionary effect on aggregate demand brought about by
an increase in net exports. This increase is brought about by the fall in
the exchange rate, which itself occurs due to the rise in the imports of
commodities (which themselves result from the higher income in the
first point), as well as the fall in the net capital inflows due to the lower
domestic interest rate.
Both of these effects are incorporated in the slopes and shifts of the IS and
AD curves. Therefore, under flexible exchange rates, monetary policy is
effective in increasing aggregate demand.
Hence, both monetary and fiscal policies are effective in changing
aggregate demand, though their degree of effectiveness would differ. In
both cases, expansionary policies increase aggregate demand, while
contractionary policies decrease aggregate demand.

5.10.3 Limitations on the effectiveness of a policy of

interest rate targeting
An expansionary monetary policy does not have an impact on aggregate
demand if:
• If investment does not respond to changes in the interest rate, i.e., if di/dr
= o. This case can occur if aggregate demand has fallen below full
capacity production levels. For example, in a depreciation (such as that of
the 1930s) or a deep recession (2007–2008), firms have excess capacity,
so that they do not wish to increase their capital stock even if the cost of
borrowing funds for investment falls. Note that these are precisely the
conditions in which there is an urgent need to boost aggregate demand,
but monetary policy is unable to do so to a significant extent.
• If changes in the monetary base will not increase the money supply, the
central bank cannot bring about increases in the money supply. This can
occur if banks become reluctant to lend to the public or buy more bonds,
even when their reserves are increased, because the riskiness of bonds
and loans has increased substantially or banks have become very risk

257
 averse. Such a situation occurred in the USA during the credit
crises/crunch of 2007-2009 (see Chapter 16).
• Increases in the money supply do not lower interest rates, as in the
liquidity trap (discussed in Chapter 6), or if the central bank cannot
further reduce its interest rates, as for example if they have already been
reduced to virtually zero.
All three cases are very likely in a depression or a deep recession,
especially one due to a collapse of the financial system and markets.
However, these are precisely the conditions under which policy needs to
raise aggregate demand from its depressed level. During such conditions,
many economists are doubtful of the success of expansionary monetary
policies in increasing aggregate demand, and recommend the use of
expansionary fiscal policies.

5.11 Internal Versus External Balance

Assuming that the net exports have an elasticity greater than unity (see
Chapter 2 later for the analysis related to this elasticity), exchange rate
flexibility provides the economy with the very useful equilibrating
mechanism of changes in the nominal exchange rate brought about by
market forces for handling the balance of payments deficits/surpluses.
Together, exchange rate flexibility and domestic price flexibility eliminate
the need for monetary and fiscal policies to return the economy to general
equilibrium and full employment — as long as the authorities are willing
to give the economy enough time to make its own adjustments to
equilibrium and are willing to accommodate the possible reserve losses
during the adjustment period. If they do pursue appropriate monetary and
fiscal policies, the preceding analysis implies that expansionary monetary
and fiscal policies increase aggregate demand, while contractionary
monetary and fiscal policies decrease aggregate demand.
Internal balance is defined as the equality of domestic aggregate demand
and domestic full-employment output, while external balance is defined as
the equality of the demand and supply of foreign exchange. Flexible
exchange rates maintain the latter equality, thereby freeing monetary
policy to focus on maintaining the internal balance of the economy. By
comparison, fixed exchange rates compel the use of monetary policy,
especially in setting interest rates, to the maintenance of external balance,
so that monetary policy is then much less able to maintain internal balance.
Extended Analysis Box 5.3: The Mundell–Fleming Model of the Open
Economy

258
 The Mundell–Fleming model of the open economy is based on the
contributions of Robert Mundell and Marcus Fleming in the 1960s. It
presented the analyses of monetary and fiscal policies for the open
economy under the assumptions of (a) the central bank targets the
money supply, so that it is exogenous, (b) elastic net exports, and (c)
perfect capital mobility. Mundell and Fleming showed that, under
flexible exchange rates, monetary policy is effective in changing
aggregate demand while fiscal policy is not. Further, if it is assumed that
the economy has (d) a fixed price level but variable output in responding
to changes in aggregate demand, such changes will lead to
corresponding changes in real output.
Parts of the Mundell-Fleming model are not relevant to our analysis
under IRT because of their assumption that the central bank targets the
money supply. For this assumption of money supply targeting, Mundell
and Fleming had shown that for small open economies with flexible
exchange rates, monetary policy would be effective in changing
aggregate demand and output while fiscal policy would be ineffective.25
This result does not hold in our model: as shown above, both monetary
and fiscal policies are effective in changing aggregate demand.
Another distinctive part ofthe Mundell–Fleming model was its
assumption that the price level is constant in the short run. This
assumption implies a horizontal aggregate supply curve for commodities
(see the sticky price model in Chapter 8). Our analysis in Chapters 7 to 9
will allow the possibility of three cases: a vertical LRAS curve for the
long run, a positively sloped SRAS curve for the short-run equilibrium,
and a horizontal supply curve under deficient demand.
While the Mundell–Fleming model is no longer as important to
modern macroeconomic modeling as in earlier periods, it did make a
considerable contribution to the development of the open economy
analysis.

5.12 Conclusions
The adaptations of the closed economy model to the open economy
introduced additional complexity into the analysis of the commodity
market.
• A new good, foreign exchange, was brought into the macroeconomic
model. Its‘price‘ which is the foreign exchange rate, could be flexible
and determined by the supply and demand in the foreign exchange

259
 market, or managed by the central bank or the government or be
fixed/pegged by them.
• For the open economy, the analysis of the IS curve was modified to
include exports and imports of commodities.
• Under interest rate targeting by the central bank, expansionary monetary
and fiscal policies are both effective in increasing aggregate demand.
Conversely, contractionary monetary and fiscal policies are both
effective in decreasing aggregate demand.
• An aspect of the changes in the intercountry flows in recent decades has
been the increasing dominance of financial capital flows over the values
of the net exports of countries. The former are more volatile and more
susceptible to speculative forces than the latter. The dominance of the
magnitudes of capital flows over net commodity flows implies that,
under flexible exchange rates, the exchange rate movements will be
mainly determined, at least in the short term, by the interest rate
differences among countries.

KEY CONCEPTS

Small open economy

Balance of payments curve
Three gaps: 467, saving gap, fiscal gap,
and external gap
The open economy commodity market
equilibrium condition
The open economy IS curve
The interest rate as the operating
monetary policy target of the
central bank
Simple monetary policy rule
Feedback monetary policy
Taylor rule
IRT curve
Aggregate demand curve
and its determination by the IS
curve
and the IRT curve
Effectiveness of monetary and
fiscal policies in changing

260
 aggregate demand
Limitations on the success of
expansionary monetary
policies, and
Internal and external balance.

SUMMARY OF CRITICAL CONCLUSIONS

• A fiscal deficit crowds out some domestic investment and also induces a
balance of trade deficit.
• The open economy aggregate demand curve has a negative slope (if the
real exchange can differ from unity).
• Under interest rate targeting, the IS and the IRT curves determine
aggregate demand.
• Under interest rate targeting by the central bank, both monetary and
fiscal policies are effective in changing aggregate demand. Expansionary
(contractionary) monetary and fiscal policies increase (decrease)
aggregate demand.
• The aggregate demand (AD) curve neither shows the supply of
commodities nor determines the quantity of commodities that will be
produced or the price level at which they will be sold.

REVIEW AND DISCUSSION QUESTIONS

1. Define a‘small open economy’. Under this definition, is your country a

small open economy? Is the USA a small open economy? Explain your
answer.
2. What are the differences between the IS curves for the open and closed
economy cases with respect to changes in the nominal exchange rate?
3. Under flexible exchange rates, assuming that net exports are elastic,
what happens to the IS curve in each of the following cases:
(a) foreign output rises,
(b) domestic output rises, and
(c) foreign interest rate rises.
4. Under flexible exchange rates, assuming that net exports are elastic,
what happens to the IS curve in each of the following cases:
(a) the central bank raises the domestic interest rate,

261
 (b) foreign demand for our commodities falls, and
(c) our residents start taking much longer holidays abroad.
5. In a model with flexible exchange rates, what are the effects of an
increased (domestic) demand for foreign commodities on the IS and AD
curves?
6. “Given flexible exchange rates, elastic net exports and interest rate
targeting, an expansionary fiscal policy is likely to be quite ineffective
in increasing the level of aggregate demand.” Do you agree or disagree?
Present the relevant analysis.
7. “Given flexible exchange rates elastic net exports and interest rate
targeting, an expansionary monetary policy is likely to be ineffective in
increasing the level of aggregate demand.” Do you agree or disagree?
Present the relevant analysis.
8. Assume a small open economy with elastic net exports,; a flexible
exchange rate and interest rate targeting. Discuss the effects of each of
the following (one-at-a-time) on domestic aggregate demand:
(a) an increase in the foreign demand for the country's exports,
(b) an increase in government expenditures,; and
(c) a decrease in the central bank,s target real interest rate.
9. Assume a small open economy with elastic net exports, a flexible
exchange rate and interest rate targeting. Discuss the effects of each of
the following (one-at-a-time) on domestic aggregate demand:
(a) an increase in the world interest rates and
(b) an increase in‘world income’.
10. Suppose that a banking and financial crisis reduces credit to
households, which limits their purchases of consumer durables, and
firms, which reduces their investment. Analyze the impact of this
‘drying up’ of credit (also called a ‘credit crunch’) on aggregate demand
in the economy.
11. If a credit crisis in the USA (a‘very large economy’) reduces its
aggregate demand and output, discuss its effects on aggregate demand
in the countries (i) which export extensively to it and (ii) whose
financial institutions extensively borrow from and lend to financial
institutions in the USA.
12. [Optional, based on Chapters 4 and 5 together]. For the general form of
the macroeconomic model used in this and the preceding chapter, what
are the differences between the multiplier for the open economy and that
for the closed economy?

262
 ADVANCED AND TECHNICAL QUESTIONS

Basic Model for this chapter

[R is the nominal interest rate, r is the real one, and, by the Fisher
equation,

The central bank targets the domestic real interest rate, so that

r = rT = 0.04

Benchmark Fisher equation: R = r + πe

Assumption: πe = o
T1. For the Basic Model of this chapter, do the following questions:
(a) Derive the IS equation. [Do not use the IS equation formula for this
answer. Use step-by-step calculations. Start by calculating disposable
income. Noting that xc and zc depend on pr, calculate pr in terms of P
. Then use the equilibrium condition y = e for your further steps.]
(b) Given rT = o.o4, derive the AD equation.
(c) If exports rise by 100, what is the new aggregate demand equation?
(d) If imports rise by 100, what is new aggregate demand equation?
T2. For the given target real interest rate and assuming ne = o, use the
equality of money supply and money demand to derive the equilibrium
money supply. [Hint: the money supply should be specified in nominal
terms, so that, for equilibrium in the money market, it should equal P?

263
 md .]
Revised Model for This Chapter

The central bank sets the interest rate such that r = 0.02.
LR output: yf = 2000
T3.(a) Derive the real exchange rate for this economy.
(b) Calculate disposable income as a function of income.
(c) Derive net real exports (i.e., exports — imports in domestic real
terms) as a function of income. [Hint: imports in domestic real terms
equal zc/fr.]
(d) Derive the IS equation.
(e) Derive the aggregate demand for commodities Cyd)?
(f) Suppose that the short—run output in the economy was determined
only by aggregate demand, so that actual output equaled yd. Calculate
the budget deficit at this level of output.
(g) If the output was at the long run output level, specify government
expenditures, revenues, and the full–employment budget deficit.

1Remember that ñ falling (i.e., less £s per $) means that the domestic
currency — the dollar — has depreciated. This would make our
commodities cheaper to foreigners while making their commodities more
expensive to us. To illustrate, if the exchange rate falls to half, the price to
foreigners (in £s) of a $10 shirt will fall to half while the dollar price (to
us) of a £20 trousers will double. This will make our goods cheaper for
foreigners and increase our sales to foreigners (i.e., increase our exports),
while making foreign goods more expensive for us and decreasing our
imports.
2It is also assumed that the other endogenous variables such as incomes,
prices and exchange rates are not likely to significantly affect capital flows
from the viewpoint of a purely comparative static analysis.
3Some textbooks do not divide zc by ñr under the assumption that it

264
always equals unity. This is done under the implicit assumption that
Purchasing Power Parity (PPP) holds, which makes ñr equal to unity.
However, as shown in Chapters 3 and 13, PPP does not normally hold,
even over periods as long as 10 or 20 years. Therefore, we do not assume
that ñr = 1.
4 Note that there is no reason for this equality (i = s) to hold in the open or
the closed economy. In particular, it would not hold even in the closed
economy — let alone the open economy — if there are budget surpluses or
deficits, or if the commodity market is in disequilibrium. It almost never
holds in the open economy.
5 This equals –NR.
6 One reason for refusing to do so is based on the argument that foreign
aid was used by corrupt and/or inefficient governments with a pattern of
misusing aid. Forgiving past debts would cause such governments to
borrow more with the expectation that the debt will be forgiven, and/or
they will use the debt forgiveness as providing a fresh opportunity to
borrow again for similarly wasteful purposes.
7 Foreign Policy, November/December 2001, pp. 20–26. New borrowing
by these countries during the same periods was $41 billion.
8 Note that the past accumulation of foreign debt implies that the country
did not have enough national saving to finance its domestic investment.
9 Therefore, the IS curve will shift to the left and aggregate demand will
fall.
10 Therefore, the IS curve will shift to the left and aggregate demand will
fall.
11 If à has a value between zero and one, the increase in exports increases
aggregate demand in the economy by less than the increase in exports.
12 For this derivation, first expand the IS equation by opening the brackets
and rewriting it as:
y = αc0 – αcyt0 + αi0 – αirr +…,
where -αir is the coefficient of r. To consider the effects on y of an
exogenous change only in r, the parameters and any other variables on the
right-hand side of this equation are held constant, so that the change in
them is zero. Hence, this equation implies that y = -αir r. For small
changes in y and r, this transforms to ∂y/∂r = –αir .
13 The formal reason for this has been specified earlier in this chapter. It is
as follows. ρ P/PF is first substituted in the benchmark IS equation
reported in this chapter. Then, the equilibrium value of ρ ; from the

265
Balance of Payments equilibrium condition is substituted in the benchmark
IS equation, thereby eliminating ρ from the IS equation. The resulting form
of the IS equation will depend not only on the assumptions given earlier
for the commodity sector variables but also on the balance of payments
equilibrium condition and the assumptions on its components (especially
commodity and capital flows). Therefore, the final IS equation will
incorporate the adjustments in the equilibrium value of ρ .
14Note that an increase in r will reduce investment, which also produces a
movement up along the IS curve.
15 Unless purchasing power parity holds, in which case ñ will depreciate in
proportion to the increase P and net exports will not change.
16 The indirect transmission operates through the impact of money supply
changes on interest rates and, through them, on investment and aggregate
demand. The direct transmission mechanism operates through the
recipients of the increase in the money supply spending it directly on
commodities, which increases aggregate demand.
17 Note that the definition of an efficient market includes the absence of a
lag in its reaching equilibrium, but does not exclude lags in the interaction
from one market to another or from one variable to another.
18 In practice, central banks set the nominal interest rate R, not the real
interest rate r. However, their objective in setting R is to aim at its
corresponding value of r.
19 Note that the IRT curve replaces the usual LM curve (see Chapter 6),
which is based on the assumption that the central bank targets/sets the
money supply. The central bank can target/set either the interest rate or the
money supply, but not both, so that the LM curve is not appropriate under
interest rate targeting.
20 This decrease in net exports would cause the nominal exchange rate to
depreciate but not by enough to cancel out the initial appreciation of the
real exchange rate. Further, under perfect capital mobility, capital flows
would not change since the domestic and foreign interest rates would not
have changed.
21 The Fisher equation is R = r + π e (see Chapter 2) .
22 If m > 0, an increase in the nominal interest rate will decrease the
R
demand for money balances.
23To deal with the discrepancy between the market rates indicated by the
intersection of the IS and LRAS curves and the interest rate set by the
central bank, the central bank has to increase the money supply to maintain
equilibrium in the domestic money and bond markets.

266
24 This effect will be stronger in economies with greater capital mobility.
25 For the fixed exchange rate case and an exogenous money supply, the
Mundell–Fleming model showed that monetary policy is ineffective in
changing aggregate demand — and, therefore, ineffective in changing
output under the horizontal supply curve assumption — while fiscal policy
is effective.

267
 CHAPTER 6
Aggregate Demand Under a Money
Supply
Operating Target: IS-LM Analysis

This chapter presents the analysis of aggregate demand for the

case where the central bank is able to control the money supply
and set it at an exogenous level.With an exogenous money supply,
money market equilibrium implies an LM equation/curve. In this
monetary policy scenario, aggregate demand in the economy is
determined jointly by the IS equation/curve and the LM
equation/curve.
This chapter also presents the detailed analysis of money
demand and money supply.

Chapter 4 presented the analysis of the commodity market and derived the
IS curve for the open economy. Chapter 5 presented the determination of
aggregate demand under the assumption that the central bank sets an
exogenous interest rate target and derived aggregate demand for the open
economy by the IS-IRT analysis. This chapter focuses on the money
market under the assumption that the central bank sets an exogenous
money supply target. This assumption implies an LM equation/curve when
the money market is equilibrium. Further, with an exogenous money
supply, the IS and LM equations/curves have to be used to derive
aggregate demand in the economy. The name given to this analysis of
aggregate demand is the IS-LM analysis.
The IS-LM and the AD-AS modes of analyses together represent the
currently dominant technique for the exposition of the short-run
macroeconomics. Neither is by itself a paradigm or a theory; each is only a
mode of exposition of the underlying theory/model. Each of them
encapsulates the information assumed on the macro markets of the
economy into compact relationships — or curves in diagrammatic analysis
— and studies the equilibrium properties of the assumed theory/model.

268
6.1 Monetary Policy
In most modern, financially developed economies, the central bank has
been allocated the role of determining monetary policy. Here, as explained
in Chapter 5, the two main options of the central bank are:
1. determining the underlying interest rate in the economy and
2. determining the value of a monetary aggregate.
The central bank either follows an interest rate target or a monetary target
as its primary instrument. If it pursues an interest rate target, it lets the
markets determine the monetary aggregates; if it pursues a monetary
aggregate target, it lets the markets determine the interest rates. The
pursuit of both targets in an independent manner, rather than in a
coordinated, supportive one will mean that the central bank will most
likely not achieve either of its targets. Macroeconomic analysis for each of
these alternate targets pursued as the primary one (with the other variable
changed in a supportive but not independent manner) is quite different.
Chapter 5 presents the analysis for interest rate targeting while this chapter
does so for monetary targeting.

6.1.1 Reasons for choosing interest rate targeting over

money supply targeting, or vice versa
The main reasons for the central bank to select the use of the interest rate
as the operating target of monetary policy for financially developed
economies are:
• longer lags in the impact of money supply changes than in that of interest
rates on aggregate demand and
• the instability of money demand.
The main reasons for selecting the use of the money supply as the
operating target of monetary policy over interest rates are:
• imperfect and segmented financial markets, which occur in financially
underdeveloped economies, especially if there is a large informal
financial sector as well as black market holdings of money,
• the instability of the IS curve, which arises because of the instability of
investments, exports, and other determinants of the IS curve, and
• lags in the impact of changes in the money supply on aggregate demand
are shorter than of changes in the interest rate.
Lags in the impact of money supply changes versus those of interest rate
changes
In financially developed economies to which the IS model developed in

269
the preceding two chapters applies, an increase in the money supply
increases aggregate demand by first lowering interest rates, which
increases investment. The first part (that from money supply changes to
interest rates) of this process not only takes some time to occur but may be
uncertain and difficult to predict with sufficient accuracy. This part of the
process can be cut out if the central bank directly targets interest rates,
which implies that targeting interest rates would have a shorter lag in its
impact on aggregate demand.
Instability of money demand
Financial innovations, such as the proliferation of credit and debit cards
and Internet banking, in recent years have meant that the money demand
function has become unstable, so that money demand cannot be predicted
accurately. Since money demand and supply together determine the impact
of money supply changes on interest rates, this impact has also become
uncertain and unpredictable. This uncertainty and unpredictability can be
avoided if the central bank was to directly change interest rates rather than
to change the money supply and let the financial markets produce changes
in the interest rates.
The informal financial sector and black money in developing economies
Many countries, especially underdeveloped ones, have a large informal
financial sector (i.e., other than modernstyle banks) and large holdings of
black money. Further, much of their production is by very small firms. The
informal sector generally takes the form of lending by small moneylenders
and by friends and relatives to the small business owners. Interest may not
be charged on such loans but, where the loans are by moneylenders, the
interest rate tends to be very high with its variation dependent on local
conditions, rather on the interest rate set by the central bank or that in
formal financial markets.
In the less developed economies, the formal financial sector consists
mainly of banks. They do not have significant number of other financial
institutions such as investment banks, investment brokers, and pension
funds. Their bond and stock markets are either non-existent or relatively
insignificant.
Therefore, overall for the whole economy, the interest rate plays a very
limited role in decisions on investment and consumer expenditures, while
the money supply plays a more important role. This implies that money
supply changes are likely to have greater impact on expenditures and
aggregate demand in the economy than interest rate targeting, so that the
former is the more essential tool for monetary policy. However, such

270
economies also usually possess a somewhat developed sector which has
large firms that use bonds and bank loans to raise capital, so that interest
rate changes can play a supportive role to money supply changes, rather
than vice versa.

6.1.2 Choosing the monetary aggregate as the target

variable
Monetary aggregates are the monetary base (M0), narrow money (M 1),
broad money (M2), etc. The definitions of these aggregates were given in
Chapter 2. These aggregates are related to each other by the structure of
the economy, so that changes induced by the central bank in one of them
trigger changes in the other ones. Instead of exercising direct control over
interest rates, the central bank can choose to manipulate one or more of the
monetary aggregates. The central bank has various options in doing so.
These are:
• The central bank sets at its discretion the desired level (or growth rate) of
the money supply — and achieves it through its control over the
monetary base or/over interest rates. In this case, the money supply is said
to be exogenously determined by the central bank.
• The central bank determines the money supply according to a feedback
policy rule. An example of such a rule is a ‘Taylor-type’ rule, according
to which it would increase the money supply when there is unacceptably
high unemployment while decreasing it if there is unacceptably high
inflation.
This chapter assumes that the central bank sets the money supply as the
exogenous variable. Under this assumption, the central bank maintains the
money supply at a particular level.1 Hence, our assumption for the money
supply function is that:

where M is the exogenously determined money stock.2 As explained in

Chapter 2, the central bank controls the money supply through changes in
the monetary base. These changes are often brought about through open
market operations (i.e., purchases and sales of bonds from the private
sector). The relationship between the money supply M and the monetary
base M 0 was given in Chapter 2 as:

where α is the monetary base (to money supply) multiplier. The central

271
bank can control the monetary base M 0 but not its multiplier α, so that it
needs to know the multiplier α reasonably well to be able to control the
money supply.
Figure 6.1 shows the exogenous money supply curve by the Ms curve.
Note that in this case, the horizontal axis represents the nominal money
supply. The money supply curve Ms shown is vertical under the
assumption that it is determined exogenously and, therefore, is
independent of the interest rate. That is, an increase in the nominal interest
rate from R0 to R1 leaves the money supply unchanged at M0 . The money
supply is also independent of output and the price level in the economy, so
that the shifts in these variables will not move the Ms curve. When the
money supply does not vary with the interest rate or any other variable, it
is referred to as the money stock.

If the central bank increases the monetary base and thereby increases the
money supply, it would move the money supply curve to the right from M
s to M s in Figure 6.1.
0 1

6.2 The Demand for Money

As discussed in Chapter 2, money consists of items that serve as the
medium of payments in the economy. In practice, it is represented by M 1,
M 2, or one of the still broader monetary aggregates. The demand for
money was also discussed in Chapter 5. The presentation below is an
iteration and elaboration of the analysis of money demand presented in
Chapter 5.
Holdings of money vary during each period. For example, the consumer
may start a week with, say $100, and spend it during the week, ending the
week with zero money holdings. To avoid dealing with such variations,
macroeconomics focuses on the average money balances held during the
week or period in question. Further, the consumer is taken to be free of

272
money illusion and determines his demand for money in real terms, since
the real value of money holdings is the proper indicator of their purchasing
power in terms of commodities. Hence, the money demand variable is the
real value of average money holdings over the period in question. The real
value of these average money holdings is called ‘real balances’. To arrive
at real balances m, divide the nominal balances M held by the public by
the price level P , so that m = M/ P .
Money balances are held by households and firms for several reasons,
the most important of which are:
• Households hold money balances to facilitate their purchases of
commodities; firms hold money balances to facilitate production of
commodities and employment of inputs. Such balances depend on the
cost of holding money. Since money holdings do not pay interest, the cost
of holding them is the return foregone by not holding interest-paying
bonds. The demand for money under this category is called the
transactions demand for money. This transactions demand is a function
of national income and the interest rate. It increases if income increases
and decreases if the interest rate rises.
• Investors (including both households and firms) hold money balances for
some time while switching among investments. This component of
money demand is combined with the following speculative component of
money demand.
• Both households and firms hold money balances to diversify their
portfolios, so as to reduce their exposure to risk. For example, in riskier
environments, when stock markets are volatile, investors increase the
proportion of their portfolio held in money, as well as of other highly
liquid financial assets.3 The money balances held for this reason depend
on the return on investments and their perceived riskiness, as well as the
size of the portfolio (i.e., amount of wealth). The return on investments is
approximated by the interest rate on bonds, so that money demand
depends on interest rates and wealth. This component of money demand
is often called the speculative demand for money. It decreases if the
interest rate rises and increases if wealth rises.
• Households hold money balances as a precaution against the possibility
of sudden expenditures, e.g., due to an accident requiring sudden medical
treatment, or reduction in incomes due to a job loss. This component of
money demand is called the precautionary demand for money. It depends
on income and the interest rate and is combined with the transactions
demand for money in the first point above.

273
6.3 The Motives for Holding Money
Intuitive explanations of the demand for money often focus on the
individual’s motives for holding money. The two most important ones
among these4 are the following.
1. The transactions motive (related to income): this motive specifies the
amount needed to finance expenditures on commodities through the use
of money. The demand for money under this motive is called the
transactions demand for money, which, when expressed in real terms,
can be written as m d, tr. When it is expressed in nominal terms, we give
it the symbol M d, tr. The real transactions demand for money is, for
simplification, assumed to be a constant proportion of real
expenditures/income y, so that it equals myy. Hence, m d, tr = myy.
2. The portfolio/speculative motive (related to interest rates): this motive
specifies the amount held in money rather than bonds and is an element
of the overall desired portfolio of financial assets. It arises mainly
because of the uncertainty of the actual yields on bonds, which includes
any capital gains or losses arising from fluctuations in bond prices.
From the intuitive perspective, the portfolio/speculative motive for holding
real balances (i.e., money in real terms) arises from the choice between
money and bonds in investors’ portfolios of financial assets. Let the total
amount to be allocated between money and bonds be FW (financial
wealth), whose real value will be FW/P , where P is the price level of
commodities (not of financial assets). Since R is the return on bonds while
money does not pay interest or pays much less than bonds do, bonds will
become more attractive as the nominal interest rate R rises. Write the real
demand for bonds as bd = mR(R, FW/P). For simplification, let the amount
invested in bonds equal mR R, so that mR = ∂bd/∂R ≥ 0. The demand for
money is the part of FW not held in bonds, so that it will equal (FW0 – mR
R). An increase in the interest rate R will induce investors to invest more in
bonds, so that mRR will increase while the portfolio money holdings —
equal to (FW0 – mRR) — will decrease. Therefore, money balances held in
the portfolio are given by:

where m d, sp is the speculative demand for money. In this theory of the

demand for money, the speculative demand is the interest sensitive
component of money demand: an increase in the interest rate increases the

274
demand for bonds and decreases the speculative demand for money.
Since total money demand is the sum of the transactions and speculative
money demands, total money demand is given by:

The transaction demand for money increases with income, while the
speculative demand for money decreases when the nominal interest rate
rises.
Note that the holdings of money balances cannot really be divided into
separate holdings for each of the two motives or demand, so that any
correspondence between the two terms on the right side of the money
demand equation and motives/demand can only be rough and approximate.
For ease of discussion, the traditional practice is to call the term (myy) the
transactions demand for money, and to call the term ( FW/P – mR R ) the
speculative demand for money and associate it with the ‘speculative
motive’ related to speculation in the financial markets. Also, note that
changes in the real value of financial wealth, FW/P, will become important
if the value of financial wealth (see Chapter 2 for some information on this
value) does not change in proportion to changes in the price level of
commodities, as it usually does not do in the real world.5

6.3.1 The volatility of the speculative demand for money

In the real world, the portfolio demands for money and bonds depend on
the expectations of — rather than the actual — bond (and stock) yields.
These yields include expected capital gains and losses, which depend on
the expected future prices of bonds. These expectations are very volatile:
an example of this volatility is the volatility of stock prices. An illustration
of this occurs when an impending collapse in stock prices signals negative
yields from holding stocks and drives up — ‘a flight to liquidity’ —
money demand. This volatility induces volatility in the speculative demand
for money, though more so in M 2, M 3, and broader money supply
measures, than in M 1, since M 1 is now mainly kept for transactions
purposes. The volatility in the expectations of the yields on bonds is not
incorporated in the standard money demand function, so that the shifts in
these expectations have to be captured through exogenous shifts in the
standard money demand function. Such a shift is often referred to as a shift
in ‘liquidity preference’ .

6.3.2 Other reasons for the volatility of the demand for

275
 money
Recent years have witnessed considerable innovation in the types of
instruments (checkable accounts, savings accounts, credit cards, debit
cards, etc.) that can serve as a medium of payments and also in the
technology of the payments system (e.g., the introduction and spread of
automatic teller machines). These have shifted both the transactions and
speculative demand for money. Overall, money demand in recent decades
has proved to be unstable (i.e., shift frequently). Further, these shifts have
proved to be difficult to predict.

6.4 The Standard Money Demand Function

Therefore, the total demand for real balances md is a function of real
income y, the nominal interest rate R, and real wealth. Standard
macroeconomic analysis simplifies by ignoring wealth and assumes that
demand for real balances depends only on real income y and the nominal
interest rate R, and that the money demand
function is linear. Therefore, the demand function for real balances is
simplified to:

As discussed earlier, the component of money demand represented by myy

is usually called the transactions demand for money and the component of
money demand represented by ( –mR R) [or, more accurately, by (financial
wealth - mR R)]6 is usually called the speculative demand for money. Note
that these designations are just meant to be simplifications for facilitating
discussion of the reasons for money demand. The component ( –mRR) or,
more accurately (financial wealth - mRR), can also be referred to as the
interest-sensitive component of money demand.
Figure 6.2 shows the demand curve for real money balances by md. Note

276
that the nominal interest rate R is on the vertical axis and the demand for
real balances md is on the horizontal one. Since R is on one of the axes,
changes in R imply movements along the curve. An increase in R reduces
money demanded, which means moving up along the existing md curve: a
rise in R from R0 to R1 means going from the point a to the point b along
this curve. However, income is not on one of the axes, so that, for a given
rate of interest R0, an increase in income increases transactions demand
and shifts the md curve toward the right from m d0 to m d1.
Correspondingly, a decrease in income decreases transactions demand,
which shifts the curve toward the left from .
For financially developed economies, empirical studies indicate that, for
M 1, there are economies of scale in the transactions use of money: the
income elasticity of real balances held in M 1 (i.e., with respect to real
income) is about 0.7, so that real balances to finance expenditures increase
by 70% if real expenditures were to double. The interest elasticity of
money demand with respect to the short-term (Treasury bill) interest rate is
about 0.15, so that an increase of 1% in the interest rate decreases money
demand by 0.15%.
Extended Analysis Box 6.1: Special Cases of Money Demand
There are two theoretical extremes of the money demand function.
These are:
1. The quantity theory case/range. This occurs if mR = 0, so that:

In this case, the demand for money does not depend on the rate of
interest, which is the return on holding bonds, so that there is only a
transactions demand for money. This case assumes that economic
agents do not substitute between money and bond holdings. This is
likely to be valid for economies with rudimentary financial systems in
which bonds do not exist or are not accessible to most individuals, or
if there is a very large informal financial sector in the economy.7 It is
not likely to be valid for the modern financially developed economies.
2. The liquidity trap case/range. This occurs if mR → to, where ‘m R →
∞’ is pronounced as ‘mR goes to infinity’. In this case, the interest
elasticity of money demand is infinite, meaning that the private sector
is willing to increase its money holdings by any amount, however
large, at the existing interest rate. This case will occur if the interest

277
 rate is too low to compensate for the costs and uncertainty of holding
bonds, or if holding bonds is expected to lead to a net loss. The latter
means that the expected rate of return on bonds is negative and occurs
if the expected prices of bonds are expected to decline by more than
the fixed coupon payments on them. This case may be more
applicable in depressed economies but is not likely to be of
macroeconomic significance in normal conditions in the financially
developed economies. John Maynard Keynes, writing in the midst of
the Great Depression of the 1930s, drew attention to this case as an
analytical possibility while denying its empirical significance, and
labeled it as the liquidity trap. To distinguish this trap from a liquidity
supply trap defined below, we can more precisely call it a ‘liquidity
demand trap’ However, note that this term is not in general usage, so
that we will continue to use the term ‘liquidity trap’ for this case.
A liquidity supply trap
A ‘liquidity supply trap ’ or a ‘monetary base trap ’ occurs if an increase
in the monetary base induced by the central bank does not result in an
increase in the money supply. This situation occurs if commercial banks
refuse to use the increases in their reserves (because of increases in the
monetary base) to invest in private/corporate bonds or otherwise lend to
private firms because of a perceived high risk in such lending, and/or an
exceptionally high degree of risk aversion by lenders. This occurred in
many countries, especially the USA and UK, during the financial crisis
of 2007–2010, when the solvency of many firms (including major
financial firms) was not easily transparent and was seriously being
questioned.8
The impact of the extreme reluctance to lend to the private sector was
that although the central bank increased the monetary base very
substantially, the banks added these increases to their cash reserves or
bought government bonds, mainly Treasury bills, but not commercial
bonds. Consequently, bank deposits and the monetary base (whose main
component is bank deposits) did not increase very much, while the
excess reserves and liquidity of banks shot up. This failure of the money
supply to expand in the face of increases in the monetary base meant
that monetary policy became ineffective in changing the money supply,
and, through it, aggregate demand in the economy. We can call this a
‘money supply trap’ and define it as a situation in which increases in the
monetary base do not cause significant increases in the money supply.
Note that this is not a liquidity trap arising from the public’s
preference/demand for money

278
 holding rather than buying bonds, but arises from the behavior of the
banking sector, which prevents the increases in the money supply in
spite of increases in the monetary base.
Note that both the liquidity demand trap and the liquidity supply trap
render an expansionary monetary policy ineffective.
The potential shapes of the money demand and LM curves
Figure 6.3 incorporates the two special cases of money demand, along
with the normal case, into the money demand curve.

6.4.1 Equilibrium in the money market

Note that the definition of the money market for our macroeconomics
purposes is that it is the market for money. Hence, the money market is the
market for M 1, M 2, or a broader definition of money. It is not the market
for short-term bonds, which is the common English language meaning of
the term ‘money market’. There is, therefore, quite a difference between
the investors’ and banks’ usage of the term ‘the money market’ and the
economists’ usage.
Equilibrium in the monetary sector requires that money demand equals
money supply. Hence, for equilibrium in nominal terms, with money
supply exogenously given as M :

which gives,

Note that while the nominal stock of money is exogenously given and
assumed to be in the control of the monetary authorities, the real value of
that stock, (M s/P), depends upon the equilibrium price level P in the
economy and incorporates choices made by the public, including those on

279
its demand for money.
Disequilibrium in the money market
The converse of equilibrium is disequilibrium. The money market is in
disequilibrium if the demand for money differs from its supply. Suppose
the former is greater than the latter: the public wants to hold more money
than is available and there is excess demand for money. Applying this at
the individual level, an individual with insufficient money balances will
try to increase his money balances by selling some bonds, thereby
receiving in exchange money from bond purchasers. In the aggregate,
these sales of bonds by the public represent a decrease in the demand for
bonds, which would push down bond prices. As Chapter 2 showed, a
decrease in bond prices raises the return on them, so that the interest rate
on bonds rises. In the money demand equation, this increase in the interest
rate reduces the quantity of money that is demanded, and, therefore,
reduces the excess demand of money. Households will continue to sell
bonds until the interest rate rises sufficiently to eliminate the excess
demand for money.

6.5 The LM Equation

The preceding Eq. (4) specifies those combinations of y and R that
maintain equilibrium in the money market. It is called the LM equation or
relationship — where L stands for liquidity preference — which is a term
meaning the demand for money (i.e., liquidity) — and M stands for the
money supply. We want to rearrange the LM equation to bring y to the left
side. To do so, rearrange it as:

so that the LM equation can be rewritten as:

This is the usual form of the LM equation, though any of its other forms
are equally valid.
The impact of changes in the money supply M on income y, with interest
rates held constant, is captured by the ‘money (to income) multiplier’,
which is:

280
so that increases in the money supply increase income by multiplying the
increase in the money supply by (1/myP).

6.5.1 The LM curve

The LM equation is plotted in Figure 6.4a as the LM curve. The LM curve
is defined as the set of points at each of which there exists equilibrium in
the money market. The LM curve has a positive slope.9 The intuitive
explanation for this positive slope is: for a given money supply and price
level, an increase in y along the horizontal axis increases the transactions
component of the demand for money, thereby requiring the public to
reduce the amount of money held for speculative purposes — which the
public will only do at a higher rate of interest. That is, an increase in real
income must be accompanied by an increase in the rate of interest for
equilibrium to be preserved in the monetary sector. Note that changes in y
or R induce movements along the LM curve and not shifts in it.

6.5.2 Shifts in the LM curve

Shifts in the LM curve can be caused by changes in the money supply M
or the price level P — or by a shift in the parameters my or mR. We focus
on the former in the following.

In Figure 6.4a, suppose that at the existing price level P 0 and the
existing money supply M 0, the LM curve is LMo | M0/P0 . Now suppose
that the money supply increases from M 0 to M1 . Hence, the real money
supply increases from M 0 /P0 to M1/P0. We examine its effects at a given
level of income y0. Since income is being held constant, the transactions
demand for money does not change. Therefore, the public has extra money
balances, which it will use to buy bonds. The increased demand for bonds

281
will raise their prices — thereby lowering the rate of return on them, i.e.,
the interest rate R decreases. How far does it have to fall? It must fall by
enough to persuade the public to willingly hold (i.e., increase money
demand) all of the increase in the money supply. This will occur at the
interest rate R 1 (at y0) in Figure 6.4a. If we were to follow this reasoning
for different levels of income, we would have a new set of points that will
trace out the new LM curve as LM1 | M1/P0 and pass through the point
(R1, y0). This would lie to the right of LM0 | M0/P0 . Correspondingly, if
the money supply decreases to M2,10 the LM curve will shift left to LM2 |
M2/P0
Similar arguments can be applied to show the effects of changes in the
price level P, holding the money supply constant. The position of the LM
curve depends on the real money supply, so that an increase in the money
supply or a decrease in the price level shifts the LM curve to the right.
Conversely, a decrease in the money supply or an increase in the price
level shifts the LM curve to the left. To reiterate the impact of changes in
the price level on the LM curve, note that the price level P is a variable in
the monetary sector and, therefore, in the LM equation, so that — since P
is not on the axes of the IS-LM figures — a change in P must cause a shift
in the LM curve. A rise in P decreases the supply of real balances which
shifts the LM curve to the left while a fall in P shifts the LM curve to the
right.11 The effect of changes in the price level, with the money supply
held constant at M0, is shown in Figure 6.4b, with P2 > P0 > P1.
The LM curve shifts to the right if:
• The money supply increases.
• The price level decreases.
• Money demand decreases because of shifts in the money demand
function. From the money demand function, this occurs if the transactions
demand per dollar of income (i.e., my) decreases or if mR increases. In
these cases, money demand decreases for given values of R and y.

6.5.3 Shifts in the LM curve versus movements along it

The above discussion implies that the economy moves along the LM curve
as the interest rate or income change and alter the demand for money. But
it shifts if the money demand parameters (my, mR), the policy variable M s,
or the endogenous variable P change.
In particular, an increase in the money supply shifts the curve to the right
because this increases the real money supply (M/P) . But an increase in the

282
price level shifts the LM curve to the left because this decreases the real
money supply. A proportionate change in the money supply and the price
level leaves M/ P unchanged, so that the LM curve would not shift.

6.5.4 Final comments on the LM curve

The modern economy has a money demand that is sensitive to the rate of
interest but does not possess a liquidity trap. This was designated as the
normal range of the LM curve. Except when we specifically try to explain
the implications of the classical range or the liquidity trap, we will ignore
these cases and draw the LM curve as an upward sloping straight line, as
shown in Figures 6.4a and 6.4b. To emphasise the impact of the money
supply and price level changes on the LM curve, we label this curve as
LM1 |M0/P0. An increase in the money supply from M 0 to M 1 shifts this
curve to the right to LM0|M1/P0. An increase in the price level from P 0 to
P 1 shifts this curve to the left to LM2|M 0/P 1. A proportionate increase in
both M to M1 and from P to P 1 leaves M/P unchanged and leaves the LM
curve at LM0, with M1/P1 = M0/P0 .

Extended Analysis Box 6.2: The Potential General Shape of the LM

Curve
The potential general shape of the LM curve, incorporating the liquidity
trap range (when m R → ∞ and the quantity theory range (when m R =
0), is as shown in Figure 6.5 by the curves labeled as LM0 and LM1,
with the latter representing an increase in the money supply. The
liquidity trap is the horizontal part of the LM curve and the quantity
theory range is the vertical part of the LM curve. Between them lies the
upward sloping part, which is the ‘normal range’ of the LM curve. Our
earlier drawings of the normal part of the LM curve replaced the ‘curve’
by a straight line.
An increase in the money supply shifts the LM curve from LM0 to
LM1, thereby elongating the liquidity trap part of it.
Financially developed economies in normal operation neither have the
liquidity trap nor are in the quantity theory range for significant periods
for relevance to the macroeconomy. They are in the normal range with a
positively sloped LM curve.

283
6.6 Deriving the Aggregate Demand for
Commodities by Combining the IS and LM
Curves

6.6.1 The IS equation for the commodity market

equilibrium
Chapter 5 had derived the IS curve for commodities for the open economy.
Its derivation should be reviewed at this stage. The IS equation derived in
Chapter 5 is:

The definitions of the symbols are as given in Chapters 4 and 5. This

equation is the IS equation for the open economy. Note that y is a function
of r and ρr where ρr = ρP /P F, so that replacing ρr by ρP /P F makes y a
function of r , ρ, P , and P F, in addition to policy and exogenous variables
and parameters. Under flexible exchange rates, the equilibrium value ρ* of
the nominal exchange rate is determined by the balance of payments
equilibrium condition (Eq. (3) of Chapter 5). Replacing ρ by ρ* gives the
final form of the IS equation. In this form of the IS equation, y will be a
function of P , r , the fiscal policy variables and the exogenous foreign
variables P F, r F (and R F under the assumption that π e = 0), and y F, but
ρ will no longer appear as a separate determinant of y.
Under an exogenous money supply, the IS and LM equations constitute
the two components which together determine aggregate demand.

284
However, note that the IS equation has the real interest rate r while the LM
equation has the nominal interest rate R.

6.6.2 The relationship between the nominal and real interest

rates
The nominal interest rate R is related to the real interest rate r in perfect
bond markets by the Fisher equation:

where πe is the expected rate of inflation. The approximation to this

relationship for low values of r and πe is given by:

The analyses behind the IS and LM curves is static and is used to examine
changes in y, r, etc., when there is a one-time exogenous shift in the money
supply, fiscal variables, and other variables and parameters. This analysis
compares on equilibrium point prior to the shift with another equilibrium
after the shift, and is called comparative static analysis. It does not study
the dynamic path between the two equilibrium points. Since our analysis is
comparative static, there is no inflation and no expected inflation at the
equilibrium point prior to the shift and also none at the equilibrium point
after the shift. Note that inflation and expected inflation are aspects of the
dynamic movements in prices, which are not relevant in comparative static
analysis. Therefore, we will assume for our comparative static analysis that
πe = 0. In this case, R = r and we could proceed with a diagram on whose
vertical axis we can put either r or R . We choose to put r on the vertical
axis.
Restating the LM equation
The LM equation derived above was:

Using the assumption that π e = 0, so that R = r, we replace R in the money

demand and LM equations by r. This turns the LM equation (5) into:

Now substitute this value of the interest rate in the IS equation. The
resulting AD equation for the open economy is:

285
where:

6.6.3 Diagrammatic determination of aggregate demand

Figure 6.6a draws both the IS and the LM curves in the same diagram. The
intersection point of the IS and LM curves provides those values of r (and
R under our assumption that πe = 0) and y at which there is simultaneous
equilibrium in both the commodity and money markets. This point shows
the aggregate demand for commodities, after incorporating the impact of
the money market on the interest rate, which is needed to determine
investment in the commodity market. The demand curve thus derived —
just as for demand curves in microeconomic analysis — specifies the
quantity demanded of commodities as a function of the price of
commodities.
In Figure 6.6a, start with the IS curve and the LM curve, labeled as LM0
| M0/P0. The latter is drawn for an initial money supply M 0 and an initial
price level P 0. Their intersection at the point d0 indicates the level of
aggregate demand yd, at the price level P0. If, for the given money supply,
the price level were somehow to fall to P1 (P 1 < P 0), the LM curve would
shift to the right to LM1 | M0/P 1. Then, the intersection of the IS and the
LM1 curves would occur at the point d 1, so that the level of aggregate
demand would become y d1, at the price level P 1. That is, a fall in the
price level produces a higher aggregate demand, in real terms, for
commodities. Conversely, a rise in the price level will produce a lower real
aggregate demand for commodities.
Hence, the IS curve does not alone determine the level of aggregate
demand. The IS curve depends on investment which depends on the
interest rate, which is jointly determined with the money market. This
makes the LM curve an essential component for the determination of the
aggregate demand for commodities — as well as of the interest rate.

286
The IS, LM curves and the determination of output: a caveat
Note that the intersection of the IS and LM curves determines the
aggregate demand for commodities and determines the AD function/curve
used in Chapter 1. Their intersection does not determine the aggregate
quantity of commodities that will be produced in the economy. This
requires knowledge of both the demand and the supply functions/curves of
commodities. Chapter 1 had shown this by incorporating both aggregate
demand and supply in the AD-AS diagram.
Because the IS and LM intersection only determines the demand for
commodities but not output (i.e., the quantity produced), it is strictly
inappropriate to interpret y on the horizontal axis of the IS-LM, as well as
the AD-AS, diagrams as the output (i.e., the quantity produced) of
commodities. The symbol y on the horizontal section of both the IS-LM
and the AD-AS diagrams is to be interpreted as the ‘aggregate quantity of
commodities’, not as output (i.e., the quantity produced). This usage
means that the intersection of the IS and LM specifies the quantity
demanded of commodities — not the output of the economy — at the
specific price level for which the LM curve has been drawn.

6.7 The Impact of Expansionary Monetary and

Fiscal Policies on Aggregate Demand
In Figure 6.6b, suppose that the economy is initially in overall equilibrium

287
at the point d0 at the intersection of the IS and LM0|M 0/P 0 curves, where
the LM0|M 0/P 0 is drawn for real balances M0/P0. At d0, aggregate
demand in real terms equals y d0 at the price level P 0. Now suppose that
an increase in the money supply to M 1 increases the real balances to
M1/P0 and shifts the LM curve from LM0 to LM1|M 1/P 0. The new
equilibrium between the money and commodity markets is shown by the
point d 1 and shows an increase in the real aggregate demand from y d0 to
y d0, with both evaluated at the price level P 0. Hence, an increase in the
money supply increases aggregate demand at the pre-existing price level.
Conversely, a decrease in the money supply will shift the LM curve
upwards and cause a decrease in the real aggregate demand for
commodities at the pre-existing price level.
Now start with the given money supply as M 0 and the price level as P 0
and an initial IS curve as IS0. Suppose that fiscal policy becomes
expansionary with an increase in government expenditures. This would
shift the IS curve from IS0 to IS1 in Figure 6.6c. At the pre-existing price
level P 0, real aggregate demand would increase from from the
point d0 to the point d1 at the intersection of the IS1 and LM0|M 0/P 0
curves.
The opposite pattern of effects occurs for a decrease in government
expenditures or an increase in taxes.

6.8 Bringing Aggregate Supply into the Open

Economy Analysis, a Preview of Chapters 7 to
9

6.8.1 The impact of an increase in aggregate demand on the

quantity supplied and the price level
Since a change in the price level shifts the LM curve, further analysis
requires the determination of the price level. This determination requires
the specification of aggregate demand by the IS-LM analysis and that of
aggregate supply set out in Chapters 7 to 9.
Would an increase in aggregate demand not raise the price level from its
pre-existing level and shift the LM curve? To answer this question, we

288
need to know not only the determinants of aggregate demand but also
those of aggregate supply, which were briefly examined in Chapter 1 but
are more rigorously examined in Chapters 7 to 9. Since the model
specified so far in this chapter does not have the relevant aggregate supply
curve, we are not yet able to determine either the change in the
commodities produced or the price level at which they will be sold. The
student who wishes to get an introduction to these effects at this point can
do so by turning to the relevant analysis in Chapter 1. The more complete
analysis of the supply side of the economy is in Chapters 7 to 9.
Hence, while the preceding sections have analyzed the effectiveness of
monetary and fiscal policies in changing aggregate demand,13 we also
have to bring in aggregate supply, using the aggregate demand (AD) and
aggregate supply (AS) curves in the (P, y) space. There is a long run LAS
curve and a short run SAS curve for the open economy, as in the closed
economy case. This would give us an analysis similar to that for the closed
economy case, except that the AD curve is now susceptible to international
influences through commodity and capital flows.
As Chapters 7 and 8 show, there are two different supply responses to an
increase in AD. These depend on whether the economy is at full
employment along the LAS curve or off it. In the latter case, it may be in
short-run equilibrium and operate—as in the case of errors in price
expectations or adjustment costs—along a SAS curve or the economy may
be in disequilibrium.

6.8.2 The impact of monetary and fiscal policies on

equilibrium output and price level in the open economy
For this analysis, we bring together two results:
1. For the flexible exchange rate economy with elastic net exports, an
expansionary monetary policy is more efficacious in increasing
aggregate demand than an expansionary fiscal policy.
2. Output is at its full-employment level y f in the long run, but does
increase in the short run as the price level rise. [This is explained in
Chapters 1, 7, and 8.]
The short-run impact of aggregate demand changes on output and the
price level
In the short run, an increase in aggregate demand increases both the price
level and output. Therefore, the expansionary monetary policy, which does
increase aggregate demand under flexible exchange rates, will increase

289
both output and the price level.
Comparing expansionary monetary and fiscal policies under flexible
exchange rates, the relative efficacy of monetary policy over that of fiscal
policy in increasing aggregate demand means that the former will produce
both greater output and higher inflation in the short run. Whether this
makes monetary policy more desirable than fiscal policy depends on the
economy’s implied trade-off between inflation and the increase in output,
and the policymaker’s willingness to accept this trade-off.
The long-run impact of aggregate demand changes on output and the price
level
In the long run, an increase in aggregate demand does not change the level
of output from its full-employment level yf, so that it only increases the
price level. Therefore, in the long run, the expansionary monetary policy
will only produce inflation, without a corresponding benefit in higher
output. Hence, given flexible exchange rates, from the long-run
perspective, monetary policy is worse than fiscal policy since the former is
more efficacious in increasing aggregate demand. Of course, with
employment and output maintained at their fullemployment levels, neither
policy serves a useful long-run purpose with respect to these variables.
Therefore, neither policy should be pursued from the perspective of the
long run.

6.8.3 Disequilibrium in the domestic economy and

stabilization through monetary and fiscal policies
For the analysis of disequilibrium in the domestic economy, assume that
the country has flexible exchange rates and these ensure continuous
equilibrium in the balance of payments. Any disequilibrium, as in Figures
6.7a and 6.7b at the point d , will emerge as a difference between
aggregate demand (shown by the point d ) and aggregate supply (shown by
yf) and will be handled by the adjustment in the domestic price level.
Assuming that the economy has price level flexibility, the price level will
eventually fall to ensure equilibrium in the economy and monetary and
fiscal policies will not be needed for this purpose. But this process can take
real time.

290
 If the authorities do not want to rely upon changes in the domestic price
level to eventually restore full employment — e.g., because this process
takes too long — they can pursue an expansionary monetary policy, as
shown by shifting the LM curve to the right from LM0 to LM′ in Figure
6.7a. Alternatively, they can pursue an expansionary fiscal policy, as
shown by shifting the IS curve to the right from IS0 to IS′ in Figure 6.7b.
However, under flexible exchange rates and deficient demand
disequilibrium, an expansionary monetary policy will be preferable since
monetary policy is more efficacious than fiscal policy in increasing
aggregate demand and output.

6.8.4 Summation on the roles of monetary and fiscal

policies under flexible exchange rates
To sum up, exchange rate flexibility provides the economy with the very
useful equilibrating mechanism of changes in the nominal exchange rate
brought about by market forces for handling the balance of payments
deficits. Together, exchange rate flexibility and domestic price flexibility
eliminate the need for monetary and fiscal policies to return the economy
to general equilibrium and full employment — as long as the authorities
are willing to give the economy enough time to make its own adjustments
to equilibrium and are willing to accommodate the possible reserve losses
during the adjustment period. If they do pursue appropriate monetary and
fiscal policies, the preceding analysis implies that:
• From the perspective of the long-run equilibrium analysis, monetary
policy is less desirable than fiscal policy because the former’s efficacy
over the latter in increasing demand produces greater inflation, without
any benefit in higher output. [The analysis behind this result is in
Chapters 12 and 13.]

291
• In the short-run equilibrium analysis, an expansionary monetary policy
produces both greater output and inflation than fiscal policy. Hence, the
desirability of monetary policy over the fiscal one depends on the
economy’s trade-off between price and output increases and the central
bank’s preferences between them.
• In the case of deficient demand in the domestic economy, monetary
policy is preferable to fiscal policy: the former is more efficacious in
increasing aggregate demand and, therefore, in increasing output and
employment.
Since the first and last points given above reverse the desirability of
monetary over fiscal policy, the policymaker’s preference for one over the
other will vary with the existing state of the domestic economy and its
propensity to maintain full employment or stay in deficient demand
disequilibrium for significant periods.
Extended Analysis Box 6.3: The Mundell–Fleming Model of the Open
Economy
The Mundell-Fleming model of the open economy is based on the
contributions of Robert Mundell and Marcus Fleming in the 1960s. It
presented the analyses of monetary and fiscal policies using the open
economy IS-LM framework under the assumptions of elastic net
exports, perfect capital mobility and a fixed price level but variable
output. We can separate its analysis into two parts: that of the impact of
the policies on aggregate demand and that of changes in the aggregate
demand on real output. Chapter 12 and 13 shows that, under flexible
exchange rates, monetary policy is more effective in changing aggregate
demand than fiscal policy.

We consider here the effect of adding in the assumption of a constant

price level and variable output. Under this assumption, the aggregate
supply curve becomes horizontal. Therefore, as Figure 6.8a shows,
changes in real output would equal the policy-induced changes in

292
 aggregate demand. Hence, monetary policy would be effective in
changing aggregate output while fiscal policy is ineffective.14
Our analyses of aggregate supply in the subsequent Chapters 7 and 8
imply a vertical LAS curve for the long run and a positively sloped SAS
curve for the short run. For further exposition with these differing slopes
of the supply curves, we can truncate the Mundell-Fleming model by
confining its analysis to the determination of aggregate demand. We can
replace its assumption of a fixed price level and a horizontal aggregate
supply curve by the SAS and LAS curves. This was done in our analysis
of the impact of monetary and fiscal policies on output, and implied
conclusions that differ from those usually associated with the Mundell-
Fleming model. The biggest change occurs if the Mundell-Fleming
assumption of a horizontal supply curve is replaced by the vertical LAS
curve. As Figures 6.8a and 6.8b show, the Mundell-Fleming assumption
in Figure 6.8a implies that an expansionary monetary policy increases
real output and is desirable, while the vertical LAS curve in Figure 6.8b
implies that it will only cause inflation and is undesirable. The SAS
curve would have given an intermediate response: an expansionary
monetary policy increases both output and the price level.
Under flexible exchange rates with perfect capital mobility, an
expansionary fiscal policy does not change aggregate demand and, no
matter what the shape of the supply curve, has no effect on output.

6.9 The Central Bank’s Control Over the Money

Supply
We now turn from the derivation of aggregate demand, output, and the
price level to a detailed treatment of the determination of money supply.
The central bank controls the money supply through its instruments of
monetary policy, which are:
• changes in the monetary base,
• changes in reserves requirements, which are imposed on commercial
banks, and
• changes in the discount rate at which it lends to commercial banks.

6.10 The Central Bank’s Instruments for Changing the

Monetary Base
The central bank can change the monetary base through the following

293
instruments at its disposal:
• open market operations,
• moving government deposits between itself and the commercial banks,
and
• allowing the commercial banks to borrow from it.

6.10.1 Open market operations

Open market operations are the purchase (or sale) by the central bank of
securities in financial markets, and results in corresponding increases
(decreases) in the monetary base. Countries with well-developed financial
markets and extensive amounts of public debt traded in the financial
markets usually rely on such operations to change the money supply as
their main tool of monetary policy. They — along with changes in the
discount/bank rate15 — are the most important tools of monetary policy in
the USA, Canada, and the UK, and in most economies with well-
developed financial sectors and capitalist economies. However, open
market operations do not hold a corresponding degree of importance in
very many other countries. In particular, financially underdeveloped
economies generally do not possess sufficiently the size and depth of
domestic bond markets required for adequate open market operations. This
is also usually so in countries without a strong commitment to free
enterprise and capitalism.

6.10.2 Shifting government deposits between the central

bank and the commercial banks in Canada
The central bank usually acts as the government’s bank, i.e., keeping and
managing the government’s deposits. Increases in these deposits with the
central bank occur because of payments by the public to the government
out of their deposits in commercial banks. This reduces the monetary base
while decreases in such deposits because of increased payments by the
government to the public increase the monetary base. One way of avoiding
changes in the monetary base, because of payments to the government or
receipts from it by the public, is for the government to hold accounts with
the commercial banks and use them for its transactions with the public.
The resulting increases and decreases in the government deposits with the
commercial banks do not change the monetary base — while transfers of
these deposits to the central bank reduces this base.
In Canada, the Bank of Canada manages the distribution of the
government deposits between itself and the chartered banks in Canada as a

294
way of manipulating the monetary base and, therefore, as a tool of
monetary policy akin to open market operations. In current practice, such
shifting of balances is more convenient and has become more important
than open market operations for changing the monetary base over short
periods.
This practice is feasible for the Canadian banking system since it has a
very small number (six) of very large banks, which are considered to be
fully secure.16 It would not be practical in a country with a large number
of banks, as in the USA, nor in one where some of the commercial banks
have a significant risk of becoming insolvent. The latter is likely to be the
case in many developing countries.

6.10.3 A mechanism for commercial banks to change the

monetary base
Central banks often allow commercial banks and other designated
financial institutions to borrow from it at a designated interest rate called
the discount rate (in the USA) or the Bank Rate (in Canada and the UK).
Canada, the UK, and the USA have traditionally done so. Such borrowing
from the central bank changes the monetary base. The level of the discount
rate relative to the economy’s ones at which banks can invest determines
the willingness of commercial banks to borrow from the central bank, so
that changes in the discount rate induces changes in the amount borrowed
and thereby change the monetary base.
The central bank is not always able to accurately predict or achieve the
monetary base at which the economy’s interest rate will equal the target
interest rate. A difference between them changes the incentives for the
commercial banks to increase their borrowings from the central bank. For
example, if the monetary base is not enough to ensure that the economy’s
rate is as low as the target one, the banks will increase their borrowings
from the central bank, thereby increasing the monetary base. But if the
monetary base is so large that it makes the economy’s rate lower than the
target one, the banks will reduce their borrowings, thereby decreasing the
monetary base. Therefore, allowing the commercial banks to borrow from
the central bank serves as a safety valve — and a fine-tuning mechanism
— in adjusting the monetary base to the one required for equilibrium in the
money market.

6.11 The Central Banks’ Control Over the

295
 Monetary Base Multiplier through Reserve
Requirements
The imposition of reserve requirements17 has historically been a common
tool of controling monetary aggregates for a given monetary base. In
countries where the markets are too thin (small) for viable open market
operations, or the monetary base cannot be controled for some reason, the
monetary authorities often attempt to limit the creation of reserves by the
banking system through the imposition of, or changes in, reserve ratios
against demand deposits and sometimes also against other types of
deposits. These ratios can range from 0% to 100%, though they are often
in the range of 0% to 20%.
A decrease (increase) in the required reserve ratio increases (decreases)
the amount available for loans to the public and increases the monetary
base multiplier, so that it increases (decreases) the money supply. This tool
for changing the money supply is often employed in developing
economies. The change in the required reserve ratio is usually of the order
of 0.25% or 0.5%.

6.12 The Impact of Discount Rate Changes on the

Economy
A change in the discount rate serves as an instrument of monetary policy
in three ways:
1. The change — or lack of a change when one was expected — acts as a
signal to the private sector of the central bank’s intentions about
monetary policy.
2. The change affects the amount of borrowing from the central bank.
Changes in the amount borrowed change the monetary base and the
money supply.
3. The change cascades through various interest rates in the economy
because of the institutional lending practices of the banks and other
financial intermediaries.

6.12.1 The central bank’s discount rate and interest rate

differentials in the economy
The commercial banks and other financial intermediaries usually, though
not always, follow the lead given by the discount rate changes to alter their

296
own interest rates — such as the prime rate, the personal loan rates, and
the mortgage rates—as well as in their purchases and sales of market
instruments. This behavioral pattern results in a shift of the interest rates
throughout the economy, while leaving the spread between any pair of
rates to market forces.
The central bank’s power to set its discount or bank rate does not extend
over the differentials or spreads among the various interest rates in the
economy. In particular, the spreads between the commercial banks’
deposit rates and the short-term market rates, such as Treasury bills and
money market mutual funds, are still outside the direct influence of the
central bank and depend upon market forces. However, macroeconomics
simplifies the analysis by making the assumption that the interest rate
differentials remain unchanged for changes in the target interest rate and/or
in the money supply.

6.13 The Determination of the Money Supply

As explained in Chapter 2, the quantity of money is usually measured by
M 1 (the sum of currency in the hands of the public and the demand
deposits of the public in commercial banks), M 2 (M 1 plus savings
deposits of the public in commercial banks), or a still broader measure. A
related concept was the monetary base, for which we use the symbol M 0.
Chapter 2 should be reviewed at this point.
The generic definitions of the basic monetary symbols are:

where:
M 0 = monetary base,18
M 1 = narrow definition of money,
M 2 = broad definition of money,
C = currency in the hands of the (non-bank) public,
BR = commercial banks’ reserves (in currency held by the banks or
deposits with the central bank),
D = demand (checkable) deposits of the public with the commercial banks,
and
S = savings deposits of the public with the commercial banks.
No matter whether the money supply in the economy is defined as M 1 or
M2 or some broader measure, there are three main participants in its

297
determination. They are:
1. the central bank, which determines the monetary base and the reserve
requirements for the commercial banks, and sets the discount rate at
which it lends to the commercial banks,
2. the public, which determines its currency holdings relative to its demand
deposits, and
3. the commercial banks, which, for a given required reserve ratio,
determine their actual demand for reserves as against their demand
deposit liabilities.19
There is considerable interaction among the behavior of the central bank,
the public and the commercial banks in the money supply process. This
interaction becomes of the greatest importance in studying the behavior of
the central bank, which must take into account the responses of the public
and the commercial banks to its own actions.
The relative importance of the major contributors to changes in the
money supply varies between the business cycle and over long periods.
Their contributions can be summarized as:
• The most important cause of the long-term (known as secular) growth of
the money stock is the growth in the monetary base.
• Over the business cycle, the fluctuations in the currency ratio (C/M 1)
tend to have relatively large amplitudes, and can account for as much as
half of the fluctuations in the growth rate of M 1.
• Over the business cycle, fluctuations in the monetary base are another
major source of the fluctuations in the growth rate of the money stock.
Over the business cycle, the reserve ratio (BR/D) can also fluctuate and be
a source of the fluctuations in the growth rate of the money stock. The
extent of this fluctuation varies among economies. In Canada, the reserve
ratio is now at very low levels (less than 1%), so that the fluctuations in it
are no longer a major source of fluctuations in the growth rate of M 1.
They play a more significant role in the USA and many other countries.
Extended Analysis Box 6.4: The Creation of Demand Deposits
Banks hold reserves to meet withdrawals by their depositors and any
reserve requirements imposed by the central bank. One part of these
reserves is held as currency in the banks’ tills and vaults. Another part is
held as a checking account with the central bank, to be drawn upon as
needed. Individual banks’ deposits with other banks are also part of their
reserves. However, in the aggregate over all banks, interbank deposits
cancel out and are not part of the reserves of the banking system.
Therefore, for the banking system as a whole, bank reserves are

298
currency held by the banks (not by the non-bank public) plus their
deposits with the central bank.
Assume that the banks hold reserves against demand deposits in the
proportion ρ, so that:

where ρ (Greek letter ‘rho’) is the reserve ratio BR/D. As mentioned

earlier, reserves are held for two reasons:
1. To meet a minimum reserve ratio set by the central bank. This ratio is
called the required reserve ratio. Some countries have now
eliminated such a requirement, while others maintain one. As an
example, Canada requires banks to have a zero or positive reserve
ratio at the end of each day, while the USA has a reserve requirement
ratio of approximately 9%.
2. To meet the net withdrawals (i.e., withdrawals less deposits) by
depositors on a continuous basis.
The reserve ratio ρ is usually greater than the required reserve ratio set
by the central bank for the banking system because the banks need to
hold extra reserves to meet withdrawals by the public from their deposit
accounts. If the commercial banks have reserves equal to BR, our
preceding equation implies that they can somehow create demand
deposits equal to:

This equation is the elementary deposit creation formula for the creation
of deposits by the banks based on the reserves held by them.
What is the amount of reserves available to the banks as a whole? To
answer this question, start with the identities:

Therefore, the amount of reserves held by the commercial banks is

determined by the monetary base M 0 (provided by the central bank)
less the public’s desire to hold currency. Given this availability of
reserves, banks create deposits equal to (1/ρ)BR. As an illustration,
suppose that ρ = 0.250. In this case, banks create $4 of demand deposits
for every dollar of their reserves.
In most countries, ρ is less than unity, so that they are said to possess a
fractional reserve system. In such a system, the commercial banks are
the major player in the creation of the money supply, even when the
initiative in changing the money supply rests with the central bank
through its manipulation of the monetary base.

299
6.14 A Common Money Supply Formula for M1
To derive the quantity of M 1, start with the accounting identities and
definitions:

where:
M 1 = narrow money supply,
M 0 = monetary base = BR + C,
c = currency ratio of the public,
ρ = reserve ratio of the commercial banks,
C = currency in the hands of the public,
D = demand deposits of the public in commercial banks, and
BR = commercial banks’ reserves.
The steps in the derivation are:

where [∂M1/∂M0] is the monetary base multiplier: it specifies the amount

of M 1 created by an increase in the monetary base by $1. The above
equation for M 1 separates its basic determinants into changes in the
monetary base and changes in the monetary base multiplier.
The dominant factor influencing the long-term (secular) growth in the
money stock is the growth in the monetary base. For the long-term
changes in M 1, changes in the two ratios contribute much less than the

300
changes in the monetary base. However, for cyclical movements in the
money stock, the changes in the currency ratio are an important element of
changes in the money supply. Changes in the reserve ratio have only a
minor impact in stable, developed economies, but they can shift very
significantly during a period of financial and economic crisis, as they did
during 2007-2010 in the USA.

6.14.1 The monetary base multiplier

The monetary base multiplier is the multiple by which a change in the
monetary base changes the money supply. Hence, for M 1, it is specified
by:

This multiplier is itself determined by the reserve ratio and the currency
ratio. Of these, the reserve ratio reflects the banks’ demand for reserves.
The currency ratio reflects the public’s behavior on its demand for
currency. Hence, the three main determinants of the money supply are: the
monetary base determined by the central bank, the currency ratio
determined by the public, and the reserve ratio determined by the banks.

6.14.2 Numerical examples

Suppose we have the following information: c = 0.2, ρ = 0.1, and $M0 =
100 million. What would be the values of the monetary base multiplier, M
1, C , and D ?

However, a reserve ratio of 10% is too high for many countries. Therefore,
for another example, let c = 0.2 and ρ = 0.02, so that ∂M 1/∂M0 = 1.2/0.22
= 11.411. Then, for M0 = $100 million, M 1 = $545 million. Comparing
the multipliers and the money supply for the two values of ρ, we find that
the lower the reserve ratio of banks, the greater is the monetary base
multiplier and the expansion in M 1.
Further, assume that the currency ratio rises to 0.3, while ρ = 0.1. In this
case, = 1.3/0.4 = 3.25 and M 1 = $325 million. Hence, an increase
in the currency ratio decreased the monetary base multiplier and the
money supply.

301
 It is left to the students to calculate the currency holdings and demand
deposits in each of the above cases, and to provide the answer to the
question: what do they show about the effects of increases in the currency
ratio and the reserve ratio on currency holdings and demand deposits?

6.15 Conclusions
• In the money market, the two main elements of the demand for money
are the transactions and the speculative ones. If the central bank pursues a
policy of targeting the money supply, the nominal money supply is
assumed to be exogenous.
• The two special cases of money demand are the quantity theory and the
liquidity trap. These do not apply during normal conditions in the
financially developed economies.
• One technique of analysis used in short-run macroeconomics with an
exogenous money supply is that of grouping the relationships of the
commodity and the money markets into the IS-LM format and the IS-LM
diagram.
• Under an exogenous money supply, the intersection of the IS and LM
curves shows the real aggregate demand for commodities at the pre-
existing price level for which the LM curve was drawn.
• Increases in the policy variables of government expenditures and the
money supply raise real aggregate demand (at the pre-existing price
level). Increases in taxes reduce it.
• The division of an increase in aggregate demand between increases in the
price level and the quantity produced of commodities also requires the
specification of the aggregate supply curve.

KEY CONCEPTS

An exogenous money supply

Money stock
The demand for money
Transactions demand for money
Speculative demand for money
The LM relationship/curve
The liquidity trap
The quantity theory range
The determination of aggregate
demand by IS-LM analysis

302
Monetary base multiplier
Open market operations
Reserve ratio, and
Currency ratio.

CRITICAL CONCLUSIONS

• An increase in the demand for money shifts the LM curve to the left.
• An increase in the money supply shifts the LM curve to the right.
• An increase in the money supply is an expansionary monetary policy.
• In the normal context of the economy, both expansionary fiscal or
monetary policies increase real aggregate demand at the existing price
level.
• The intersection of the IS and LM curves determines the amount of
commodities demanded at the price level for which the LM curve is
drawn. It shows neither the supply of commodities nor its ability to
determine the quantity of commodities that will be produced or the price
level at which they will be sold.

Appendix

Determination of the LM curve

This appendix assumes an exogenous money supply. That is:

where M is the exogenously determined money stock. The assumption on

the money demand function is:

Money market equilibrium requires that:

which gives,

Note that while the nominal stock of money is exogenously given and in
the control of the monetary authorities, the real value of that stock, (Md/P),
depends upon the equilibrium price level P in the economy and

303
incorporates choices made by the public, including those on its demand for
money.
The preceding equation is called the LM equation and its corresponding
curve is known as the LM curve. Its usual form is obtained by rearranging
the above equation as:

This LM equation shows that y depends on positively on R and M, but

negatively on P.

Derivation of aggregate demand for the closed economy

from its IS and LM equations
Using the IS equation for the closed economy from Chapter 4, the above
LM equation and the Fisher equation relating the real and the nominal
interest rates, we have the complete model for determining aggregate
demand.
This model is:
IS equation for the closed economy:

LM equation:

Fisher equation:

The Fisher equation introduces an additional variable πe, so that its

determination needs to be specified. The simplest assumption for the
mathematical derivation is one where πe = 0.20 We will make this
assumption, so that for the following derivations, R = r.
To combine the IS and LM equations, first solve the LM equation for R
in terms of y and P , which gives:

Using the assumption that R = r, we replace R in the money demand

equation by r. This turns Eq. (21) into:

304
Now substitute this value of r (= R, under our assumption that πe = 0) in
the IS equation. The resulting AD equation for the closed economy is:

where:

This preceding AD equation specifies the aggregate real demand for

commodities as a (negative) function of the price level P . Since P is in the
denominator on the right side, an increase in P reduces yd, so that ∂yd/∂P
< 0. That is, the aggregate demand (AD) for commodities is inversely
related to its price — which is the price level — so that we now have the
equation with the usual form of a downward sloping demand curve.
For diagrammatic analysis, the above AD equation shows those
combinations of (yd, P) which simultaneously maintain equilibrium in the
commodity and monetary markets.
Intuition
An increase in the price level reduces aggregate demand by reducing the
real money balances (M/P) available in the economy. A reduction in real
balances held by individual consumers implies that they cannot finance
their previous level of purchases. In order to do so, they will try to sell
some of their bonds. This will lower bond prices and raise their return —
which is the interest rate. As the interest rate rises, investment falls, which
reduces the expenditures on (demand for) commodities. For equilibrium to
be maintained, this fall in expenditures reduces real income y.
Simplifying the AD equation for further analysis
We can rewrite the closed-economy AD equation as:

There are only two endogenous variables, y and P , in this equation. There
are also the policy variables of M , g, and the tax parameters (t0 and t1).
Note that the interest rate is not a variable in this equation since it was
dropped out in combining the IS and LM equations. We can now easily
derive the effect of changes in the monetary and fiscal policy variables on

305
aggregate demand.
We can plot this AD equation in a diagram with y on the horizontal axis
and P on the vertical one, as in Figure 6.A.1. Since P is in the denominator
in the above equation, an increase in P reduces yd, which explains why the
AD curve is downward sloping, as specified in Chapter 1.

Determination of the long-run price level under an

exogenous money supply
Figure 6.A.1 shows the downward sloping AD curve. From the AD
equation, we find that an increase in the money supply and/or in the fiscal
deficit will shift the AD curve to the right from AD0 to AD1.
Figure 6.A.1 also shows the determination of the price level in the long
run, that is, with a vertical LAS curve. The shift of the AD curve to the
right from AD0 to AD1 increases demand at the existing price level P0 to
the point d. This demand exceeds the supply of commodities, which
remains at y f0, so that the market pushes up the price level from P 0 to P 1.
However, this price increase does not change the long-run output, which
stays at yf.

Derivation of aggregate demand for the open economy from

its IS and LM equations
Using the IS equation for the open economy from Chapter 5, the above
LM equation and the Fisher equation relating the real and the nominal
interest rates, we have the complete model for determining aggregate
demand in the open economy. This model is:
IS equation for the open economy:

Assuming that πe = 0 in the Fisher equation, we arrive at the conclusion

that R = r.

306
 To combine the IS and LM equations, first solve the LM equation for R
in terms of y and P , which gives:

Using the assumption that R = r, we replace R in the money demand

equation by r. This turns Eq. (26) into:

Now substitute this value of the interest rate in the IS equation. The
resulting AD equation for the open economy is:

REVIEW AND DISCUSSION QUESTIONS

1. What assumptions are made for the money supply in deriving the LM
curve?
2. Suppose that firms and consumers increase their demand for real
balances because of an increase in the income elasticity (use m y as the
approximation for this) of this demand. Show its effect on the LM
curve. Is there a change in the slope of the LM curve?
3. What assumptions are made for the money demand function in deriving
the normal shape of the LM curve? Show diagrammatically the normal
ranges of the demand for money curve and the LM curve for a modern
economy with well-developed financial markets?
4. Define the quantity theory (classical) range of the demand for money
and show it diagrammatically by its versions of the demand for money
curve and the LM curve.

307
5. Show the liquidity trap ranges of the demand for money curve and the
LM curve? What justifies the liquidity trap? Discuss the likelihood of its
occurrence for significant periods in modern economies.
6. Can we specify from the IS-LM analysis (i.e., without the commodity
supply analysis) the impact on real output of changes in the money
supply?
7. Using the IS-LM diagram, analyze the effects of a contractionary
monetary policy (i.e., a decrease in the money supply) on the interest
rate and aggregate demand in the economy.
8. Using the IS-LM diagram, analyze the effect of a contractionary fiscal
policy (i.e., a decrease in government expenditures or increase in taxes)
on the interest rate and aggregate demand in the economy.
9. If the central bank uses the money supply as its operating target, can it
predict its impact on the interest rate? Or should it try to set the interest
rate independently of the money supply? Discuss.
10. It is often contended that governments in LDCs have no choice but to
fund at least some of their deficits by increasing the monetary base.
Why and how does this happen? What are the effects of doing so on
aggregate demand? Illustrate your answer by appropriate diagrams.
11. “While central banks in financially developed economies may be able
to effectively control aggregate demand by only using the interest rate
as their operating target, those in LDCs with under-developed financial
markets need to use both the interest rate and money supply as their
operating targets”. What is the nature of the economies that justifies this
argument?

ADVANCED AND TECHNICAL QUESTIONS

T1. The following equations describe the commodity market of an open

economy:

Additional information: P = 2, P F = 4, p = 4 (i.e., the exchange rate

has been fixed by the central bank at this level). [Hint: use this information
to first derive the real exchange rate. Substitute this value in the export and

308
import equations. Chapter 5 can provide the guidance on this.]
(a) Derive the IS equation? [Do not use the IS equation formula for this
answer. Use step-by-step calculations. Start by calculating disposable
income. Noting that xc and zc depend on ρr, calculate ρr in terms of P .
Then use the equilibrium condition y = e for your further steps.]
(b) Assuming r = 2, what is the value of y?
(c) Assuming r = 2, if the government increases its expenditures to 1000,
what is the value of y?
(d) What is the fiscal multiplier?
T2. The following equations describe the money market of the economy:

(a) Derive the LM equation?

(b) Given the commodity market model of question T1 and the money
market model of this question, derive the aggregate demand equation.
(c) What is the autonomous investment multiplier for aggregate demand?
T3. Given the preceding information on the commodity and money
markets and if the money supply were kept at 1000, derive the domestic
interest rate determined by the economy? (Hint: this interest rate is
determined jointly by the domestic commodity and money markets, i.e., by
the intersection of the IS and LM curves. It will depend on y.)
1If needed, it uses its control over the interest rates as an instrument to
support the desired money supply.
2The actual determination of the money supply by the central bank is
somewhat more complicated. The central bank directly controls only the
monetary base, but can manipulate it to determine the money supply.
3When the return on bonds and stocks becomes risky because of the
volatility of bond and stock prices, investment brokers usually advice
holding larger percentages of portfolios in money and treasury bonds, etc.
4Advanced macroeconomic analysis also considers two other motives for
holding money balances. These are the precautionary motive and the
buffer stock motive. For our analysis, these will be encompassed under the
transactions motive.
5This will occur if the valuations of bonds and stocks in the bond and
stock markets change more or less than the commodity price level.
6(mRR) is the part of the financial portfolio which is held in bonds, so that
speculative money demand is obtained by subtracting this part from the
size of the financial portfolio. As the interest rate rises, investors will
increase their demand for bonds, so that m R ≥ 0. Doing so decreases their

309
demand for money balances.
7In such a context, land and houses often become the alternative to holding
money, especially over long periods.
8Even if old loans were rolled over, there was an increase in interest rates
to such borrowers, often accompanied by a decrease in the quantity
provided (i.e., quantity rationing). New loans to the private sector were
hardly being given. However, such lack of confidence in lending to firms
did not apply to government Treasury bills and bonds, so that they
continued to be acceptable alternatives to holding money. Therefore, there
was a switch from loans to the private sector to loans to the government.
9Since ∂y/∂R = m /m > 0.
R y
10If the money supply decreases, the public tries to replenish its real
balances by selling bonds, thereby driving down their prices and raising
the interest rate.
11Note, however, that the price level is not an exogenous variable, so that
a change in it must be determined and explained by reference to changes in
some exogenous variables.
12This is the IS equation without ‘Ricardian equivalence’, which assumes
that private saving rises, and private consumption is cut back, exactly by
the amount of any increase in the fiscal deficit. We do not make such an
assumption in this chapter or book. A brief introduction to this concept can
be found in Chapter 11. The form of the open economy IS equation that
assumes Ricardian equivalence is different from the above equation but is
not presented in this book.
13Note that an increase in aggregate demand due to an increase in the
money supply or an increase in government expenditures has to be
interpreted as a real increase in the aggregate demand for commodities at
the existing price level.
14For the fixed exchange rate case, the Mundell–Fleming model showed
that policy is effective in changing aggregate demand–and, therefore,
output under the horizontal supply curve assumption–while fiscal policy is
ineffective. Chapter 13 will present our analysis of the fixed exchange rate
case.
15In Canada, the shifting of government deposits between the central bank
and the commercial banks is also used as a substitute for open market
operations.
16Besides security of deposits, there is also the question of equity in the
distribution of government deposits among the commercial banks. This
can be ensured by distributing the government deposits in proportion to the

310
individual banks’ holdings of private deposits. Note that commercial banks
increase their revenues by investing increases in government deposits with
them in interest-bearing bonds.
17In most countries, the required reserves have to be held by the bank in
currency or in deposits with the central bank. These deposits normally do
not pay interest.
18It is also called the reserve base or high-powered money.
19These are not the only actors in the money supply process. In particular,
in open economies, the balance of payments surpluses (deficits) of a
country can increase (decrease) its money supply. This relationship
between the balance of payments and the money supply is discussed in
Chapter 13.
20Another simple alternative is to assume that it is exogenously given.

311
 CHAPTER 7
Full-Employment Output and the
Natural Rate of Unemployment

The study of real output in the economy requires the analysis of

the labor market and the aggregate supply of commodities for the
economy. The main variables of interest in this analysis are the
long-run levels of employment and output, and the natural rate of
unemployment. Their derivation is distinct from that of the short-
run behavior of the economy (covered in Chapter 8) and of the
disequilibrium behavior of the economy (covered in Chapter 9).
The long-run behavior of the economy, captured under the
notions of the full-employment level of output and the natural rate
of unemployment, is presented in this chapter.

The counterpart of the aggregate demand for commodities is their

aggregate supply. This chapter presents the derivation of the long-run (LR)
equilibrium levels of employment and output. This expression is usually
shortened to the ‘long-run levels’, thereby dropping the word ‘equilibrium’
but implying it nevertheless. The LR aggregate output in the economy is
also known as the full-employment output.
As Chapter 1 explained, the full-employment output in macroeconomics
is to be interpreted as the optimal output that can be produced on a
sustainable basis over long periods from the efficient use of the economy's
available factors of production and its available technology — given the
economy's organizational and market structures,1 the wishes of the owners
of the factors of production,2 as well as its economic, social, and political
structures. This full-employment output is often more loosely referred to
as the ‘potential long-run output’ that can be maintained with the
economy's current resources and structures. In more formal terms, the full-
employment level of output is defined as the long-run equilibrium level of
output, where the term ‘long run’ refers to that analytical period when
there are no rigidities, adjustment costs or expectational errors.

312
 The long run (equilibrium) requires the assumptions:
• If there exists uncertainty, the expected values of the variables are
identical with the actual values.
• There are no labor contracts between firms and workers, and nominal and
real wages adjust instantly and fully to reflect market forces.
• There are no costs of adjusting prices or employment or output, so that
the adjustments in these and other variables to their desired levels are
instantaneous.3
• There exists equilibrium in all markets.
Given these assumptions, the economy's resulting employment level is
said to be the full-employment one and its output is said to be full-
employment output. Our symbol for this output level is yf.
Alternatively, if there do exist rigidities or adjustment costs in the
economy, the long run is that analytical period by which, following any
shocks to the economy: (a) any discrepancies between the expected values
of the variables (aggregate demand, prices, etc.) and the actual values have
been eliminated, (b) all contracts based on such a discrepancy have
expired, (c) all adjustments by markets and economic agents
(households/workers, firms, and government) have been completed, and
(d) all markets have reestablished equilibrium.
Note that the actual output in the economy can differ from its full-
employment level because of the existence of rigidities, adjustment lags,
expectational errors, and other factors. These can make the short-run
output to be greater or lesser than its full-employment level.4

7.1 The Production Function

In industrialized economies, capital and labor are the dominant inputs in
production, while land plays only a minor role and is normally not
included in macroeconomic analysis. With this assumption, the remaining
inputs in the production of commodities are labor and capital. The
production function for the economy — as represented by that for ‘the
representative firm’ — can then be written as:

where:
y = output,

313
 K = capital stock,
n = labor employed,
yn = marginal product of labor (MPL),5 and
yK = marginal product of capital (MPK).6
In the short term, the flows of physical capital, labor and technology
among countries are likely to be relatively limited, or limited in their
impact on the economy, so that they are ignored in the short-run open
economy analysis.7 Further, since our analysis is short term, we assume
that the domestic labor force, capital stock, and technology are constant.8
With technology held constant for short-run analysis, the production
function remains unchanged in going from the closed economy to the open
one.
The stock of physical capital is also held (i.e., assumed to be) constant in
the short-run macroeconomic analysis,9 so that , where the line over
the symbol indicates ‘constancy’ or ‘exogeneity’. Hence, we have:

With this modification, labor is left as the only variable input, so that we
can rewrite the production function as:

We assume that the marginal product of labor (MPL) is positive and

diminishing, i.e., successive increments of labor yield smaller and smaller
increments of output.
Fact Sheet 7.1: Diminishing Marginal Product of Labor
This Fact Sheet illustrates the shape of the output curve usually assumed
in economics and relates it to the marginal product of labor. Marginal
product is given by the slope of the production function at any one point.
The slope is assumed to be always positive, meaning that marginal
product of labor is always positive and that an increase in labor supply
will always produce an increase in output. Notice the slope at point C is
lower than at point B and lower still than at point A. This is indicative of
the diminishing marginal product of labor, since at higher levels of
labor, additional units will produce less and less output.

314
 Figure 7.1a shows the usual shape of the production function. Its positive
slope indicates that it has positive MPL as employment increases. Its
concave shape reflects diminishing MPL. The shape of the MPL curve
corresponding to the shape of the production function in Figure 7.1a is
shown in Figure 7.1b. This shape is often simplified in textbook
macroeconomics to the downward sloping straight line shown in Figure
7.2a.

315
 The general shape of the MPL curve is shown in Figure 7.2b. Along this
curve, the MPL first increases and then decreases. The production function
corresponding to this shape of the MPL curve has initially increasing
MPL, followed by a segment with diminishing MPL. Such a production
function is often used in microeconomics, and we will use it in the growth
theory Chapters 14 and 15, but do not show it diagrammatically in this
chapter.
Mathematical Box 7.1: Examples of Production Functions and the
Derivation of the Marginal Product of Labor
The general procedure for deriving the MPL from the production
function
Suppose we are given a production function of the form:

where α0, α1, and β are parameters. The procedure for deriving the MPL
(which in calculus corresponds to taking the first derivative of y with
respect to n ) is:

(a) Take the first term (α0n) on the right-hand side of the equation. Note
that the exponent on n is 1. Multiply the coefficient in this term by the
exponent on n, and change the exponent on n by subtracting 1 from the
initial exponent. This gives (α0 • 1n1-1), which simplifies to α0. This is
the increase in output contributed by the first term.
(b) Take the second term (—α1nβ ) on the right-hand side of the
equation. Note that the exponent on n is f. Multiply the coefficient α1 in
this term by the exponent on n, and change the exponent on n by
subtracting 1 from the initial exponent. This gives [− α1 • βnβ−1 ]. This
is the increase in output contributed by the second term. [If there are

316
more than two terms on the right side, follow these steps for each term.]
(c) Add the increases in output contributed by the two terms to arrive
at the MPL for the above production function. This gives:

For the application of this procedure, we consider three production

functions and derive their corresponding marginal product of labor.
i.A linear production function:

where α is a constant. The marginal product of labor for this

production function is:

so that the marginal product of labor is constant for all levels of

employment. This violates the usual assumption of diminishing MPL, so
that it is not appropriate to assume a linear production function.
ii.A quadratic production function:

where α 0 and α 1 are constants. The MPL for this production function
is:

so that the MPL varies with employment. The MPL is positive but
diminishing, as required.
For a numerical illustration of the quadratic production function, let α0
= 1,000 and α1 = 0.75. Then the production function is:

and its MPL is:

This is plotted in Figure 7.2a and shows a downward-sloping straight

line for the MPL. Note that the MPL declines as employment increases.
The quadratic production function has the severe disadvantage that the
exponent on the second term (−α1n 2) is 2, which is much too high. This

317
 exponent should be about 0.6 or 0.7. However, using such values makes
for greater mathematical difficulty in solving the model, so that the
exponent of 2 is used for mathematical convenience — not realism.
iii.The Cobb-Douglas production function [optional]
A Cobb-Douglas production function, with labor and capital as inputs,
is specified by:

where A and α are constants. This function has the special property
that the income shares of labor (at α) and of capital (at 1 — α) are
constant. With capital constant at , rearrange this function as:

and write it as:

where B equals (A (1—α)) and is treated as a constant in short-run

analysis. The MPL for this production function is:

so that the MPL declines as employment increases. For many

countries, the value of α lies in the range 0.6 to 0.7, so that their MPL is
positive but diminishing, as required. If this MPL curve had been plotted
(but was not), it would be a downward sloping concave curve rather than
a straight line.
iv. The general shape of the MPL curve
A still more general shape of the MPL curve occurs if the MPL
initially increases and then decreases. This is often the shape shown for
the MPL curve. It is drawn in Figure 7.2b, and shows a concave curve
with an initial upward sloping segment followed by a downward-sloping
one. In this case, the profit-maximizing firm would employ workers up
to the point where the MPL equals the wage rate on the downward —
not the upward — sloping part of the MPL curve. This is explained in
the next section.

7.2 The Labor Market

The specification of the labor market requires specification of the demand

318
and supply functions of labor. The demand function for labor is derived
from profit maximization by firms and the supply function for labor is
derived from the utility analysis of households/workers. We here present
the simplified functions used in standard macroeconomic models. These
assert that both the demand and supply of labor depend only on the real
wage rate.

7.2.1 Demand for labor

It is assumed that the representative firm maximizes profits in perfectly
competitive markets and takes the price of its product and the nominal
wage rate W as set by the market. The former is represented in aggregate
analysis by the price level P . The microeconomic theory of perfect
competition implies that the profit-maximizing firm would employ labor
until the value of its marginal product (i.e., P times MPL) equals its
nominal wage rate. The reason for this is that if the value of the MPL is
more than the nominal wage, the firm will increase its profits by hiring an
additional worker, since that worker's output contributes more to the firm's
revenue than his wage. Conversely, if the value of the MPL were less than
the nominal wage, the firm would increase its profits by reducing its
employment. Therefore, profit maximization implies the equality of the
value of the MPL to the nominal wage rate. The firm achieves this by
choosing the appropriate level of employment. That is, employment will
be carried to the point at which

where:
P = price level and
W = nominal wage rate.
Dividing both sides of this equation by P implies that, in perfect
competition and in the aggregate, profit maximization requires that the
representative firm employ labor up to the point where the real value of its
marginal product equals its real wage rate. That is, MPL = W/P = ω, so
that:

where ω is the real wage rate and the MPL depends on the level of
employment n . Solving the above equation for employment n , and
designating this value as the demand for labor n d by firms, we have:

319
Since ω = W/P , the demand for labor can also be written as:

Diagrammatic analysis
The diagrammatic derivation of the labor demand curve is shown in Figure
7.2a for the linear MPL and in Figure 7.2b for the more general case where
the MPL first increases and then decreases. To maximize profits, the firm
increases employment until the MPL equals the market-determined real
wage rate. Figure 7.2b shows two points of intersection at a and b between
the market-determined wage w0 and the MPL.10 Of these points, b is the
point of profit maximization at ω0, so that the firm would demand n0
workers.11
In Figures 7.2a and 7.2b, the firm's initial labor demand is n0 at wage ω0.
If the wage rate rises to ω1, the firm's labor demand would fall to n 1
workers. The intuitive explanation for the fall in labor demand is that a rise
in the wage rate makes labor more expensive, so that firms will hire less
workers. Therefore, in Figures 7.2a and 7.2b, as the wage rate fluctuates
up and down, it will trace out the demand for labor by the downward
sloping part of the MPL curve, so that the labor demand curve is identical
with the MPL curve if the upward-sloping part of the MPL is omitted. This
labor demand curve is shown by the curve marked nd in Figure 7.3. Hence,
the demand for labor is negatively related to the wage rate: if the latter
rises, the former falls.

7.2.2 Supply of labor

In microeconomics, an individual worker's supply of labor is derived by
constrained utility maximization for a given real wage rate determined by
the market and his time constraints (24 hours a day to be divided into work
and leisure). For a worker, the real wage rate is the marginal return to an
extra hour of work, as well as the opportunity cost of taking an extra hour
of leisure. An increase in the wage rate causes the return to work to rise.
Conversely, it causes the opportunity cost of leisure to rise. Therefore, at
higher wage rates, the worker will substitute more work for leisure and
increase the supply of labor. As the wage rate continues to rise, the
increase in the labor supply is likely to become lesser and lesser. Box 7.1
discusses the possibility that the labor supply of a given worker may

320
eventually begin to fall, as in Figure 7.4b.12 Short-run models assume that
this fall does not happen. The justification for this assumption is given in
Box 7.1.

The preceding paragraph implies that the worker's supply of labor

depends upon the real wage rate, which is the marginal return to an extra
unit of work, as well as being the marginal opportunity cost of leisure. For
the economy, as the wage rate increases, some of the employed workers
may increase their labor supply by working longer hours, some of those
about to retire may postpone their retirement and continue to work, and
others who were not in the labor force (students, housewives, etc.) may be
induced by the higher wage to enter it. In addition to the real wage rate w,
the economy's aggregate supply of labor over all workers would also
depend on the labor force. Therefore, the labor supply function is:

where:
ns = labor supply,
ω = real wage rate, and
L = labor force.
The labor force L depends on the population and the labor force
participation rate. The latter depends on the age, gender, geographic, and
skill composition of the population. This participation rate can be altered
by both social and economic changes, e.g., the labor force participation

321
rates of women increased very considerably in the 1960s and 1970s in
most countries. However, such shifts are treated as exogenous in the short-
run macroeconomic analysis.
The cross-border flows of labor can be significant and endogenous for
some countries — i.e., depend upon their unemployment rates and real
wages — so that, for such countries, the labor force is not really
exogenous.13 However, with immigration strictly controlled by most
countries, the annual inflows and outflows of labor to most economies are
very small relative to their existing labor force, so that their labor force is
not significantly affected by such movements in the short run. For such
economies, the labor force is taken to be exogenous in the short run.
Therefore, the labor supply function adopted for our open economy
macroeconomic model is the same as it would be if the economy were
closed. This function is specified as:

so that the supply of labor n s depends positively on the real wage rate ω.
Both n s and ω are for the current period. Since ω = W/P , we can also
write the supply of labor as:

Note that the supply of labor depends upon the real rather than the nominal
wage. That is, workers are free from price and inflation illusion. This
illusion is the distortion caused when money wage rates and the prices of
commodities rise by identical proportions but the workers, looking at the
rising nominal wage rates, believe that they are better off, although the
purchasing power of wages has remained unchanged. A proportionate
increase in the nominal wage rate and the price level would leave the real
wage unchanged, so that, without price illusion, the supply of labor also
does not change.
Figure 7.4a shows the likely short-run shape of the labor supply curve. It
shows that an increase in the real wage rate increases labor supply. We
have drawn the labor supply curve with an increasing slope at higher wage
rates since, as wages increase, the increases in the labor supply tend to
become lesser and lesser. This is especially likely to occur as the labor
supply becomes closer to the labor force level. Since the labor supply
curve is upward sloping, it is often simplified for analytical convenience
by a straight line with a positive slope.
Extended Analysis Box 7.1: The Intertemporal Analysis of Labor Supply

322
Economic theory implies that if the choices over time are considered (as
in intertemporal analysis), the individual is likely to substitute between
work in the current period and future periods — if the employment
patterns of firms allow such a choice. In this analysis, the current
period's supply of labor depends on the current wage rate and the present
discounted value (PDV) of the future wage rate. In the two-period
analysis for periods t and t + 1, the current wage will be designated as
ωt, and the PDV of the future wage rate is ωt+1/(1 + rt).
In the aggregate over all workers in the labor force, the supply of labor
also depends on the labor force. Therefore, the supply function for labor
can be expressed as:

where:
n t = labor supply in the current period t ,
ω t = real wage rate in t,
ωt+1 = real wage rate in t + 1 (i.e., future wage rate),
r t = real interest rate in t, and
L = labor force.
In this labor supply function, an increase in the expected future (but not
the current) wage or a decrease in the interest rate will induce the
individual worker to reduce labor supply and increase leisure in the
current period, while increasing labor supply and reducing leisure in the
future. However, an increase in both the expected future and the current
wage does not change the market rate of substitution between work in
the current and future periods. Therefore, a temporary increase in the
real wage will bring about a stronger increase in labor supply than a
permanent increase.
Note that this tilt in the hours worked from the present to the future, or
vice versa, requires the flexibility in working hours to be exercised at the
employee's (not the employer's) option. For most workers in the
economy, such a tilt is not an option allowed to them by their employers,
at least not in the short run, so that the intertemporal substitution of
labor may not be significant. However, there may be some workers such
as self-employed ones, students working part time, those close to
retirement, etc., who may possess some flexibility.
Empirical evidence usually reports that, over a quarter or a year,
changes in the expected real wage lead to very limited changes in the
hours worked, so that many economists do not consider the

323
intertemporal substation of labor to be really significant. However, some
other economists dispute this assessment of the empirical evidence and
consider the dependence of current labor supply on future wage rates
and the interest rate to be significant enough to use it in their explanation
of the variations in unemployment over the business cycle.14
The Backward Bending Labor Supply Curve Over Long Periods
For many jobs in the 19th and early 20th centuries, work hours were as
long as 60 hours or more per week. As wages and the standards of living
rose in the 20th century, work hours per week fell. This phenomenon is
the outcome of two effects of higher wages on labor supply. One of
these is the substitution effect, which specifies the substitution of work
for leisure as wages rise and makes leisure relatively more expensive.
The other is the income effect, which specifies that as wages rise,
incomes rise and workers can afford to purchase more commodities as
well as more leisure. In fact, increased purchases of entertainment goods
usually require more leisure to enjoy them. As the purchase of leisure
per week increases, the labor supplied per week falls. The overall effect
of rising wages on labor supply represents the opposing pull of the
substitution and income effects, which could produce an increase in
labor supply for some increases in wages but a fall for larger increases in
wages. This possibility is captured in the backward bending labor curve
shown in Figure 7.4b.
Short-run macroeconomic theory usually simplifies by assuming that
the net effect of rising wages is to increase the labor supply, i.e., the
labor supply curve is upward sloping. This is a reasonable assumption
for relatively small percentage increases in real wages that occur over
short periods such as the business cycle. However, it cannot be taken for
granted for large increases in wage rates over very long periods. In fact,
the very long-term historical experience, starting with low wages and
incomes, has been one of a gradual decrease of work hours per week.
This pattern is also likely to occur in the very long term in those
developing economies that currently have very high work hours per
week.15 Whether this pattern will continue to hold for further increases
in wages in the high income, developed, economies remains to be seen.
If we look beyond the individual worker and focus on the family, the
social norms on work by all its members — including children —
become important. For given norms on the labor force participation rates
of men, women, and children, the long-term impact of higher wages
often means that the family as a unit decreases labor supply as wages
and incomes rise. For instance, in recent decades, on average, children

324
 have continued their education longer and reduced their work hours per
year. However, these norms can shift over time. One such shift has been
the dramatic rise in the labor market participation of females since the
1960s, which has led to a dramatic increase in the average number of
hours supplied per family.

7.3 The Long-Run Equilibrium Levels of

Employment and Output
The equilibrium level n* of employment is determined by the equality of
the demand and supply of labor. Hence,

where n is employment and the asterisk (*) indicates its equilibrium value.
Since this equation has only one variable, ω, solving it would yield the
equilibrium wage rate ω*. This wage rate, substituted in either the demand
or supply function, yields the equilibrium level of employment n *. This
level of employment substituted in the production function would yield the
equilibrium level of output y * for the economy. Hence, we have:

These equilibrium levels of output have been derived for the long-run case
where there are no adjustment costs, expectational errors or rigidities.
They are, therefore, the long-run equilibrium (LR) levels of employment
and output. To reflect this, we will write y * as y f and n * as n f.

7.3.1 The impact on long-run output of an increase in

325
 the price level or inflation rate
Suppose that the price level or inflation increases for some reason. These
variables have already been shown not to affect the long-run output level,
so that the increase in them will not change output in the long run.

7.3.2 The diagrammatic analysis of employment and

output
Figure 7.5a plots the demand and supply functions of labor, with the usual
slopes for the demand and supply curves. LR equilibrium occurs at (n f,
ωf).
Figure 7.5b plots the production function: it plots output y against
employment n and is labeled as y (output). This curve has a positive slope
and is concave, representing the assumption of the diminishing marginal
productivity of labor. The equilibrium level of employment n f determined
in Figure 7.5a is carried over into Figure 7.5b and implies the LR
equilibrium — and profit-maximizing — output yf.
Suppose the price level increases. It does not shift any of the curves in
these diagrams, so that the LR levels of ωf, n f, and y f will not change.
Therefore, if we plot y f against P , as in Figures 7.5c and 7.6b, we would
have a vertical curve. This curve was called the long-run supply (LRAS)
curve in Chapter 1 and is shown by LRAS0 in Figure 7.5c.

The effect on output of an improvement in labor productivity

Suppose that an improvement in the productivity of labor or an increase in
labor supply increases the LR output of the economy, thereby shifting
LRAS curve in Figure 7.5c to the right from LRAS0 to LRAS1. Hence, an
increase is labor productivity increases the long-run level of output but
decreases the price level.

7.3.3 The impact of aggregate demand on long-run

output
Chapter 5 presented the analysis of aggregate demand under an interest
rate target. The open economy AD equation derived in that chapter was:

where the symbol α stands for the (autonomous investment) multiplier

326
and has the meaning:

The meanings of the symbols are as given in Chapters 4 and 5. This

aggregate demand equation implies that the aggregate demand curve AD
has a negative slope, as shown in Figure 7.6.
The pursuit of fiscal policies
The IS-IRT analysis in Chapter 5 showed that an increase in government
expenditures increases aggregate demand and shifts the AD curve to the
right, as shown in Figure 7.6. With output at the full-employment level, the
increase in aggregate demand will cause the price level to rise, but without
any increase in output.
The pursuit of expansionary monetary policies
A cut by the central bank in its target rate constitutes an expansionary
monetary policy. Suppose the bank pursues such a policy. This would
increase investment and aggregate demand, as shown by the rightward
shift of the AD curve in Figure 7.6. This would create excess demand. Its
effects will be similar to the above ones for an expansionary fiscal policy:
i.e., increase demand and lead to a price level increase, but without any
increase in output.

7.3.4 The ineffectiveness of monetary and fiscal policies

in LR equilibrium
Note that the LR levels of employment and output do not depend on any
variables other than encapsulated in the production function (which
determined labor demand) and the labor supply function (which depend on
workers’ work-leisure preferences). Changes in aggregate demand, prices
and inflation do not change the production function and work-leisure

327
preferences, so that they do not affect the LR employment and output.
Since the impact of monetary and fiscal policies occurs through aggregate
demand, these policies cannot change LR employment and output. This is
an important conclusion. It asserts the uselessness — and therefore, the
inadvisability of pursuing — monetary and fiscal policies for changing LR
output and employment. Its message is reinforced by the following
Mathematical Box 7.2.
Mathematical Box 7.2: The Derivation of the Demand for Labor,
Employment, and Output
The quadratic productionfunction for the representative firm in our
earlier example was:

Its MPL was:

Since the profit-maximizing firm equates the MPL to the marker-

determined real wage rate w, we have:

Solving for the demand for labor nd, we get:

If we let a0 = a0/(2ai) and ai = i/(2ai), we get the labor demand function:

Assume that the supply function of labor is:

In equilibrium, with nd = ns,

so that,

328
Substitute this wage in the labor demand (or labor supply) function to
derive the LR equilibrium level of employment as:

Substituting this level of employment in the production function

provides the LR level of output as:2

The role of aggregate demand in determining output and the price level,
an illustration
Since the long-run supply of output ys LR depends only on parameters
(whose values are exogenously given) and not on variables, its value can
be specified by a number (a constant). Designate this value as y0. That
is.

Let the AD equation be simply of the form:

This AD equation implies that a reduction in the target interest rate and
an increase in government expenditures increase aggregate demand.
Further, an increase in the domestic price level P decreases aggregate
demand (by causing consumers at home and abroad to switch from
domestic commodity purchases to foreign ones). However, we can see
from the LR aggregate supply equation that such increases in aggregate
demand cannot change LR output, which will remain at y0. Therefore,
monetary and fiscal policies cannot increase LR output.
If the long-run AS and AD equations above are combined by
substituting LR output into the AD equation, we get:

so that:

Hence, P would increase if the monetary policy variable rT was to

decrease, the fiscal policy variable g were to increase, or if output y

329
 decreased.
The impact of expansionary monetary and fiscal policies
Suppose the interest rate target is reduced by 10% from its previous
level. This reduction does not change any of the parameters in the LR
employment and output functions. Therefore, it does not change LR
output and employment. However, the price level would increase to
adjust the larger aggregate demand to an unchanged output.
Suppose government expenditures are increased by 10% from their
previous level. This increase also does not change any of the parameters
in the LR employment and output functions. Therefore, it does not
change LR output and employment, but would increase the price level.
This conclusion on the ineffectiveness of monetary and fiscal policies
to change LR output and employment holds not only for the illustrative
AD equation used above but also for any AD function. It is a very
important analytical and policy conclusion on the long-run performance
of the economy.

7.4 The Effects of an Increase in the MPL on

LR Output and Employment
Increases in the MPL occur because of an improvement in technology
and/or the increased use of capital and other inputs by firms. Assuming
that the increase in the MPL is accompanied by an increase in the average
productivity of labor, the output curve relating output to employment shifts
up. An increase in the MPL increases labor demand, which means that the
labor demand curve shifts up for any given level ofemployment. This
increases both the real wage and employment. LR output rises for two
reasons: higher employment (from the labor market) and higher labor
productivity (from the positive shift in the production function).
Conversely, decreases in the MPL occur because of a regression in
technology and/or decreased use/availability of capital and other inputs by
firms. Assuming that the decrease in the MPL is accompanied by a
decrease in the average productivity of labor, the output curve relating
output to employment shifts down. A decrease in the MPL decreases labor
demand, which means that the labor demand curve shifts to the left. This
decreases both the real wage and employment. LR output falls for two
reasons: lower employment (from the labor market) and lower labor
productivity (from the negative shift in the production function).

330
 While our general knowledge indicates that improvements in technology
occur frequently over time, it is harder to envisage a regression in
technological knowledge. It is much easier to envisage decreases in
capital, which can occur through the destruction of factories and other
capital facilities in a war, through the bankruptcies of firms because of a
fall in demand, and through depreciation or obsolescence of older
machines without adequate replacement by new ones.
A still more common possibility for a fall in the MPL is a reduction in
the availability of inputs other than labor and capital, or a rise in their
prices so that the firm reduces their usage. These inputs include raw
materials and intermediate goods. Among these, energy and imported
inputs are important elements. A rise in the price of oil, gas, or electricity
would lead to a reduction in their usage per unit of labor and reduce labor
productivity. Further, many developing countries often have to curtail
imports of inputs, including intermediate goods and spare parts for
machines, due to foreign exchange problems, which reduces labor
productivity. These reductions in labor productivity decrease labor
demand, the wage rate, employment, and output.

7.4.1 Diagrammatic analysis of the impact of an

improvement in technology on output and
employment
For this analysis, suppose that positive technical change in the economy
shifts the production function up from y0 to y1 in Figure 7.7a and the labor
demand curve right from n$ to nf in Figure 7.7b. The latter shift increases
LR employment to n1 which, in Figure 7.7a, increases LR output to y1. In
this case, equilibrium output has increased for two reasons: greater
employment and higher labor productivity. Therefore, as in Figure 7.5c,
the LRAS curve will shift to the right from y0 to y1. As shown in Figure
7.7b, the increases in employment and output are accompanied by an
increase in the real wage at full employment. This is related to the increase
in labor productivity (= y/n).

7.5 The effects of an Increase in Labor Supply

on L R Output, Employment and Price
Level

331
Labor supply depends not only on the wage rate but also on the labor
force, which depends on the population size and its participation rate. The
labor supply out of a given labor force depends on social norms, workers’
preferences between work and leisure, expected future wages and
consumption needs — which depend on family size and the number of
dependants, longevity, etc. — and other sources of income such as the
return on stocks and bonds. It also depends on factors such as
unemployment insurance benefits, welfare, and minimum income support
by the government, etc. Among examples of major increases in the labor
supply out of a given labor force is the increasing participation in the last
four decades of women in work outside the home, and significant
immigration levels.

An example of a decrease in the labor supply is the decreasing

participation rate of younger persons because of an increase in the number
of years spent on education, with larger proportions of high school leavers
continuing on to college and university education.
An increase in the labor supply shifts the labor supply curve to the right.
It decreases the LR real wage rate but increases employment — which
increases the LR output. As illustrated in the earlier Figure 7.5c, such an
increase in output will be accompanied by a decrease in the price level.
Conversely, a decrease in labor supply shifts the labor supply curve to the
left. This increases the LR real wage rate but decreases employment —
which decreases the LR output. This decrease in output will be
accompanied by an increase in the price level.
Comparing the effects of a productivity increase with those of an
increase in labor supply, increases in the labor supply or productivity raise

332
the LR levels of employment and output. But the former decreases the LR
real wage while the latter increases it.
Mathematical Box 7.3: The Effect of an Improvement in Technology
To see the effects of productivity improvements on employment and
output, assume that initially the production function is as given in the
Mathematical Box 7.2. Let there be a productivity increase such that the
production function shifts to:

whose MPL is:

That is, the MPL (compared with that derived in the Mathematical Box
7.2) has increased for any given level of employment by 5%. The profit-
maximizing firm would now equate this MPL to w, so that:

which can be solved for the new demand curve for labor. This would be:

If the original and the new labor demand curves are plotted, it will be
seen that the labor demand curve would have shifted to the right. Since
the increase in labor productivity shifts the labor demand curve to the
right for any given wage, it increases the equilibrium levels of both the
wage rate and employment. The latter increases the equilibrium level of
output through the production function. The derivations of the solutions
of this case for w, n, and y are left to the student.
The effect of an increase in labor supply
Suppose labor supply increases — due to a higher labor force
participation rate or immigration — by 5% so that the labor supply
function of the Mathematical Box 7.2 shifts to:

Assuming the initial production function, MPL, and labor demand to be

as in the Mathematical Box 7.2, the increase in labor supply would
increase LR employment to:

333
 Since employment increases, the production function implies that output
will also increase. But the LR real wage decreases to:

Diagrammatically, the increase in labor supply has shifted the labor

supply curve to the right, reduced w but increased n — which increased
y.

7.6 Conclusions on Changes in the Equilibrium

Levels of Employment and Output
The preceding analysis shows that the long-run equilibrium levels (wLR*,
nLR*, and yLR*) of the real wage rate, employment, and output depend on:
• the production function and
• the supply of labor.
Therefore, the factors that will change the LR values wLR*, nLR*, and
yLR* are shifts in the production function — with implied shifts in the
demand for labor — and in the supply of labor. As established earlier, an
increase in the MPL — which is illustrated in Mathematical Box 7.3 by
increases in a0 and/or a1 — will increase the demand for labor and thereby
increase the real wage rate, employment, and output. A decrease in the
supply of labor — in Mathematical Box 7.2, because of a decrease in b1 —
will increase the wage rate but reduce employment and output.
Conversely, an increase in the labor force participation rate would increase
labor supply, decrease the wage rate while increasing employment and
output.
These LR values wLR*, nLR*, and yLR* do not depend upon:
• the price level or inflation,
• the aggregate demand for commodities, and
• monetary and fiscal policies.
Therefore, shocks to the economy which bring about shifts in aggregate
demand and the price level will not alter the LR values wLR*, nLR*, and

334
yLR*. Among these shocks are changes in the policy variables, such as the
target interest rate, money supply, and fiscal deficits, which are in the
overall macroeconomic model but do not appear as arguments of the
production function and the supply function of labor. This is a very strong
implication for the equilibrium states of the above model. It implies that
the aggregate demand policies such as monetary and fiscal policies cannot
affect the LR levels of wages, employment and output in the economy.
Similarly, the levels or shifts of consumption and investment do not alter
the LR levels of output, employment, and real wages, but will change the
price level.

7.7 Full Employment and the Full-

Employment Level of Output: Definitions
and Conditions
‘Full employment’ was defined in Chapter 1 and earlier in this chapter as
the employment level that exists under certainty and in the absence of
adjustment costs and lags, when all the (qualified and in the right
locations)16 workers who want jobs at the existing wage rate are employed
while the firms can get all the workers that they want to employ at this
wage rate. That is, full employment corresponds to the long run (i.e., once
all adjustments have been completed) equilibrium level of employment
under certainty. To reiterate, the conditions for the existence of full
employment are:
• The absence of uncertainty about demand and prices. Full employment
would also occur in the presence of uncertainty if the expected prices and
expected demand corresponded to the actual ones — so that there were
no errors in expectations.
• The absence of adjustment costs. Examples of adjustment costs are those
of adjusting employment, output, and prices. Alternatively, if there are
adjustment costs, as is normally quite likely, the economy's long-run
responses captured in the analysis are only those after all the adjustments
are over and the economy has again become stationary.
• The absence of labor contracts.
• Equilibrium in the labor market.
Since the analysis of this chapter has assumed certainty and the absence of
adjustment costs, the condition for full employment is met at its
equilibrium employment level nLR*. To emphasise this property, we

335
designate it as nf. Its corresponding equilibrium output level yLR* is the
full-employment level of output yf.
Our conclusions on the full-employment levels of output and
employment are:
• These levels are uniquely determined by the production function and
labor supply.
• They do not depend on aggregate demand, monetary, and fiscal policies,
or the price level in the economy.
• They shift only if there are shifts in the production function and/or in the
labor supply function.

7.8 The Role of Supply-Side Policies in

Changing Full-Employment Output
As we have seen above, positive shifts in the production function and the
supply of labor increase the full- employment level of output. Policies that
do so are called supply-side policies. Among the examples of such policies
are:
• Shifts in the production function induced by government subsidies to
research and development (R and D).
• Policies that promote investment in the domestic economy — through
government subsidies and tax breaks to firms to build factories or
modernize their equipment — increase and/or make more efficient the
capital stock and thereby increase labor productivity.
• Policies that increase labor supply include reductions in unemployment
insurance benefits, thereby increasing the incentive to look for and accept
job offers sooner. Other labor-oriented policies that increase employment
include retraining schemes and those that increase the participation rate
(which increases labor supply) or lower the minimum wage (which
induces firms to hire more workers).17 LRAS0 LRAS1

336
7.9 Stagflation
The condition in which output is static (stagnant) or falling while prices
are rising is called stagflation. The increases in oil prices in 1973-1975 and
1980-1981 produced a negative shift in the production function and were
among the sources of the fall in output, while the increases in aggregate
demand induced by expansionary monetary and fiscal policies to offset the
fall in output and employment were the main cause of stagflation during
1973-1975 and 1980-1981. The impact of rising oil prices on the economy
is analyzed in Box 7.1 and Figures 7.8a and 7.8b. The following Fact
Sheet provides information on the movements in the price of oil in recent
decades.
Fact Sheet 7.2 : The Price of Oil in the USA, 1960-2008
This Fact Sheet shows fluctuations in the price of oil, which is an
important source of energy in the production of commodities. The price
of crude oil is stated in terms of dollars per barrel. As we see in the
graph below, both real and nominal prices of oil have fluctuated greatly
since 1960. The 1973 and 1979 OPEC oil embargoes caused surging oil
prices after 1973 and for most of the 1980s. However, the increases in
oil prices since the early 2000s have also been considerable. High
energy prices usually have a contractionary effect on output in oil
importing countries since it is an essential input in production, but not
for large oil exporting countries, which benefit from high oil prices.

337
 Increases in aggregate demand brought about by expansionary fiscal or
monetary policies do not offset long-run decreases in output but will add to
inflation. Such policies were attempted during 1973-1975 and again in
1980-1981 in an attempt to offset decreases in output due to the oil price
increases. Their result was to increase the inflation rate in subsequent
years, so that these policies contributed to the stagflation initiated by the
oil price increases.
Box 7.1: The Impact of Oil Price Shocks on Full Employment and Full-
Employment Output
in Oil Importing Countries
Oil is a major source of energy in the production processes of the
economy. Oil prices in real terms in the world economy increased
substantially in 1973-1975, in 1980-1981 and again in 2000. At other
times, they increased less dramatically. There were also many years
during which they decreased. Our analysis is only concerned with the
real or relative price of oil. By this, we mean the price of oil divided by
the price level.
How do increases or decreases in oil prices impact on the economy?
For the LR analysis, we will assume that the increase in oil prices is
permanent and was fully anticipated, so that it was taken into account in
the firms’ production and employment decisions. We will also ignore
any effects of this shift on investment and aggregate demand, and any
adjustment costs and lags.
To properly incorporate the impact of oil prices on production, we
need to modify our production function to:

where O designates the usage of oil. The rationale for its inclusion is
that oil is an essential energy source that affects the output produced by

338
a given quantity of employment. Assume that both labor and oil are
complementary in production, so that a reduction in oil usage will
reduce the marginal product of labor. Since the MPL now depends on
both employment and oil usage (as well as the capital stock and
technology), we write it as yn(n, O). Therefore, the profit-maximizing
condition for employment becomes:

An increase in the price of oil will reduce its usage and shift the MPL
curve to the left. In Figure 7.8a, the oil price increase decreases its usage
and shifts the labor demand curve to the left from n0 to n^1. The LR
employment falls from n0 to n\. Hence, from the production function,
LR output will also fall. This fall in output occurs for two reasons: the
fall in employment and the decrease in labor productivity. It is shown in
Figure 7.8b by a leftward shift of the LRAS curve from LRAS0 to
LRASi. This causes the long-run output yf to fall from y0 to yi and the
real wage to fall. The price level is also likely to rise (the analysis for
this is given in Chapters 8 and 9). Since output falls, while the price
level rises, an oil price increase produces stagflation.
Oil is only one of the elements of a factor of production that can be
labeled as ‘resources’ (raw materials). The above analysis can also be
used to find the impact of changes in the costs and availability of other
resources on output, employment, prices, and real wages.
Stagflation (i.e., stagnant inflation) is the term used for a situation in
which output is constant or falling, while the economy has inflation. It is
a dynamic situation that can be caused, as explained above, by continual
oil price increases, which, by shifting the LRAS curve left, keep
reducing output and increasing the price level over several quarters or
years. The resulting inflation is likely to be at low levels. It would
increase further under expansionary monetary and fiscal policies, for
which see below.
Note that rising oil prices will also worsen the balance of payments of
oil importing countries, and are likely to cause decreases in their
exchange rates against the currencies of oil exporting countries.
The impact ofoil price increases on oil producing countries
Note that the preceding analysis strictly applies to oil importing
countries. By comparison, countries that produce their own oil and even
export some are likely to increase their oil production levels and thereby
increase their GDP and their employment. If they are mainly oil

339
producing countries, with most of their GDP originating in the oil sector,
the raising oil prices are likely to shift their LRAS curve to the right, not
to the left, thereby rising their GDP.
Can expansionary monetary and fiscal policies eliminate this
stagflation?
As oil prices rose considerably in 1973-1975 and again in 1980-1981,
they set up the potential for the fall in output. Many countries tried to
offset this by expansionary monetary and fiscal policies. To see the LR
effects of such a response, assume that aggregate demand is raised by
such policies and shifts the AD curve to the right in Figure 7.8c from
AD0 to ADi. This shift does not increase or change the (post-increase in
oil prices) LR output from yi but provides a stronger impetus to the price
increases. Therefore, in the long run, there are two causes/components
of the price level rise:
• The rise in the price level due to the oil price increase.
• The further raise in the price level due to the expansionary monetary
and fiscal policies.
The LR analysis implies that since output falls from y0 to yi, while the
price level rises, these policies only made stagflation worse in the long
run.

Extended Analysis Box 7.2: The Role of Demand-Side Policies in

Changing Full-Employment Output
The preceding analysis has shown that nf and yf depend only on the
parameters of the production and labor supply functions, and not on any
of the parameters and exogenous policy variables that determine
aggregate demand. In particular, they do not depend on the target
interest rate, money supply, government expenditure, or deficits. That is,
the long-run equilibrium effects are specified by:

and

where rT stands for the target interest rate. That is, the output level under
full employment is independent of any demand influence, whether it is

340
from investment, fiscal policy, or monetary policy, or the parameters of
aggregate demand in the economy.
An implicit assumption in the above conclusions
Since the full-employment output is independent of the demand side of
the economy, the preceding analysis has the implicit assumption: the
economy has adequate and sufficiently fast reacting equilibrating
mechanisms to force aggregate demand yd into equality with the full-
employment output yf immediately or in a short enough time through
price changes only, so that the deficit or excess demand do not affect
production and employment by firms, nor do they affect the consumer
demand and labor supply of household or investment by firms. A
corollary of this assumption is that the firms do not react to changes in
their current and expected future demand directly but only to the prices
established by the markets for their products and assume that they will
sell the amounts consistent with the full-employment levels of demand.
Further, workers/consumers do not react — by changes in their
consumption patterns or in the search for jobs — to changes in their
current employment and incomes and future prospects for jobs and
incomes directly but only to their market-established wages, and assume
that the market will in fact ensure that the full-employment number of
jobs will be available at this wage. These represent very strong
assumptions and not all economies in all possible stages of development
or of the business cycle meet them. When these assumptions are not
met, we need to rely on the short-run equilibrium and/or disequilibrium
analyses, which are presented in Chapters 8 and 9.
The LR analysis of full employment assumes that the above
assumptions are met. In it, the equilibrating forces adjusting aggregate
demand to the aggregate supply (at the full-employment level)
determined in the production-employment sector bring about, in the
long-run equilibrium, appropriate changes in the interest rate, real
wages, and the price level. Therefore, we can assert that "in the long-run
equilibrium of this model, the equilibrium level of supply creates its own
demand” — through the equilibrating changes in prices, wages, and
interest rates.

7.10 Crowding Out of Investment by Fiscal

Deficits, Given the LR Supply of Output in

341
 the Closed Economy
’Crowding out’ is a term which is applied to the decrease in investment
and net exports by fiscal deficits. There are two reasons for crowding out.
They are:
1. Interest-rate crowding out. This occurs if fiscal deficits financed by
government borrowing in the form of new bond issues increase the
interest rate in financial markets, which reduces investment. This form
of crowding out occurs in the determination of aggregate demand if the
money supply is being held constant by the central bank. This case of
money supply targeting was discussed in Chapter 6. However, note that
if the central bank targets the interest rate (as in the analysis of Chapter
5) and holds it constant in the presence of fiscal deficits, interest-rate
crowding out does not occur.
2. Output-crowding out. This case occurs when output does not change in
response to fiscal deficits, even if the deficits do increase aggregate
demand. This occurs only if the economy was already producing the
full-employment output and maintains this level of output in the
presence of fiscal deficits. This is the context of the long-run analysis of
this chapter. [However, it does not apply in the determination of output
in the short run (explained in Chapter 8) or in the disequilibrium
analysis of Chapter 9.]
The following analysis is that of output-crowding out based on the analysis
of long-run output in this chapter. To start, first suppose that the fiscal
deficit is caused by an increase in government expenditures directly on
goods and services. In the long run, with y = yf, the equilibrium between
the LR output and the commodity sector specifies that:

Since the exogenous increase in g does not change the left side, the
increase in g must reduce either c or i and/or net exports (xc — zc/pr). But
the increase in g also does not alter disposable income, so that c does not
change. Therefore, the increase in g must be matched by a corresponding
decrease in either i or net exports, or both. This crowding out is being
enforced by the LR supply constraint set by full-employment output on the
economy, so that it belongs to long-run analysis.
Now suppose that the fiscal deficit is caused by an increase in
government transfers to households. Hence, the public's disposable
income increases, so that consumption c rises. Therefore, on the right side

342
of the preceding equation, while government expendituresg do not increase
(or change), c does. Therefore, the sum of investment and net exports must
decrease by a corresponding amount.
Hence, government deficits, whether due to the increase in the
government's provision of commodities for the public's consumption or in
transfers to households, can be quite inimical to LR private investment and
net exports by crowding out either one or both of them. On the crowding
out of investment by government deficits, if the deficit is due to increases
in government expenditures on capital projects, then it is preferable to
focus on total investment (sum of public plus private investment). It is the
crowding out of total investment by the consumption expenditures of
either the government or the households that is of concern for the future
growth of the economy's capital stock and wealth.

7.11 The Actual Level of Output in the Economy

Actual output in an economy in a given period can differ from the full-
employment one. The components of the actual level of output can be
differentiated in the form:

so that the actual level of output has three components:

1. yf: the full-employment level of output. This is the LR equilibrium level
— under certainty (or in the absence of price misperceptions under
uncertainty) and once all adjustments have been completed.
Alternatively, if there is uncertainty, economic agents do not make any
errors in their expectations.
2. (y* — yf): the deviation of the short-run equilibrium level of output
from the full-employment one due to errors in expectations under
uncertainty and adjustment costs.
3. (y — y*): the disequilibrium level of output, which would exist if actual
output was not even in short-run equilibrium.
Therefore, the actual output in the economy can differ from the full-
employment level for one or more of the following three reasons:
1. Under uncertainty, the difference between the short-run equilibrium and
the long-run (full-employment) output due to errors in expectations.
2. The difference between the short-run equilibrium and the long-run full-
employment output due to adjustment costs and lags.
3. The possibility that the economy may not even be in equilibrium.

343
Cases (1) and/or (2) lead to the short-run aggregate supply (SRAS) curve
analyzed in Chapters 8. Case (3) represents disequilibrium and is analyzed
in Chapter 9.
This chapter has only examined the determination of the LR (full-
employment) level of output. Its analysis was under the assumption of
certainty so that expectational errors were ruled out. Further, adjustment
costs and lags were not allowed in the analysis. Therefore, by assumption,
(y* — yf) = 0. Furthermore, this analysis was under the assumption
ofequilibrium in the labor market, firm's production, and workers’ labor
supply, so that disequilibrium was also ruled out by assumption, with the
result that (y — y*) = 0. However, we cannot take it for granted that the
real-world economies will meet these assumptions for each period of our
study, when these periods are as short as, say, a month, a quarter, or a year.
We, therefore, resorted to the analytical notion of the long run, which is
defined as the stage in which these assumptions are met, so that the
economy can be said to be at the full-employment (LR) level of output.
Hence, the aggregate supply curve for the full-employment level of
output is called the LR aggregate supply (LRAS) curve. It is applicable in
the case of zero expectational errors, zero adjustment costs and lags, and
equilibrium in the labor market and in production. It is strictly not
applicable when these conditions are not met. However, the LR levels of
output and employment can be used as a benchmark or reference state
toward which the economy will tend to move. If it does not do so of its own
volition or does not do so fast enough, macroeconomics (see Chapters 8
and 9) examines the policies that can induce such a movement.

7.11.1 Changes in the actual rate of output over time

The actual level of output alters over time because of changes in any or all
of its three components since:

As we have argued earlier, the full-employment level of output does

change over time due to shifts in technology and in labor supply. The other
two components of the actual level can also change and do change over the
business cycle. In particular, they are positively related to the business
cycle and tend to have positive values during a boom than during a
recession. They can be changed by monetary and fiscal policies.

344
7.12 The Rate of Unemployment
This chapter has focused on the long-run analysis of output and
employment. Changes in unemployment and its rate are the converse of
those in employment, so that we now derive the implications of the long-
run analysis for unemployment. Chapter 10 will present the treatment of
unemployment in greater detail.
The level of unemployment was defined in Chapter 1 as:

where:
U = level of unemployment and
L = labor force.

Since n < L, unemploymentisalwaysnon-negative.18 For our analysis at

this point, we will make the simplifying assumption that the labor force is
exogenously given and is independent of the wage rate. Figure 7.9 shows
the initial demand for labor by, the supply of labor by n0 and the labor
force as L. LR employment is given by the intersection of labor supply and
demand and equals n0 and LR unemployment equals (L — n0).

7.12.1 The LR equilibrium (natural) rate of

unemployment
The long-run equilibrium (full-employment) level of unemployment U * is:

The rate of unemployment is specified by:

Therefore, the long-run equilibrium rate of unemployment, also called the

345
full-employment rate of unemploy-

The long-run equilibrium (full-employment) rate of unemployment is

called the ‘natural rate of unemployment and often designated as un.
Alternatively stated, unemployment is at its natural rate when the economy
is at full employment. Hence:

where uf stands for the ‘full-employment rate of unemployment’. To

reiterate, the natural unemployment rate un is the equilibrium rate of
unemployment under certainty (or without errors in price expectations
under uncertainty), and once all adjustments are over.
For the initial and n0 curves, Figure 7.9 shows the labor force L and full
employment n0. The difference between them is the level of natural
unemployment U01. This unemployment level divided by the labor force
gives the natural rate of unemployment u0.
Shifts in the natural rate of unemployment The natural rate of
unemployment changes if there is a shift in the production function or in
the labor supply. To illustrate, an improvement in technology that
increases the marginal product of labor and shifts the labor demand curve
right from n 0 to n\ in Figure 7.9 increases the full-employment level from
n0 to n\ — which causes a reduction in the natural unemployment level to
U f. Hence, the natural rate of unemployment is not a constant. Nor is it
likely to be a constant in real-time, real-world economies in which changes
in productivity and labor supply occur frequently.
The natural rate of unemployment also varies among countries since the
labor demand and supply functions tend to differ among countries. The
lower the capital stock and the less efficient/advanced the technology, the
greater is the likelihood of a higher natural rate. It would also be so if the
labor force is less educated or skilled relative to the technology used by
firms. It would also be so if the labor force has a higher preference for
leisure. In general, LDCs (less developed economies) have higher natural
rates of unemployment than the developed economies.
The lack of impact of monetary and fiscal policies on the natural rate of
unemployment
Our earlier analysis showed that, for a given economy, nf is independent of

346
monetary and fiscal policies, as well as of any other determinants of
aggregate demand, so that the natural rate of unemployment is also not
altered by these policies. That is:

and

We want to reiterate this important conclusion: for a given economy, both

the level of full employment and the natural rate of unemployment are
independent of changes in aggregate demand and therefore, of monetary
and fiscal policies.

7.13 The Long-Run Equilibrium (Natural)

Rate of Interest
In the long-run equilibrium, the interest rate is determined by the
intersection of the LRAS and IS curves. As such, it is the rate of interest at
full employment and is called the LR or full-employment rate of interest.
Its determination is shown in Figure 7.10. In this figure, the initial LRrate
of interest is shown as being rLR. An increase in investment or in
government expenditures which shifts the IS curve from IS0 to ISi will
increase this rate of interest from rLR to ri

Since the LRrate of interest is not only determined by the supply

structure of the economy but also changed by shifts in the commodity
market components, applying the designation of’natural’ to it is less
appealing than for the LR unemployment rate. However, some economists

347
do apply the term ‘the natural rate of interest’ to the LR rate of interest.
The ineffectiveness of monetary policy and inflation in changing the long-
run real interest rate
In the long run, shifts in fiscal policies can alter the long-run equilibrium
real interest rate. A monetary policy change in the interest rate can only
cause the short-run interest rate to differ from the long-run one, but not
change the long-run rate itself.19
The nominal interest rate will change because of changes in the real rate
and the inflation rate. For the distinction between the real and the nominal
interest rate, see the Fisher equation on interest rates presented in Chapter
2.

7.14 Conclusions
• The long-run prosperity of the economy depends on its ability to utilize
its factors and technology at the most efficient sustainable level. The
level of employment generated at this level of utilization for the given
structure of the economy is referred to as its full-employment level and
its corresponding output is the full-employment level of output.
• For a given social, political, and economic structure of country, the main
determinants of the full- employment level of output are the production
function and factor inputs. In the short-run analysis, the capital stock is
assumed to be fixed, so that the only factor included in the determination
of the full- employment level of output is labor. Shifts in the full-
employment level of output occur due to shifts in the production function
or in the supply of labor — or shifts in the availability or prices of
resources such as oil.
• The economy does not always produce at the full-employment level of
output so that the next two Chapters study the sources of deviations from
this level.
• The rate of unemployment at the full-employment level of output is
called the natural rate of unemployment. Its main determinants are the
production function, the supply of labor and the labor force. Shifts in
these alter the natural rate of unemployment, so that it cannot be taken to
be constant over time. The actual rate of unemployment can — and
usually does — differ from the natural rate.
• The long-run equilibrium level of the interest rate is determined by the
intersection of the LRAS and the IS curves, and is independent of
monetary policies, but not of fiscal policies.

348
• There are two types of crowding out of private investment and net
exports by fiscal deficits. One is caused by the rise in interest rates due to
bond-financed fiscal deficits and is called interest rate crowding out. This
can occur if the central bank targets the money supply and holds it
unchanged in the presence of fiscal deficits (see Chapter 6), but not if the
central banks targets the interest rate (see Chapter 5). The other type is
output-crowding out. This occurs in the long-run analysis when output is
maintained at its full- employment level and is not changed by fiscal
deficits. Full crowding-out is unlikely to occur in short-run analysis
(Chapter 8) or disequilibrium analysis (Chapter 9).

KEY CONCEPTS

The demand for labor

The supply of labor
Equilibrium in the labor market
Full employment
Full-employment output
The employment-output relationship/curve
The long-run aggregate supply (LRAS) relationship/curve
The natural rate of unemployment
The full-employment (natural) rate of interest
Ineffectiveness of monetary and fiscal policies in terms of long-run output
and employment
Interest rate crowding out; output crowding out.

SUMMARY OF CRITICAL CONCLUSIONS

• The long-run equilibrium levels of full employment and full-employment

output are under the assumptions of (a) certainty or the absence of
expectational errors, (b) the absence of adjustment costs, or after the
economy's adjustments are over, and (c) equilibrium in all markets. These
levels are independent of aggregate demand — and of monetary and
fiscal policies. Shifts in the private or public components of aggregate
demand cannot change them.
• The natural level of unemployment and the natural rate of unemployment
occur when there is full employment in the economy.

349
• The long-run (equilibrium) level of the interest rate occurs when there is
full employment in the economy. It is determined by the LRAS and IS
curves. Fiscal deficits increase it and fiscal surpluses reduce it.
• The long-run real interest rate is independent of money demand and
money supply, as well as of the interest rate target rule adopted by the
central bank.
• In the long-run with the full-employment output unchanged by fiscal
deficits, fiscal deficits crowd out investment or net exports, or both.

REVIEW AND DISCUSSION QUESTIONS

1. What is the general form of the production function commonly used in

short-run macroeconomics? What assumption does the short-run
production function make with respect to physical capital? What is the
justification for this assumption?
2. Show diagrammatically the relationship between output and labor, and
specify its assumptions. Show in this diagram the effects of (a) an
increase in the capital stock and (b) an increase in employment.
3. Show diagrammatically the downward-sloping curve for the marginal
productivity of labor (MPL). Show the effects on this curve of (a) an
increase in the capital stock and (b) an increase in employment.
Discussthe derivation of the demand curve for labor (nd curve) from the
MPL curve and show the demand curve for labor.
4. Show diagrammatically the effects of an increase in the marginal
propensity to consume on (a) aggregate demand and (b) LR output.
5. Show that the LR equilibrium rate of interest does not depend on (a) the
demand or supply of money and (b) the interest rate target rule.
6. Does the natural rate of unemployment depend upon government
expenditures on goods and services or on government transfers to the
public? Give reasons for your answer.
7. Suppose an expansionary monetary policy is pursued. Would it change
the natural rate of unemployment? Discuss in the context of the analysis
of this chapter.
8. Would you expect full employment and full-employment output to (a)
always exist in the economy and (b) hold in 2008 and 2009 in the USA,
Canada, and other economies hit by the financial and economic crisis?
Give reasons for your answer.
9. What is meant by an expansionary monetary policy in the context of
interest rate targeting by the central bank? Does such a policy alter the
following?

350
 (a) full-employment output and
(b) the natural rate of interest.
Explain your answer, using the appropriate diagrams.
10. Does an increase in government expenditures alter the following?
(a) full-employment output and
(b) the natural rate of interest.
Explain your answer, using the appropriate diagrams.
11. Analyze, using diagrams, the impact of an increase in labor
productivity on the natural rate of unemployment, full-employment
output and the long-run equilibrium interest rate.
12. Show the long-run effects of a permanent increase in oil prices on the
long-run equilibrium levels of output, employment, real wage, and
interest rate.

ADVANCED AND TECHNICAL QUESTIONS

Basic AD Model for Chapter 7

Assume that the following equations describe the commodity markets of
an economy and the interest rate targeted by its central bank.

T1. Given the Basic AD model, now add the assumption that the long-run
aggregate supply function is given by:

(a) Derive the IS equation. [Do not use the IS equation formula for this
answer. Use step-by-step calculations. Start by calculating disposable
income. Noting that xc and zc depend on p1 , calculate p in terms of P
. Then use the equilibrium condition y = e for your further steps.]

351
 (b) Derive the AD equation.
(c) Derive the equilibrium price level Pq .
Basic AS Model for Chapter 7
For the supply structure of the economy, assume the following. The
production function:

The labor supply function:

T2. For the Basic AS model, derive:

(a) The marginal product of labor.
(b) The labor demand function.
(c) Equilibrium real wage and employment.
(d) Equilibrium output.
T3. For the combined Basic AD and AS models, derive:
(a) The equilibrium levels of employment and output.
(b) The equilibrium price level P|.
(c) The equilibrium values of the nominal and real wage.
T4. Starting with the combined Basic AD and AS models, assume that
there is an increase in the labor force participation rate, which shifts the
labor supply function to:

Recalculate your answers to the preceding two questions.

Revised Model for Chapter 7
Assume the following:
The production function:

The labor supply function:

Labor force:

The aggregate demand (AD) function:

352
T5. For the revised model, calculate:
(a) The AD function.
(b) The long-run equilibrium wage.
(c) The natural rate of unemployment.
(d) The equilibrium price level.
(e) What will be the impact on the natural unemployment rate of (a) an
increase in the target interest rate to 0.05 and (b) an increase in
government expenditures by 10%?
T6. What do the answers to the preceding technical questions imply for the
impact of on the long-run equilibrium levels of output, employment,
unemployment rate, real wage rate, and the long-run equilibrium interest
rate of:
(a) Increases in the target interest rate.
(b) Increases in government expenditures.
1 Whether the economy's industries have perfect competition,
monopolies, or oligopolies clearly affects their production and
employment levels.
2 These wishes are relevant because the owners of the factors of
production may choose to supply less than the full amount that exists in
the economy. This means that, in the case of labor, labor supply may be,
and usually is, less than the labor force. The owners of capital also usually
do not choose to keep all their machines functioning all the time.
3 Note that the capital stock, technology, and the labor force are still
being held constant so that the overall context continues to be that of the
short-run macroeconomic models.
4 Hence, the long-run – or full-employment – output is not really the
maximum output that the economy could produce at any time, e.g., if all
its resources were used 24 hours a day, or even the equilibrium output that
will be produced in the short run (see Chapters 1 and 8) or the actual
output that might be produced when the economy is not in equilibrium (see
Chapter 9).

353
\
5 ynn < 0 means that the MPL is diminishing.
6 yKK < 0 means that the MPK is diminishing.
7 Over a longer term, domestic physical capital is increased by direct
foreign investment and domestic technology improves by the adoption of
better production methods from abroad.
8 Both these could be affected by international flows in an open
economy in the long run.
9 The assumption is that though investment occurs, such investment
takes time to translate into physical capital in actual production. These are
due to the ‘gestation lag’ for capital, caused by the time-to-build factories,
machines, etc. and bring them into producing goods for sale. Further, as
indicated earlier, the inflows of physical capital through direct foreign
investment by foreigners are assumed not to alter the domestic capital
stock in the short run. The impact of these inflows will then have to be
treated as a shift in the production function.
10 Point a represents the minimum level of profits. At this point,
increasing employment by one worker would increase output more than
the wage rate, so that profits would rise. Therefore, the firm would
increase employment beyond the point a, and continue hiring more
workers until it reaches point b. Hence, the firm would never operate (i.e.,
employ workers) on the upward-sloping part of the MPL curve.
11 If the firm were to increase employment beyond n0, the increase in
output due to the higher employment would be less than the wage rate, so
that profits would fall.
12 This did occur over the 20th century. As the real wage rate rose from
very low levels at the end of the 19th century to their present substantially
high levels, workers chose to buy more leisure along with buying more
consumer goods. The former translated into a decrease in the hours worked
per week. However, such an effect occurs over long periods and for very
substantial, rather than marginal, increases in wage rates.
13 Examples of these are some of the countries of the Arabian Peninsula,
the southern region of the USA with immigration from Mexico, etc.
14 The theories doing so are called real business cycle theories. These
are explained in Chapter 6.
15 For Canada, the average number of hours worked per week declined
from 34 hours and 22 minutes per week in 1979 to 33 hours in 2003,
mirroring a similar trend in other industrial countries. The annual average
hours worked in 2003 were 1,718 for Canada, 1,792 for the USA, 1,354

354
for the Netherlands, and 2,400 for South Korea.
16 These qualifications are clarified in Chapter 8.
17 This assumes that the minimum wage was initially above the
equilibrium one for workers possessing the relevant skills.
18 If L can be assumed to be exogenously given as L, so that it does not
vary with the real wage, the labor force will be the sum total of workers
who are able and willing to work at any wage. It will be the maximum
amount of potential employment in the economy. If the number of workers
willing to work increases as the real wage rises, the labor supply function
will be L = L(w), with ∂L/∂w > 0. This is further explored in Chapter 10.
19 For guidance, see the analysis relevant to this point in Chapters 4 and
5.

355
 CHAPTER 8
Output in the Short Run: The Role of
Expectations and Adjustment Costs

This chapter presents the short-run analyses of the labor market

and the aggregate supply of output for the closed economy. There
are several reasons for the deviation of the short-run equilibrium
output from the full-employment one.
On the supply side, one of these reasons is the existence of
uncertainty so that the economic agents have to form, and act, on
their expectations of future prices. Expectational errors
incorporated in wage contracts imply the Friedman supply rule,
while those in the relative prices of commodities imply the Lucas
supply rule. Both lead to the same conclusion: unanticipated
inflation (deflation) increases (decreases) output above (below) its
full-employment level. However, these deviations from the full-
employment level are likely to be transient and self-correcting .
Actual output may also deviate from the full-employment one
due to the existence of adjustment costs. These costs include those
incurred by the firms in (i) adjusting prices in commodity markets,
(ii) adjusting employment, and (iii) adjusting output. These costs
can cause deviations in output and employment from their long-
run levels .

This chapter studies the short-run equilibrium behavior of employment and

output. The short-run equilibrium levels of output and employment can
differ from their long-run (full-employment) levels for several reasons. Of
these, macroeconomics focuses on:
• The existence of uncertainty, with errors in price expectations. The two
main theories on this imply the Friedman and Lucas supply rules.
• Costs of adjusting prices in response to changes in demand and
productivity. These lead to sticky prices for some products. The theory on
this is called the menu cost theory.

356
• Costs of adjusting output in response to changes in demand and
productivity.
• Costs of adjusting employment in response to changes in demand and
productivity. These lead to variations in the work effort of employees.
The theory on this is the implicit contract theory, which implies Okun's
rule.
These factors imply that the short-run aggregate supply (SRAS) curve
becomes different from the long-run aggregate supply (LRAS) curve: the
SRAS curve has a positive slope while the LRAS one is vertical in the
AD-AS diagram.
On (i), when there is uncertainty about the future value of an economic
variable — for instance, the price level, the rate of inflation, the demand
for products or employment, etc. — firms and individuals have to form
expectations on its likely value. The two major hypotheses in economics
for constructing the expected value of a variable are the adaptive
expectations hypothesis (AEH) and the rational expectations hypothesis
(REH). The former is a statistical procedure, while the latter represents an
economic theory of expectations. On the role of expectations in
macroeconomic theory, this chapter focuses on (a) expectations on the
price level and the rate of inflation — rather than on demand and expected
sales (which are considered in Chapter 9) — and (b) the use of the REH
for modelling these expectations.
This chapter examines (i) the two main theories on the impact on output
of errors in price expectations (Sections 8.1 to 8.11), (ii) the theory on the
costs of adjustment of prices (Section 8.12), and (iii) the theories on the
costs of adjusting employment and production (Sections 8.13 and 8.14).
These theories separately and collectively imply that the short-run impact
of changes in aggregate demand on output and employment will differ
from the long-run impact and that, in the AD-AS diagram, the SRAS curve
will have a positive slope, even though the LRAS curve is vertical. This
chapter does not examine the role of a fall in aggregate demand below its
full-employment level, which would cause a demand deficiency in the
economy. Since such a situation in macroeconomics is considered a case
of disequilibrium, rather than of short-run equilibrium; this analysis is left
to Chapter 9, which focuses on disequilibrium in the economy. In addition,
it does not present the analysis of the impact of a credit crisis on output,
which is covered in business cycle theory (Chapter 16).
Note that actual output in the economy (i) can be at the long-run
equilibrium (full-employment) level, (ii) not be at the long-run equilibrium
(full-employment) level, but only at the short-run equilibrium level, or (iii)

357
not be at the long-run or short-run equilibrium levels, but at the
disequilibrium level, for numerous reasons. This chapter studies (ii), while
Chapters 7 and 9 study (i) and (ii) respectively. To reiterate a point made
earlier in this book, the long run and short run are analytical states (i.e.,
under specific assumptions about the economy), not real-world states
existing in real, chronological time, while the data on the economy is on
actual real output.

8.1 The Role of Uncertainty and Errors in

Expectations

8.1.1 The theory of rational expectations

The theory of rational expectations was proposed during the 1960s, but
was popularized in macroeconomics by Robert Lucas, among other
economists, only in the 1970s. The REH is currently the dominant
approach in macroeconomics to modeling expectations. This approach
asserts that:
• Expectations are formed on the basis of all available information. This
information includes not only past experience but also anything known or
expected about the future. This knowledge will include knowledge of
economic theories and the policy rules followed by the monetary and
fiscal authorities.
• Systematic errors in expectations are costly; therefore, economic agents
try to eliminate them by acquiring better information. They revise the
theories and data on which they are based, until the remaining errors in
expectations are purely random ones. Random errors have a zero mean
(average value) and are not related to information known at the time the
expectations are formed; therefore, they are not predictable. [However,
the REH does not specify how long it will take to acquire and process
adequate information for the errors to become merely random ones. This
period can be exceedingly long, e.g., possibly even decades.]
• For simplification for analytical purposes, we can extend the assumptions
of the long run in Chapter 7 by adding the assumption that it has no
expectational errors (not even random ones). We will call this the long-
run (or expectational) equilibrium, the important aspect of which is that it
does not have errors in expectations.
Applying the theory of rational expectations to future inflation rates, the

358
actual inflation rate — which will become known only in the future — can
be decomposed into two components. These are:
1. The rationally anticipated (or expected) inflation rate (π e), which has
only random errors.
2. The unanticipated (or unexpected) inflation rate (π u), which equals the
actual inflation rate less the anticipated one. That is π u = π – π e. The
unanticipated inflation rate can also be called ‘errors in expectations’ or
‘expectational errors’.
Under rational expectations, the rational expectation of the future inflation
rate will equal the ‘mathematical expectation’ (average) of the actual
inflation rate. The rationally expected inflation rate will be sometimes
above and sometimes below the actual one, but without any systematic —
i.e., persistent — positive or negative bias in the difference between these
rates. Therefore, the (currently) unanticipated component, i.e.,
expectational errors, will be random, which makes them unpredictable
with a zero mean.

8.1.2 Random errors and their predictability

Some of the changes that occur are based on known information. If they
are uncertain, the economic agent—i.e., households/workers, firms, and
policy makers — forms expectations on them and attaches subjective
probabilities (i.e., personal guesses on their likelihood) to the potential
outcomes. These outcomes are said to be expected or anticipated.
Random errors are not based on any known information. Therefore, there
is no basis on which to expect such an error to be negative or positive;
there is also no basis for forming estimates of their magnitudes for a future
period. Hence, random errors are inherently unpredictable — even to the
extent that a subjective probability cannot be assigned to the actual value
of the error in the next period. However, we can specify two properties for
them. These are (a) the expected value of random errors is zero and (b) the
sign and magnitude of the actual error that will occur are unpredictable.

8.1.3 Application of rational expectations to monetary and

fiscal policies
Rational expectations implies that the systematic value of— and changes
in — any variable relevant to economic agents will sooner or later be
learnt by them so that they will anticipate any changes in it. Only the

359
random variation of the variable will remain unanticipated. Applying the
REH to monetary and fiscal policies implies that any systematically
pursued monetary and fiscal policies will become anticipated by the
public. Therefore, the future impact of such systematic policies on output
and the actual inflation rate will also be incorporated into the public's
expected values for them, so that such policies cannot cause unanticipated
changes in production and employment. Only random alterations in
monetary and fiscal policies — just as for shifts in consumption,
investment, or net exports — could possibly cause unanticipated changes
in production and employment.

8.2 Price Expectations and the Labor Market:

Output and Employment in the Context of
Wage Contracts
In industrialized economies, firms and workers negotiate the nominal wage
rate through explicit contracts or implicit arrangements. This nominal
wage rate is established in advance of the production and employment
decisions by the firm and before the actual price level is known. The
following considers the effects of this pattern of wage negotiations and
production on employment and output. It defines the two stages of this
process as:
1. The first stage during which wage negotiations take place and the
nominal wage is set.
2. The second stage during which firms take the nominal wage as given
and determine their employment and output.

8.2.1 Labor supply in wage negotiations during the first

stage
To start, suppose that there is uncertainty about the price level P in the
period ahead. Since workers want to know what the purchasing power of
their nominal wage rate W is likely to be, they have to form expectations
on P . Let the household's expected price level for the period ahead be P
eh. Workers supply labor on the basis of their expected real wage W/P eh,

so that their supply function can be written as ns(W/P eh), where W is the
nominal wage. Similarly, let the firm's expected price level for the period
ahead be P ef, and their labor demand curve based on their expected real

360
wage, equal to their marginal product of labor, be n d(W/P ef). Their
negotiations yield a nominal wage given by equilibrium between labor
demand and supply, i.e., by ns(W/P eh) = nd(W/P ef). The nominal wage
thus established becomes the wage contract between workers and firms.
This wage remains unchanged for the following period, during which
production and employment take place.
An increase in the firm's expected price level increases its willingness to
agree to higher nominal wages, and an increase in the household's
expected price level makes workers demand higher nominal wages, so that
either or both of them will result in a higher nominal wage set in the wage
contract.1
The analysis further assumes that: (a) the contracted wage rate W c is set
for the duration of the labor contract and (b) workers will be willing to
increase or decrease the amount supplied of labor without asking for a
change in the contract wage during the contract period. This implies that
workers will supply any amount of labor demanded by the firms at W c.
Hence, for the duration of the wage contract, the labor supply is as if it was
horizontal at W c in the (n, W) diagram,2 so that the ex-ante labor supply
curve is set aside in the analysis of employment for the contract period.

8.2.2 Employment during the contract period (the

production stage)
At the time of production, the j th firm would know the actual price of its
own product as an element of its joint production and pricing decisions, so
that its actual demand for labor will be based on the nominal wage divided
by the actual price of its product, rather than on the price which had
previously been expected by it.
Hence, over the two stages,
1. As discussed earlier, an increase in the expected price level establishes a
higher contractual nominal wage during the first stage's wage
negotiations. This increase in the contracted nominal wage, ceteris
paribus, increases the actual real wage, which makes labor more costly
and reduces labor demand, during the second stage.
2. For the given contractual nominal wage, any increase in the actual price
level during the second stage will lower the actual real wage, thereby
making labor cheaper and increasing the employed.
Hence, the introduction of uncertainty and wage contracts into the labor
market analysis implies that employment will depend upon: (i) the

361
duration of the wage contract, (ii) the price level expected by firms and
households during wage negotiations, and (iii) the actual price level when
employment occurs. Since output and employment are positively related
through the production function, the above analysis implies that output
will depend positively on the ratio P/P e, which can be written as y = f(P/P
e). For this function, a rise in P e would raise the contracted nominal wage
and make labor more expensive, thereby reducing employment and output,
while a rise in P lowers the real wage for the pre-determined contracted
nominal wage and makes labor cheaper.

8.2.3 Diagrammatic analysis

Figure 8.1a presents the labor demand n d(W/P ef0) and the labor supply n
s(W/P eh ) curves for the coming period. These curves are drawn for the
0
particular price levels expected by the firms and the households. Note that
the vertical axis in this figure is the nominal wage rate W . The negotiated
nominal wage will be set at the equilibrium level W c0, and has the
expected employment level of n e0. An increase in P ef will shift the labor
demand curve to the right3 and a rise in P eh will shift the labor supply
curve to the left,4 so that each will raise the nominal wage. However, the
former will increase the expected employment level and the latter will
decrease it. If both P ef and P eh increase proportionately, the two curves
will shift proportionately and the nominal wage will increase in the same
proportion, without a change in the expected employment level.
Henceforward, to simplify further analysis, we will assume that the price
level expected by both firms and households is identical and is designated
as P e0.

362
 Actual employment is determined not in Figure 8.1a but in Figure 8.1b,
now with the actual real wage w — equal to W/P — on the vertical axis.
For the contracted nominal wage W c0 from Figure 8.1a, and a given price
level P 0 equal to the expected one P e0 (i.e., P 0 = P e0), employment in
Figure 8.1b is n 0 — so that n 0 = n e0. However, with the contracted
nominal wage still at W c0, a lower price P 1, with P 1 < P e0, will raise the
actual real value of the contracted nominal wage to W c0/P 1 and decrease
employment to n 1 with n 1 < n 0. Conversely, if the actual price level is
above the expected one (e.g., P 2 > P e0), real wages would turn out to be
lower (at W c0/P 2) and employment higher (at n 2) relative to their
expected levels.
Note that no explicit labor supply curve was drawn in Figure 8.1b for the
contract period. But workers remain willing to supply any amount of labor
at the negotiated wage, irrespective of changes in the price level.
Therefore, we can think of an implicit labor supply curve that is horizontal
at W c (in Figure 8.1a with the nominal wage W on the vertical axis and
labor supply on the horizontal one), and at the actual real wage (=Wc/P)
(in Figure 8.1b with the real wage w on the vertical axis and labor supply
on the horizontal one, as shown).

363
 If there were no errors in expectations — that is, if P ef = P eh = P ,
actual employment n will equal ne0 , as determined in Figure 8.1a;
therefore, we can take this to be the full-employment level n f or the
‘expectational equilibrium’ level n *. If P is higher than both P ef and P eh,
n > n e0, and vice versa. Therefore, the deviation — i.e., (n – n e0) — in
employment from its expected level n e in Figure 8.1a is positively related
to the errors (P – P e) in expectations.
The effect of a proportional increase in the expected and actual price
levels
An identical increase in both the expected and actual price levels will not
change the real wage, even though the nominal wage will rise
proportionately, so that it will also not change employment. The
diagrammatic analysis of this case is shown in Figures 8.1c and 8.1d. In
Figure 8.1c, since the price level expected by both firms and workers rises
by the same amount, both the labor demand and supply curves will shift up
in such a way that at the equilibrium employment level n e0, the contracted
nominal wage will rise (from W c0 to W c1) proportionately with the
increase in the expected price level. Although the contracted wage has
risen, the actual price level has also risen in a proportionate manner, so
that labor does not become either cheaper or more expensive and the labor
demand curve does not shift from its initial position. In Figure 8.1d, the
real wage also remains unchanged at w 0 (with W e0/P0 = W e1/P 1).
Therefore, if nominal wages and prices rise in the same proportion,
employment does not change in spite of the increase in the nominal wage
rate.
Conclusions from the contract wage analysis of production
Hence, our results for the SR equilibrium employment in the context of
nominal wage contracts are that:
(i) If the actual and expected price levels are identical, the economy would
have full employment and produce the full-employment level of output.
(ii) If the actual price level proves to be higher than the one expected at the
time of the signing of wage contracts, the employment and output levels
would rise above their LR (full-employment) ones.
(iii) If the actual price level proves to be lower than the one expected at the
time of the signing of wage contracts, the employment and output levels
would fall below their full-employment levels.

364
(iv) If the actual and expected inflation rates remain equal, with both
changing proportionately, employment and output will not change.
The expectations augmented employment and output supply curves
The above implications of the contract-based analysis are captured by
writing the equations for employment and output as:

These equations assert that if the equilibrium price level P * is greater than
the expected price level P e, both employment and output would be higher
than their LR equilibrium (full-employment) levels. Remember that this
result occurs because the unanticipated rise in the price level reduces the
real wage below its anticipated level and makes workers cheaper to
employ. Once the contracts are re-negotiated and embody the
unanticipated higher price level, the increases in employment and output
above their full-employment levels disappear — unless new positive errors
occur in the price level expectations for the following period.
The above two equations are respectively the expectations augmented
employment and output/supply equations. These are important equations
describing the short-run equilibrium response of the economy to price
changes and inflation. This output supply rule is also called the Friedman
supply rule (FSR).
Implications of the expectations augmented employment and output supply
equations for variations in employment and output over the business cycle
The curve for the FSR is shown in Figure 8.2. The LRAS curve results
under the FSR if P = P e — and π = π e. In this case, y = y f. At a given
expected price level P e0 equal to P 0, the FSR generates the SRAS curve
as SRAS0. If P > Pe, the economy is on the SRAS curve to the right of y f,
but if P < Pe, the economy is on the SRAS curve to the left of y f. If the
expected price level rises to P e1 equal to P 1, the SRAS curve moves up to
SRAS1. That is, for a given LRAS curve, there is a separate SRAS curve
for each expected price level embodied in the labor contracts.

365
 For applications to business cycles, we need to replace the price level P
by the inflation rate π . In the upturns of most business cycles, while the
rate of inflation is rising, its expected value tends to lag behind its actual
value, so that the expected inflation rate becomes lower than the actual (ex
post) one.5 This causes employment to rise above its full-employment
level. In the downturns of most business cycles, the inflation rate is falling,
and its expected value tends to lag behind (i.e., become higher than) its
actual value, which causes employment to fall below its full-employment
level. Expectational errors especially tend to emerge when inflation is
accelerating or decelerating, as compared with when the inflation rate is
static/constant.
These cyclical variations in employment produce corresponding cyclical
movements in output over the business cycle.
Box 8.1: Errors in Price Expectations, the Duration of the Wage Contract
and Cost-of-Living Clauses
The deviation in employment from its expected level n e — envisaged
at the time of the signing of the wage contract — will only occur during
the duration of the wage contract since the past errors in expectations
will be eliminated when the wage contract is renegotiated. This is
usually done through the ‘catch-up’ clause for cost-of-living increases in
labor contracts. Therefore, continuously new errors will be needed to
maintain employment above n f. While this can occur for some time —
the ‘people can be fooled some of the time’ syndrome — it cannot
continue indefinitely — ‘they cannot be fooled all the time’. The former
usually has to take the form of accelerating inflation rates. The latter
usually occurs in two ways:
1. The future expectations of inflation tend to ‘jump’ beyond the past
— experienced — inflation rates in an attempt to capture the potential
future acceleration in inflation.

366
 2. In order to reduce or eliminate the loss in purchasing power through
inflation, the duration of wage contracts is reduced or cost-of-living
clauses6 are built into them.
Therefore, while the errors in expectations can induce increases in
employment — and do so during accelerating inflation — such increases
can be only a short-term but not a long-term phenomenon in practice.
Further, errors in expectations caused by inflation cannot be relied upon
to occur over lengthy periods or at very high and persistent rates of
inflation. Hence, over the longer term, the economy will revert to the
long-run employment level n f.

8.3 Friedman's Expectations Augmented

Employment and Output Rules
The above expectations augmented employment and output equations were
based on Milton Friedman's, and other economists’, ideas for contract-
based labor markets and we have named them after him. The Friedman
employment rule for an inflationary context is:

where the inflation rate π has replaced the price level P . Note that the SR
equilibrium employment level is predicated on labor market clearance in
wage negotiations, so that this level is not a disequilibrium one.7Therefore,
n* is the SR equilibrium employment level, with * standing for SR
equilibrium, while the superscript f stands for the long-run (full-
employment level).
Correspondingly, since changes in employment produce changes in
output, the Friedman supply rule (FSR) for output in an inflationary
context is:

where all the variables are in logs. The asterisk symbol (*) on a variable
indicates that its value is the SR equilibrium one, while the superscript f
indicates its LR equilibrium value. Note that:
• The expectations of inflation in these equations refer to those
incorporated in wage contracts, and the economy deviates from its full-
employment level due to errors in these expectations: if π > π e, real
wages are lower while employment and output are greater compared with

367
 the full-employment level, and vice versa. The mechanism of the FSR
relies upon wage contracts established in advance of production, so that
an increase in prices/inflation above the expected ones lowers real wages,
thereby making it attractive to hire more labor than in the error-free LR
equilibrium.
• Several assumptions were needed for deriving the FSR. Among these
was the assumption of market clearance in the labor market at the wage
contract stage. Further, at the production stage, given the contracted
nominal wage, labor willingly supplies the amount of labor demanded by
firms. Hence, the FSR does not allow for disequilibrium in the labor
market, in which some workers may want jobs at the existing wage but
cannot get them or firms want more workers at this wage but cannot get
them. Further, the commodity market also clears. Therefore, the FSR
represents the short-run equilibrium behavior of the economy.
• These deviations from full employment require a lack of adequate
knowledge about the future and the existence of nominal wage contracts.
The duration of their effects depends on the length of the wage contracts.
In general, this length is reduced by firms and workers if inflation rates
are quite high — e.g., during the late 1970s and 1980s, as inflation rates
rose to double digits, the duration of many wage contracts was reduced
from what used to be three years during the low-inflation periods of the
1950s and 1960s to one year. In some cases, cost of living clauses came
to be inserted into the contracts.
Extended Mathematical Analysis Box 8.1: Labor Demand when There
are Expectational Errors in the Context of Nominal Wage Contracts
To provide a mathematical illustration for the case of wage contracts
with price uncertainty, let the labor market functions at the time of wage
negotiations be:

In equilibrium:

Hence, the contractual nominal wage will be:

so that ∂W c/∂P eh, ∂W c/∂P ef > 0. Note that a proportionate increase in

both expectation increases the nominal wage rate in the same proportion.

368
The expected level of employment obtained by substituting this equation
in the labor supply function is given by:

which implies that a proportionate increase in both expectations does not

change π e .8
At the time of production, with the nominal wage set by the wage
contract at W c, the actual real wage and employment will be:

Both w and n are independent of changes in P , P ef, and P eh. If P

exceeds both of its expectations, the real wage will turn out to be less
than its expectation in the wage contract and n will be greater than ne
and vice versa.
We now simplify the preceding equations by assuming that firms and
households hold the same expectations, so that P ef = P eh = P e, where
P e is the commonly held price expectation. This simplifies the above
equation for employment to:

The expectational equilibrium level of employment

If there are no errors in expectations, P = P ef = P eh, so that:

This expectational equilibrium level of employment n e* is independent

of the price level and is, therefore, the full-employment level. Positive
expectational errors (P – P e) induce actual employment to be greater
than this full-employment level and vice versa.
The impact of expectational errors on employment
To find this impact, subtract equation (2) from (1). This gives:

where α = a0a1/(a1 + b1) > 0. Assuming that actual employment by

369
firms will always equal their demand for labor, replace n d by n.
Therefore, the preceding equation implies that employment is given by:

This equation is usually simplified and written in a linear or log-linear

form as:

Since α > 0, positive expectational errors (defined as P = P e) raise

employment above n f, while negative expectational errors (P < P e)
reduce employment below n f. This is the Friedman employment rule.
The Implications for Employment and Output of Wage Contracts
with Expectational Errors in Prices
Our arguments have shown that, given nominal wage contracts
conditional on expectations, firms would employ workers according to
their labor demand. Substituting this level of employment from equation
(4) in the production function implies that output will also depend on
expectational errors in the price level. Hence, from (4), output is given
by:

This equation is often written as:

This equation is the Friedman supply rule (for output). These equations
imply that:

Hence, the Friedman supply rules imply that inducing unexpected

positive errors (i.e., P* — P e) in the price level raises both employment
and output above their full-employment levels. But unexpected negative
errors in the price level will lower both employment and output below
their full-employment levels.

8.4 Lucas Supply Rule with Errors in Price

370
 Expectations in Commodity Markets
In the 1970s, Robert Lucas proposed a short-run output function that came
to be labelled as the Lucas supply rule. Like the Friedman supply rule, it
implies that output will be above its full-employment level if the actual
price level proves to be above the expected one. However, the Lucas
supply rule is not based on nominal wage contracts but on firms’
production behavior when they misinterpret the nature of changes in their
product prices. The following exposition gives a simple and very brief
glimpse into this analysis.
Lucas’ analysis shows that at the beginning of the period, the
representative firm will start with an initial expected price level P e.
During the period, it will observe the price of its own product, which on
average over the whole economy is represented by the price level P. Part
of the difference (P – P e) could be due to the rise in the firm's relative
price (i.e., relative to the prices of all other commodities) but part of it
could be due to a general inflation or deflation affecting all firms. The
firm's opinion on the latter leads it to revise its initial expected price level.
Lucas assumed that this revision cuts down the actual error in price
expectations to α(P – P e), where 1 > α > 0. α is the revision factor of the
expected price level on the basis of a change in its observed product price.
To reiterate, (P – P e) is the initial error in expectations based on prior
information and α is the revision of this initial error on the basis of new
information on changes in product prices. Under rational expectations, the
magnitude of the revision factor α is heavily influenced by the firm's
knowledge of the relationship between past variations in product prices
and those in the price level — as well as such information about the future.
Intuitively, if we were to focus only on the past information for
simplification, if the past inflation rates were close to zero, firms are likely
to expect that the price level is not going to rise even though they observe
that their own product price is rising. That is, the increase in product prices
will be interpreted as an increase only in relative prices (i.e., the product
price divided by the price level), with α having the value one or close to it.
But if the past experience was of hyper-inflation, an increase in product
prices will be interpreted as an increase in the price level, so that the
expected price level will be revised to equal the actual one. In this case, α
= 0. Experience indicates that the value of α will lie between zero and one.
The Lucas supply rule then specifies the economy's short-run supply
function as:

371
where γ is the response of output to a unit error in the expected price level,
while α is the revision of the prior error based on new information on
product prices. If α is positive, rather than zero, a ‘positive expectational
error’ (with P > P e) will raise output above its full-employment level. The
Lucas supply rule specifies the short-run deviations in output from the full-
employment level if firms make errors in the expectations of their relative
product prices. By the definition of the long run, such errors will not occur
in the long run, so that the long-run equilibrium output level is given by:

One of the short-run extreme cases occurs when α = 1. In this case, αγ = γ

and the Lucas supply rule becomes:

The second extreme case (usually if recent experience is one of very high
inflation rates or hyperinflation) occurs when α = 0, so that γα = 0. In this
case, the Lucas supply rule, even for the short-run analysis, becomes y tSR
= y tf. In this case, given a past history of very high inflation, expectational
errors will not produce any deviations of output from full employment.
The Lucas supply rule implies that barring the case where α = 0, the
SRAS curve will have a positive slope, while the LRAS will be vertical, as
shown in Figures 8.2, 8.3a, and 8.3b.
In terms of empirical applications, the Lucas analysis implies that:
• Countries going through hyperinflation are not likely to experience an
increase in output due to inflation or further increases in inflation.9
• Countries with a history of stable prices will benefit from emergent
inflation10 because their output and employment will rise. Hence, starting
from a period of stable prices, if a central bank pursues policies that
increase the inflation rate, the economy's output will rise. However, as
firms observe the higher inflation rates, two adjustments occur in the
Lucas supply rule. One is that P e will increasingly become closer to P ,
so that (P – P e) will become smaller. The other is that α will decrease in
value. Both these adjustments will decrease the benefits from inflation;
eventually the benefits will disappear at very high and persistent inflation
rates.
• In the LSR just as in the FSR, the benefits from unanticipated inflation
disappear as soon as accurate information becomes available to the public
on the actual price level. In developed economies with speedy data
collection and release, this period may be only one or two quarters;

372
 therefore, the increase in output from an inflationary policy would be
quite transient. This is especially so for the LSR since it does not rely on
labor contracts or rigidities in prices that could have induced lags in the
impact of new information on output and employment.
Since any expectational errors become corrected by the accrual of new
information, the deviations of output from the full-employment one under
both the FSR and LSR will be automatically eliminated. Therefore, such
deviations from full employment are short-run, transient and self-
correcting.

8.4.1 Firms’ responses to errors in expectations

Central banks in economies that start with stable low-inflation rates can
increase output beyond full employment by expansionary monetary
policies, provided that such policies generate unanticipated inflation.
However, as inflation picks up, the experience of rising inflation rates
causes two changes:
1. As soon as the actual inflation rate, and the error in the expected
inflation rate, becomes known, firms learn that they had made an error
in their price expectations. They cancel its effects by revising their
expectations, which reduces or eliminates the error and the effect on
output.
2. For future inflation or the acceleration in it, firms lower their value of
their response factor α. Hence, the benefit of increasing inflation rates in
raising output lessens. It disappears in hyperinflation.
Extended Analysis Box 8.2: Diagrammatic Analysis: Comparing the FSR
and LSR Curves (FSR and LSR) and the LRAS Curve
Both the Lucas supply rule (LSR) and the Friedman supply rule (FSR)
are aspects of the short-run behavior of equilibrium output under
expectational errors. Figure 8.3a illustrates both the SRAS and LRAS
curves under the FSR and the LSR. For both hypotheses, the LRAS
curve comes about if there are no errors in expectations. The aggregate
demand curve AD has a negative slope, as derived in Chapter 5 for the
open economy.
For the SRAS, assume that the initial equilibrium price level P 0 is
known — and, therefore, fully expected with P e0 = P 0. If expected
prices remain equal to this initial price level, an increase in P will create
positive errors in expectations, implying, under both the FSR and the

373
LSR curves, that the economy will increase output in the short run.
Therefore, the SRAS curve has a positive slope at the point a. In Figures
8.3a and 8.3b, start by assuming that an increase in aggregate demand
from AD0 to AD1 shifts the actual price level to P 1, but without a shift
in the expected price level, which still equals P 0. With P 1 > P e0 = P 0,
in the short run, the output supplied will be y 1 along the SRAS0. Sooner
or later, the public will learn about the positive error in its expectations,
and revise upward the expected price level to P 1, i.e., P e1 becomes
equal to P 1 . This will cause nominal wages to rise under the FSR. It
would also cause the perception of relative price increases to decrease
under the LSR. If the price level remains at P 1, so that the actual and
expected prices become identical, output will revert to y f0, i.e., return to
the LRAS curve. The resulting fall in output from y 1 to y f0 is due to the
elimination of expectational errors.
Now, if we were to start with a price level equal to P 1 and if the
expected price level remains at P e15, equal to P 1, while the actual one
varies, a new SRAS curve will be generated through the point (P 1, y f0).
Note that each upward shift in the expected price P e produces a higher
SRAS curve — e.g., SRAS2 at P e2 = P 2. In the limiting case, if P e
always equals P , only the LRAS curve will be observed. Therefore, the
LRAS curve in Figure 8.3a was designated as LRAS|P e = P . Further,
under the LSR, the SRAS curve will become steeper (though this is not
shown in Figure 8.3a) as higher prices are experienced.
Note also that the separate SRAS curves for the FSR and the LSR are
likely to have different slopes. For the economy, the relevant SRAS
curve will incorporate both influences of expectational errors on output,
and, is therefore, an amalgam of the FSR and the LSR curves.

374
8.5 The Implications of the FSR and LSR for the
Impact of Anticipated Demand Increases
Assume that the economy is initially in full employment and that
aggregate demand increases in an anticipated manner. This increase could
come from the private components of aggregate demand (e.g., through
exogenous increases in consumption, investment, exports, etc.) or from the
public components (government expenditures and taxes, or monetary
policy). Since it has been assumed that the public anticipates such an
increase, we will also assume that its impact on prices is anticipated.
Therefore, the increase in demand will not produce any error in the
expected price level or in relative prices, so that, according to both the
FSR and the LSR, the economy will maintain full employment. Hence, the
demand increase will not bring about any increases in output and
employment, but merely cause proportionate price and nominal wage
increases.
Figure 8.3b shows an increase in aggregate demand by the shift from
AD0 to AD1. The economy is initially in equilibrium at the point a, with
the price P LR0 and output y f0. With the demand increase fully
anticipated, the error in price expectations will be zero, so that the
economy would operate according to the LRAS curve.11Along the LRAS
curve, the shift from AD0 to AD1 takes the economy to the point c , with
increases in the price level to P LR1, while leaving output unchanged at y
f . Conversely, an anticipated decrease in aggregate demand will merely
0
lower all product prices, the price level and the nominal wage rate without
changing the output of individual firms or aggregate output.

8.6 The Implications of the FSR and the LSR for

the Impact of Unanticipated Demand Increases
In the case of unanticipated demand increase, the public would not be able
to anticipate its impact on the price level, so that any price increase will
also be unanticipated, thereby generating a positive error for the expected
price level. Hence, in both the FSR and the LSR, output will increase.
Under the FSR, this would be due to the negotiated nominal wage being
too low, so that the real wage will fall and make labor cheaper. Under the
LSR, the increase in output will occur because some of the observed

375
product price increase will be attributed by individual firms to an increase
in their relative price, which will induce them to increase their output.
Diagrammatically, the supply curve relevant to an analysis of
unanticipated demand increases is the SRAS and not the LRAS curve. In
Figure 8.3a, given the unanticipated increase in aggregate demand from
AD0 to AD1 (with P e = P 0) takes the economy to the point b along the
SRAS0 curve, thereby increasing the price level along SRAS0 to P 1 while
increasing output from y f to y 1. Therefore, expansionary monetary and
fiscal policies that cause unanticipated demand increases can push output
and employment beyond the full-employment level. Conversely,
unanticipated contractions in aggregate demand can push the economy into
employment and output below their full-employment levels.

8.6.1 The impact of revisions in anticipations and real time

As explained earlier, after some delay, the public will come to learn of any
demand increase. Further, the observed price increase to P 1 will sooner or
later force the public to revise their price expectations. As this occurs and
nominal wages rise to reflect the price P 1 and the expected rise in the
relative prices decreases, the SRAS curve shifts up from SRAS0 to
SRAS1, which raises the price level to P 2. This process will shift the
SRAS curve up to SRAS2 in Figure 8.3a, with P e2 = P 2. Eventually, the
price increase resulting from the demand increase will have become fully
anticipated. This would make the expected price equal to its long-run level
at P LR1, thereby shifting the short-run supply response of the economy to
the long-run supply at y f1 and ending in the new long-run outcome at (P
LR , y f ).
1 0

8.6.2 Practical aspects of the FSR and LSR analysis

The above arguments raise the important practical question: how long does
it take in real time to go from the point a to c? During the analytical
interval in the shift from the short run to the long run, the economy has
higher output — though with rising prices — from the demand increase.
This analytical interval allows a positive short-run impact of aggregate
demand increases on output and employment in the economy. The real-
time counterpart of this analytical interval varies with the nature of the
economy, the publication of accurate data and its dissemination, the

376
alertness of the private sector, the stage of the business cycle and
adjustment costs.
The FSR and LSR imply four aspects of the real time impact of increases
in aggregate demand:
1. Under the FSR and LSR, any impact on output and employment occurs
through changes brought about by the markets in prices and nominal
wage rates, but not through any other channel, such as reacting directly
to the changes in the quantities demanded or supplied. In particular, if
the commodity and labor markets do not change prices and nominal
wages in an unanticipated manner, there would be no impact on output
and employment. Further, if these markets are slow in adjusting prices
and nominal wages, economic agents may react (faster) by changing
quantities demanded and supplied. The FSR and LSR do not provide the
analyses of such possibilities, so that there is an implicit assumption that
in terms of real time, the markets react instantly to bring about the
changes, required to restore equilibrium, in prices and nominal wages
for any anticipated shifts in aggregate demand and supply, and
economic agents react subsequently to these changes.
2. In both the FSR and the LSR, the expected price level is replaced by its
actual value as time passes and the information on the actual value
becomes available, so that the impact on output and employment is
transitory (except under the FSR where the duration of the wage
contracts has to be considered). Therefore, any deviations of the
expected price level from the actual price and, therefore, from full
employment are self-correcting. Consequently, in the well-informed,
modern, developed economies, this impact is unlikely to last for more
than a few quarters.
3. Further, as the inflation rate accelerates from low single digits to higher
ones, economic agents become more alert to the inflation rate and
changes in it. They shorten the duration of their contracts and try to
become better informed about the determinants of the future inflation
rate. Hence, the relative magnitude and duration of the real-time impact
of unanticipated inflation becomes shorter. Therefore, not only are the
deviations from full employment transitory and self-correcting, their
real-time duration becomes shorter in inflationary periods.
4. There is no benefit from increases in demand, prices, and inflation
during hyperinflation.
Hence, in terms of real time for the modern economies, the FSR and the
LSR imply that the fluctuations in investment, exports, and other
components of demand should only cause transitory (only a few quarters)

377
and self-correcting deviations in output from its full-employment level.
Such deviations should, then, be hardly of much concern to the public or
policymakers who are more concerned with real time rather than with
analytical time.
Extended Analysis Box 8.3: The Implications of the FSR and LSR for the
Impact of Monetary and Fiscal Policies
We have to differentiate between two different roles of monetary and
fiscal policies, depending upon the stage of the economy prior to their
pursuit. These roles are:
i. The stabilization role for monetary and fiscal policies
Stabilization policies are ones that are pursued to eliminate or reduce the
economy's deviations from full employment. These deviations occur due
to prior shifts in the private components of aggregate demand or shifts in
aggregate supply. They are intended to stabilize output and employment
at their full-employment levels.
The FSR and LSR imply that such deviations can occur. If these
deviations take significant real time before the economy eliminates
them, the policymakers can step in and pursue the appropriate policies to
speed their elimination. Such policies are known as stabilization
policies.
ii. The proactive role for monetary and fiscal policies
Policies intended to raise the levels of output and employment above
their full-employment levels are said to play a proactive role. As we
have shown above, based on the analysis of the FSR and the LSR,
starting with a position of long-run equilibrium, expansionary monetary
and fiscal policies can increase employment and output in the economy
provided that they are not anticipated ones and create positive errors in
the expected rate of inflation. Hence, the proactive role of policies under
the FSR and LSR requires the creation of unanticipated inflation.
However, in some ways, this represents an attempt to ‘fool the public’.
Focusing only on the proactive role of central bank policies —
irrespective of whether expectations are rational or not rational or
whether systematic policy is fully anticipated or not — there are benefits
and costs to any unanticipated expansion of the money supply. The
benefits occur in the present in terms of higher employment and output.
The costs come later. There are several types of costs.
One of these costs is the damage to the policymaker's credibility with
the public since the central bank pursued policies that misled the public.

378
This loss of credibility means that the public will become sceptical about
the central bank's announced future targets for money supply growth and
inflation.
Another cost occurs later when the central bank becomes concerned
about high inflation and wants to switch to a policy of low inflation and
uses a contractionary monetary policy, but the public refuses to adjust its
inflationary expectations as fast. The result is unanticipated negative
errors in the expected inflation rate, which pushes employment and
output below their full-employment levels. Creating unanticipated
inflation becomes a policy of ‘benefit now, pay later’, which raises
doubts about the net benefits of such a policy.
Uncertainty about the future pursuit of policies and the rates of
inflation and uncertainty increases, which makes business planning for
investment more risky, and affects the long-term growth of capital
accumulation and output growth.
If the economy already starts from a state of full employment and if
the expansion in the money supply is anticipated, long-run equilibrium
implies that it will have no effect on employment and output, but will
only change the price level and the nominal wage — but not the real
wage.
To conclude, under the FSR and LSR, monetary and fiscal policies can
have a valuable stabilization role to play in the economy if the
economy's deviations from full employment, caused by non-policy
shifts, take significant real time before the economy can eliminate them.
Their role in a proactive form has both benefits and costs.
The Implications of Rational Expectations for Systematic Monetary
and Fiscal Policies
Rational expectations imply that the public will anticipate a systematic
policy, whether monetary or fiscal; therefore, its implied price increase
will also be anticipated. Further, the only type of policy that will not be
anticipated will be a random one. Therefore, the FSR and the LSR imply
that systematic monetary and fiscal policies will not shift the economy
away from full employment, while random ones will do so.
However, a random policy does not make sense. To illustrate the
nature of a random policy, a random monetary policy will increase the
target interest rate or decrease the growth of the money supply in
magnitudes specified by a table of random numbers. Using money
supply as an example, if the random number for the current quarter turns
out to be 3%, the money supply would be increased by 3%; next quarter,
the random number might turn out to be –10%, so that the money supply

379
would be cut by 10%; and so on. In following this pattern of random
numbers for growth rates, the central bank would not be pursuing a
monetary policy related to the perceived needs of the economy. Hence, a
purely random policy would not serve the interests of the economy.
Since the only types of policies that could change output and
employment in the economy are random one and these do not make
sense, adding rational expectations to the FSR and the LSR implies that
the monetary and fiscal policies should not be pursued. The conclusion
is that the monetary and fiscal authorities should leave the economy
alone in the sense of not using their policies in a futile attempt at
improving on its performance.
This is a very strong conclusion on the applicability and pursuit of
monetary and fiscal policies, and, at the practical level, not all
economists subscribe to it. Most central bankers do actively reduce or
raise interest rates as they think fit, so that they do not follow the
implications of the above conclusion.

The Scope for Monetary and Fiscal Policies for Stabilization when
Private Demand is Volatile

Several of the private components of aggregate demand are volatile —

i.e., change over time for reasons not encompassed in the model. The
short-run analysis shows that this can induce movements in
unemployment, output, interest rates, and prices that may not be
desirable. Therefore, one of the roles for policies is the stabilization one
aimed at offsetting any undesirable changes in aggregated demand
coming from the volatility of its private components.
The preceding analysis implies a distinction between the effects of
unanticipated, possibly random, changes in private demand and those of
anticipated ones. In the latter case, the change would only be in prices
and possibly interest rates, but not in unemployment and output, which
will remain at their full- employment levels. If the policymakers are not
bothered by the changes in prices and interest rates, they need not take
any action. If they are bothered by even these changes, they can take
counteraction designed to stabilize aggregate demand.
Unanticipated changes in the private components of aggregate demand
will induce short-run changes in unemployment and output, as well as
changing prices and interest rates. The policymakers may choose to
tolerate them since they can be expected to be transient ones.
Alternatively, the policymakers may consider them to be undesirable
and try to offset them by appropriate counteracting policies. In practice,

380
an unanticipated demand reduction is more likely to be considered
undesirable since it increases unemployment and reduces output, than an
unanticipated demand increase since this reduces unemployment and
increases output in the short run. Therefore, our analysis implies that
policymakers are more likely to try to offset unanticipated decreases in
demand than other types of demand changes.
However, note that unless the policymaker has better information than
the private sector, unanticipated demand changes in the private sector
will also be unanticipated by the policymakers so that they can only try
to offset them once they become known — by which time they will also
be known to the private sector — and have become anticipated.
Therefore, the stabilization role for policies has to be based on either:
a. Superior information available to the policymakers than to the private
sector — which is an aspect of asymmetric (i.e., between the
policymaker and the public) information, and/or
b. Long-time lags in the impact of private demand changes on the
economy along with a faster impact of policy induced changes in
demand.

The Scope for Monetary Policies and the Political Economy of the
Government's Budget
Aggregate demand depends on the fiscal deficit. The size of the fiscal
deficit is often determined by the expenditures needs of the government,
including the size of the public debt and the payments on it, and the
taxes it can raise, both of which are heavily influenced by the economic,
political, and social context — including the desired levels of military
spending, anti-poverty programs, wars, disasters, etc. This context may
produce continuous deficits over many years and continuous surpluses
over other years.
Therefore, the size of the deficit is rarely determined by the
stabilization needs of the economy, so that monetary policy has to often
assume the main — or the sole — burden of economic stabilization. In
performing this role, it may not only have to offset undesirable
fluctuations in the private components of demand but also redress the
undesirable impact of the fiscal position on aggregate demand.
Therefore, monetary policy is the pre-eminent stabilization policy in
most developed countries. However, even in these countries, an
expansionary fiscal policy is commonly used to supplement monetary
policy in severe recessionary conditions induced by a fall in the private
components of aggregate demand.

381
8.7 FSR and LSR: The Impact of Anticipated
Permanent Supply Changes on the Economy 12
As in the case of aggregate demand changes, the impact of anticipated
supply changes differs from that of unanticipated ones. For anticipated
supply shifts, assume that there is an anticipated permanent increase in
productivity. This shifts the LRAS curve to the right, from LRAS0 to
LRAS1 in Figure 8.4a.

For the FSR, assume that the output supply increases and its effect on the
price level was known when the wage contracts were signed, so that the
wage contracts anticipated the new long-run price level. Therefore, under
the FSR, the relevant SRAS curve would now go through the point c at the
intersection of LRAS1 and AD0. For the LSR, assume that the firms know
the impact of the LRAS shift on the price level, so that they know that
there is no alteration of relative prices. Hence, there will not be any errors
in relative price expectations. Therefore, under the LSR, the relevant
SRAS curve would also shift so as to go through the point c at the
intersection of LRAS1 and AD0. The SRAS curves were shown in this
figure to elucidate the assumptions being made on expectations but are
clearly not relevant to the long-run impact of the productivity increase on
output.
Hence, for the anticipated positive supply increases, the long-run
equilibrium of the economy will move from the point a at the intersection
of LRAS0 and AD0 to the point c at the intersection of the LRAS1 and
AD0 curves, with output increasing from y f0 to y f1 and the price level
falling from P 0 to P 1.

8.8 The Impact of Unanticipated but Permanent

Supply Changes on the Economy
382
Now consider the impact of an unanticipated but permanent increases in
supply due to a positive productivity shift. For the FSR, we will assume
that this shift was not known at the time the wage contracts were signed. In
the new LR equilibrium, as seen in the preceding section, the MPL is
higher but the equilibrium price level is lower, so that the value (P.MPL)
of the MPL could become higher or lower at the new full-employment
level (at the point c in Figure 8.4a) than at the old full-employment level
(at the point a in Figure 8.4a). In the short run, under the FSR, the
contracted nominal wages (Pw) proves to be (a) lower than justified by the
new (unexpectedly higher) marginal product of labor, but (b) higher than
justified by the new (unexpectedly lower) LR price level. The former
increases employment while the latter decreases it. The net effect on
employment could be positive or negative. But note that the increase in the
MPL will raise output for any given level of employment, so that the SR
employment could become lower while its accompanying output becomes
higher.
For the LSR, the LR price level has fallen because of the positive supply
shift. Assume that this positive supply shift introduces errors in relative
price expectations: firms mistakenly expect a reduction in relative prices
when it is a general price reduction due to the supply increase; therefore,
they decrease output relative to the new LR full-employment level. Hence,
while output tends to increase because of the productivity increase, the
price fall acts under both the FSR and the LSR to hold back output and
employment relative to their new long-run levels: employment would fall
under both the FSR and the LSR.
In Figure 8.4b, the initial curves are LRAS0, SRAS0, and AD0. The
positive productivity increase shifts the LRAS curve to LRAS1, with the
new long-run output as y f1 and the corresponding LR price level as P 2.
The new SRAS curve (incorporating the old expected price level but the
new productivity level) is SRAS1. It lies below SRAS0 but above the
eventual SRAS2 curve through the point c . The equilibrium between
SRAS1 and AD0 has been shown as (y 1, P 1). As contracts are
renegotiated and the expectational errors are rectified, the SRAS curve will
move down to SRAS2 to go through the long-run point c. Starting from (y
0, P 0), this movement produces continuing increases in output and
reductions in the price level until the new LR equilibrium is reached at (y
f , P ).
2 2
Hence, positive supply shocks tend to produce, over several periods,
gradually increasing output and falling prices — or at least act to moderate

383
the rate of inflation that would otherwise occur due to demand increases.
However, over some of these periods, employment could fall even when
output is rising.
We showed in Chapter 7 that oil price increases produce long-run
declines in labor productivity and aggregate supply. The preceding
analysis can be extended to derive the short-run effects of anticipated
versus unanticipated changes in oil and other energy costs on the economy.
This extension is left to the students to pursue.
Box 8.2: Cost of Living Clauses
Many wage contracts contain clauses to ensure that the nominal wage is
automatically increased by the rate of inflation. Such clauses are called
cost-of-living clauses. Their effect is to ensure that the negotiated wage
rate embodied in the contracts becomes effectively the real wage rate.
Other contracts in which a payment is agreed upon can have similar
clauses.
Cost-of-living clauses are mostly inserted in contracts in periods of
high, especially hyper, inflation rates. They serve to eliminate the effects
of unanticipated inflation. Given such a clause in the wage contracts, the
Friedman supply rule ensures that employment and output cannot be
changed by inflation and will remain at the full employment levels.

8.9 The Short- and Long-Term Relationships

between Output and Inflation
For the LSR, high and persistent (which approximates the long-run) rates
of inflation over time will lead firms to expect that all price increases in
their individual product prices are a reflection of the general price increase,
so that they will not increase output. For the FSR, high and persistent rates
of inflation will lead workers to adopt better mechanisms for predicting the
price level or reduce the duration of wage contracts sufficiently.
Alternatively, they may put cost-of-living clauses in wage contracts, so
that nominal wages rise automatically with the price level. These reduce or
eliminate the errors in price expectations and their impact on real wages
will be reduced. As a consequence, under high and persistent inflation
rates, the SRAS curve will be vertical or virtually so. The end result under
both supply rules will be that high and persistent inflation will not produce
significantly higher output than the full-employment one even in the short
term.

384
 As a corollary, high and persistent rates of increase in the money supply
that produce high and persistent rates of inflation will also not increase
output. Therefore, the economy's output will not show much variation in
response to changes in the inflation rates and money supply growth rates in
periods of high and persistent inflation.
Since the SRAS curve has a positive slope (because of both the FSR and
the LSR) at low rates of inflation, unexpected increases in the inflation
rates and money supply can produce some increases in output for some
time. Therefore, at low rates of inflation, the economy will tend to show a
positive relationship between output and inflation over the short term.
In an article published in 1973, Robert Lucas estimated a version of his
output equation for a cross section of countries and found that countries
with low rates of inflation (such as the USA in the 1950s and 1960s)
showed evidence of a positive relationship between output and demand
increases, while those with hyperinflation (such as Argentina in the 1950s
and 1960s) did not, thereby concluding that this relationship shifts as
inflation increases. This finding is consistent with the preceding analysis.

8.10 Empirical Validity of the FSR and the LSR

The major economic crisis of 2007–2009 in the USA and other countries
was due to a fall in aggregate demand and a decline in credit availability
brought about the mortgage-based financial crisis. It was not due to any
significant initial decrease in the inflation rate, so that it was not due to
errors in price or inflation expectations. Therefore, the FSR and the LSR
cannot account for the decline in output and employment during this crisis.
For the overall validity of the FSR and LSR, we draw upon the ‘Nobel
Lecture’ given by Robert Lucas (1996)13 on his receipt of the Nobel Prize
in economics. Robert Lucas wrote that:

“…anticipated and unanticipated changes in money growth have very

different effects” (1996, p. 679). However, on the models which
attribute this non-neutrality to unanticipated or random changes in the
price level, the evidence shows that “only small fractions of output
variability can be accounted for by unexpected price movements.
Though the evidence seems to show that monetary surprises have real
effects, they do not seem to be transmitted through price increases ,
as in Lucas (1972).” (1996, p. 679; italics added).14

This quote on the empirical validity of the FSR and the LSR downplays

385
their importance as a major contributor to valid explanations of the real-
world deviations of output and employment from their full-employment
levels. This assessment is consistent with the point emphasized several
times in this chapter that deviations in output caused by errors in price
expectations tend to be transient and self-correcting. Therefore, we need to
look for additional sources of such deviations. This is done in the rest of
this chapter and the next chapter.

8.11 Types of Adjustment Costs and Their Impact

on Output
The preceding analysis in this chapter has focused on adjustments in prices
and their expectations. However, there can also be other sources for
deviations in output and employment from their long-run levels. Among
these are costs of adjusting prices, employment, and production. Briefly,
the three types of adjustment costs are:
1. costs to the firm of changing prices under imperfect competition,
2. costs to the firm of changing employment and the capital stock, and
3. costs to the firm of changing output at the pre-existing levels of
employment and the capital stock.
Next, we consider these three types of adjustment costs in greater detail.
1. If firms operate under some form of imperfect competition and
determine the price that they set for their products, there may be a cost
to changing these prices. The costs of changing product prices are called
‘menu costs,’ which refers to the cost of reprinting restaurant menus to
indicate new prices. The theory of menu costs implies that prices will be
sticky — i.e., left unchanged for some time — under certain
circumstances, even though they are not rigidly set but would change
under other circumstances.
2. Firms also face adjustment costs if they wish to change employment and
the capital stock. To change employment, firms have to incur hiring and
training costs. Further, firms usually need workers with some firm-
specific skills, which can only be gradually acquired by workers while
working in the firm in question.
These factors lead to implicit or explicit contracts between the firm and
its workers for long-term employment and labor hoarding in response to
short-run fluctuations in aggregate demand and production. In a sense,
the employment of workers and their hours worked become sticky,
while that of the employee's work effort is made variable. To increase

386
 (decrease) the capital stock, the firm has to undertake investment
(disinvestments), which the firm would be willing to do if the change in
aggregate demand was expected to be permanent but not if it was
expected to be temporary. Note that any change in the capital stock
involves planning, investment, and integration of the new
machines/capital into the existing ones, so that there are adjustment
costs of changing the capital stock, which makes the capital stock also
sticky. Firms get around this stickiness by changing the rate of
utilization (e.g., by using them over more hours per day) of their capital
stock.
Given the costs of changing employment and the capital stock, firms
may try to increase their output with their existing employment and
capital. They do so by varying the rates of utilization of their existing
capital and the speed and work efficiency of their employees. However,
in doing so, they again incur adjustment costs.
3. Costs to the firm of changing its output with the existing levels of
capital and workforce. In the short term, firms can vary the level of
effort of its employees, as well as capital utilization.
The following two sections consider the analysis of sticky prices for the
commodity market and that of sticky employment for the labor market.
These produce short-run commodity supply behavior that is different from
the long-run one (when, by assumption, there are no adjustment costs).
Note that for their analysis, the long run is being defined as that analytical
period which has zero adjustment costs and lags.

8.11.1 Firms’ responses to increases in the demand for their

products
The preceding arguments imply that, when demand increases, firms have
the following options:
• Raise prices, while leaving output unchanged (as under the long-run
aggregate supply with full-employment output).
• Increase output to meet the demand increase, while leaving prices
unchanged (the sticky price hypothesis).
• Not change output or prices, thereby leaving demand unsatisfied at the
prices charged.
• Some mix of the above options.
In general, adjustment costs of various types individually and together
imply that the short-run-term supply response of the economy to aggregate

387
demand changes is slower than the long-run one15 and slower than that
posited by the Friedman supply function (FSR) and the Lucas supply
function (LSR).

8.12 Menu Costs and Price Stickiness as the

Explanation of the SRAS Curve
While some commodities in the economy are homogeneous and traded in
perfectly competitive markets, many commodities — especially at the
retail level — are differentiated by firms in some way or other. Such
differentiation is often in the form of differences in color, packaging,
location, associated services, or just established brand loyalty. Such
differentiation in practice is usually not enough to create a monopoly for
the firm but enables it to function in a monopolistically competitive
manner. Profit maximization by a monopolistically competitive firm
implies that it is not a price taker, as firms are under perfect competition,
but a price setter with a downward sloping demand curve for its product.
Consequently, increases in the price it sets neither reduce its sales to zero
nor do reductions in it allow it to capture the whole market for its industry.
As a price setter, the firm sets the profit-maximizing price and supplies the
output demanded at this price.
Changing the set price imposes a variety of costs, collectively known as
menu costs. Examples of these are: reprinting price lists and catalogs,
informing customers, re-marking the merchandise, etc. These costs, though
often relatively small as a percentage of the price of the firm's product, can
still be greater than the gain in revenue from a small price change. Further,
even if there is a net gain from changing the price following an increase in
demand, it may not be enough to persuade the firm to immediately raise its
price since the inconvenience and costs to the firm's customers of frequent
price changes are likely to be resented. Consequently, the firm may not
find it optimal to respond to demand changes with price changes unless the
demand changes imply large enough price changes. Over time, as demand
increases accumulate, the optimal price change becomes large enough for
the firm to be willing to incur the menu costs and change the actual price
of its product. These arguments imply — though for a monopolistically
competitive context rather than a perfectly competitive one — that:
• In the short run/term, the (monopolistically) competitive firm will change
its prices infrequently, but will accommodate intervening changes in
demand by changing its output at the existing price. In the limiting case

388
 of no change in the price but only a change in output to meet a change in
demand, prices are said to be ‘sticky’ and the firm's supply curve
becomes horizontal.
• In the long run/term, the competitive firm will adjust its prices to reflect
demand changes.
• Even in the short run/term, if the demand change is large enough, the
price adjustment will become profitable.
The aggregate economy is a mix of firms in perfect competition and
monopolistic competition. Further, when aggregate demand increases,
while some firms may be experiencing sticky prices, others may be
coming out of such a phase. Hence, at any given time, the economy will
only have some firms with sticky prices. An increase in aggregate demand
will cause some sectors and firms, especially those with more competitive
markets, to adjust their prices faster while others will not immediately do
so but will respond to demand changes by changing output. Consequently,
an increase in aggregate demand will be partly met by an increase in the
price level and partly by an increase in aggregate output, so that the short-
run supply curve will have a positive slope, as shown in Figure 8.5b, but
not as steep as when none of the firms is suffering from sticky prices.
Conversely, a decrease in the demand for the products of the firm in
monopolistic competition need not immediately cause it to lower its price
unless the implied optimal price reduction was sufficiently large.
In the extreme case, if all firms in all sectors simultaneously have sticky
prices, the average price level in the neighborhood of the initial
equilibrium would become constant. In this extreme case, shown in Figure
8.5a, the price level is assumed to be constant at its initial level P 0. Prices
are sticky at P 0, so that we can specify a short-run aggregate supply curve
SRAS which is horizontal at P 0, while the LRAS curve still remains
vertical at y f. The increase in the aggregate demand from AD0 to AD1
raises the supply of output from y f to y 1 at the sticky price P 0.
Conversely, the decrease in the aggregate demand from AD0 to AD2 leads
to the supply of output y 2 but again without an accompanying change
from the sticky price P 0.
Therefore, the sticky prices theory implies that:
• Transient and small changes in aggregate demand are accommodated by
a change in output.
• Cumulative or persistent changes in the same direction — or large
aggregate demand changes — will, however, make it optimal for all firms

389
 to increase prices, so that the long-run response to such changes is taken
to be along the LRAS curve.

• Caution should be exercised in the use of a horizontal SRAS curve for

the aggregate economy since this requires assuming that all firms will
simultaneously have sticky prices. As pointed out earlier, the economy
has a mix of perfectly competitive, monopolistically competitive,
oligopolistic, etc., markets, so that the sticky prices argument does not
apply to all commodities. Even for a given commodity, some producers
may be in a sticky prices phase, while others are coming out of it, so that
the adjustment of price lists is staggered. Therefore, even under the
sticky-prices argument, the normal case for the aggregate economy is an
upward-sloping SRAS curve rather than a horizontal one.
• Another caution, from the implications of the menu cost approach to
price stickiness, is that prices will adjust faster the greater the increase in
demand, so that the greater the demand increase, the more quickly will
prices adjust and the less will be the effect of aggregate demand increases
on real output. Hence, larger increases in aggregate demand and,
therefore, potentially greater inflationary pressures, will produce smaller
real effects. Hence, the sticky-price SRAS curve bends backward above a
certain point.

8.12.1 Aggregate supply in the sticky-price hypothesis

The supply equation under the sticky-prices hypothesis can be illustrated
by:

where β > 0 if |(y dt – y t-1) | is ‘small’, while β = 0 if |(y dt – y t-1) | is

‘large’.16 Note that the variation in output is from last period’s level, not
from the full-employment level. This sticky-prices short-term equation
differs from the FSR and LSR by making the increase in demand, not in
the price level, the engine for increases in output. Therefore, in the sticky-
price model, expansionary monetary and fiscal policies have a direct
impact on output without necessarily first producing an increase in the

390
price level. This implication accords well with the stylized facts on the
impact of changes in the money supply on output.
Corresponding to the output adjustment, there is a price adjustment
equation, which can be illustrated by:

Totally, sticky-price level would require α = 0. However, more

realistically, 0 ≤ α ≤ 1 in the short run.

8.12.2 Monetary policy and the sticky-price SRAS curve

The sticky-price analysis implies that, for the analysis of monetary policy,
relatively small reductions in the target interest rate — and, hence,
relatively small increases in aggregate demand — would increase output
but larger reductions in the target interest rate — and, hence, relatively
larger increases in aggregate demand — would produce smaller increases
or no increase in output. Further, there is no direct causal relationship
between from inflation to output; rather, the causal relationship runs from
aggregate demand to output — as well as (later) to inflation. From the
perspective of inflation, only relatively large increases in aggregate
demand would necessarily cause significant inflation in the short term.
These conclusions are very different from those of the theories (FSR and
LSR) based on expectational errors in prices. However, the implications of
the sticky-prices argument that concur with those of the theories based on
expectational errors are:
• Any increase in output induced by an expansionary policy will be
temporary. Prices will sooner or later adjust and output will return to its
full-employment level.
• There is no clear-cut and strong case for the continuous or massive
pursuit of proactive expansionary policies as a route to a lasting increase
over several quarters and years in output and employment beyond their
full- employment levels.

8.13 Costs of Adjusting Employment: Implicit

Contracts as an Explanation of the SRAS
Curve
The production function implies a unique relationship between
employment and output. This is based on an implicit — and rather

391
unrealistic — assumption that work effort and, therefore, the output per
hour of work is constant at any given level of employment. Under this
assumption, as employment and output increase, the marginal productivity
of labor (MPL) declines, so that the real wage rate also declines.17 Figure
8.6 shows the standard (long-run) labor demand curve as n d. It has the
downward slope derived earlier from the production function in the
preceding chapter: the MPL (and average product of labor) decline as
employment n increases from its long-run equilibrium at n f. This decline
implies that higher employment will be associated with lower real wage
rates.

8.13.1 Variations in work effort over the short term

In the short term, aggregate demand increases usually induce increases in
the work effort of employees relative to their long-run norm at full
employment. This moderates the decline in the MPL, which would have
otherwise occurred as employment increases. Conversely, short-term
decreases in demand induce a reduction in effort, so that the MPL does not
rise as much as implied by the production function. Therefore, in Figure
8.6, instead of the standard (long-run) MPL curve labeled as MPLLR, the
short-term MPL curve becomes MPLST. This short-term MPL curve —
and, therefore, the short-term nd curve — is flatter than the long-run one
for variations in employment induced by aggregate demand changes.
This theory, allowing variations in work effort, is known as the implicit
contract theory. An implicit contract between two parties is an
understanding or a verbal arrangement rather than a written, legal
document. The implicit contract theory deviates from the standard
production function in two ways.
1. It allows for variations in work effort, with variations in the MPL as

392
 aggregate demand changes.
2. It allows for long-term implicit contracts between firms and workers,
with labor hoarding (i.e., retaining workers but using them at less
intensity than at their normal long-run level) in demand-deficient
periods and increased work effort during excess demand periods.
Box 8.3: Work Effort in Restaurants During the Day
Restaurants provide a good example of variations in work effort during
the day. Restaurants experience spurts of demand during the lunch and
dinner hours, with very little demand during other periods. They work at
a faster pace during the high-demand hours while doing very little work
during the low-demand hours. The average work intensity over the day
can be thought of as the long-run work intensity for that type of
restaurant, with short-run variations in work intensity induced by the
fluctuations in demand. In particular, the restaurant does not resort to
laying off some workers during the slack periods while hiring more
during the busy ones, since these would increase hiring and training
costs, poorer service by temporary employees, as well as a loss of staff
morale. Holding onto excess staff during the slack periods and reducing
their effort demanded of them is called labor hoarding.
If, during a particular week, the restaurant experiences an increase in
its customers, it usually accommodates them by further increases in the
work intensity of its existing employees, even if this exceeds the long-
run norms on work intensity for that restaurant or class of restaurants.
However, this level cannot be sustained on a long-run basis, otherwise
employee morale will suffer, leading to increases in absenteeism, poor
service, and resignations. If this demand remains exceptionally high
over the longer term, the restaurant would increase its staff or raise its
prices, or both. The former is often done whether the increase in demand
is a relative one or a general one for the economy.
If the number of customers decreases, the restaurant is likely to
respond by reducing the work effort of its staff without laying-off an
appropriate number or changing its prices. Once it is realized that the
decrease in demand is a longer-term one, the restaurant has the options
of reducing its prices and laying-off some employees. The latter is often
pursued, whether the decrease in demand is a relative one for the
restaurant or a general one for the economy.

Work Effort by Students over the Term and the Calendar

Year

393
 Students follow similar patterns of variations in study effort during the
academic term and the calendar year. At one extreme, there are vacation
(and other!) days that are sometimes spent without any studies being
done. At the other extreme, there are intense study days and nights near
examinations. Students do not load up with extra courses on a short-term
basis in slack periods and drop some when pressure builds up. Rather,
they follow some notion of a long-run norm for the average work effort
over the term and the year. Variations around this norm occur, leading to
increased effort and productivity near examinations and reduced ones
during slack periods.

8.13.2 The Flexibility of the Employment-Output Nexus in

the Short Term
Chapter 7 had specified the relationship between output and employment
by the production function, which is strictly a one-to-one relationship, so
that a given level of output is associated with a unique level of
employment. This relationship assumes a constant ‘work-effort’, which is
to be interpreted as the intensity of work. However, the existence of costs
of adjusting employment implies that firms may sometimes find it less
costly to increase output by increasing the work effort of their employees
rather than by increasing their employment, or do so by some combination
of the two.

Long-term labor contracts and labor hoarding

Firms and workers often find it in their interests to enter into long-term
implicit (non-written) and explicit (written) employment contracts when
there are costs of hiring and training workers and/or workers possess some
firm-specific skills, usually acquired through training and learning on the
job. For such workers, the productivity of an existing skilled employee
will be greater than that of new hires. The employed worker also benefits
from this higher productivity through higher wages in his/her existing firm
than if he/she was to quit and join other firms. This mutual benefit from
continued employment implies that the firm would try to retain its skilled
workers if it can do so through a period of reduced demand for its output,
rather than laying them off immediately. The firm, therefore, finds it
optimal to (temporarily) lay off less workers than justified by the fall in
demand, leading to a form of labor hoarding during recessions. Such
hoarded labor works less hard during recessions because there is less work
to do or is often diverted to low-productivity tasks such as maintenance

394
etc. In the case where a worker is laid off, the worker also has an incentive
to wait to be recalled by his/her old employer rather than immediately
accept a job with another firm in which his/her productivity and wage will
be lower. The firm and its workers are then said to have an implicit
contract for long-term employment under certain types and magnitudes of
fluctuations in demand, with short-term fluctuations in work effort to
permit output to respond to demand fluctuations. Therefore, reductions in
aggregate demand in the short term partly lead to labor hoarding, with a
consequent fall in average productivity, and partly to an increase in
unemployment, with some of the laid-off workers being put on recall and
voluntarily waiting to be recalled rather than actively searching for jobs.18
An implicit agreement between firms and its workers also means that
workers accommodate increases in the demand for the firm’s product with
increased effort, even in the absence of wage increases, so that moving
along the SRAS curve away from full-employment output will be partly
accommodated by changes in work effort. Hence, increases in aggregate
demand will increase the average product of labor (APL)19 relative to its
LR trend, and, therefore, increase aggregate output. The APL will be
procyclical and output will fluctuate more than employment over the
business cycle.
The short-term production function and the output supply function under
the implicit contract hypothesis are:

where e stands for work effort, which varies with the excess or shortfall of
demand relative to last period’s output. Its implications are similar to those
of the sticky-price output function. However, it differs from the latter in
stressing that, proportionately, the variations in employment will be less
than in output.

8.13.3 Okun’s Rule: The Relationship between

Unemployment and Output Changes
Okun’s rule — named after Arthur Okun — is the statement that, in the
short term, increases in the level of aggregate output are accompanied by
less than proportionate decreases in the level of unemployment. To
illustrate this with an example from one country, over the business cycle,
one estimate of this relationship for Canada is that a 5% increase in the
economy’s output rate is accompanied by approximately a 3% decrease in

395
the unemployment rate.
Okun’s rule runs counter to the assumptions made for the production
function, under which increases in output are accompanied by more than
proportionate increases in employment (because of diminishing marginal
productivity of labor) and more than proportionate decreases in the
unemployment rate. The explanations for Okun’s rule lie in (a) the short-
run variations in labor effort and labor hoarding, and (b) the discouraged
worker effect. (a) implies that as demand increases, the average
level/intensity of effort increases so that the average productivity of labor
also increases. Hence, employment rises — and unemployment falls —
less than proportionately with output.20
Another reason for Okun’s rule is the discouraged worker effect. During
recessions, some workers are discouraged after looking some time for jobs
and meeting rejections of their applications. In such circumstances, some
workers close to retirement may choose to retire earlier. Both these drop
out of the labor force and the measured unemployment rate. As demand
increases and firms increase employment, the prospect of landing a job
rises. This induces some of the discouraged workers to re-enter the labor
force, so that the labor force increases, making the decrease in the
unemployment rate less than proportionate to the increase in output.
Therefore, over the business cycle, the unemployment rate fluctuates less
than the growth rate of output around its trend.

8.14 The Implications of Adjustment Costs for

Persistence of Output and Employment
Fluctuations Over the Business Cycle
The long-run analysis (in Chapter 7) is based on the assumption of instant
adjustment of the equilibrium price level to shifts in the demand and long-
run supply functions. The analysis of this chapter shows that adjustments
in output and prices by firms are not instantaneous but gradual because of
the existence of costs in changing employment, prices, and output. Further,
firms face costs in adjusting their capital stock, which, combined with the
uncertainty of future sales, make their adjustment of physical capital much
slower than the instantaneous one assumed in the long-run analysis.
Consumers also have valid reasons for slower than instantaneous
adjustments of their consumer demand and labor supply.
What these arguments translate into in terms of our macroeconomic
diagrams is that the movements along the various demand and supply

396
curves can be quite slow ones spread over several quarters, possibly years,
rather than an almost instantaneous one. To illustrate, the slow adjustment
of the capital stock implies that the change in investment specified by the
investment function will occur over many quarters. Movements along the
AD and AS curves will be slow ones. Movements along the labor demand
and supply curves will also be slow. In fact, movements along any of the
curves used for macroeconomic analysis are likely to be slow and gradual
ones, rather than the instantaneous ones assumed in the AD-LRAS
analysis.
In practical terms, starting with the economy at the peak of the business
cycle, if aggregate demand or supply were to decrease because of
exogenous shifts, the declines in output and/or prices will not come about
instantaneously. If this were to happen, the economy would have gone
instantly from the peak of the business cycle to its trough. Instead, output,
prices, and employment tend to decline gradually in a stretched-out
recessionary movement. Conversely, starting with the economy in the
trough of the business cycle, if aggregate demand or supply were to
increase because of exogenous shifts, the increases in employment, output,
and/or prices will not come about instantaneously. If this were to happen,
the economy would have go instantly from the trough of the business cycle
to its peak. Instead, output, employment, and prices tend to increase
gradually in a stretched-out upturn. Therefore, the existence of adjustment
costs implies that the economy would not display sudden highs and lows
of output and employment, but is likely to follow business cycle upturns
and downturns of several years’ duration.
A corollary of this argument is that there is a high degree of persistence
in output and employment fluctuations over the business cycle. This
persistence in the sign of output changes means that an increase in output
(or employment) in one period is more likely to be followed by another
increase rather than a decrease, and vice versa. By comparison, the
instantaneous adjustment assumed in the long-run analysis would produce
random changes in output and employment in response to random shocks
to aggregate demand and supply. ‘Random changes’ in output means that
an increase in output (or employment) in one period is as likely to be
followed by another increase as a decrease, and vice versa. Business cycles
clearly display persistence of the signs of changes in output and
employment, rather than random changes, except at their peaks and
troughs.
Therefore, the incorporation of the adjustment costs of employment and
variations in work effort under implicit contracts implies that over the
business cycle:

397
• Variations in output occur due to demand fluctuations, whether
anticipated or not.
• Output fluctuates more than employment.
• The economy produces output higher (lower) than the full-employment
one in response to demand increases (decreases).
• The APL increases as output rises in booms and decreases as output falls
in a recession. Hence, the APL would be procyclical.
• Under Okun’s rule, employment increases less than proportionately as
output rises, especially in the early parts of an upturn. Conversely,
employment declines less than proportionately as output falls, especially
in the early parts of a downturn.

8.15 Implications of Adjustment Costs for the

Impact of Monetary and Fiscal Policies
The essential implication of the analysis of this chapter is that the full
impact of a significant shock, whether to demand or supply and whether
from private or policy decisions, on output and employment is likely to be
spread over some time. The behavior of the economy during this period is
not necessarily identical with that which equilibrium analysis leads us to
expect. In particular, the theories of sticky prices and implicit contracts
imply that the responses to demand and supply shocks are likely to include
changes in quantities, e.g., in output and employment, as well as in prices.

8.16 Stagflation and the Recessions of 1973–1975

and 1980
Stagflation occurs when the economy has inflation while its output is
stagnant (i.e., not changing). In the more general case, the term stagflation
means inflation with constant or falling (or at least not rising) output.
Stagflation was also discussed in Chapter 7, Section 7.9. Fact Sheet 7.2
and Box 7.1 in Chapter 7 related the experience of stagflation in 1973–
1975 and 1980 in many countries to increases in the price of oil.
As shown in Chapter 7, output would be stagnant in an economy
(without growth) in the long run of a static economy (i.e., without
technical change or capital or labor force growth), since output is constant
at the full-employment level. In such a context, continuous increases in
aggregate demand would produce an increasing price level, so that there
would be inflation. Frequent causes of continuous demand increases are

398
continuing expansionary monetary and fiscal policies.
A special case of stagflation occurs when output is falling (and not
merely constant) while the economy has inflation. The fall in output can be
due to a declining marginal product of labor, due to a negative shift in the
production function, a decreasing physical capital stock or rising prices of
inputs. On the last cause, Chapter 7 discussed the declines in output
brought about by oil price increases in 1973–1975 and again in 1980–
1982. These periods were marked by expansionary monetary and fiscal
policies, meant to increase aggregate demand in the belief that the increase
in the demand for commodities would increase output. In fact, output fell
in many oil-importing countries because of the impact of the rising oil
prices while their expansionary demand policies caused inflation. These
two periods provide clear cases of stagflation.
Box 8.4: Empirical Evidence on Price Changes, Wage Contracts and
Output Response
In a relatively low inflation context, one survey revealed that about 10%
of firms change their prices less frequently than once a year, about 39%
change their prices once a year, about 28% change them up to four times
a year, while about 21% change them more often. Of the latter, about
10% change more frequently than once a week. Therefore, about 59% of
firms change their prices four or more times a year. The most common
reasons given for infrequent price changes were waiting for other firms
to go first (about 60%), not changing prices until costs rise (about 55%),
fairness to customers (about 50%), etc. The cost of changing prices was
cited by only about 30% of the surveyed firms.21 This seems to suggest
limited support for the sticky-price hypothesis.
Empirical studies show that the average duration of wage contracts
was about one year in the 1970s. All wage contracts are neither
renegotiated on the same date nor do all price changes occur on one
date. Both wage contracts and price changes are staggered and
overlapping in the economy, so that the average nominal wage rate and
the price level are never constant. Hence, even over the short term, the
aggregate supply curve for output is never horizontal.

8.17 Stylized Facts and Intuition on the Theories

Explaining Variations in Output
Note that this chapter has presented three major approaches (price

399
expectations, sticky prices, and implicit contracts) and four theories (FSR,
LSR, sticky prices, and implicit contracts) to explain variations in output
from the full-employment level. How do these theories compare with the
empirical evidence? For low-inflation economies, the stylized facts on the
impact of aggregate demand on output are:
• Positive (negative) demand shocks, including both anticipated and
unanticipated ones, have a positive (negative) impact on current
aggregate demand and output. Demand shocks sometimes have a smaller
impact on output than at other times. [This finding is consistent with the
sticky prices and implicit contracts hypotheses. The variations in output
in response to anticipated demand shocks are not consistent with the FSR
and the LSR. The variations in output in response to unanticipated
demand shocks are consistent with all four theories.]
• The impact of increases in aggregate demand or of the money supply on
output does not always or necessarily require a prior impact on the price
level or inflation. Not all of the impact of changes in aggregate demand
on output filters through changes in prices or inflation. [This finding is
consistent with the sticky prices and implicit contracts hypotheses, but
not with the FSR and the LSR.]
• Current inflation is positively related to current output and an increase in
the inflation rate causes it to exceed expected inflation for the current
period. [This finding is consistent with the FSR and the LSR but not
necessarily inconsistent with the sticky prices and implicit contracts
hypotheses.]
• Past inflation is negatively related to current output. Past inflation leads
to higher expected inflation rates for the current period. [This finding is
consistent with the FSR and the LSR but not necessarily inconsistent with
the sticky prices and implicit contracts hypotheses.]
• For economies with very high inflation rates, further increases in
inflation or in aggregate demand do not tend to increase output and
employment. [This finding is consistent with the FSR and the LSR but
not necessarily inconsistent with the sticky prices and implicit contracts
hypotheses.]
• Cutting back on aggregate demand or lowering the inflation rate
significantly from past levels sends the economy into a recession and
reduces output and employment. [This finding is consistent with all four
theories.]
• Employment fluctuates less than output over the business cycle. [This
finding is only explained by the implicit contracts hypotheses.]

400
8.18 Conclusions
• The FSR based on nominal wage contracts and the LSR based on errors
in perceived relative prices imply that output and employment increase
when there is an unanticipated increase in the price level.
• The FSR and LSR imply that expectation errors in prices generate the
positive slope of the SRAS curve. Under rational expectations, only
random policies can generate such errors. Systematic policies will not
generate these errors and will not produce deviations from full-
employment output.
• In the FSR, expansionary demand policies induce an increase in output
through a reduction in real wages induced by an unanticipated increase in
the price level. Any such reduction will be temporary, with a
readjustment when wages are renegotiated or, even earlier, under
pressure from disappointed workers and at the volition of firms. The LSR
does not require a prior reduction in real wages.
• The FSR and the LSR assume optimization by both firms and households
and clearance of the labor market in one way or another. Therefore, these
theories rely on the assumption of equilibrium to derive the SRAS curve
and its policy conclusions. In this equilibrium, firms can sell all that they
want to sell and there are enough jobs for all the workers who want jobs
at the existing wage. Note that involuntary unemployment does not occur
along the LRAS curve, or along the SRAS one.
• The deviations from full-employment output due to errors in price
expectations under the FSR and LSR are transient and self-correcting.
They account for only some of the deviations that occur in the real world.
• The costs of adjusting prices lead to sticky prices, so that changes in
demand are reflected in output changes. In the sticky-prices model, menu
costs imply that firms’ profits could fall if they were to adjust their prices
by small amounts. Rather than change prices, firms find it profitable to
accommodate changes in aggregate demand by changing output and
employment in the short run. For these firms, the supply curve would be
horizontal in the vicinity of their long-run position.
• At any given time, since not all firms in the economy find it profitable to
maintain sticky prices, the sticky- price model does not really imply a
horizontal SRAS curve for the aggregate economy. Further, more rapid
increases in aggregate demand will produce smaller rather than larger
increases in output.
• For firms experiencing sticky prices, an increase in aggregate demand
increases their output without a prior or accompanying increase in prices.
Therefore, the impact of monetary policy on output is not a function of

401
 the change in their prices but rather of the change in the aggregate
demand — and through it, in the money supply — itself. In fact, any
increase in prices reduces the impact of monetary expansion on output.
• In the sticky-price model, the impact of aggregate demand and the money
supply on output is not uniform since relatively small increases in
aggregate demand and the money supply increase output but very rapid
increases in them induce larger increases in aggregate demand which
produce more rapid price adjustments, with consequently smaller
increases in output.
• Costs of adjusting employment explain the existence and nature of
implicit contracts, which allow adjustment of work effort, with labor
hoarding in recessions, so that labor productivity changes as aggregate
demand changes.
• Changes in work effort imply that employment fluctuates less than output
in response to demand fluctuations and that the marginal product of labor
and real wages do not necessarily have to fall in upturns or rise in
recessions, as they do under the long-run production function.
• In the FSR, expansionary demand policies induce an increase in output
through a reduction in real wages (induced by the increase in the price
level). Any such reduction will be temporary, with a readjustment when
wages are renegotiated or, even earlier, under pressure from disappointed
workers and at the volition of firms. The LSR and the sticky-prices
models do not require a reduction in real wages.
• Each of these theories is based on optimization by both firms and
households and clearance of the labor market in one way or another.
Therefore, these theories rely on the assumption of equilibrium for their
short-run/term analysis and its policy conclusions. In this equilibrium,
firms can sell all that they want to sell and there are enough jobs for all
the workers who want jobs at the existing wage under the existing
explicit and implicit contract arrangements.
However, real-world economies do go through recessions and booms, and
sometimes through depressions. Unemployment and output fluctuate very
significantly over time. These casual observations suggest the possibility
that there may be periods (of deflationary demand) during which the firms
collectively cannot sell all the output that they produce if they offered full
employment to workers in the economy. These may also be periods during
which some of the unemployed workers — in spite of possession of the
appropriate skills for jobs and the willingness to accept the existing wage
rates for their skills — cannot get jobs. That is, there exists the possibility
that the economy may experience levels of employment and output that are
neither at the full-employment level nor on the short-run supply curve of

402
the economy. The next chapter will explore this possibility.

KEY CONCEPTS

Price expectations
Adaptive expectations
Rational expectations
Expectational errors
Expectational equilibrium
Wage contracts
The short-run employment-output equation/relationship
The short-run aggregate supply(SRAS) curve
Friedman supply rule (FSR)
Lucas supply rule (LSR)
Imperfect competition
Implicit contracts
Firm-specific skills
Labor hoarding
Okun’s rule
Sticky prices
Persistence in output and employment fluctuations, and
Stagflation.

SUMMARY OF CRITICAL CONCLUSIONS

• Nominal wage contracts allow the impact of the errors in the expected
rate of inflation on the real wage, employment, and output. This effect is
captured in the Friedman supply rule/curve, which explains the deviation
of output from its full-employment level due to the reduction in the real
wage under unanticipated inflation. This rule is based on the errors in the
expectations – of households and firms – embodied in nominal wage
contracts.
• The Lucas supply rule/curve for output is based on the errors made by
firms in their expectations of the relative prices of their commodities.
• The deviations in output from its full-employment level due to errors in
price expectations are transient and self-correcting.
• The costs of adjusting prices, output and employment are also significant

403
 determinants of the short term deviation of output and employment from
their long-run levels.
• Some firms, especially monopolistically competitive or monopolistic
ones, find it best to maintain their prices in response to limited demand
increases for their products, so that they are said to have sticky prices in
the neighborhood of the full-employment level of output.
• In the short run, firms are able to vary the work effort of their employees,
so that employment fluctuates less than output. Since the average
productivity of labor rises in booms and falls in recessions, the APL
moves procyclically over the business cycle.
• In the theories of sticky prices and implicit contracts, in the short term,
changes in demand directly impact on output and employment rather than
indirectly through prior price changes brought about by markets. In the
FSR and LSR, demand changes must alter prices in order to have any
impact on output even in the short run.
• The sticky-price hypothesis relies on some form of imperfect competition
with firms able to manipulate their prices rather than under the strict
assumptions of perfect competition.

REVIEW AND DISCUSSION QUESTIONS

1. Do rational expectations allow errors in the expected rate of inflation

relative to its subsequent actual value? If so, explain the nature of this
error.
2. Explain the two-stage process in the FSR for the determination of
employment.
3. Specify the equation for output under the FSR. Why should an actual
inflation rate higher than the expected one cause output to rise under the
FSR? Explain.
4. Discuss the following in the context of the FSR: “The central bank has
to keep on accelerating the inflation rate in order to continuously
maintain employment and output above their full-employment levels”.
5. “A boom due to accelerating inflation is bound to bust and end in a
recession when the central bank is fed up with the ever-rising inflation
and decides to lower it”. Explain how this would happen in the context
of the FSR.
6. Explain the contributions of Robert Lucas on the dependence of output
and employment on errors in expectations. Why would the response of
the economy to a rise in the inflation rate differ between economies that

404
 follow low-inflation strategies versus those that maintain very high-
inflation rates?
7. [This question is based on interest rate targeting and, therefore, on the
IS-IRT analysis]. Suppose the government cuts income taxes, show its
impact in the long-run and the short-run output (i.e., along the LRAS
and SRAS curves based on the FSR and the LSR) under two alternative
assumptions:
(i) The central bank uses interest rate targeting and keeps the target
interest rate constant.
(ii) The central bank uses interest rate targeting and changes the target
interest rate to maintain aggregate demand equal to long-run supply.
8. What is stagflation? Describe a situation that could produce stagflation?
Could the stagflation in the situation described by you have been
avoided through an expansion of the money supply?
9. What is a cost-of-living clause in wage contracts? If all wages were
indexed to the inflation rate, how would this affect the applicability of
the FSR? Discuss the short- versus long-run effects of an initially
unanticipated but permanent increase in oil prices on output,
employment, real wage, real interest rate, and the price level.
10. In the early 1990s, the Bank of Canada pushed up the economy’s
interest rates (and decreased the growth rate of the money supply) to
reduce inflation in the Canadian economy. Analyze the effect of such a
contractionary monetary policy on the interest rate, aggregate demand,
and output in the economy:
(i) If this monetary policy was unanticipated.
(ii) If this monetary policy was anticipated.
Use the IS-IRT and AD-AS diagrams for your answer.
11. Explain menu costs. What implications do they provide for the shape
of the SRAS in industries with menu costs? What implications do they
provide for the shape of the SRAS if the economy has some industries
with menu costs and others without any such costs?
12. “The sticky-price hypothesis implies that the aggregate supply curve is
horizontal in the short term”. What is the sticky-price hypothesis? Is this
statement likely to apply to modern economies in the short term?
Discuss.
13. Discuss the following in the context of the sticky-price hypothesis:
“The central bank has to keep aggregate demand in excess of long-run
aggregate supply order to continuously maintain employment and output
above their full-employment levels”.

405
14. Suppose that firms can vary the work effort of their workers in the
short run in response to variations in the demand for its products. How
does this affect the marginal product of labor in the short run versus the
long run? Explain.
15. Assume that firms face adjustment costs in changing their prices and
employment. Suppose the government cuts income taxes. Show its
impact in the long run and the short run under two alternative
assumptions:
(i) The target interest rate remains unchanged.
(ii) The central bank adjusts its target interest rate to keep aggregate
demand constant.
16. Explain the implicit contract theory.
17. In 2004 and 2005, the US government ran large fiscal deficits while
the Federal Reserve System of the USA raised the Federal Funds rate
several times to curb but not eliminate excess demand. Using only the
general nature of this information (rather than the specific situation in
the USA at the time), discuss the initial cause of excess demand as
specified in this question? Using the analysis of adjustment costs,
discuss the likely effects of the remaining excess demand on aggregate
demand and output in the short run.
18. The demand expansion of 2004-2005 in the USA initially increased
output without a corresponding significant impact on employment, but
subsequently also led to a strong growth in employment. How can this
pattern be explained by the relevant adjustment cost theories presented
in this chapter?
19. What is Okun’s rule? Provide the justification for this proposition.
20. What does Okun’s rule imply for the relative variations in employment
and output over the business cycle? What does Okun’s rule imply for
the relative variations in unemployment and output over the business
cycle?

ADVANCED AND TECHNICAL QUESTIONS

Basic structural AD-AS model for this chapter [with interest rate
targeting and an endogenous money supply]:

c = 0.8yd

t = 0.2y

406
 i = 1000 – 50r

g = 800

x c = 200 – 0.5pr

Z c/ρ r = 500 + 0.1yd

m d = 0.25y – 60R

Output:
Long-run output: y f = 1,000
Short-run aggregate supply function: y = 1,000 + 0.1(P – Pe)
[Note: depending on the context, the short-run aggregate supply can be
stated as a positive function of the error in the expected price level or of
the error in the expected inflation rate.]
Foreign trade sector:

PF=1

ρ=1

The central bank targets (i.e., fixes) the domestic real interest rate, so
that:

r = r T = 0.04

T1. For the basic structural AD-AS model above,

(a) Derive the IS equation? [Do not use the IS equation formula. Do the
calculations step by step. To do so, first calculate the real exchange rate
in terms of the domestic price level P. Then, calculate disposable
income. Substitute these in the consumption, import, and export
functions. Then, use the equilibrium condition y = e to derive the IS
equation for the information in this model.]
(b) What is the level of aggregate demand y d0?
(c) Equating demand and long-run supply, what are the long-run levels
of real output and the price level? Designate this output as y f0 and the
price level as P *0.
T2. For the basic AD-AS model, assume that the economy was in long-run

407
 equilibrium in period 0. In period 1, the government raises its
expenditures to 1,000.
(a) Calculate the new level of aggregate demand (y d1).

(b) Since y d1 > y f, calculate the new long-run equilibrium price level P
LR and output y LR .
1 1
T3. [This question uses the rational price expectations hypothesis in which
economic agents know the new long-run price level and their expected
price level is equal to the appropriate long-run price level.] Using the
basic AD-AS model and your answers to the preceding question, answer
the following. Assume that, in period 1, the expected price level P e1
equals the long-run price level P LR1, but, in fact, the actual price level
in period 1 rises by only 50% of the increase required to move to P LR1.
(a) Using the SRAS function, what is the price level P SR1 and the level
of output y SR1? Designate the latter as y SR#1. (b) Explain the reasons
provided by the FSR and the LSR for the change in output from y f to y
SR# .
1
T4. [This question uses the static price expectations hypothesis, according
to which the expected price level for the period ahead is the actual price
level one period earlier. ] Using the basic AD-AS model and your
answers to the preceding question, answer the following. Now, assume
that in period 1, the public expects that prices will stay at P *0 (i.e., P e1
= P *0). Using the short-run supply function, calculate the output in
period 1.
a. For period 1, calculate the increase in the short-run levels of real
output.
b. What is the short-run fiscal (government expenditures) multiplier for
real output in period 1? Is it zero? Explain your answer.
c. What is the long-run fiscal (government expenditures) multiplier for
real output? Is it zero? Explain your answer.
T5. Comparing your results in questions T2 and T3. (a) Which provides a
smaller deviation from full- employment output? (b) Which seems a
more realistic expectations hypothesis as far as your own behavior is
concerned? Discuss.
T6. [Optional] Note that in the preceding three questions, the target real

408
 interest rate had remained at r LR0, even though fiscal expenditures
increased. Now assume for period 2 that the central bank realizes that
aggregate demand has risen. It adjusts the target rate r T to 20. Calculate
the new equilibrium price level P *2 and output y *2. Compare these to P
* and output y * . Are they different? Why or why not? Explain your
2 0
answer.
Basic reduced-form AD-AS model:
For the basic reduced-form AD-AS model, assume that the following
equation describes the aggregate demand function for the economy. IS
equation:

y t = 100,000 + 0.2gt - 10,000rt - 100Pt t = 0, 1,…,gt = 1,000

The central bank’s real interest rate targets are:

In period 0, r T0 = 0.04

In periods 1, 2, and 3, r T1 = r T2 = r T3 = 0.05

The money supply adjusts to be consistent with money market

equilibrium at the interest rate target.

m d = 0.25y - 60R

Fisher equation:

R=r+πe

π e0 = π e1 = π e2 = π e3 = 0

Long-run output:

yf = 90,000

Because of adjustment costs, the short-run aggregate supply function is

given by:

y t = 90,000 + 0.2(y td – y t–1)

409
 [Note: in this specification of the SRAS, SR output is a function of full-
employment output and the deviation of the demand for output from last
period’s output. This SR output function is in the Keynesian tradition, in
which firms adjust output in the short run in response to changes in the
demand for their products. Such an SR function is consistent with the menu
cost theory, which is in the Keynesian tradition. For comparison, note that
the classical tradition would make the derivation of SR output from the
full-employment level an function of the error in prices, as in the FSR and
the LSR.]
Output in period t – 1: The economy is in long-run equilibrium in period t
– 1 with y t – 1 = y f = 90,000. Price level: The economy maintains the
price level in period 0 that is consistent with long-run equilibrium.
T7. Given the basic reduced-form AD-AS model, assume that the
economy is in long-run equilibrium in period t – 1 with y t–1 (=10,000)
being the initial long-run output. The current period is t = 0.
(a) What is the level of aggregate demand in period 0? What are the
levels of actual and long-run output in periods 0, 1,2, 3, and 4?
Discuss the impact of demand on the time pattern of actual output.
(b) [Optional] Suppose that g rises to 1200 in period 1. Calculate the
output level for periods 1, 2, 3, and 4.
(c) [Optional] Using the above answers, calculate the fiscal (government
expenditures) multipliers for actual output? What is the long-run
fiscal (government expenditures) multiplier for output?

1Wc will increase proportionately if both P ef and P eh increase by the

same amount but not if only one of them increases.
2Clearly, this would only be so in the neighborhood (i.e., applying only to

small changes) of the expected equilibrium employment level at W c in the

(W, n ) diagram.
3As Pef rises, labor is expected to become cheaper, so that firms will
increase their labor demand for a given value of W , say W 0. This higher
labor demand at the given W 0 implies that the n d curve shifts right in the
(W, n ) diagrams.
4As P eh rises and lowers the expected lower wage for a given nominal
wage W 0, labor will require a higher nominal wage W to supply a given
amount of labor, say n 0. The higher value of W for the given n 0 implies

410
an upward shift of the labor supply curve in the (W, n ) diagrams.
5The expansion of 2004–2006 in the USA and Canada was remarkable in
that the inflation rate remained low (1% to 3%) and there seemed to be no
perceptible errors in inflation expectations. This may have been due to the
successful pursuit of interest rate targeting monetary policy during the
period.
6Cost-of-living clauses in wage contracts ensure that nominal wages rise
automatically with the price level.
7Disequilibrium would occur if the market — or wage negotiations — do
not bring about the equality of the demand and supply of labor. For this
case, we would need the disequilibrium analysis presented in the next
chapter.
8Note that the employment level is not set in the wage contract and can
deviate from n e.
9This is so because, in a hyperinflationary environment, firms are likely to
view any inflation as merely due to changes in the price level, rather than
due to changes in their relative prices.
10This is so because, with a history of stable prices as the past experience,
firms are likely to view (though mistakenly) any inflation as being due to
an increase in their relative prices rather than an increase in the general
price level.
11It will not operate according to the SRAS curve for the anticipated
demand increase since that has been assumed not to induce errors in price
expectations.
12This would include knowledge of past shifts in demand and supply and
past variations in local and general prices, but can also include any
available information about the future.
13Lucas, Robert E. Jr. (1996). Nobel lecture: Monetary neutrality. Journal
of Political Economy, 104, 661– 682.
14Lucas, Robert E. Jr. (1972). Expectations and the neutrality of money.
Journal of Economic Theory , 4, 103–124.
15The long run is that analytical period when there are zero adjustment
costs and lags, and no uncertainty.
16The magnitude of ‘small’ and ‘large’ depends on the firm’s behavior.
17Hence, they behave counter-cyclically over the business cycle.
18Note that the economy is quite diverse with some sectors resorting to
labor hoarding while others not.
19This occurs since all employed workers, and not merely the marginal

411
one, work harder. In this scenario, real wages are likely to be mainly
determined by the long-run MPL, though short-term changes in wages are
likely to be somewhat responsive to the short-run movements in the APL.
20Conversely, in the short run, as demand falls, firms decrease the average
effort of their employees and resort to labor hoarding, in order not to lose
valuable employees while waiting for the pick-up in demand. If this
happens, average labor productivity falls.
21Blinder, Alan S. (1994). On sticky prices: Academic theories meet the
RealWorld. In N.G. Mankiw (ed.), Monetary Policy . Chicago: University
of Chicago Press.

412
 CHAPTER 9
Actual Output, Disequilibrium, and
the Interaction among Markets

The economy is often not in short-run equilibrium, let alone in

long-run equilibrium. If either the commodity or the labor market
is not in full-employment equilibrium, the study of the interaction
between these markets becomes important. The resulting
disequilibrium analysis gives different results from the equilibrium
ones on the determination of output and employment.
There can be numerous causes and symptoms of
macroeconomic disequilibria. This chapter presents the analyses
of two cases: when the economy develops a demand deficiency
and when the real wage is too high, relative to their equilibrium
levels. In each of these cases, the failure of a market to clear at the
full-employment level (i.e., to have adequate demand to buy the
full-employment output) impacts on the other markets. This
chapter derives the impact of the resulting disequilibrium in the
commodity and labor markets on output and employment, and the
appropriate role of demand and supply side policies in such
conditions.

The analysis of full-employment output under the long-run supply curve

(see Chapter 7) and of deviations from it along the short-run supply curve
(see Chapter 8), is based on the assumption of market clearance (i.e., the
equality of demand and supply) in all markets, with economic decisions by
households and firms taking it for granted that market clearance will, in
fact, immediately occur following any shifts in demand and supply. Hence,
both households and firms believe that they can buy and sell as much as
they want at the existing market prices and act accordingly in making their
decisions on consumption, investment, labor supply, labor demand,
production, etc. This chapter examines the interaction between the
commodity and labor markets when they are not in equilibrium (i.e.,
demand and supply are not equal, that is, they do not clear), so that

413
consumers and firms, knowing this non-clearance and behaving rationally,
modify their demand and supply for commodities and labor accordingly. It
derives the implications of this interaction between the commodity and
labor markets for output and employment.
However, the economy sometimes enters a situation in which there are
not enough jobs for all the workers who are willing to accept jobs at the
existing wage rate. In this case, the difference between the available
number of jobs and the number of workers at the existing wage represents
involuntary unemployment. The involuntarily unemployed workers are
unable to sell the labor that they want to sell at the existing wage, even
though they possess the requisite skills and are in the right locations, as
evidenced by their employment just prior to the fall in aggregate demand.
Unable to earn incomes, they reduce consumption expenditures, which
reduces aggregate demand. Therefore, from household behavior, there is a
special nexus, running from employment to the demand for commodities.
This nexus is not fully apparent in the AD-AS analysis studied so far.
Similarly, the economy sometimes enters situations in which firms want
to sell more commodities than they are able to sell. In particular, the
demand for commodities becomes less than what the firms can produce if
they were, collectively, to hire the number of workers specified by the full-
employment level. Such a situation is known as that of deficient demand.
In this case, since firms will not want to pile up unsold inventories, they
will reduce employment below the full-employment level. Therefore, from
firm behavior, there is a special nexus between the demand for
commodities and employment, which translates into a special nexus
between aggregate demand and unemployment. This nexus is also not
fully apparent along the LRAS and SRAS curves.
Given the preceding arguments, we need to study the causes of the
persistence of disequilibrium and the behavior of the economy during
disequilibrium. This chapter does so. The worldwide economic crisis and
recession of 2008–2010 attests to the empirical relevance and importance
of the analysis of this chapter for the economy.

This chapter focuses on the potential deviations of actual output from the
output shown by the LRAS and SRAS curves. The relevance of these
curves depends on whether the economy is always in equilibrium or not.
Some economists claim that the economy is always, or almost always, in
or near this state, and, if it gets moved away from it, reverts quite fast and
on its own to this full-employment state. Most economists and economic
policy makers, however, believe that the economy can be out of

414
equilibrium for several quarters, if not years. This occurs because,
following a shock, the economy’s required adjustments to reach the new
equilibrium are varied and take time. The fluctuations in unemployment
and the existence of recessions and booms provide casual empirical
validity to the economy being sometimes out of its equilibrium state.1
When the deviations from equilibrium occur, the roles of monetary and
fiscal policies become quite different from those derived so far from the
LRAS (see Chapter 7) and the SRAS (see Chapter 8). In particular, since
such deviations can be brought about by shifts in the private components
—such as consumption, investment, and money demand — of aggregate
demand and the economy has lags in restoring equilibrium, monetary and
fiscal policies acquire the very useful and important role of stabilizing the
economy at or close to its full-employment level. This stabilization role of
demand-management policies does not occur if the economy is always in
full-employment, as in Chapter 6.

The Three Components of Actual Output

As a reminder, note that Chapter 7 had divided actual output into three
components. These are repeated in the following identity.

where:
yf : long-run (full-employment) level of output (see Chapter 7).
(y* the deviation of the short-run equilibrium level of output from the
– full-employment one due to errors in expectations under uncertainty
yf) : and adjustment costs (see Chapter 8)
(y – the deviation, in disequilibrium, of output from its short-run
y*): equilibrium level (this chapter)
This chapter focuses on the last component: that is, on the deviations of
output from even the short-run equilibrium level. Real-world economies
do not continuously maintain equilibrium in the commodity and labor
markets. In many instances, demand and supply shocks cause the economy
to initially move away from its equilibrium. While it may soon start
adjusting towards equilibrium, this process takes real time during which
the economy is in disequilibrium and away from its equilibrium
employment and output levels. During this adjustment period, which can
take several quarters or even years, the powerful conclusions on the
ineffectiveness of monetary and fiscal policies in long-run equilibrium
derived in Chapter 7 do not apply. Further, the channels of adjustment may

415
not be as specified in the short-run supply hypotheses in Chapter 8. This
chapter focuses on the likely adjustment scenarios and the roles of
monetary and fiscal policies during this adjustment period.
Fact Sheet 9.1 shows the relationship between money supply growth and
real output growth in the USA. This relationship occurs because an
increase in money supply growth causes aggregate demand to increase,
which stimulates output, thereby creating a positive relationship between
money growth and real output growth, so that increases in money supply
growth are followed by increases in output growth, and decreases in
money supply growth are followed by decreases in output growth. This
relationship is better explained by the disequilibrium analysis of this
chapter than by the short-run analysis of Chapter 8.
Fact Sheet 9.1: Money Growth and Output Growth in USA, 1960–2008
This Fact Sheet illustrates the relationship between money supply
growth and real GDP growth. The graph below illustrates this positive
relationship using US data: both variables follow a similar zigzag
pattern during the 1970s, increase in economically prosperous times
such as the early 1980s and late 1990s, and fall during the late 1980s and
early 2000s. Further, peaks and troughs in money supply precede peaks
and troughs in real GDP growth, so that the former cause the latter.
As a corollary, this graph shows that in real time and for real-world
economies, money supply shifts do not leave either real output or
velocity unchanged/constant, as asserted by the quantity theory of
money.

9.1 The Relevance of Expectations on Variables

Other Than Prices
The short-run analysis of Chapter 8 had considered the impact of firms’
and households’ expectations on prices and that the optimal supply

416
responses of firms to changes in demand take adjustment costs into
account. This chapter argues that there are also other reasons for firms to
form expectation on the future demand for their products, as well as for
workers to form expectations on the future demand for their labor.
Therefore, the expectations formed by firms are on:
(i) the quantity that they would be able to sell,
(ii) the price at which their product will be sold, and
(iii) on wages to be paid.
Similarly, workers form expectations on jobs, wages, and nominal and
real incomes. These expectations, just as those on prices, play important
roles in the decisions of firms and households/workers. For example,
firms’ expectations of future sales strongly influence their investment and
the number of jobs they want to offer. Households’ expectations of being
employed or becoming unemployed, as well as their expected incomes,
strongly influence their consumption.
The role of the expectations formed on the ‘quantities’ of the (real)
variables and their nominal values was excluded from the short-run
analysis provided by the Friedman and Lucas supply rules (Chapter 8) in
constructing the SRAS curve. However, their role is important enough for
economic analysts to construct indices of ‘business confidence’ and
‘consumer confidence’ to explain what is happening to output and
employment in real time in the real world. Some aspects of this role of the
expectations formed on the nominal and real values of the real variables
are captured through the disequilibrium analysis specified in this chapter.
Another component of this chapter’s analysis is the interaction between
the commodity and labor markets when one of these fails to achieve
equilibrium. This analysis will be irrelevant to economies if prices and
wages change instantaneously to ensure their equilibrium values. It
becomes relevant when they do not do so. In the latter case, there is
spillover of disequilibrium between markets.
This chapter explores the role of expectations on variables other than
prices and of deviations from the equilibrium levels of employment and
output when there is interaction between the labor and commodity
markets.

9.2 Shocks to Aggregate Demand and Supply

The shocks to—that is, shifts in—aggregate demand can come from the
domestic private sector, the foreign sector, the central bank, and the

417
government. Shocks from the central bank and the government are
discussed as aspects of the pursuit of policy. Shocks to aggregate demand
from the domestic private sector occur because of shifts in consumption,
investment, net exports, or the demand for money. Each of these variables
tends to vary over time, often on a daily basis. We illustrate some of the
major sources of shocks to these variables.
Examples of shocks to consumption
First, consider the shocks to consumption. Clearly, the seasonal variations
in climate and in holidays cause variations in consumption over the
calendar year. More important for macroeconomics is the impact of
changes in wealth on consumption. The major components of household
wealth are houses, bonds, and stocks. The prices of bonds and stocks are
determined in the bond and stock markets and do not consistently mirror
either the rate of inflation or the changes in the physical stock or the sum
of the two. To illustrate the shifts in wealth that can occur, Table 9.1
presents the change in the Dow Jones Industrial Index — which serves as a
rough indicator of the stock market prices in the United States — during
the 20th century and up to 2008.2
Table 9.1 provides a very rough indication of the changes in the nominal
and real values of the financial wealth held in stocks over the last century
by examining changes in the (USA) Dow Jones Industrial Index for the
New York Stock Exchange. These variations affected consumption,
pushing up consumption if the stock markets were doing particularly well,
and pushing it down if they were doing badly, relative to income. They
also affected firms’ ability to raise funds for investment. The resulting
variations in consumption and investment expenditures led to variations in
aggregate demand.
Table 9.1 Changes in the Dow Jones industrial index (DJII)

418
 The largest single item of wealth in household portfolios is the value of
owner-occupied houses. House prices fluctuate considerably, often rising
for several years and sometimes falling for several years. Increases in
house prices increase household wealth, which for a given level of income,
persuades households to increase their consumption expenditures and
reduce their saving.
Another source of shocks to consumption comes from changes in
consumer confidence. Consumers go through episodes of optimism
(especially in economic booms) or pessimism (especially in recessions)
about their future job prospects and wages. Optimism encourages a
reduction in precautionary saving which is partly a hedge against future
job loss or fall in the value of accumulated wealth, so that consumption
expenditures increase. This is represented by either an increase in
autonomous consumption or in the marginal propensity to consume, or in
both. Conversely, if consumers become pessimistic about the future,
precautionary saving rises and consumption expenditures fall.
At current levels of income and wealth in the developed economies, the
impact of changes in wealth on the expenditures on durable consumer
goods, such as kitchen appliances, TVs and other entertainment units,
automobiles, etc., tends to be much greater than those on non-durables,
such as food and rental units.
Examples of shocks to investment and net exports
Investment is often the most volatile component of aggregate demand.
Changes in it are caused by shifts in:
• Technology, of which changes in information technology (IT) have been
very noticeable in the last few decades.
• Firms’ existing physical capital relative to current aggregate demand.

419
• Availability of qualified workers.
• Costs and availability of external funds. The cost of such funds is
approximated in the investment function by the interest rate.
• Expected future demand for commodities and workers.
• Business confidence.
• Firms’ ability to raise funds from external sources. Firms use short-term
credit and long-term bonds and stocks for their external sources of funds.
The availability of such funds is also an important consideration in
determining the actual levels of investment (see Chapter 16). Rapid
increases in the stock market prices of a firm’s shares improve its ability
to raise additional funds for investment, while rapid decreases in the
stock market prices of a firm’s shares are detrimental to its ability to raise
additional funds for investment. The availability of short-term credit is
also important for investment in the short term.
The uneven pace of investment over time, due to shifts in one or more of
its determinants above, is the most frequent and significant cause of
fluctuations in aggregate demand.
After investment, exports are the next most volatile component of
aggregate demand. Shifts in exports occur due to changes in the real
exchange rate, foreign incomes, and domestic supply constraints. Their
determinants are discussed in Chapters 3, 5, 12, 13, and 16.
Shocks to money demand and supply
Money demand is influenced by many factors, among which are the
returns on stocks and bonds, which are affected by speculation in the stock
and bond markets. Many economists believe that the speculative
component of money demand is very volatile since it depends on the
subjective expectations of returns in the highly speculative bond and stock
markets. These expectations are often based on inadequate information and
are influenced by ‘herd behavior’ and ‘contagion’.3
Shocks to the cost and availability of credit
Credit provided by banks and other financial institutions to households is a
major determinant of their purchases of durable goods and houses, etc.
Credit provided to firms is a major determinant of their investment,
especially in inventories, and of exports (since there is a time lag between
production of exported goods and payment for them by buyers abroad).
Therefore, reductions in credit, whether due to a reduction in the money
supply or a perception by financial institutions of an increase in the risk of
lending to consumers and firms, reduce consumption, investment and
exports, so that aggregate demand falls.

420
9.2.1 Shocks to the aggregate supply of commodities
Shocks to aggregate supply occur through:
• shifts in the production function,
• the supply behavior of labor and the supply of other inputs, and
• the supply of credit and other funds to firms.
Of these, the shifts in the production function depend upon changes in
technology, which is rarely a smooth and even process over time.
However, in general over the whole economy, shifts in technology tend to
be less volatile in terms of the frequency of occurrence and relative
magnitudes than shifts in the private components of aggregate demand.
In addition, shocks can also occur to the prices of most of the inputs,
including the wage rate, energy prices, and the interest rate, which can
affect the cost of production and, therefore, the supply of output.
The financial crisis of 2007–2010 in major economies also showed that
reductions in credit limit the ability of firms to finance production.
Production firms employ labor and buy other inputs in advance of sales of
their products, so that they have to bridge the time gap between their cost
disbursements and receipt of revenue. This is partly done through credit
borrowed for short periods from financial institutions directly in the form
of loans or through sale of short-term bonds in bond markets. Therefore,
reductions in credit supply reduce the production that can be undertaken,
thereby reducing aggregate supply in the economy.

9.3 An Analysis of the Disequilibrium Following

a Demand Shock

9.3.1 The assumptions for disequilibrium analysis

The analysis of disequilibrium for any given comparative static
(equilibrium) model is tricky since there can be many possible
disequilibrium patterns. Hence, we should not expect a consensus on the
particular pattern that will hold. In particular, the disequilibrium patterns
depend on the specific assumptions made for behavior in disequilibrium.
To limit these patterns, we will maintain the assumptions of the rationality
of economic behavior by households and firms in the light of the
information available to them. 4 However, we will dispense with the
assumption of instant market clearance in each individual market and
instant restoration of general equilibrium in the economy, i.e., our analysis

421
will allow the possibility of some disequilibrium for some time, since these
assumptions are quite unrealistic ones.5
To start this analysis, take the economy to be initially in full-
employment equilibrium. Now suppose that aggregate demand falls for
some reason such as an exogenous decrease in consumption, investment,
fiscal expenditures, or an increase in taxes — or an increase in the central
bank’s target interest rate. Such a fall in aggregate demand is shown in
Figure 9.1 by a shift from AD0 to AD1. We need to analyze the plausible
dynamic responses to it by rational economic agents in the modern
economy.
Effective aggregate demand and supply
Effective demand is the actual demand that exists in the economy. It is
based on the actual consumption of workers, actual investment by firms,
actual government expenditures, and actual net exports. It differs from the
level of demand that would exist if the economy was in full employment,
and expected to continue functioning in full employment. This level of
demand is called notional demand. Effective demand will be less than
notional demand if there is not sufficient demand for full-employment
output. This is called ‘deficient demand ’. If effective demand is greater
than notional demand, the economy is said to have ‘excess demand’. An
economy with excess demand will usually be in a boom (see business
cycle theory in Chapter 16).

The effective supply of commodities is the actual supply that exists in the
economy. It differs from notional supply , which exists when the economy
is functioning at full employment. Effective supply will be less than
notional supply in recessions caused by a fall in aggregate demand, since
firms then produce enough to meet actual demand. Effective supply is
usually greater than notional supply in booms caused by an increase in
aggregate demand, since firms then try to produce enough to meet the
higher demand.
In general, during disequilibrium in the economy, effective demand will

422
be less (recessions) or more (booms) than notional demand and supply will
be less or more than notional supply. This chapter studies the behavior of
the economy and the need and impact of monetary and fiscal policies in
such situations (recessions and booms), since disequilibrium is usually the
most common state in which real-world economies normally function.
Extended Analysis Box 9.1: The Dynamics Required for the Maintenance
of a Full-Employment Scenario
First, consider the response pattern if the economy (without any action
on monetary or fiscal policy pursued by the central bank and the
government) were to adjust to the fall in aggregate demand while
constantly maintaining full-employment output. Note that since the
central bank has been assumed to not take any action, the money supply
remains constant at its pre-existing level and exerts an influence on
aggregated demand, so that the LM curve and the price level for
calculating the real money supply become relevant. Therefore, the
relevant AD curve is the downward-sloping one.
In Figure 9.1a and b, the market is initially at (P0f, yf), but is suddenly
faced with the fall in demand to AD1. To outline the required responses,
start with the assumption that firms continue to keep employment at nf
and maintain their output at yf even though demand for their products
has fallen.6 Since demand will have become less than the output
supplied, which is at yf , prices and nominal wages will fall in response
to the lower demand, while the marginal product of labor (MPL) and the
real wage will remain at their full-employment levels.
In Figure 9.1a, the price fall required to clear the commodity market
will be from Pf0 to Pf1 . At Pf1 , there is adequate aggregate demand in
real terms to buy the full-employment output. This occurs because the
decline in the price level sufficiently increases the real value of the given
money supply, so that the LM curve in the IS-LM diagram would have
shifted to the right. With real aggregate demand restored to its initial
level, the firms’ policy of maintaining full-employment output will have
been validated. However, the dynamics of this full-employment analysis
require that the economy go instantly from P 0f to Pf0 , without changing
output in the process. Therefore, firms never become doubtful about
their ability to sell all of their full-employment output and do not resort
to a reduction in employment and output. The realism of these dynamics
is highly questionable for real-time, real-world economies.

423
9.3.2 A plausible dynamic scenario
Initial effects of the emergence of a demand deficiency
Firms do at times develop doubts about their ability to sell all the output
that they can produce with their existing equipment and employment, and
worry about mounting inventories of unsold products. When they do so,
they do usually reduce output and employment. Therefore, consider the
following components of an intuitively quite plausible dynamic scenario.
As aggregate demand falls, the rational firm in the economy shares in this
experience and believes, or even finds, that it cannot sell its former level of
output, so that with production in advance of sales, it ends up
unintentionally accumulating inventories. To run these down and to adjust
to its reduced sales, its optimal response is to cut back on production,
rather than wait for the economy’s aggregative markets to clear. It is also
likely to reduce the price for its product.7 This reduction in output by the
representative firm takes the economy below the equilibrium output yf ,
with jobs offered falling below full employment.
Secondary effects of the demand deficiency
As output falls below its full-employment level, so does employment. This
fall in employment results in a reduction in the incomes received by
workers who have become unemployed, so that they cut their consumption
expenditures. Further, some of the employed workers find or believe their
jobs to be at greater risk than normal at the full-employment level, so that
the employed workers on average will increase their precautionary saving
and reduce their consumption expenditures. This process accentuates the
initial decrease in aggregate demand, leading to further reductions in
output and prices by the representative firm.
Facing a fall in sales, the firms are also likely to cut back on their
investment plans, thereby contributing further to the fall in aggregate
demand. This is especially so since the reduction in sales and sales revenue
would reduce their ability to service their existing debt and maintain their
share prices, as well as reduce their ability to raise funds for investment
through new bond and share issues.
Let us now assume that the economy eventually returns to equilibrium.
However, the economy would have spent some real time at less than full
employment. In Figure 9.1a, the economy will not have gone from the
initial equilibrium at (yf, P0f) to the new equilibrium at (yf, P1f) instantly.
Output will loop from the point a to the point b, with the output in the loop
falling for some period below the full-employment level. This loop

424
occurred not because ofprice misperceptions or adjustment costs, for none
were assumed, so that it represents disequilibrium positions.
The transition from the old equilibrium to the new one
If the economy does not move directly to its long-run equilibrium but
instead moves to its short-run one, it will move from the initial position a
in Figure 9.1b to its short-run equilibrium at c (at the intersection of the
AD1 and SRAS curves), but through disequilibrium positions. These
positions are depicted by the loop from a through c in this figure. In this
case, output would have fallen for some time even below its short-run
equilibrium level y 1*.
Hence, the rational production responses of firms and the rational
consumption responses of households work in ways that do not always
make for an instant restoration of general equilibrium but could accentuate
the departure from equilibrium for some time. Note that this scenario
allowed prices and wages to be flexible, so that there was no assumed
rigidity of prices and either nominal or real wages.8 What was relevant
was the firms’ and households’ information set. Firms did not observe or
know either the new long-run price level (Pf1) or the new short-run
equilibrium price level (P2) in Figure 9.1b, corresponding to the fall in
aggregate demand, since the markets did not establish these instantly.
Their observation was that their sales were lower due to a decline in
aggregate demand.
Note that if the long-run equilibrium output is stable, both the short-run
equilibrium position and the disequilibrium ones are ‘transient ones’. The
chronological time taken by this transition is usually significant and
depends on many factors — among which are the severity of the fall in
aggregate demand, firms’ and households’ expectations on the magnitude
and duration of the fall and their reactions to these expectations .
Output-price adjustment function [Optional]
The above arguments suggest that firms adjust both prices and output
whenever demand differs from fullemployment capacity. To capture this
effect, we posit the output adjustment in a simple linear form as

The price adjustment can also be posited as a function of the fall in

demand relative to that in the previous period. A simple linear form of the
price adjustment function is

425
where P is the price level. For both adjustment functions, the adjustment
has been posited to depend on the deviation of current demand from the
preceding period’s output level.9 The relative adjustment ratio b/a will
depend on many factors, including the extent of the decrease in aggregate
demand and the expectations of its duration. These differ among
recessions, so that the adjustment in output relative to that in prices will
differ among recessions.
Box 9.1: The Nature and Critical Role of Expectations in Dynamic
Adjustments
As we have seen in the above arguments, expectations play a critical
role in the dynamic process that is actually followed by the economy.
We have already discussed two different types of expectations that are
relevant to the macroeconomic analysis. These are:
(a) Expectations of future prices
(b) Expectations on the future values of the real and nominal values of
the relevant variables, such as sales, incomes and employment
prospects, interest rates, etc.
The macroeconomic analysis of Chapter 8 had captured the role of price
expectations through the Lucas and Friedman supply rules. The
macroeconomic analysis of this chapter captures the role of expectations
held on the real values of the variables in the disequilibrium analysis.
In planning production and investment expenditures, firms form
expectations on their expected sales. In planning consumption
expenditures versus saving, households form expectations on their
expected employment and incomes. These expectations are reflected in:
• Business confidence on expected demand and sales
• Consumer confidence on jobs and incomes
We illustrate their very important role for the economy’s dynamic
adjustments by starting with full employment and examining two polar
expectation scenarios:
(a) A fall in aggregate demand that is expected by firms to be mild and
temporary and occurs in an overall intertemporal pattern of enough
demand to absorb the full-employment output in the aggregate
economy. In this case, the representative firm’s profit-maximizing
response is likely to be to maintain full employment rather than lay
off some trained employees whose replacement, when the expected
demand pickup occurs, would impose hiring and training costs. With
employment by firms maintained at its full-employment level, labor

426
 incomes would not fall, so that consumers will not reduce their
consumption. In this scenario, output may remain at full employment,
with some accumulation of inventories, or be decreased somewhat
through changes in the work effort of existing employees. There is
also likely to be some softening of prices (e.g., in the form of more
than the usual sales and discounts, etc.) but no significant price
reduction. In retrospect, the initial fall in demand is not worsened by
decreases in consumer expenditures and firms’ investment
expenditures, since these do not occur in this scenario. The economy
remains at or close to full employment and its initial price level.
Aggregate demand is soon restored to its full-employment level, thus
validating firms’ expectations.
(b) A significant fall in demand expected by firms and households to last
for a significant period. In this case, often backed by the experience of
earlier recessions in the firms’ information set, business and consumer
confidence declines. Firms reduce their employment and output —
which takes the economy onto a dynamic path below full
employment. As workers lose jobs, they reduce their consumption
demand. Other workers may also do so in order to increase their
precautionary saving. The consequent further fall in demand validates
and strengthens firms’ earlier pessimistic expectations and provides
ex post justification for their cuts in employment and production. It
may make firms even more pessimistic about future sales.
What are the critical differences between these two scenarios? The first
scenario requires firms to continue producing yf output and employing nf
workers at a wf wage rate, even when they observe the initial fall in
demand. They are only likely to do so if they are, individually and
collectively, convinced that the fall in demand is very temporary and
will be reversed very soon, so that they can temporarily accommodate
any output that is not sold. However, there are many instances of
reductions in aggregate demand which are not so transient or at least
firms generally do not take them to be such, so that the actual response
of firms would be to reduce output, thereby reducing employment and
leading the economy into a recession. Hence, aggregate demand shocks
can have none or minor effects on output and employment, or large
ones, depending on business and consumer/worker expectations and
confidence.
From the policy perspective, the irony of the above scenarios is that
the deliberate pursuit of aggressive policies to maintain aggregate
demand at the full-employment level leads to expectations by consumers

427
 and firms which maintain full employment, while a policy of leaving the
economy alone may, for larger demand shocks, bring into play dynamic
responses which take the economy below full employment.

9.4 Diagrammatic Analysis Following a Fall in

Aggregate Demand
Assume that the demand and supply functions for labor are as shown in
Figure 9.2b and that initially the economy is at full employment nf and
full-employment output yf in Figure 9.2a and b. Now assume that a shock
reduces aggregate demand to y 0d in Figure 9.2a so that a demand
deficiency — such that the firms are not able to sell the full-employment
output yf at the existing price level — emerges in the economy. Also
assume that firms do not consider this decrease in demand to be very
transitory. The actual aggregate demand can be supplied by the
employment of n0d workers.
In Figure 9.2b, the marginal product of labor for employment equal to n
d is MPL , which is above the full-employment wage wf . However, if
0 0
firms were to employ more than n 0d workers, they would not be able to
sell the extra output so that their marginal revenue product would be zero.
Hence, if aggregate demand falls to y 0d, firms would cut employment to
only n0d workers in order not to exceed their desired levels of inventories.
This causes the emergence of involuntary unemployment ui, one measure
of which is (nf – n 0d), which equals the deficiency of jobs relative to the
full-employment level. These workers want jobs for which they were fully
qualified in the old equilibrium and would prove to be fully qualified once
the aggregate demand needed for full employment is restored — but in the
mean time they cannot get jobs at the existing wage.10

428
 In Figure 9.2b, these n0d workers can be paid a nominal wage rate which
can change, as can the price level, with the resultant real wage being
anywhere in the range w ′0 and w ′′0, without a change in the firms’
employment of n0d workers, so that the real wage rate could drift up or
down from the initial equilibrium level of w *.11 Hence, the decrease in
employment (from n* to n0d) can be accompanied by either an increase or
a decrease in the real wage rate of the employed workers.12
The above effects are only partial or initial ones. Since the unemployed
workers do not receive any incomes, they cut back on their consumption
demand, which worsens the recession in output. The higher risk of losing
the job leads the still employed workers to increase their precautionary
saving by cutting back on their consumption.
Would a cut in the real wage rate restore full employment? If wages
were cut below wf , the lower incomes of the employed workers would
also lead to a reduction in their consumption.13 The resulting fall in
aggregate demand further shifts the effective demand curve to the left14
and worsens the recessionary effects derived in the preceding paragraph.
Hence, a reduction in the real wage rate due to a fall in aggregate demand
could drive the economy further from full employment, rather than
towards it.
Effective demand for labor
Firms want to hire the amount of labor whose output can be actually sold.
Therefore, the effective demand for labor is that level of employment for
whose output there is adequate demand. In Figure 9.2a, if the demand for
commodities is only y0d, the effective demand curve for labor in Figure
9.2b will be n0d. This curve differs from the usual (notional) demand curve
marked nd, which is the labor demand curve if firms had no problem
selling all their output.

9.5 Disequilibrium with Flexible Prices and

Wages
The above scenario is not the only one that can be sketched but it does
seem to capture the most common response patterns of actual firms,
consumers, and workers to a significant fall in aggregate demand. Critical
to this scenario is not any rigidity in prices and wages (either nominal or

429
real), for none has been assumed, nor the absence of an eventual return of
the economy to equilibrium, but the critical role of expectations (or
optimism versus pessimism) on the real and nominal values of sales,
incomes and jobs, the absence of a mechanism for instantly restoring
equilibrium,15 and at least some adjustment in production by firms and in
consumption by households.
Extended Analysis Box 9.2: The Relative Rapidity of Commodity
Markets in Adjusting Prices versus those by Firms and Consumers in
Adjusting Production, Employment, and Prices
The above disequilibrium analysis placed particular focus on the failure
of immediate/rapid clearance in the commodities and labor markets. In
particular, the main initial impulse was a fall in aggregate demand due to
a fall in investment, in consumption or government expenditures, or an
increase in economy’s interest rate. Firms responded to this demand
deficiency not solely by reducing prices but also by reducing output and
employment, with the latter, in turn, reducing labor incomes and
inducing a reduction in consumption demand. This fallback in
consumption demand further increases the demand deficiency in the
commodity markets, which is likely to persuade firms to further reduce
output and employment. While the economy may eventually return to
equilibrium, this process may take a sufficiently long time, during which
the economy would have less than full employment.
As the demand for commodities falls, why does the price level not fall
instantly to restore equilibrium in the commodity market? The reason is
that the commodity market is not a unified market with a homogeneous
product. It is really a collection of very diverse markets with an
enormous variety of products. Further, not all of the markets for
products operate with perfect competition or perfect efficiency (i.e., the
immediate adjustment of prices to equate demand and supply). In a
large number of industries, in fact for most industrial goods, firms sell
differentiated products and set their own prices in price lists, rather
than taking the prices of their products as given by the market. The
result is that there is no mechanism for an orchestrated simultaneous
reduction in all prices in response to the aggregate demand reduction.
Therefore, firms operating along upward sloping marginal cost curves
respond by both reducing their prices as well as reducing their output
supply.
If the firms’ reduction in output also leads to a reduction in
employment, the latter lowers labor incomes and also increases the
uncertainty of future expected incomes of those still employed. This

430
induces a reduction in consumption and therefore in aggregate demand.
Hence, the representative firm’s response to a reduction in aggregate
demand is a multi-pronged one — including reductions in price,
employment, and output — and could be a destabilizing one that
worsens or prolongs the effects of the initial fall in aggregate demand.
The sluggishness or failure of the market mechanisms in adjusting the
price level and the real wage relative to the faster responses of the firms
in adjusting their employment, output, and prices lie at the heart of the
real-world dynamic adjustments in the modern economy. Many
additional factors, including the reactions of the stock markets and the
financing of investment and consumer durable goods purchases,
contribute to this scenario.
The failure or sluggishness of the labor market in adjusting nominal
wages versus the faster responses of firms and workers in adjusting
employment and consumption
In the labor market, as unemployment emerges, why does the real
wage not instantly fall to restore full employment or at least short-run
equilibrium? The reasons are that:
(a) The labor market is also not a unified market with a homogeneous
good. It is really a collection of very diverse markets in skills,
occupations, and locations.
(b) The labor market does not possess a mechanism for a simultaneous
general reduction in real or nominal wages.
(c) In most jobs, wages are usually negotiated in nominal, not real, terms
and are often fixed in contracts.
Therefore, if a reduction in real wages is required for the return to full
employment, the labor market usually takes too long for the required
reduction in real wages. Moreover, even when real wages do fall, this
could reduce labor incomes, which further reduce consumption demand
and worsen the demand deficiency in the economy. This, in turn,
induces firms to further cut back on output and employment. Hence,
while the reduction in real wages is required by theory as a critical
market equilibrating mechanism, it could in reality become a potentially
destabilizing mechanism that could worsen and prolong unemployment.

431
9.6 If the Real-World, Real-Time Economy is not
on its LRAS or SRAS Curve, Where will it be?
If actual output that can be sold differs from its long-run equilibrium level,
the economy would not be on the LRAS curve but could be on the SRAS
curve. But if actual output that can be sold differs from its short-run
equilibrium level, the economy also cannot be on the SRAS curve.
Therefore, in disequilibrium, the economy must be off both the SRAS and
LRAS curves.
Suppose that a decrease in aggregate demand has driven the economy
into disequilibrium. The AD-AS Figure 9.3 illustrates the possibilities. Out
of both the long- and short-run equilibrium, the actual economy is neither
at a, b or c. If we assume that there continues to be equilibrium in the
income–expenditure process, the economy will be on the aggregate
demand curve. This assumption limits the number of possible positions in
Figure 9.3 to those between d and c . If prices were rigid, the economy
would be at the point d. If the price level adjusted fully to its short-run
equilibrium level, the economy would be at the point c . But it is more
likely that prices do fall somewhat — or rather soften through sales
discounts, refund coupons, etc. — and do so gradually, but the required
short-run reduction in prices does not occur fast enough. Under this
scenario, the economy would move to a lower price level than P0, as well
as a lower output than given by the point c . Such a point is shown as e .
Note that, as mentioned earlier, that both points e and c in Figure 9.3 are
transient ones for a stable economy. However, the movement of the
economy from the point e towards its short-run equilibrium at c can take
real time. Further, the movement from the point c to the long-run
equilibrium at b can also take real time. The duration of these periods will
depend on the causes and magnitudes of the initial fall in aggregate
demand and the actual dynamics of business and consumer confidence.
Therefore, this duration will differ between recessions for a given
economy and among economies for a given initial fall in aggregate

432
demand.

9.7 Can the Economy Get Stuck Below Full

Employment?
If the real-world, real-time economy stays below full employment for
some time and does not seem to be moving rapidly towards full
employment, we might designate its state as that of a temporary
equilibrium or of under-full-employment equilibrium, where ‘equilibrium’
is being defined as the state from which there is no significant and inherent
tendency to change. Alternately, observing some (though relatively slow)
movement in the economy, we can say that the situation is one of
disequilibrium; though the economy’s inherent forces or mechanisms are
weak, they will eventually get the economy to full employment. These are
two different ways of describing a slowly adjusting economy, analogous to
describing a glass as being ‘half full’ or ‘half empty’. How long is
‘eventually’ in real time? In practical terms, leaving aside semantics and
dogma, an economy in such a disequilibrium situation might spend several
quarters or even years below full-employment output.16 In such situations,
policy could play a very useful role in speeding up the return to full-
employment equilibrium.
Impact of the disequilibrium on the long run: hysteresis
Can the dynamic path during disequilibrium affect the short-run and long-
run equilibrium positions for the future? This would occur if it alters the
accumulation of skills and physical capital while the economy is away
from equilibrium. Less of these are accumulated in recessions than in
booms, so that drawn-out recessions and booms can leave their impact on
the economy beyond their own duration. Such an effect of the short-run
adjustment on the long-run path is called hysteresis. Mild recessions and
booms would not have a significant hysteresis effect, while long and deep
recessions and strong booms are likely to leave behind their impact on the
long-run of the economy.

9.8 Optimal Monetary and Fiscal Policies for the

Demand-Deficient Economy
The policy cure for an economy suffering from the hangover of a demand
deficiency requires a policy-induced expansion in aggregate demand. This

433
could come from expansionary fiscal and monetary policies.17 However,
note that fiscal policies can be notoriously slow to implement as a
stabilization tool in some countries, while monetary policy — through
reductions in interest rates supported by appropriate increases in the
money supply — can be implemented much faster.
The use of expansionary monetary policies
To derive the optimal role and impact of monetary policy in a demand-
deficient economy, assume that the economy is already at n d0 of
employment and y0 of output in Figure 9.2a and b. Further, suppose that
the central bank believes — on the basis of its information, including its
past experience of recessions — that the economy if left to itself would
take several quarters to return to equilibrium. Therefore, to speed up this
process, the central bank decides to pursue an expansionary monetary
policy (through a cut in the target interest rate, supported by the
appropriate increase in the money supply), so that there is a reduction in
the economy’s interest rate and an increase in its aggregate demand. To
see its effects, we need to supplement Figure 9.2a and b with Figure 9.4a
and b, and start with the assumption that the central bank had originally set
the interest rate target at rT0. Figure 9.4a shows the pre-existing position of
the economy at (yd0, rT0), so that there exists deficient demand (i.e., yd0 <
yf ). For this level of demand deficiency, Figure 9.4b shows the effective
amount of labor demanded (i.e., the level of employment whose output
can, in fact, be sold, so that it determines the pre-existing employment) as
n d0. Now suppose that the central bank lowers the target interest rate from
rT0 to rT1. This reduction in the interest rate stimulates investment, and
through the multiplier, increases aggregate demand at the pre-existing
price level from yd0 to yd1. Firms confronted with the increase in demand
would increase output, thereby shifting the effective amount of labor
demanded (though not the labor demand curve labeled nd) in Figure 9.4b
from n d0 towards nf (say to n d1) and increasing output closer to its full-
employment level.

434
 In this process, since the pre-existing situation was one of deficit demand
and involuntary unemployment, output increases in response to the
increase in aggregate demand. Since firms also make some adjustment in
prices in response to excess demand, there may also be some increase in
prices. This price increase will depend on how deficient the earlier demand
was and how large was the earlier excess capacity in the economy, but, in
any case, the expansionary monetary policy would have succeeded in
increasing real output in the economy. The real wage may rise or fall,
depending upon where it was earlier in relation to the full-employment
wage w *.
Limits to the effectiveness of monetary policies in a demand-deficient
recession
The above policy recommendations of an expansionary monetary policy in
a demand-deficient economy would not be appropriate if the fall in
aggregate demand had led to excess productive capacity—that is, firms
were left with more physical capital than required to meet actual demand
for their products. In this case, the decreases in interest rates are less likely
to stimulate investment — thereby reducing the ability of monetary policy
to revive aggregate demand in the demand-deficient disequilibrium.
During the financial and economic crisis of 2007–2010, the central banks
of many countries found themselves in such a position. While they reduced
the interest rates, in some cases close to zero, and increased the monetary
base, investment, consumer demand and aggregate demand did not pick up
fast enough (see Chapter 16).
The use of fiscal policy: increases in government expenditures and/or cuts
in taxes
In a demand-deficient situation, an expansionary fiscal policy (i.e., a cut in
taxes and/or an increase in government expenditures) would also raise
output and employment. Its analysis is similar to the above one for an
expansionary monetary policy. However, there is one major difference: a
fiscal deficit increases aggregate demand without requiring a prior increase

435
in investment and makes fiscal policy especially valuable if the economy
has excess productive capacity. In the economic crisis of 2007–2010, the
governments of many countries, led by that of the USA, pursued massive
fiscal deficits in their attempts to revive aggregate demand.
The policy dilemma
The above arguments raise the question: if aggregate demand falls and the
market adjustments in prices and nominal wages are sluggish, which of the
following approaches is preferable.
(i) Leave the market alone to make the needed adjustments. These require:
(a) Decline in the price level, so that the real value of the money supply
increases sufficiently to raise aggregate demand to the appropriate
level
(b) Decline in the market interest rate (without a pre-emptive reduction
by the central bank in the interest rate that it had set, i.e., the central
bank follows rather than leading the market in changing interest rates)
sufficient to increase effective demand
(c) Fall in nominal wages corresponding to or more than the fall in the
price level
(ii) The central bank takes the initiative to increase aggregate demand. It
cuts the interest rate (or increases the money supply) and/or the
government increases its deficit to increase aggregate demand
From the perspective of the time taken for each of the steps, (ii) is clearly
more rapid, especially for the central bank’s pursuit of monetary policy.
Further, (a) requires all or the majority of firms to cut their prices. Firms in
modern industrial economies hardly ever do so. Further, (ib) requires
reductions in nominal wages, which workers resent, so that workers’
reactions induce delays, industrial unrest, and productivity losses.
Consequently, firms are reluctant to attempt reductions in the nominal
wages paid, especially to existing employees. (ii) avoids these major
barriers to market-determined downward adjustments of prices and
nominal wages, and can be much faster as a means of restoring aggregate
demand. However, (ii) requires the central bank to have adequate
knowledge to act in time and have the willingness and the will to pursue
the appropriate monetary policy by the right amount. These requirements
are only met sometimes, so that some economists believe that the central
bank should pursue an expansionary monetary policy only in clear cases of
a very significant and clearly observable fall in demand, but not attempt to
offset small variations in aggregate demand. The latter is sometimes

436
referred to as fine-tuning the economy, and the recommendation is to
avoid such fine-tuning, since the possibility of errors in observations and
the wrong policy being pursued is greater for small variations in aggregate
demand.
Fact Sheet 9.2: Economies in Disequilibrium: USA During the Great
Depression
The graphs of this Fact Sheet provide an illustration of changes in
consumption, investment and aggregate demand, and the roles of
monetary and fiscal policies, during the Great Depression of the 1930s.
After the stock market crash in 1929, the USA, just as many other
industrialized countries at the time, entered into a period of deep
economic recession marked by significant drops in output, personal
consumption, and investment. As seen below, the lowest point of the
depression occurred in 1933, though a second smaller decline can be
seen in 1937. As unemeployment rose sharply, aggregate demand fell
dramatically, which caused reductions in output, saving, and investment.
The 1930s started with a strongly held belief in the USA in ‘fiscal
responsibility’, which requires that governments should keep their
budget balanced. This implied that as tax revenues fell with rising
unemployment, government spending was also cut. However,
government spending did rise with Roosevelt’s New Deal, initiated in
1933 in an attempt to provide jobs and care for the unemployed.
Between 1929 and 1933, many banks failed in the USA, so that money
supply fell. The deep contraction of the money supply allowed by the
Federal Reserve System until 1933 is considered by many to have
exacerbated the economic downturn. The eventual end of the Great
Depression in the USA, as in European countries, was due less to
economic policies but to the start of the Second World War, which led
to rapid increases in employment in war-related production and
conscription into the armed forces, and to large fiscal deficits and rapid
money supply growth to finance the war efforts.

437
Recapitulation of the roles of monetary and fiscal policies in a demand-
deficient economy
The preceding arguments imply that in demand-deficient economies with y
= yd < yf , the results for monetary and fiscal policies are likely to be:

These effects of expansionary monetary and fiscal policies on aggregate

demand and output in the demand-deficient economy differ from the zero
values of these multipliers for the full-employment equilibrium along the
LRAS curve derived in Chapter 7. Given these differences in the multiplier

438
values between the LR equilibrium and the disequilibrium phases of the
model, the policy maker has to be concerned with the real time the
economy will stay at recessionary levels, and what it can do to speed up
the return to full employment.
These arguments imply that:
• For real-time, real-world economies, there is no straightforward
relationship between expansionary monetary and fiscal policies and real
output. This relationship depends upon the initial state (full-employment,
inflationary or recessionary) of the economy and the extent of the
monetary and fiscal expansions.
• The transmission of the impact of the monetary and fiscal policies on
output does not always require—and need not mostly go through — price
level increases.
• The over-riding concern for policy makers becomes one of the
determination of the actual existence of equilibrium or disequilibrium in a
specific period—even if the economy is inherently one which can
eventually adjust on its own to full-employment output — and with the
duration which will have to be spent below full employment if the
policies to revive aggregate demand are not undertaken.
• Expectations, reflected by consumer confidence on jobs and incomes and
business confidence on sales and ability to obtain funds for investment,
have a significant impact on the time path of employment and output.
• Deep recessions and depressions, as in the 1930s and 2007–2010, can
reduce the effectiveness of an expansionary monetary policy. In such
circumstances, an expansionary fiscal policy is more likely to revive
aggregate demand and output.
Extended Analysis Box 9.3: Disagreements Among Economists on the
Appropriate Policies in Real Time for a Real-World Economy
The analysis of this chapter provides considerable scope for
disagreements among economists on the particular stage of the economy
at any given time and the policies that are appropriate to it. These
disagreements can be broadly classified as of two types:
(a) There are often disagreements on whether the economy, at a given
time, is in disequilibrium, in SR equilibrium or in LR equilibrium.
Unfortunately, there is no convincing statistical procedure for
determining this.
(b) Even when it is agreed that the economy is in disequilibrium, there is
considerable scope for disagreement on the extent of the departure
from full employment, how long it would take the economy to reach

439
 it, and on the strength and timing of the policy doses that are
appropriate. Again, the statistical techniques do not provide generally
accepted answers to these questions.
(c) There are often disagreements on where the new long-run
equilibrium is.
Hence, economists’ opinions tend to differ considerably on practical
issues. The policies that are recommended by economists as well as
those that are followed by policy makers usually depend on judgement
calls in a context of inadequate and imperfect information. These issues
are further discussed in Chapters 11 and 16.

9.9 Excess Demand18 in the Economy and

Appropriate Policies
We have so far considered the dynamics of demand declines from a full-
employment level. But suppose aggregate demand increases when there is
already full-employment output. This demand increase could again come
from a change in consumption, investment, money demand, or net exports.
Most firms seeing an increase in the demand for their products would,
besides raising their prices, tend to increase the supplies of their
commodities. This could occur through increases in employment, overtime
worked, as well as increases in the work effort of employees and their
efficiency. Each of these is feasible in the short term. The increase in
employment beyond the initial full-employment level could come about
through increased working hours of part-time workers, through students
delaying resumption of studies and through increases in the overtime put
in by employed workers, through potential retirees postponing their
retirement and through increases in effort, etc. While such increases in
employment and output above their full-employment levels can and do
occur in the short term and yield increases in output for several years, their
scope becomes eventually limited and constrained by workers’ long-run
preference for work (workers eventually get tired of putting in undesired
overtime, etc.). Hence, the (analytical) long run assumes that such
increases in employment and output beyond the full-employment levels
cannot occur in the long run but the disequilibrium and short-run analyses
allow such increases.
When the private components of demand have been expansionary
enough to pose serious risk of inflation, contractionary and fiscal monetary
policies can be used to reduce aggregate demand and thereby reduce the

440
inflationary pressures, though such a policy would also forego the
associated disequilibrium and short-run increases in output beyond the
full-employment level. Examples of such contractionary monetary policies
are the commonly observed increases in interest rates and decreases in the
money supply by the central bank when the economy is working above its
capacity level. Contractionary fiscal policies require a budget surplus.

9.10 Asymmetry in the Economy’s Responses to

Deficient and Excess Demand
Note that there is a significant difference in the economy’s constraints—or
in their short-term flexibility—and responses between increases and
decreases in aggregate demand. While the increases in aggregate demand
can increase output and employment for some time beyond their full-
employment levels, the increase in prices is more likely and faster than the
fall in prices in response to a decline in demand from the full-employment
level.19Usually, firms are readier to increase prices than to reduce them.
Consequently, the dynamic response patterns are different between the
demand-deficient and excess-demand cases. This asymmetry is vital to
differentiating between the stabilization roles of monetary policy when the
private economy generates demand below full employment versus when it
generates demand above full employment. But there is a role in both cases.

9.11 An Analysis of Disequilibrium Following a

Supply Shock
The sources of supply shocks are usually specified as changes in
productivity, the capital stock, and the labor supply. However, the
worldwide recession of 2008–2009 showed that a shock to supply can also
come a decline in credit to firms.

9.11.1 Supply shocks from labor productivity

Assume that there is a positive shock to the marginal productivity of labor.
For an anticipated shock, long-run analysis implies that this will cause the
full-employment output of the economy to increase, and actual output will
rise correspondingly. Further, employment and the real wage will also rise,
as shown in Chapter 6. However, these changes rarely happen smoothly

441
and we need to look into the plausible behavior patterns to derive the
likely dynamic patterns.
There are clearly many possibilities for the possible dynamic path
followed by the economy after a positive supply shock. Two different
scenarios of this path are:
(i) The markets adjust all prices instantly by the required amounts for
taking the economy to the new full-employment equilibrium, the firms
know these prices and respond by adjusting their output and
employment accordingly.
(ii) Firms set prices, while markets are slower in establishing the market-
clearing price level that equates aggregate demand to the new level of
the full-employment output. The following expands on a plausible
pattern for this scenario.
Suppose that the positive productivity shock shifts the LRAS curve from
LRAS0 to LRAS1 in Figure 9.5a. This implies that the long-run price level
will have to fall from P0 to P1. But firms will not observe/know the new
LR price level since the market has not yet established it.20 Assume that
the firms experiencing the increase in productivity do not fully and
instantly lower prices to the new equilibrium price level for the economy,
but lower their prices gradually and marginally in order to increase their
market share. Hence, let the actual price level only fall to P2. At P2, the
aggregate demand is only yd2, which is inadequate to buy the new long-
run output of yf1 . In Figure 9.5b, aggregate demand at yd2 implies
employment of n 2, which, in Figure 9.5c, is below the new full-
employment level nf1 . Note that the new employment level n 2 has been
drawn to show an increase in employment relative to nf0 . However, there
exists the possibility that the new employment level could have been to the
left of this point, so that employment could have fallen from its initial level
yf0 . Hence, in this scenario, following a positive productivity shock, firms’
gradual groping towards a new set of prices could increase output, while
decreasing employment or increasing it less than the increase in the full-
employment level. If this happens, involuntary unemployment emerges
during the dynamic adjustments even though the nation’s output is rising.

442
 This scenario was based on the assumption that the firms experiencing
an increase in productivity gradually lower their prices. This ‘groping’
process is a common pattern in periods with productivity increases and
falling costs of production. The economy may eventually reach its new
full-employment level, but there could be an interval of involuntary
unemployment and possibly even of employment below the old full-
employment level. This is a seemingly paradoxical result: a positive
productivity shock can increase output while creating involuntary
unemployment for some duration, even among the otherwise qualified
workers.
Another reason for an increase in unemployment is that some of the
previously employed workers will not possess the right skills for the new
technology and will become redundant. Further, workers employed in
declining industries will need to shift to expanding ones, which would take
time and re-training, which may be too late for older workers and not
practical for others.
The scope for monetary and fiscal policies
Now consider the policy implications of this scenario. The critical element
of the emergence of involuntary unemployment was the failure of the price
level to decrease sufficiently — and to do so sufficiently fast — to
increase the quantity demanded21 of commodities to meet the increase in
their long-run supply. In this scenario, if the monetary and fiscal
authorities were to increase aggregate demand sufficiently through
expansionary monetary and fiscal policies to match the increase in supply,
the new long-run supply would occur at the pre-existing price level. Its
result would be that the economy would have made a smoother transition
to its new long-run position.
Box 9.2: Demand Shocks Emanating from Shifts in Long-Run Supply
In some periods (but not always), a demand response emerges from a
positive shift in the long-run equilibrium output of the economy. An
instance of this occurred in the late 1990s when the rapid technical
change occurring in the computer, internet, and telecommunications

443
 sectors both increased the productivity of the economy and also caused a
mania for the stocks of companies in these sectors. This enthusiasm for
their stocks spread to the stocks of companies in other sectors, and
caused a stock mania. With the increase in stock prices, there was an
increase in the ability of companies to raise funds for new investment
and an increase in stockowners’ wealth, which increased their
consumption. These caused increases in aggregate demand. These
increases were greater than in the productive capacity of the economy,
so that employment rose and unemployment rates fell through much of
the 1990s.
Hence, technical change can have two effects: increasing the long-run
level of full employment and increasing aggregate demand. Their net
effect can then be of a decrease in the unemployment rate.
However, such a response of aggregate demand to technical change
cannot be taken for granted. The collapse in stock prices from 2000 to
2002 produced a fall in investment and in aggregate demand. This fall,
in a period of continuing positive shifts in technology, sent the
economies into recessions and raised unemployment rates in many of
them. The subsequent recovery during 2002–2006 produced increases in
output but did not, for the first couple of years, produce increases in
overall employment because the productivity of labor had increased in
the meantime while the increase in aggregate demand was not strong
enough to require firms to increase employment. In the USA, the very
large fiscal deficits following the Iraq war and the bubble in house
prices during 2003–2006 eventually led to a sufficiently strong increase
in aggregate demand for employment to rise and unemployment to fall.
However, such an increase in aggregate demand did not occur in
western Europe whose economies struggled with high unemployment
levels well into 2006. Unemployment rate rose further during the
worldwide recession of 2007–2010.

9.11.2 The impact of a decline in credit supply on

the effective supply of commodities
Firms borrow extensively to finance their operations. In developed
financial economies, the amounts borrowed are usually from other firms in
the form of trade credit, banks in the form of loans and bond markets. We
can classify these amounts into two categories: (short-term) credit and
long-term bonds, of which credit encompasses amounts borrowed through

444
inter-firm trade credit, bank loans, and short-term bonds, such as through
the issue of ‘commercial paper’ (usually three month to one-year bonds).
Firms use credit to finance their inventories, payroll expenditures, and
other rotating needs for funds. A sudden unexpected cutback in funds
forces them to cut back their orders for inputs and lay off some workers, so
that aggregate supply of commodities falls. The credit crisis of 2007–2010
illustrates the impact of a credit restraint on the production of
commodities: it shifted the effective supply of commodities to the left of
the long-run and short-run supply curves, thereby causing a recession in
output.
The appropriate policies in this scenario require a credit increase to meet
the production needs of full-employment output. However, monetary and
fiscal authorities do not normally directly lend to firms, so that their
policies have only an indirect and delayed effect on the supply of credit. In
the interval before the reduction in credit is eliminated, the reduction in
credit would cause a reduction in output. The 2007–2010 period shows the
impact on output and employment of the reduction in credit from banks
and other lenders. Further, it shows that even though the expansionary
monetary policy drastically increased the monetary base and governments
ran deficits to prop up the credit supply, the restoration of credit to normal
levels was too slow to restore output to a full-employment level for many
quarters.
Chapter 16 provides more detailed analysis of these issues.

9.12 ‘Crowding-Out’ or ‘Crowding-In’ of

Investment by Fiscal Expenditures in a
Demand-Deficient Economy?
Crowding-out is defined as the crowding out (i.e., reduction) of investment
by a fiscal deficit. Such crowding-out can occur due to (1) interest rate
increases, caused by increased borrowing to finance the deficit by issues of
government bonds, and (2) the output constraint of full-employment output
in the long-run analysis. These were also discussed in Chapters 7 and 8.
On (2), the long-run analysis of output in Chapter 7, with y given by yf ,
implies that firms’ purchases of commodities for investment will be fully
crowded out in a closed economy by increases in the government’s
purchases of commodities — because output does not increase while the
government, with a deficit, takes a bigger cut out of it, so that less is left
for investment (and/or net exports in the open economy analysis). The

445
short-run analysis along the SRAS curve moderates this degree of
crowding out since it allows some increase in output beyond yf .
In deficient-demand conditions, the increases in government
expenditures would reduce the degree of demand deficiency and raise
output, thus lessening the degree of crowding out. Further, the increases in
aggregate demand and sales due to the increase in government
expenditures will tend to boost business confidence and raise the expected
demand for their products. This would induce increases in firms’
investment to raise their production capacity. Therefore, instead of any
crowding out, an expansionary fiscal policy could increase private
investment. This will mean ‘crowding-in’ (i.e., increase in) investment
because of the fiscal deficit.
Hence, for the real-world economies in recessions, there could be
positive effects of increases in government expenditures on private
investment, rather than the strong negative one that full crowding out
implies for the analytical long run or the partial crowding out that the
short-run analysis implies.
Similar effects would follow a cut in the central bank’s interest rate
target. In this case, the increase in investment would occur for two reasons:
(i) For the given investment function, the cut in the interest rate increases
investment.
(ii) In a demand-deficient recession, the expected increase in aggregate
demand due to (i) boosts business confidence for future sales and causes
them to further increase investment. There would also be an increase in
consumption due to the boost to consumer confidence. These increases
in investment and consumption would further increase aggregate
demand and output in the demand-deficient context.

9.13 Disequilibrium (Involuntary Unemployment)

in the Labor Market due to a High Real Wage
Assume that a shock (such as a fall in the price level or in aggregate
demand, or union action, etc.) has pushed (or left) the real wage rate above
its full-employment level. Since firms employ along their demand curve
for labor, their employment would be less than the full-employment one
and the economy would have involuntary unemployment . In this case, the
involuntary unemployment would not be due to a demand deficiency but
due to the real wage being too high.
Further, assume that this high real wage is maintained even if the price

446
level were to rise, i.e., the factors that brought it about cause nominal
wages to rise in proportion to any inflation. In this case, an expansionary
monetary policy cannot reduce the real wage or decrease the involuntary
unemployment.
Hence, there are two reasons for employment and output to be below
their full-employment levels:
(i) A demand deficiency
(ii) A high real wage
The policies appropriate to each of these possibilities differ, so that there
has to be a prior diagnosis of the causes of the involuntary unemployment
and the responsiveness of the real wages to inflation, before the adoption
of policies to address involuntary unemployment.
Expansionary monetary policy as a means of lowering the real wage and
restoring equilibrium in the labor market
Suppose that the current real wage exceeds its full-employment level and
that nominal wages do not increase proportionately with the price level so
that inflation will reduce the real wage and involuntary unemployment. In
this case, an expansionary monetary policy can be pursued to achieve these
goals. The central bank will have the following three options:
(a) Leave the economy alone to move on its own to lower the real wage to
the full-employment one.
(b) Pursue expansionary monetary and fiscal policies that cause a rate of
inflation sufficient to reduce the real wage to the full-employment one.
(c) Pursue a policy of changing labor market structures to lower labor’s
resistance to wage cuts.
Each of these policies involves some time lags and has side effects.
Judgements on these differ among economists. Some claim that the labor
market adjusts fairly fast to lower the real wage whenever there is
involuntary unemployment, so that (a) is the best policy. Others claim that
there are significant impediments and lags in the economy’s adjustment to
a fall in real wages. Further, for this adjustment to occur, many economists
claim that it is not the market, which lowers nominal wages but that
individual firms have to do so, not only for new employees but also for all
employees. When individual firms try to cut the nominal wages of their
employees below the existing ones, workers’ resentment over the cutbacks
leads to strikes, work slowdowns, vandalism, and other types of disruptive
‘industrial action’, which makes firms reluctant to cut nominal wages.
Therefore, a less painful and speedier adjustment of real wages would be
through expansionary monetary policy, which creates sufficient inflation

447
to bring about the appropriate reduction in the real wage for the existing
nominal wage,22 so that some economists favor (b). However, in practice,
this policy prescription may only work for relatively small differences
between the actual real wage and the full-employment one, and
correspondingly appropriately small inflation rates. In particular, it is
unlikely to work if the policy creates hyperinflation — in which case
workers are likely to both be very alert to the actual rate of inflation and
promptly negotiate nominal wage increases to match the inflation rate. In
this case the long-run remedy for high real wages is (c), not (b).
The third approach, as in (c) above, was followed by President Ronald
Reagan of the USA.
Extended Analysis Box 9.4: Is the Preceding Demand-Deficient Analysis
a Reflection of Market Failure or of Markets’ Sluggishness in Adjustment
of Prices and Nominal Wages?
The underlying theme of the dynamic analysis presented in this chapter
can be interpreted as one of the ‘failure of the market’ — or of the
competitive forces in the economy — to adjust the price level instantly
to levels consistent with full employment. However, such instant
adjustment seems to be asking too much for a very heterogeneous and
dispersed economy, even if the economy had high levels of competition.
Our analysis really did not assume that the markets did not adjust prices
when demand or supply shift or that they do not initiate changes to move
towards equilibrium. Our arguments were based on the relatively slower
adjustment of prices by the markets relative to the faster adjustments by
firms, consumers, and workers. These economic agents adjust
production, employment, consumption, and money demand — in
addition to the adjustment of prices by firms. Further, governments and
central banks may adjust their expenditures, taxation, and the target
interest rate (or the money supply) faster than the market adjusts prices
and interest rates. These adjustments by economic agents — especially
the quantity adjustments in production, employment, and consumption
— lie at the core of the dynamic analysis of this chapter. They become
especially significant when aggregate demand changes and firms and
consumers adjust output and employment in response.
Some economist would call these results as those of ‘market failure’,
while others see them as merely — though dragged out in real-time—
dynamic adjustments on the way to eventual equilibrium at full
employment. No matter what they are called, they provide an important,
stabilizing role for monetary and fiscal policies in the real-time, real-
world economies.

448
9.14 The Dual Implications of a High Saving Rate
The saving rate plays a dual role for the closed economy:
(i) An increase in the saving rate promotes the growth of productive
capacity and, therefore, improvements in future output and standards of
living. This would shift the future long-run aggregate supply curve
(LRAS in Chapter 1) to the right and raise the long-run levels of future
full-employment income.
(ii) An increase in the saving rate reduces consumption expenditures and
aggregate demand for commodities. This demand can cause the
economy to reduce output below its full-employment level by moving it
along its short-run aggregate supply curve (see the SRAS curve in
Chapters 1 and 8). In other cases, as shown in this chapter, the fall in
aggregate demand below the current productive capacity of the
economy can push it for some time into a demand-deficient
disequilibrium, so that the economy goes into a demand-deficient
recession with high unemployment. This disequilibrium effect can be of
serious concern in the context of real-time, real-world economies.
The paradox of a high saving rate — also known as the paradox of thrift
Note that the amount saved depends on the saving rate multiplied by
national income/output. Suppose that the economy does not adjust fast
enough to prevent a demand-deficient disequilibrium and production falls
because of inadequate demand. If output falls more than in proportion to
the rise in the saving rate, the amount actually saved will fall. Therefore,
we could have the seeming paradox: an increase in the saving rate can
cause a decrease in output and the amount saved. Note that this paradox
does not occur if output does not fall or falls less than in proportion to the
rise in the average saving rate.

9.15 Conclusions
For the study of disequilibrium and dynamics to be a potentially useful
exercise requires the belief that the shocks from the private sector or by the
monetary and fiscal authorities can drive the economy away from its full-
employment equilibrium state for significant periods of time. Intuitively,
continuous general equilibrium requires the belief that the economy
responds to a shock by remaining in — or fairly soon returning to —

449
equilibrium, whether along its LRAS or SRAS curve. To illustrate this
from the use of monetary policy, this belief requires that an expansionary
policy would immediately cause a proportionate increase in the price level
and that a decrease in it would not cause a recession in output and
employment. There is considerable evidence to show that these
requirements are not always, or are not most of the time, met in most real-
world economies.
The usefulness of monetary and fiscal policies clearly depends on the
lags in the response of the economy to the shocks coming from
investment, consumption, and money demand, and on the lags in the
formulation and impact of monetary and fiscal policies.
The disequilibrium/dynamic analysis of this chapter has shown that:
• There are numerous types of shocks that constantly impinge on the
economy. They can be from the demand or the supply side of the
economy.
• For the private sector, the shocks from the demand side can come from
shifts in consumption, investment, money demand, and exports, etc.
• The shocks to supply come from shifts in the production function due to
technological change, shifts in industrial structure, in the supply of labor,
minimum wage laws, etc.
• Adjustments in response to shocks are made by markets (to prices), by
firms and households (to output, employment, and prices) and by policy
makers (to fiscal and monetary policies). Their relative speeds of
response are essential to understanding the dynamic adjustments of the
economy and how long the economy stays away from full employment,
and whether the pursuit of monetary and fiscal policies is useful or not.
• Once a shock hits the economy, competition, flexible wages, and prices,
and the economic rationality of firms and workers are not enough to
always ensure that the economy will immediately revert to full
employment after an aggregate demand shock. It may do so for some
types and some magnitudes of shocks but not for others.
• The expectations of firms and households — represented by business and
consumer confidence — on the likely duration of deficient or excess
demand play very significant roles in whether the economy will move
away from equilibrium or not, and even on whether it may do so in a
stabilizing or a destabilizing fashion. The emphasis in these expectations
is not on the expectations of prices but of sales, incomes, and jobs.
• The economy sometimes behaves in a manner that takes its output to a
level below full employment, with movement back to full employment
sometimes being very hesitant and imperceptible. Such a situation can be
interpreted as one of disequilibrium or equilibrium — where equilibrium

450
 is defined as the state from which there is no inherent tendency to change.
From a policy perspective, it does not matter what name is assigned to
this situation.
• It is not enough for the economy to eventually bring nominal wage rates
and prices to their equilibrium levels consistent with full employment.
These processes must occur fast enough, since, otherwise, the firms will
act on the fall in their demand by cutting their output and employment
and workers will react to the loss of employment by cutting consumption.
Further, the interest rates must also change appropriately and the increase
in investment must occur adequately and fast enough.
• The likely reactions of competitive markets, firms, and households to
demand decreases can include cutbacks in nominal wages and prices, as
well as in production, jobs, investment, and consumption. The former can
be of the type that could take the economy to equilibrium, but they could,
alternatively, be of the type that worsens the demand deficiency—and
increase involuntary unemployment. The latter would move the economy
away from full-employment equilibrium — rather than towards it.
• In demand-deficient economies, government deficits can cause crowding
in.
A caveat is necessary here. Some economists are likely to disagree with
the position that the economy can be in disequilibrium for significant
periods of time following some types of shocks. Other economists may
disagree with the above dynamic analysis because they may consider other
scenarios to be more plausible. Econometric studies have not
unambiguously answered these questions. Intuition should be used in
establishing one’s own responses to them. The arguments of this chapter
were intended to provide some guidance on this intuition.
If the actual economy under consideration has a strong tendency towards
equilibrium and can be expected to reach it in a short while, focusing the
analysis and policy prescriptions on the equilibrium properties only would
be justified. However, economic theory and econometric tests provide
some but not full guidance for determining this. Consequently, the policy
makers are also forced to rely on their ‘feel’ and ‘perception’ of the state
of the economy. Further, they can never be sure of the intensity and timing
of the impact of their policies. Therefore, even though economics is a
science, the formulation and pursuit of economic policies becomes ‘an art
rather than a science’.

KEY CONCEPTS
Disequilibrium

451
Dynamic analysis
Effective demand, excess demand
Demand deficiency
Business confidence
Consumer confidence
Involuntary unemployment
Market failure
Real time versus analytical time
Real-world economies versus
analytical/model economies
Economics is a science
Crowding-in versus crowding out
The pursuit ofeconomic policies is an
art rather than a science

SUMMARY OF CRITICAL CONCLUSIONS

• The usefulness of dynamic disequilibrium analysis occurs when real-
world economies take a significant amount of real time in reaching
general equilibrium. Casual observation of booms, recessions, and
depressions and the pronouncements and behavior of central banks—as
well as many empirical studies—indicate that this is so.
• The dynamic behavior of real-world economies takes account of the
relevant speeds of reaction of markets, firms, households/workers, and
policy makers to exogenous shifts.
• In particular, markets may be relatively sluggish in changing prices to re-
establish equilibrium after a shock, compared with the adjustment speeds
of firms and consumers in changing output, employment, consumption,
etc.
• If firms cut employment in response to a fall in aggregate demand,
involuntary unemployment emerges. The latter can reduce consumption
and further reduce aggregate demand.
• If involuntary unemployment has emerged due to a demand deficiency,
expansionary monetary and fiscal policies may be able to — depending
on their speed of formulation and implementation and the speed of
response of the economy to them — shorten the period during which the
economy stays below full employment. This is a stabilization role for
these policies.
• Involuntary unemployment can also occur if the real wage is too high
relative to its full-employment level. Expansionary monetary policies,
which cause inflation, may serve to lower the real wage and, therefore,

452
 reduce involuntary unemployment — faster than the (sluggish) market-
induced reduction in the real wage.
• Competition incorporates the ‘invisible hand of competition’ to guide the
economy towards equilibrium through market-induced price changes. But
this hand may be slow enough to leave the economy out of equilibrium
for significant periods of real time — in which case, assuming a faster
impact on aggregate demand, production and employment, the pursuit of
policies may become preferable.
• In a deficient demand context, expansionary fiscal and monetary policies
can boost private investment and result in ‘crowding-in’.
• Monetary and fiscal policy makers rarely possess accurate and timely
information on the actual state of their real-world economy. They also do
not accurately know the timing and impact of their policies, or the
strength/intensity of the policies that would be best. They have to
formulate their policies on the basis of their limited knowledge,
guesswork, and impressions of the economy, so that their pursuit of
economic policies is an art rather than a science.

REVIEW AND DISCUSSION QUESTIONS

(1) Present the diagrammatic analysis of the impact on output caused by a
fall in aggregate demand, when nominal wages and prices remain
unchanged.
(2) Present the diagrammatic analysis of output caused by a fall in
aggregate demand, with prices falling but the nominal wage remaining
unchanged.
(3) Assume that the economy was initially at full employment. The central
bank raises the target interest rate (and appropriately reduces the money
supply), which causes a reduction in aggregate demand. Discuss, using
diagrams, its effects on employment and output.
(4) What is meant by ‘business confidence’ and what can cause changes in
it? Why should changes in it alter investment? List three other types of
shocks or shifts in the economy that can cause shifts in investment.
(5) Discuss the likelihood that real-world firms respond to a fall in the
demand for their products by reducing their production but only
softening (through temporary discounts etc.) their prices. If they do so,
(a) Discuss, using diagrams, the effect on unemployment.
(b) In the context of interest rate targeting, what monetary policies are
appropriate in this situation?
(6) Discuss how a demand-deficient disequilibrium can come about even
when prices and nominal wages are flexible.
(7) What are the main reasons for disagreements among economists on the

453
 appropriate policies to follow in a recession?
(8) Start with the full-employment equilibrium position at (yf, nf, r LR, P
LR), where LR indicates long-run equilibrium values. Suppose that a
reduction in investment reduces aggregate demand, and that the markets
are sluggish in adequately adjusting prices downwards. Answer the
following:
(a) Analyze the behavior of firms hit with a fall in the demand for their
products. If this analysis shows that employment is reduced below nf ,
discuss the likely consumption responses of households and show
diagrammatically their impact on aggregate demand.
(b) If these responses of firms and households imply a movement away
from (yf, nf, rLR, PLR), what equilibrating mechanisms will come into
play to bring the economy back to (yf, nf, r LR, P LR).
(9) Which do you think is more powerful and has a faster response to
changes in the demand for firms’ products: the market-based
equilibrating mechanism of price adjustments for the commodity market
as a whole or the adjustment responses of firms and households? What
difference do the relative speeds of adjustment make to the continued
maintenance of full employment when there are fluctuations in
aggregate demand?
(10) Discuss the reasons why an economy may develop ‘unduly high’ real
wages and the barriers to its reduction.
(11) If a recession is caused by a fall in investment and generates
unemployment, would this be due to demand deficiency or a high real
wage? Would monetary policies be appropriate in this case? Use
diagrams for your analysis.
(12) If a recession is caused by an increase in oil prices and generates
unemployment, would this be due to demand-deficiency or a high real
wage? What monetary policies would be appropriate in this case? Use
diagrams for your analysis.

ADVANCED AND TECHNICAL QUESTIONS

Basic AD-AS model I, with interest rate targeting
Assume that the following equation describes the aggregate demand
function for an economy.
IS equation:

454
 The central bank’s real interest rate targets are:

The money supply adjusts to be consistent with money market equilibrium

at the interest rate target

Fisher equation:

Long-run aggregate supply function:

T1. For the Basic AD-AS model,

(a) Assume that the economy was in the long run in period 0. Derive the
aggregate demand equation for period 0.
(b) For period 0, derive the long-run price level P 0 and the aggregate
amount demanded yd0 at P 0.
T2 In the Basic AD-AS model, let autonomous investment so that for
periods 1, 2 and 3, it becomes:

For period 1, derive the aggregate demand equation. Then, solve for the
long-run levels of P and y. Designate these respectively as P LR1 and yf1
.
T3. Assume that, in period 1, instead of the price level adjusting to its
long-run equilibrium level, the price level in period 1 remains
unchanged from its level in period 0. That is, P 1 = P 0. Calculate the
actual level of aggregate demand. Designate it as y d0.
At the actual level of demand in period 1 at P 0,
(a) What is the extent of demand deficiency?
(b) Now assume that firms adjust their production so as not to have to add
unsold commodities to their inventories. Derive the level of output
under this behavior. Designate it as y 1.
T4. Given your answers to the preceding questions, following an
unexpected fall in investment,
(c) What adjustments in employment and output are profit-maximizing

455
 firms likely to make when they are faced with a fall in aggregate
demand, which the markets have not cleared because of their
sluggishness?
(d) What adjustments are the markets likely to make to the fall in
demand?
(e) Compare these and discuss their likelihood in the short term.
(f) Will the market response take the economy to full employment?
Discuss.
(g) If not, what will take the economy to full employment? Discuss.
T5. In period 2, in a delayed response to the fall in aggregate demand in
period 1, the central bank lowers the interest rate target to 0.02. Given
that the price level remains unchanged at P0, what are the actual levels
of aggregate demand yd0 and real output y2.
T6. For period 3, again assume that the price level has not changed from
that in period 0. Calculate the real interest rate that the central bank
needs to set as its target if actual output were to equal the full-
employment one. If this interest rate turns out to be negative, discuss
whether a negative nominal interest rate is possible in the economy.
T7. [Optional.] This question assumes that the deviation of demand from
the preceding period’s output is split by firms into an output change and
a price change. Further, periods 0 and t – 1 are assumed to have the
long-run solutions for output and the price level (i.e., full-employment
output and the price level which equates aggregate demand to this
output). Calculate the long-run output and price level for the Basic
model before answering this question.
For the output function of the Basic Model in a dynamic context,
assume that the firms’ output adjustment pattern depends on excess
demand according to

where, for t = 0, yt-1 = yf = 80, 000. The economy is in full employment

in periods 0 and –1.

Further, assume that the economy adjusts the price level each period
according to the following price adjustment equation:

where for t = 0, P 0 is the market clearing price at which ytd = yf .

456
 Calculate the time path of output, the price level, and inflation for the
first five periods (0, 1,…, 4) and the inflation rate for periods 1,2,3,4.
[Hint: first calculate for period 0; then use it to derive the level of output
y0 under the output adjustment equation above; then use and y0 to
determine the price P 0 according to the price adjustment pattern. Use a
similar procedure to calculate the required values for periods 1, 2, 3, 4.]
Arrange your answer in the form of a table with the columns as year, y,
P , and π.

Basic AD-AS model II

[Note: The aggregate demand parts of the following model were also
used in Chapter 8 ]

Aggregate demand function for the economy is specified by the

following information.

IS equation:

The central bank’s real interest rate targets are:

The money supply adjusts to be consistent with money market

equilibrium at the interest rate target .

Fisher equation:

Long-run output:

457
 In periods 0, 1, 2, 3, and 4, the economy has deficient demand. The
economy’s markets are sluggish in adjusting the price level, so that
firms respond to deficient demand by changing the economy’s output.
The short-run aggregate supply function when there is deficient demand
in periods 0,1,2, 3,4, is given by

Price level: before period 0, the economy has the price level that is
consistent with long-run equilibrium. The economy maintains this price
level for all subsequent periods. That is, P 0 = P 1 = P 2 = P 3 = P 4 = P
LR.

T8. [Optional.] Given the Basic AD-AS model II, assume for this question
that the economy has deficient demand equilibrium in periods 0, 1, 2, 3,
4. The short-run output function for periods 0, 1, 2, 3, 4, is

The economy has full employment before period 0. [Note: this

assumptionallows the long-run price level to be calculated. Start
answering this question by first calculating this price level, and using it
for all periods.]
(a) For the preceding information, what is the level of aggregate demand
in period 0? What are the levels of actual output in periods 0, 1, 2, 3, 4?
(b) What is the shortfall of actual output from the full-employment one in
periods 0, 1, 2, 3, 4?
(c) For period 1, suppose that the central bank, observing the demand
deficiency in period 1, reduces its target interest rate r T to 0.02. What
becomes the level of aggregate demand in period 1? What are the levels
of actual output in periods 1, 2, 3, 4? What is the shortfall of actual
output from the full-employment one in periods 0, 1, 2, 3, 4?
T9. Discuss the impact on actual output of the expansionary monetary and
fiscal policies under the assumption that following a change in demand,
markets are slow in adjusting prices while firms react faster by changing
their output: (a) when output starts by being at the full-employment
level, (b) when output is initially at a deficient demand level.

1However, not all economists agree that recessions are evidence of labor
market disequilibrium.
2For the decades up to 1940, the percentages have been rounded to the
nearest 5%. Data for each year is for 31 December.
3‘Herd behavior’ occurs when economic agents tend to follow each other,

458
as can be observed in stock market manias and collapses. ‘Contagion’ is
the spread of a shock in one market, or one country, to others, as happens
when a foreign exchange crisis occurring in one country spreads to other
countries in the region. Another example is when the collapse of the price
of the shares of one major corporation also brings about a collapse in the
share prices of other corporations in the same industry. Herd behavior and
contagion were very evident in the stock market boom in the late 1990s,
the collapse from 2000 to 2002 of the prices of Internet and
telecommunications stocks, and the worldwide financial crisis of 2007–
2010.
4Hence, for uncertainty scenarios, the rational expectation hypothesis (see
Chapter 8) will be maintained Further, we maintain the usual assumptions
on the absence of money illusion and adequate flexibility of prices and
wages to eventually (but not instantly) restore equilibrium following any
shifts or shocks in their demand and supply functions in their respective
markets.
5In addition, we dispense with the assumption that firms can sell all that
they want at the existing price. This means that firms are in monopolistic
competition, which is the dominant mode of industrial economies.
6This assumption is made so that the SRAS curve is excluded from the
analysis at this stage. It implicitly assumes that there are no price
misperceptions or adjustment costs. If either of these exists, the relevant
curve would be the SRAS curve.
7This implicitly assumes some degree of monopolistic competition rather
than perfect competition.
8There has also been no assumption of money illusion by either firms or
workers.
9In a dynamic context, P is usually replaced by the inflation rate π.
10Involuntary unemployment and its different measures are discussed in
the next chapter.
11Real wages will rise if the price level falls faster than nominal wages;
they will fall if the price level falls more slowly than nominal wages, and
will stay constant if both prices and wages fall in the same proportion.
12Wages may, therefore, follow a pro-cyclical or counter-cyclical pattern:
some recessions and some parts of a given recession could show wages
falling while others show them to be rising. If wages rise, it could be
claimed that the rise in wages is the cause of falling employment, when
this rise is itself only an effect while the true cause was the initial fall in
aggregate demand.

459
13For many employed workers, the fall in employment usually increases
the subjective risk of staying employed, so that such workers also tend to
cut back on consumption in order to increase precautionary saving to
provide for the eventuality of becoming unemployed.
14The new demand curve is not the classical notional one nd but an
effective one, and one cannot proceed with analysis using nd. A notional
demand function for labor assumes that firms can sell all the output they
find profitable to produce, which was implicit in the derivation of the nd
function and curve. An effective demand function for labor is one for
which output there is enough demand.
15Note that the assumption of perfect competition does not a priori specify
the chronological time needed by the ‘invisible hand of competition’ to
return an economy, follow a shock, to its full-employment equilibrium.
Because of this, some models resort to a process known as tâtonnement
with recontracting, which solves the problem analytically. However, since
this process does not exist in the real-world economies, its usage
represents in some ways an escape from answering the relevant question of
how long does it take the real-world economies to return to equilibrium.
16It did so in the Great Depression of the 1930s and in many deep
recessions.
17In open economies, it could also come from a policy-induced
devaluation of the exchange rate to increase net exports. This is discussed
in the open economy Chapters 12 to 13.
18Excess demand occurs when the demand for commodities exceeds their
supply, especially when the latter is at full employment.
19Further, workers are always happy to get nominal wage increases while
resisting cuts in them.
20What the firms observe/know is that the labor productivity has
increased.
21This occurs through an increase in the real value of the money supply for
a constant money supply and lower price level. The resulting shift of the
LM curve to the right in the IS-LM diagram increases aggregate demand.
22This policy will not work if nominal wages rise by the rate of inflation,
which will mean that real wages do not fall. This could happen if the high
real wage is due to labor market imperfections (including excessive union
power), higher minimum wages set by the government, etc.

460
 CHAPTER 10
Employment, Unemployment, and
Inflation

This chapter provides a detailed treatment of the types of

unemployment. Of these, structural and frictional unemployment,
but not involuntary unemployment, can exist in the long-run or
short-run equilibrium. Involuntary unemployment, in addition to
the other two types, can also occur in disequilibrium.
This chapter also presents the analysis of unemployment and
inflation. Their relationship is usually expressed by the Phillips
curve and the expectations augmented Phillips curve. However,
neither may be a stable curve since each is somewhat deficient .
This chapter also discusses the costs of unemployment and
inflation .

This chapter studies the determinants of unemployment and inflation in the

economy and the relationship between them. The analytical basis for this
study has already been laid out in the preceding chapters, so that much of
the analysis in this chapter represents a re-arrangement and elaboration of
earlier arguments.
A particular focus of this chapter is on the relationship between
unemployment and the rate of inflation. This relationship is usually
encompassed in the Phillips curve. Related to the Phillips curve, though
somewhat distinct, are Friedman’s and Lucas’ versions of the expectations
augmented Phillips curve.
Among other topics presented in this chapter are the costs of
unemployment and inflation.

10.1 Definitions of the Labor Force and Labor

Supply
Workers in the labor force have an enormous variety of education and

461
skills, live in a variety of locations, or otherwise differ in the
characteristics relevant for the labor market. Similarly, jobs also differ in
their characteristics. Further, a given worker — with a given set of
characteristics — is often able to get several offers with different wages
and working conditions. Our earlier definitions of the labor force and labor
supply did not clearly recognize this variety. The following definitions do
so.
(i) The maximum labor force (Lmax)
This is the number of workers who have a job or are looking for a job —
no matter how high the wage rate that they want and what skills they
possess.
(ii) The labor force (L)
This is the number of workers who already have jobs or are looking for
work at the current wage, irrespective of their own productivity. It does
not include workers who want a higher wage than the current one,
though they are included in (Lmax). The unemployment (Lmax–L) of
such workers is called voluntary unemployment .
(iii) The labor supply (ns )
This is the number of workers who already have jobs or would be
willing to accept a job, if offered one, at the current wage appropriate to
their skills. It excludes those workers who do not have a high enough
productivity for employment at the current wage.
If all workers and all jobs were identical, the supply of labor would equal
the labor force. However, workers are heterogeneous, and differ in
physical characteristics, education, aptitudes, and technical, management,
and interactive skills, etc. Some workers are more productive than others,
and only some have productivity equal to or higher than the current wage.
The number of workers in the labor force who do not have the requisite
skills and other characteristics, etc., for the jobs available at the current
wage, constitute structural unemployment, which arises from the structure
of the economy in terms of the requirements of the available jobs and the
workers’ own characteristics.1 Structural unemployment includes the
seasonal unemployment of workers who work in seasonal industries, such
as fishing, logging, and farming, which offer substantially more
employment in some seasons than others.
Therefore, labor supply excludes structural (including seasonal)
unemployment, which equals (L–ns ):
The above concepts are compactly expressed in the following

462
relationships:

L = Lmax – voluntary unemployment, and

ns = L – structural (including seasonal) unemployment.

The above three concepts are illustrated in Figure 10.1a to c. The

maximum labor force (i.e., at any wage) is shown as Lmax. The labor force
is shown by the L curve. The labor supply is shown by the ns curve. The
labor market is in long-run equilibrium (i.e., with ns = nd) at the market-
clearing wage w 0* and long-run employment nf.

10.1.1 Frictional unemployment and actual employment in

labor market equilibrium
The labor market is not a single market with identical jobs, identical
workers, trade taking place in one spot and with perfect information.
Workers differ in the characteristics that firms require for the available
jobs and jobs differ in the characteristics that workers want. Some of these
characteristics are objective and can be precisely specified, while others
are much more subjective and have to be determined through interviews.
Further, there is imperfect information on both the characteristics of the
jobs (including the nature of the managers, co-workers, the work place
environment, etc.) and those of the workers. These factors make the labor
market a complex one, in which the matching of workers and jobs takes
time.
The labor market is also a dynamic market because of continuous
changes to the available jobs and workers. Jobs change because of the
emergence of new industries and skill requirements while some of the
existing jobs get eliminated. The turnover of workers in the labor force
occurs because some of the existing workers retire or otherwise leave the
labor force, while new ones enter the labor force. The new entrants
primarily consist of young workers who have recently completed their
studies. The entrants to the labor force differ in educations, skills, and
experience compared with those leaving it.
The labor market is, therefore, both complex and dynamic, which
together imply that the matching of workers to the available jobs takes
time. Assuming that vacancies equal the number of workers looking for
jobs, unemployment occurring because of the time required for finding the
appropriate job is called frictional unemployment. That is, frictional

463
unemployment consists of the number of workers who remain unemployed
for some time even when enough suitable job vacancies are available.
Frictional unemployment will be higher if the economy is more dynamic
(i.e., with greater turnover of jobs and workers), has greater differences in
the location of workers and jobs and has less accurate information on
vacancies and available workers. One component of frictional
unemployment is search unemployment, which consists of workers who
have already received a job offer but have decided to remain unemployed
while searching for a better job offer. The remainder would be workers
who have not yet received job offers.

Labor market equilibrium, which is the equality of labor demand and

supply, is defined to allow for frictional unemployment. On the labor
supply side, such unemployed workers have the appropriate characteristics
and are willing to accept jobs at the current wage. However, they are
temporarily unemployed because finding the right job takes time. On the
labor demand side, a corresponding number of jobs remain vacant because
finding the right worker takes time. If such unemployed workers could be
allocated immediately to the vacant jobs, frictional unemployment would
be zero. Therefore, labor demand (supply) refers to the number of jobs
(workers) if the cost and time taken to find the appropriate job/worker
were zero and the match between vacancies and the relevant workers were
instantaneous. However, this cost is not zero, so that matching workers and
jobs involves frictional unemployment.
Note that frictional unemployment does exist even within the analytical

464
concept of full employment but is counted as part of unemployment. In
general, for any individual worker searching for a job when there is an
appropriate job vacancy for his skills and location, it is expected to be of
short duration.
The notions of matched labor supply and matched labor demand in the
presence of frictional unemployment
Since the match between vacancies and the matched workers is not
instantaneous, actual employment is strictly less than the number specified
by the equilibrium between labor demand and labor supply by the amount
of frictional unemployment. As a corollary, ‘full employment’
incorporates frictional unemployment, with job vacancies corresponding to
the amount of frictional unemployment. Suppose that, for the given
amount and commitment devoted by workers and firms on finding
appropriate workers/jobs, we deduct the amount of frictional
unemployment, from labor supply and call it the ‘matched labor supply
(ns, matched). The matched labor supply curve in Figure 10.1a will be to the
left of the ns curve. Further, we deduct the amount of frictional
unemployment from labor demand, and call it the ‘matched labor demand
(nd, matched). The matched labor demand curve in Figure 10.1a will be to
the left of the nd curve. The two ‘matched curves’ will intersect at the
same wage as the nd and ns curves. However, employment will be less
than that given by the intersection of the nd and ns curves by the amount of
frictional unemployment. Note that the intersection of the nd and ns curves
is what is designated in macroeconomics as the full employment level nf,
even though it is the intersection of the nsmatched and ndmatched curves that
specifies the numbers of workers who will in fact be employed in
equilibrium.
Hence,

matched labor supply = ns, matched = ns – frictional unemployment and

matched labor demand = nd, matched = nd – frictional unemployment.

The number of workers actually employed in equilibrium is that when

matched labor supply equals matched labor demand.

10.1.2 Discouraged workers

465
A discouraged worker is one who does not currently have a job, had
searched for one in the past but stopped looking after some time. He does
not meet the requirement of currently ‘looking for a job’ and, therefore, is
neither included in the labor force nor in the labor supply.
Over the business cycle, discouraged workers play a volatile role by
dropping out of the labor force in downturns when jobs take too long —
and become too difficult — to find, while re-entering the labor force in
upturns as jobs become easier to find. The existence of such workers is
most evident when the economy is below full employment.2 Therefore, the
labor force and supply curves shift over the business cycle because of
changes in the number of discouraged workers.

10.1.3 Other determinants of the equilibrium level of

unemployment
The equilibrium level of unemployment not only depends on technical and
economic factors but also on institutional and political factors. One of the
institutional causes is encompassed in the theory of insider-outsider
unemployment, which claims that existing job holders (insiders) exert
pressure on firms to push the wages for the employed workers above the
market-clearing level. Such pressure is exerted not only by unions3 but
also by non-unionized employees. Firms accede to such pressure to
maintain a more harmonious work place. Pressures on the government, as
well as social and political considerations, can also result in setting or
raising the minimum wage above the full-employment one.
Workers differ in efficiency and can vary their work effort by either
shirking or speeding up. Firms can provide an incentive for better
performance by offering wages higher than the market wage. Doing so
enables firms to select workers from larger numbers of applicants and also
to retain better workers, since they know they are being ‘well paid’. The
theory which focuses on this reason for the wage being higher than the
market-clearing one is called the efficiency wage theory. It implies that
unemployment will be correspondingly higher than at full employment.
Among the political factors affecting the equilibrium unemployment are
the legal and political environment which governs the bargaining power
between firms and workers, and the non-wage costs of employment.
Among the latter are the laws governing the security of employment, such
as the legally acceptable conditions for hiring or layoff and severance pay.
Equilibrium unemployment is also affected by the terms (amounts,
duration, eligibility, penalties, etc.) of employment/unemployment

466
insurance benefits and income support (social assistance) programs
provided by the government to the unemployed since these affect the
attractiveness of accepting a job readily versus not accepting one.4
Reducing such payments to the unemployed reduces their reservation
wage — which is the minimum wage a worker wants in order to accept a
job offer — and increases their willingness to accept lower-paying jobs.
Further, social attitudes and norms such as the acceptability of not
having a job, of mothers of young children working or not working, of the
‘normal’ retirement age, etc., also affect labor force participation and are
among the determinants of the labor force. They also affect the pressures
to accept a job sooner rather than later and the unemployment rate. These
social attitudes and norms tend to differ among countries and among
different social, sex, and age groups.5 Another factor that affects labor
force participation is the time needed to achieve desired educational levels.
The desired educational level has been rising in most countries in recent
decades and has been a factor in reducing participation rates. However,
increasing longevity has persuaded many workers to retire later than in
earlier decades, which has tended to increase participation rates.
Box 10.1: The Labor Force Participation Rate
The labor force participation rate specifies the labor force — which
consists of the employed and unemployed workers — as a percentage of
the working age population. The latter is the sum of the labor force and
those working age people who are not in the labor force. The working
age is usually specified to be between the ages of 15 to 24 years for
youth, 24 to 65 years for adults, and 15 to 65 years for the overall
classification. Cultural factors, family needs and health, number of
children and their ages, the availability of time-saving home appliances
and nurseries, job opportunities, and wages are the most important
determinants of the participation rate. These determinants vary among
countries and even cultural groups within a country, so that the labor
force participation rate of women tends to differ among countries and
even cultural groups within a country.
The participation rate of working-age females has been of special
significance for changes in the labor force in recent decades, since this
rate was quite low until the 1960s and then began to increase rapidly. In
Canada, the participation rate for the 15 to 65 years age group was
78.6% for men and 49.7% for women in 1979, changing to 72.5% for
men and 58.9% for women in 1999. The overall participation rate for
both sexes taken together was 64.0% in 1979 and 65.6% in 1999.

467
 The non-employment rate
The use of the unemployment rate for cross-country comparisons has
some disadvantages since different countries are likely to use different
ways of measuring unemployment. Further, the discouraged worker
effect can give an erroneous picture of the true number of workers who
would like to have jobs but are not employed. Therefore, for cross-
country comparisons, some economists prefer to compare the non-
employment rate, which is the percentage of the population in the
working age bracket that does not have jobs. This rate is measured by
the number of unemployed workers plus those not in the labor force as a
percentage of the working age population. The non-employment rate in
western countries is usually between 35% and 55% . Among countries
and even among groups within a country, besides economic factors,
culture and traditions play an important role in determining this rate.
The use of the non-employment rate has its own disadvantages. One of
these is that it includes persons who choose not to seek paid work
outside the home and, therefore, are voluntarily without a job.
Historically, many women chose to be housewives and would have been
counted as non-employed. Since this pattern has changed in recent
decades, the use of the non-employment rate for time series comparisons
poses problems as does its use for cross-country comparisons.

Shifts in labor force participation rates in recent decades

Labor force participation rates vary over time and depend on real wages,
which is the potential return, i.e., when employed, for being in the labor
force, and on many cultural, technological, and economic factors. While
both male and female participations can change, the labor force
participation rate of women, especially in the developed economies, has
increased considerably since the early 1960s when the birth control pill
was invented and became easily available to the public. Over the same
period, the male participation rate decreased significantly. Part of the latter
is due to the increase in the family incomes resulting from the rising
participation rate of women.
Fact Sheet 10.1 provides a glimpse at the shifts in the participation rates
in USA and Canada since 1960. In both Canada and the USA, over these
48 years, while this rate decreased by about 10% for males, that for
females increased by about 20% for the USA and about 32% for Canada.
The total participation rate for both males and females increased
significantly in both countries.
Fact Sheet 10.1: Participation Rates in Canada and the United States,

468
 Male and Female, 1980-2008.
The following table classifies participation rates in Canada and the
United States as the fraction of the total working age population (over
15 years old) currently in the work force.

Source: For data 1960–1970 Constance Sorrentino, International

comparisons of labor force participation, 1960–1980. CANSIM
V2062816 for Canada 1980-2008. Bureau of labour statistics for US
1980–2008

10.2 The Components of Unemployment

The above definitions of the labor force and labor supply imply several
concepts of unemployment. An unemployed worker can belong to any one
of four categories:
Voluntary unemployment (U voluntary = L max – L ): those workers who
do not have jobs because they want a wage higher than the current one for
their skills. Macroeconomics does not consider such workers as really a
part of unemployment, so that such persons are not included in the
measures of unemployment.
Structural unemployment (including seasonal one6) (Ustructural = L – ns):
those workers who are in the labor force but do not have jobs because they
do not possess the requisite characteristics (skills, location etc.) to get jobs
with a productivity which would cover the going wage.
Frictional unemployment (Ufrictional = ns – ns’matched): those workers
who have the requisite characteristics and for whom there are enough jobs
in the economy but who are temporarily unemployed while they search for
an appropriate job.7
Involuntary unemployment (Uinv = ns’matched – nd’matched): those

469
workers who are in the labor force and the labor supply but for whom the
economy does not have enough jobs. Such workers have the appropriate
characteristics for jobs for the current nature of the economy and do seek
jobs at the current (or even lower) wage, but cannot get jobs because firms
do not have enough jobs to offer. The usual reason for the inadequacy of
jobs is deficiency of aggregate demand in the economy. Involuntary
unemployment is in addition to frictional and structural unemployment.
Note that it is zero if labor demand equals labor supply at the current wage
– i.e., the labor market clears.
Macroeconomics focuses on only the last three components of
unemployment and defines the level of unemployment U as:

U = U structural + U frictional + U inv

Note that voluntary unemployment is excluded from this definition of

unemployment.
The rate of unemployment (u ) is defined as the level of unemployment
(U ) divided by the labor force (L ). That is, u = U/L. u is often expressed
as a percentage of the labor force.
In practice, structural unemployment is long term, since it is often related
to skills no longer consistent with the evolving technology or shifts in the
location of jobs. For an individual unemployed worker, it could last many
years, even the rest of his life. Frictional unemployment is short term,
since it is related to the time taken to find a job, when there are sufficient
job vacancies to absorb the workers currently unemployed. It usually lasts
a few months. The duration of involuntary unemployment depends on
whether it is due to a high real wage or a demand deficiency. The latter is
related to the duration of recessions, which normally last about two years.
The diagrammatic depiction of the components of unemployment
The difference between the labor force and the labor supply is structural
unemployment. At the equilibrium wage rate w*0 in Figure 10.1b,
structural unemployment is shown by (L0 – nf).
At the equilibrium wage rate w*0, the difference between the actual
employment and that given by the equality of labor demand and the labor
supply ns is frictional unemployment. It occurs because of delays in
matching workers and jobs. In Figure 10.1b, at the wage w*0, frictional
unemployment equals (nf – nf, matched).
Natural unemployment is the sum of structural (including seasonal) and

470
frictional unemployment at the LR equilibrium (i.e., at full-employment).
At the wage w*0 and employment nf0, it equals (L –n f, matched) 8
For our further arguments and diagrams, we ignore the distinction
between labor demand and supply and their matched versions, and take the
intersection of the labor demand and supply curves as indicating the
number of workers actually employed at full employment. This is the
standard practice in expositions of the labor market in macroeconomic
theory.

10.3 Involuntary Unemployment

Involuntary unemployment (Uinv) is the difference between actual
unemployment U and natural unemployment U n. That is, U inv = U – Un
.As discussed in Chapter 9, involuntary unemployment can occur due to
two reasons:
(1) Involuntary unemployment due to a ‘high real wage’ but without an
aggregate demand deficiency.
(2) Involuntary unemployment due to the emergence of a demand
deficiency even without a real wage above the LR equilibrium level.
In some cases, the economy can experience involuntary unemployment
due to the simultaneous existence of both a demand deficiency and a high
real wage.
Involuntary unemployment due to a high real wage
Note that the labor market will not be in equilibrium if the real wage is
above the full-employment one. This is shown in Figure 10.1c at the wage
w1; which is above the full-employment wage w*, so that involuntary
unemployment at w 1 will equal (n s1 – nd1 ). Therefore, one cause of
involuntary unemployment is a ‘high real wage’ (i.e., one above the full-
employment one).
Involuntary unemployment due to the emergence of a demand deficiency
The second and more common cause of involuntary unemployment is
aggregate demand deficiency which makes firms employ less than the full-
employment number of workers. For this, start with the assumption that
the economy’s effective aggregate demand for commodities has fallen and
is met by the effective labor demand (and employment) at n ’1d, effectlve.
This possibility was covered in Chapter 9 and should be reviewed at this

471
stage. For this scenario, suppose that the economy continued to maintain
the full-employment wage w *, even though employment had fallen
because of a demand deficiency to only nd, effectlve1.9 This is shown in
Figure 10.1d. In this case, involuntary unemployment will depend only on
the demand deficiency and will equal (nf – nd’effectlve1).
However, since, at nd, effectlve, the marginal productivity of labor
exceeds the real wage w*, firms can afford to raise the real wage of their
employed workers. But, with more workers unemployed and greater
scarcity of jobs, the increase in unemployment would put downward
pressure on real wages. The real wage could, therefore, rise or fall relative
to the initial full-employment one at w*. The greater and the longer-lasting
the rise in unemployment, the stronger will be the tendency of the real
wage to fall. However, the (up or down) change in the real wage would not
change involuntary unemployment since it arises from inadequate demand
for commodities, rather from the real wage being too high.
Involuntary unemployment due to a credit crisis
As analyzed in Chapter 9 (and later in Chapter 16), a credit crisis, defined
as a fall in credit supply sufficient to finance full-employment production
levels, can reduce output and employment so that employment falls. The
resulting increase in unemployment is involuntary unemployment due to a
credit deficiency. It would be eliminated (‘in the long run’) once credit is
restored to adequate levels but this process can take several quarters.
Note that, in some periods, the economy could have a high real wage, a
demand deficiency, and a credit deficiency, so that each could contribute
to the causes of its involuntary unemployment.
The variation in the number of discouraged workers with the state of the
economy
Discouraged workers are not included in the above categories of
unemployed workers, but need to be considered in judging the
performance of the economy since such workers were in the labor force
and will re-enter it as economic conditions improve.10 From the statistical
perspective, such workers dropped out of the labor force, so that they are
not in any of the unemployment categories. Yet, from their own personal
viewpoint, they consider themselves as unemployed, even though not
actively looking for a job. Further, from the social viewpoint, discouraged
workers without a job represent loss of some output, just as for the
unemployed but not discouraged workers. Further, in assessing the
performance of the economy, its need for job creation and the loss from

472
unemployment, we have to focus on the sum of unemployed workers (as
defined above and in official data) and discouraged workers. Hence, a
more realistic statement of the economy’s performance is provided not by
u (the unemployment rate) but by (u + d) (i.e., the sum of the
unemployment rate and the discouraged worker rate).11

10.4 The Actual Rate of Unemployment

The relationship between the unemployment rate and employment was
derived in Chapter 1. It was specified as:

As discussed in Chapters 7 to 9, the actual rate of unemployment in the

economy in any given period can be separated into three components in
the following manner:

so that the actual rate of unemployment has three components.

(i) The natural rate of unemployment. This is the rate that exists in LR
equilibrium (full employment). It is the sum of the structural and
frictional unemployment rates at the full-employment levels.
(ii) The short-run deviation of the equilibrium unemployment rate from the
natural rate due to errors in expectations under uncertainty and/or the
existence of adjustment costs.
(iii) The disequilibrium rate of unemployment. One way of measuring it
was earlier specified as the difference between the labor supply at the
full-employment wage and the labor demand at the current wage. Such
unemployment was also labelled as the involuntary unemployment rate.
The actual unemployment rate varies across countries, as is illustrated in
Fact Sheet 10.2. However, the data plotted in this Fact Sheet is not strictly
comparable across countries since they tend to measure unemployment in
different ways and with different degrees of accuracy. Unemployment
rates should be high in developing economies with a large migration of
rural labor to the urban areas. However, this is often not reflected in the
reported measures, which are often based on different definitions of
unemployment, so that accurate comparisons across countries become
difficult. The unemployment rate in any given country also varies over
time, as is shown in Face Sheet 10.2 and more clearly in Fact Sheet 10.3
for the USA. The latter reports movements in the unemployment rate over
the past century. This rate was especially high in the USA, as in most

473
industrialized economies of the time, during the Great Depression years
from late 1929 to 1939, when it reached more than 20% and stayed
exceptionally high for a decade.
Fact Sheet 10.2: Unemployment Rates in Selected Countries, 1985-2008
This Fact Sheet reports unemployment rates in five countries, two (USA
and Canada) of which are developed ones and three (China, Malaysia,
and Thailand) are developing ones. The unemployment rate in each
country varies over time, rising in recessions and falling in booms (and
wars). With truly free movement of labor between countries,
unemployed workers would migrate from their own countries to
countries with lower unemployment, so that the unemployment rates
among countries would be approximately the same. However, the labor
market of each country is especially partitioned from that of others by
very limited and controlled legal immigration. USA and Canada, two
otherwise considerably integrated neighbors, follow a similar pattern of
unemployment but still do not show equal unemployment rates.
For Malaysia and Thailand, the early 1990s brought economic
prosperity and with it decreases in their unemployment rates. However,
the Asian Crisis of the mid-1990s meant the loss of many jobs,
especially in Thailand, causing a substantial rise in their unemployment
rates during the following decade. China has been going through an
economic boom, especially for the last decade, so that it should show a
fall in its unemployment rate. However, its graph shows otherwise,
probably because the reported unemployment rate is for urban areas and
there has been a rapid influx of workers from the rural areas to the cities.
Unemployment rates across countries are defined and measured in
different ways, making accurate international comparisons difficult, so
that it would be erroneous to conclude from this graph that the
developed countries have higher unemployment rates than developing
ones. Intuitively, the opposite is quite likely.

474
 This Fact Sheet illustrates the variations in the actual unemployment
rate over time by examining those for the USA. Since 1980,
unemployment in the United States has exhibited both a downward trend
and a cyclical pattern with peaks in the early 1980s, 1990s, and 2000s.
There is likely to be another peak in 2009 or 2010 resulting from the
financial and economic crisis, which started in 2007. Since the
unemployment rate has a countercyclical pattern, it is relatively low
during booms in the economy and higher during recessions.
Looking at unemployment rates over the past century demonstrates how
much the unemployment rate has varied and its sensitivity to real-world
events and policy shifts. The United States experienced its highest
unemployment between 1929 and 1940 during the Great Depression
when the unemployment rate reached 23%. Other episodes of high
unemployment were due to the dislocation following two World Wars,
the oil supply shocks of the 1970s, and the stagflation of the late 1970s
and early 1980s. Exceptionally low unemployment rates occurred during
the First and Second World Wars due to military needs and
expenditures.

10.4.1 The division of unemployment into natural and

475
 cyclical unemployment
Some treatments of unemployment divide actual unemployment into two
components:
(1) Natural unemployment
(2) Cyclical unemployment
Cyclical unemployment is defined in terms of the difference between
actual unemployment and the natural one, where the latter is the sum of
structural and frictional unemployment in full-employment
equilibrium.12Cyclical unemployment is the sum of involuntary
unemployment and the short-run variations in structural and frictional
unemployment from their full-employment levels.13 That is,
Cyclical unemployment = involuntary unemployment + the deviation of
short-term equilibrium unemployment from the natural unemployment = U
inv + (U * – Un) .
In practice, cyclical unemployment is usually identified with involuntary
unemployment due to fluctuations in aggregate demand. Cyclical
unemployment decreases during an upturn and increases during a
downturn of the economy. Hence, it is countercyclical (i.e., higher in
recessions and lower in boom) and makes total unemployment also
countercyclical.

10.4.2 Changes in the actual rate of unemployment over

time
The actual rate of unemployment over time can change because of changes
in any or all of its three components,14 since,

where ∂u/∂t is the change in the unemployment rate over time t. Hence,
changes in the actual rate of unemployment can occur due to:
(i) Changes in the natural rate of unemployment
(ii) Changes in the deviation of the SR equilibrium rate from the natural
rate due to variations in the errors in expectations under uncertainty
and/or adjustment costs
(iii) Fluctuations in the disequilibrium rate of unemployment, i.e., changes
in involuntary unemployment
The natural rate of unemployment does change over time. The other two
components of the actual rate can also change and do change over the

476
business cycle. In particular, they are negatively related to the business
cycle and tend to have smaller — sometimes even negative — values
during a boom than during a recession.
Short-run and long-run variations in the natural rate of unemployment
The natural rate of unemployment depends, among other variables, on the
available technology and the supply of the factors of production. Short-
term shocks to non-labor inputs make the natural rate deviate from its
long-run level. For instance, a decline in the price of imports, especially of
energy, reduces the natural rate; an increase raises it. Thus, the oil price
shocks of 1973–1975 and 1980–1981 raised the natural rate, while the
decline in the relative price of oil over the following two decades lowered
it. A war which destroys part of the country’s capital stock will also raise
the natural rate. Such variations can be designated as short-term variations
in the natural rate, as against the long-term variations resulting from
permanent changes in technology and the secular growth of the capital
stock.
Fact Sheet 10.4:

Source: OECD Source for OECD members, IMF IFS for rest .

477
 Source: OECD Source for OECD members, IMF International
Financial Statistics for the rest .

Estimating the natural rate of unemployment

The actual rate of unemployment can differ from the natural one, so that it
is very difficult to find a proper measure of the latter. A simple procedure
used for calculating the natural rate uses the assumption that the economy
maintains its full-employment level on average over the business cycle,
with symmetrical cyclical fluctuations around this level.15 Under this
procedure, the average unemployment rate over one cycle is said to be its
short-term natural rate,16 with the difference between the actual rate and
this rate specifying the cyclical variation. The average rate over several
cycles is a proxy for the long-term natural rate.
Fact Sheet 10.5: Unemployment and Trend Unemployment in the United
States, 1980-2008
This Fact Sheet uses USA data to illustrate movements in
unemployment and in trend unemployment. Unemployment had a
cyclical pattern (with a period of approximately 10 years), and was
greater than the trend rate in the first half of the 1980s, the early 1990s,
early 2000s and 2008-10.
The trend can be estimated over each business cycle or over several
business cycles. The former procedure yields three distinct trends over
the three business cycles since 1980. The three period trends differ in
their slopes and intercepts, but with a clear downward movement in each
of the trends during the three periods and in the overall trend over the
whole period. If the business cycle had been perfectly symmetrical, the
period trends would have been perfectly horizontal. The decline in the
trend rate of unemployment indicates the influence of technological
change, shifts in labor supply and factors other than purely cyclical ones.

478
10.5 Policies to Reduce Structural Unemployment
Structural unemployment consists of workers unemployed by reasons of
having inappropriate skills and education or being in inappropriate
locations, etc. (relative to those required to produce the equilibrium
marginal product of labor) in the current state of the economy. Structural
unemployment results from differences between the characteristics of
workers and the available jobs. A reduction in such unemployment
requires the reduction of such differences, through either changes in job
requirements or in workers’ characteristics. For example, if the jobs are in
one part of the country, while the unemployed workers are in another part,
either firms with the jobs are moved to the location of the workers or the
workers are moved to the location of the jobs. Changes in educational
patterns and re-training schemes can improve the match between job
requirements and workers’ skills.
Seasonal unemployment consists of unemployed workers whose jobs are
in seasonal industries but who do not currently (in the off-season) have
jobs. Creating offsetting seasonal industries to provide jobs in off-seasons
can reduce seasonal unemployment. Alternatively, migration of workers in
the off-seasons to other locations and industries could help. We have
incorporated seasonal unemployment in structural unemployment.
Extended Analysis Box 10.1: Changes in Structural Unemployment
Structural unemployment depends upon the supply structure — that is,
the labor market relationships and the production function — of the
economy and will change as the supply structure shifts. Technical
change and changes in educational and skill requirements, the level of
education of the labor force, the availability of information on jobs and
workers, the location of industries, etc., are thus likely to change the rate

479
of structural unemployment. This rate is, therefore, itself a variable and
not a constant. As already pointed out, shifts in the supply side of the
economy can change the structural rate.
Structural unemployment can also be changed by shifts in aggregate
demand (in addition to shifts in the distribution of this demand among
the sectors of the economy), including those brought about by the
pursuit of monetary and fiscal policies. A decline in demand will hit
some industries and job categories harder than others. Conversely, an
increase in aggregate demand will benefit some industries and some job
categories more than others.17
The impact of fiscal and monetary policies on structural unemployment
The impact of fiscal policy on structural unemployment has to be
examined in the following manner. The public sector has different
educational and skill requirements than the private sector. Therefore,
changes in the relative size of the government sector shift the demand
for different types of education and skills and can impact on the level of
structural unemployment. Monetary policy would not have such an
effect.
Increases in fiscal spending or tax concessions usually occur
selectively among industries, locations, and firms. They can occur as
increases in the subsidies for: education, re-training unemployed
workers, the relocation of industries or workers, the employment of
targeted groups with high unemployment rates, such as those of youth,
handicapped, or workers discriminated against in employment. They can
subsidize R&D expenditures or the expansion of certain industries. They
can also increase (or decrease) unemployment insurance benefits,
welfare benefits such as to mothers of young children so that they do not
have to seek outside employment, etc. All such policies affect either the
demand or the supply of labor and, therefore, affect structural
unemployment.
Fiscal deficits change aggregate demand, which can alter structural
unemployment because of changes in the structure of demand among
occupations and industries.
Therefore, for economies which stay at full employment on their own,
fiscal policies mostly affect structural unemployment through changes in
the relative size and efficiency of the government vis-à-vis the private
sector and through their selective nature.
Monetary policies tend to be general in nature rather than selective.
General monetary policy acts on the economy through aggregate
demand and only affects structural unemployment outside the full-

480
 employment state.

10.6 Policies to Reduce Frictional Unemployment

Frictional unemployment consists of workers who are unemployed
because of the time involved in matching workers to appropriate jobs.
Such unemployment is short-term, usually lasting a few months on
average. Speeding and improving the information flows on available jobs
and workers looking for jobs can reduce frictional unemployment. These
can be affected by selective fiscal policies, such as those which improve
the flow of information through the financing of job centers, internet
posting of vacancies and available workers, financing the travel of workers
to interviews, etc.

10.7 Measuring Involuntary Unemployment

Two ways of measuring involuntary unemployment
There are two ways of measuring the level of involuntary unemployment,
which we denote by U inv, with the rate of involuntary unemployment
designated by uinv. As we have defined earlier, involuntary unemployment
consists of qualified workers — i.e., those with the requisite skills,
location, etc. — who are willing to accept jobs at the current wage but do
not have jobs or job offers. This implies that involuntary unemployment
could be defined as the difference between labor supply (that is, the
number of workers with jobs or job offers which they are willing to accept
at the given wage) and labor demand (that is, the number of jobs at that
wage). For this definition, Figure 10.2 shows this measure of involuntary
unemployment by (ns — n1d), where ns depends on the real wage that has
come into being. If the real wage had become w1, labor supply would
equal n\, so that involuntary unemployment would be (ns1 — nd1).
Involuntary unemployment could also be defined as the difference
between full employment and actual employment, with the latter being
equal to labor demand. Figure 10.2 shows this measure of involuntary
unemployment by (n f — n d1) for this definition and given employment at
n d1 (either because of a demand deficiency or because of a high real wage
equal to w 1).
The rationale for the second definition is that as the wage falls, labor

481
supply will fall with it, so that (ii) measures the amount of unemployment
that needs to be eliminated to get the economy to full employment. We
will use the second definition as the more relevant one for measuring the
need for policy to take the economy to full employment. Note that
involuntary unemployment is zero when there exists full employment.18

Many economists consider involuntary unemployment to be synonymous

with the disequilibrium unemployment, so that it is used as an index of the
extent of disequilibrium in the latter.
Monetary and fiscal policies to reduce involuntary unemployment
The ability of monetary and fiscal policies to ameliorate (or cause)
involuntary unemployment depends on which of the two types of
involuntary unemployment exists in the economy. Therefore, we have the
following results.
• Expansionary monetary and fiscal policies can reduce an existing
demand deficiency and, thereby, reduce involuntary unemployment.
• Conversely, contractionary monetary and fiscal policies can create
involuntary unemployment and have been known to do so for several
quarters and even years. Many recessions are known to have been caused
by sharp contractions in the money supply. Part of the rise of
unemployment — and the amount of involuntary unemployment — in
such recessions is of a short-run equilibrium type until expectations
adjust to the lower actual inflation rate, but part of it would also be a
disequilibrium one while the economy adjusts to a lower level of demand.
• If involuntary unemployment is due to a high real wage without a
demand deficiency, and if nominal wages rise proportionately with price
increases, expansionary monetary and fiscal policies cannot address the
root cause of such unemployment. Expansionary policies will merely
create excess demand for commodities and inflation.
Box 10.2: Causes of High Unemployment in Canada and Europe Relative
to the USA

482
This box discusses the variations in the unemployment rate in the USA,
Canada, and a few European countries. Canada and most countries of
the European Union (EU) have had persistently much higher
unemployment rates during the last two decades than the United States.
The average unemployment rate for the EU started to increase rapidly in
the mid-1970s and had reached over 10% by the mid-1980s. It started to
fall after about 1992, but was still above 8% in 2000. The
unemployment rate in Canada also started to increase from about 6% in
the mid-1970s to a high of 11.2% in 1992 and then started to decrease. It
was about 7.5% in 2000. By comparison, the US unemployment also
started rising in the mid-1970s, peaked at about 9.7% in 1982, and then
decreased steadily, reaching about 4.5% in 2000.
The reasons for the rise in the unemployment rates in the mid-1970s
were due to three shared factors: the two sharp hikes in oil prices (1973-
1975 and 1980-1982), slowdown of productivity growth in the early
1970s and 1980s, and high real interest rates in the 1980s. But why did
the unemployment rates in Canada and the European Union in the 1980s
not decline as much as they did in the United States?
Generous unemployment insurance benefits in the former were one
reason, since these reduce the incentive to return to work. Extensive
regulations on layoffs and dismissals were another. Note that the longer
the workers stay out of work, the greater the deterioration in their skills,
and the greater the reluctance of employers to hire them, so that
sustained high unemployment tends to perpetuate itself.
Canada cut back drastically on its unemployment insurance benefits in
1996, both in terms of its duration and coverage. These seem to have
had the effect of decreasing overall unemployment by decreasing
frictional unemployment and even structural one, by forcing
unemployed workers to accept jobs earlier, seeking retraining for new
skills or shifting to less suitable jobs.
In recent decades, several European countries similarly reduced
income support for the unemployed workers. Among these was Ireland,
which cut back the ratio of unemployment insurance benefits to aftertax
wages from 77% to 64% in 1994, as well as reducing other income
support for the unemployed. Netherlands also cut back on such benefits
in 1986-1987, and gradually reduced the real minimum wage. Other
initiatives included wage moderation, reduction in income taxes and
taxes paid by firms for each worker, barriers to part-time work, etc.
These programs reduced the unemployment rates in both Ireland and
Netherlands over the 1990s to less than 5%, well below the European
Union’s average rate of about 8% in 2000.

483
10.8 The Costs of Unemployment
The main economic costs of higher unemployment are:
(1) Lost output and lower incomes, since those who become unemployed
do not produce output nor do they receive any income. The cost to the
economy of higher unemployment in lowering output is specified by
Okun’s rule, which was explained in Chapter 8.
(2) There may also be a long-term hysteresis effect, for which see below
(and Chapter 9).
(3) Changes in unemployment are also usually accompanied by changes in
the distribution of income. The impact of higher unemployment on the
distribution of income occurs in three ways:
(i) Unemployment among the less educated and less skilled members of
the labor force increases more than proportionately, as compared with
that among the more educated and more skilled workers, whom firms
are more reluctant to lose.
(ii) The wages of the workers who are still employed are more likely to
suffer downward pressure, translating into wage decreases, less
promotions, etc.
(iii) Frictional unemployment increases, especially for certain groups in
the economy. In particular, it increases for new entrants into the labor
force (such as recent graduates) and those subject to discrimination,
who find it harder and take longer to find jobs suitable to their
qualifications — and when they do get jobs, they may have to settle
for inappropriate occupations and lower wages than otherwise.
Most economists believe that the distribution of income — and not
merely that of labor income — becomes more unequal in recessions as the
unemployment rate increases: the poor suffer relatively more in lost
incomes than the better paid workers and labor as a whole suffers
relatively more than the owners of capital. However, some economists
dispute this, claiming that the wealthy also suffer since the stock markets
do badly, so that the return on capital falls and may fall more than labor
incomes in some recessions, while falling less in other recessions. Further,
the government’s social programs — such as the income-support,
unemployment insurance ones, etc. — provide a safety net for the fall in
labor incomes. Therefore, which groups lose more in recessions cannot be
established on a purely theoretical basis. It is likely to depend on the
causes and the effects of the recession, such as a collapse of stock market

484
prices, a decrease in aggregate demand due to contractionary monetary and
fiscal policies, transitory increases in unemployment due to shifts in the
industrial structure of the economy, etc.
The hysteresis effect
The impact of higher or lower current unemployment on the future long-
run unemployment is called hysteresis. There can be a long-term
cost/effect for the economy of high unemployment in recessions. This
occurs through the deterioration of the skills of the unemployed workers.
Further, workers slotted into lower skilled jobs do not get the chance to
learn the skills appropriate to better and more productive jobs, so that
when the economy recovers, the economy’s productive capacity is
lowered.
Non-economic aspects of hysteresis include the social and psychological
costs of not having a job or accepting an undesirable job during a recession
for the sake of a pay check but getting locked into otherwise undesired
careers for long years.

10.9 Relationship Between Unemployment and

the Inflation Rate from the AD-AS Analysis
The AD-AS analysis, in combination with the production function,
determines the level of employment. This, in combination with the labor
force specification, determines the rate of unemployment.
The determinants of the aggregate supply of commodities were specified
in Chapters 7 to 9. They include the technology of production and labor
supply. The labor supply depends on the population and its age and skill
distribution, as well as the participation rate. The determinants of the
aggregate demand for commodities were specified in Chapter 5. They
include the central bank’s target interest rate, money supply and fiscal
deficits, as well as net exports in the open economy. Therefore, the AD-AS
model implies that there are many determinants of the actual rate of
unemployment.
Further, as shown in Chapter 8, under uncertainty, the SRAS curve based
on the Friedman and Lucas supply rules implies that output rises when
there is an unexpected increase in the price level. Therefore, in the short
run, as the price level unexpectedly rises, unemployment falls.
Furthermore, as shown in Chapter 9, the dynamic disequilibrium analysis
implies that as aggregate demand increases, output and the price level rise
while unemployment falls. Hence, the AD-AS analysis implies that

485
where P is the price level. In the usual dynamic context of real-world
economies, we replace the price level by the inflation rate and write the
relationship as

where u is the actual unemployment rate and π is the inflation rate. This
relationship asserts that the unemployment rate falls as the inflation rate
increases. This negative relationship between the actual unemployment
rate and the rate of inflation for a given country is shown by plotting its
unemployment rate on the horizontal axis and its inflation rate on the
vertical one. The resultant curve is called the Phillips curve shown in
Figure 10.3b.

10.10 Phillips Curve (PC)

In 1958, A.W. Phillips plotted the rate of change of nominal wages against
the rate of unemployment for the U.K. over several periods from 1861 to
1957, and found that the data showed a downward sloping curve. His
original plotted relationship was of the form

where W″ is the rate of increase of the nominal wage rate, u is the actual
rate of unemployment, and dW′/du 0 indicates that the rate of increase of
nominal wages falls as unemployment rises. This occurs because a higher
unemployment rate weakens workers’ bargaining power relative to that of
firms. Therefore, the unemployment-wage relationship is of the form
shown in Figure 10.3a, which has the rate of change in nominal wages W ″
on the vertical axis and the actual rate of unemployment u on the
horizontal one. The estimated forms of this curve proved to be convex to
the origin, i.e., decreases in unemployment cause successively greater
increases in nominal wages.19 This relationship soon evolved into its
inverse — that is, u = f(W″) — and then into a relationship of the form

486
 where π is the rate of inflation and du/dπ 0. This equation is called the
Phillips curve (PC) equation and is drawn as the curve PC in Figure 10.3b,
with the rate of inflation π on the vertical axis. The PC states that increases
in the inflation rate reduce the unemployment rate. Its convex shape
indicates that at higher inflation rates, further increases in inflation
produce smaller decreases in unemployment.
The transition between the above two equations comes from the link
between the nominal wage rate and inflation: nominal wages represent the
main element of the cost of production so that an increase in nominal
wages will induce firms to increase their prices; alternatively, an increase
in prices causes labor to ask for compensation in the form of wage
increases. Hence, there is a positive relationship between W″ and n, and a
negative one between these variables and unemployment.
In the 1950s and 1960s, numerous studies for many countries, including
Canada, Britain, and the USA seemed to confirm the validity of the PC.
Even though the relationship seemed to shift between periods and
countries, the general form of the relationship seemed to be valid for the
1950s and 1960s. During the late 1950s and the 1960s, many economists
came to embrace this curve, incorporating it in their macroeconomic
modeling in preference to a structural specification (labor demand and
supply functions and equilibrium) of the labor market.

10.10.1 Use of the Phillips curve as a policy trade-off

Assume that the PC has been constructed for a country A and that it is
known to be stable, especially with respect to changes in the inflation rate
and the unemployment rate. The PC shows that lower unemployment — a
desirable objective — can only be obtained at the cost of higher inflation
— an undesirable side-effect. The central bank of the country can then use
its PC to choose the levels of inflation and unemployment that it considers
more desirable. Suppose that country A has the PC curve shown in Figure
10.3b and that it chooses the combination (u A, πA). It then adjusts the
target interest rate and the money supply to ensure the inflation rate u A,
and the economy delivers the unemployment rate usubA. Using the PC for
making this policy choice means that it is being used as a policy trade-off
curve between inflation and unemployment.
Different countries or policy makers may make different choices along a
given PC. For example, if country B has the same PC as in Figure 10.3b,
its central bank could choose the combination (u B, πB). Comparing B’s

487
choice with that of country A, country B’s choice indicates a relatively
greater dislike (disutility) for inflation than for unemployment. In reality,
different countries would most likely have different Phillips curves.
Therefore, the actual levels of inflation and unemployment in each country
would differ for two reasons: different π-n trade-offs as embodied in the
PC and different choices made by the policy makers.
In the 1960s and early 1970s, the monetary and fiscal authorities of
many countries assumed that the Phillips curve was stable and used it as a
policy trade-off between inflation and unemployment. Using this curve as
a policy trade-off, the policy makers tried to manipulate aggregate demand
to achieve the desired inflation rate and its concomitant desired
unemployment rate. In pursuing this agenda, the authorities wanted to
achieve better levels of output and employment than the economy would
have generated if left on its own. They seemed to be successful for some
years in lowering the unemployment rate. However, continued success in
this endeavour required that the PC be stable with respect to changes in the
inflation rate. Around the mid-1970s, it became clear that this was not the
case and that the PC shifts up as inflation accelerates.

10.10.2 Instability of the Phillips’ curve

subsequent analysis showed that the PC is not stable if inflation increases,
so that the use of the PC for policy choices had severe pitfalls. The
fundamental flaw of the PC is that both its shape and position depend on
the expected inflation rate πe, which in turn depends on the actual inflation
rate π. It is this impact of π on πe which shifts the PC and renders it
unstable. Therefore, a decision of the policy makers to raise the inflation
rate so as to get lower unemployment would sooner or later increase the
expected inflation rate and shift the PC to the right. The instability of the
PC for this reason is brought out in the Friedman and Lucas supply rules
and their implied versions of the expectations augmented Phillips curve.

10.11 Deviations of the SR Equilibrium

Unemployment Rate from the Natural One
These deviations can occur because of errors in expectations or adjustment
costs. The impact of the errors in expectations on output and
unemployment can occur through the labor market or the commodity
market. Milton Friedman focused on the former and Robert Lucas on the

488
latter, as explained in Chapter 8.
Fact Sheet 10.6: Historical Phillips Curves in the United States, 1960-
1995
This Fact Sheet uses US data to illustrate the Phillips curve and its
shifts. Note that the Phillips curve shifts with the expected inflation rate,
which shifts with the actual one.
In the following graph, at least three periods can be easily identified as
portraying a negative relationship between inflation and unemployment,
characteristic of a downward sloping Phillips curve. The first period,
spanning the 1960s, occurred in relatively low and stable inflationary
environment. As expectations began to adjust to the higher inflation rate
in the 1970s, the Phillips curve shifted. The expansionary monetary
policies of the 1970s resulted not in lower unemployment but the
stagflation of the mid- to late-1970s. Although a Philips curve can also
be seen from 1976 to 1981 and from 1985 to 1991, it is obvious these
are only short-term patterns. Looking over the whole period of the graph
with numerous changes in the actual and expected inflation rates, there
is no clear evidence of a downward sloping Phillips curve.

10.11.1 Friedman and the expectations augmented Phillips

curve20 (EAPC)
The Friedman supply rule, presented in Chapter 8, argued that if the
inflation rate were perfectly anticipated, the labor contracts would reflect
it so that the nominal wage would increase by the expected rate of
inflation. Consequently, the expected rate of inflation would not affect the

489
real wage rate, so that it will not affect employment or output. Hence, at
the expected rate of inflation, the rate of unemployment would be the
natural rate. However, the unanticipated rate of inflation21 was not
incorporated into the nominal wage set in wage contracts, since the
contracts were negotiated in advance of the knowledge of the
unanticipated inflation rate. Positive unexpected inflation, that is (i.e., (π
— πe > 0)), lowers the real wage. Since this reduces labor costs,
employment and output rise while unemployment falls. Conversely,
negative unexpected inflation (i.e., (π — πe 0)), raises the real wage, and
causes a short-run increase in unemployment. This analysis deals with
short-run deviations of unemployment from the long-run one. Therefore,
only unexpected inflation can cause deviations in the short-run
unemployment rate u * from the natural rate u n, so that, according to the
FSR and LSR, the correct equation for the unemployment rate is

This equation asserts that the proper relationship between u* and n is not
the PC but between the deviation of unemployment from its natural rate
and the unanticipated inflation rate.22 This relationship is known as the
expectations augmented Phillips curve (EAPC). The EAPC can be drawn
in two different ways, as shown alternatively by the curves marked EAPC
in Figure 10.4a and b. Note that the horizontal axis of Figure 10.4a has (u*
— un), while that of Figure 10.4b has only u*. The vertical axis of both
figures measures unanticipated inflation (π — πe).
The EAPC in Figure 10.4b differs from the PC in two ways:
(i) The horizontal axis for the PC diagrams has the actual unemployment
rate u while the EAPC in Figure 10.4b has the short-run equilibrium rate
u* on this axis.
(ii) The vertical axis for the EAPC figures is (π — πe), while that for the
PC is n.

490
Empirical research and the widespread experience of stagflation in the
mid- and late-1970s in the industrialized economies seemed to show that
the PC was unstable in a period of accelerating inflation. Further, the
EAPC seemed to perform much better in such periods, especially at high
and accelerating rates of inflation. Many economists after the 1970s
preferred dropping the PC/EAPC curve from their models, and replacing it
by full labor market specification (i.e., with labor demand and supply
equations and an equilibrium condition).
Similar to the arguments for going from the FSR for output to the EAPC,
the Lucas supply rule (LSR) for output also implies that:

which has the same generic form as Friedman’s version of the EAPC.
We, therefore, have two types of EAPC relationships. One of these is
Friedman’s version which is based on errors in expectations in labor
market and contractual nominal wage rigidities during the duration of the
labor contract. The other is based on the Lucas supply rule which is based
on errors in the expectations of relative prices in commodity markets. We
will call the latter as the Lucas version of the EAPC. In both versions of
the EAPC, the decrease in unemployment is a consequence of the
unanticipated increase in inflation. Hence, the increase in aggregate
demand does not reduce unemployment unless it first causes an
unanticipated increase in the inflation rate.

10.11.2 Expectational equilibrium and the natural rate

Define expectational equilibrium as one when there are no errors in
expectations — which is an assumption of the long-run equilibrium. In this
LR (error-free) equilibrium,

Therefore, the EAPC implies that in the expectational equilibrium (i.e.,

with Pe = P ):

which assert that, in expectational equilibrium, employment and the

natural unemployment rate are the full- employment ones and are
independent of the inflation rate or the price level. Deviations from these
occur because of expectational errors, with positive errors [i.e., (π — πe) >
0] causing an increase in employment and a decline in the unemployment

491
rate. Conversely, negative errors [i.e., (π — πe) < 0] cause a decrease in
employment and an increase in the unemployment rate.
Extended Analysis Box 10.2: The Friedman and Lucas Supply Rules
Chapter 8 presented the Friedman supply rule (FSR) based the price
expectations incorporated in wage contracts. It had also presented the
associated concept of Friedman’s expectations augmented employment
function. The FSR is:

where all the variables are in logs, y* is the SR equilibrium output and
yf is the LR equilibrium output.
Correspondingly, the expectations augmented employment function
was

where n* is the SR employment and nf is the LSR full-employment

level. The diagrammatic analysis of these equations was given in
Chapter 8.
Note that the relationship between the short-run equilibrium level of
employment n* and the short-run equilibrium unemployment rate u* is:

where L is the labor force. Hence, equation (12) can be converted from
employment to unemployment and implies that

where f is a functional symbol. In a dynamic context, (P — Pe) is

replaced by (π — πe), so that we have

Note that several assumptions were needed for deriving Friedman’s

EAPC. Among these were:
a. EAPC assumes market clearance in the labor market at the wage
contract stage. At the production stage, labor willingly supplies the
amount of labor demanded by firms. Further, commodity markets
clear. Hence, the EAPC does not allow for disequilibrium in the labor
market — in which some workers may want jobs at the existing wage
but cannot get them or firms want workers at this wage but cannot get
enough — or in the commodity market — in which firms cannot sell
all that they want to produce.

492
 b. Friedman’s EAPC captures the impact of one type of errors in
expectations: that is, those embedded in wage contracts. The Lucas
supply rule captured the impact of expectational errors in commodity
markets and leads, as shown below, to a similar EAPC.
Expectational errors and the commodity markets: The Lucas supply
rule
The Lucas supply rule presented in Chapter 8 considered the impact of
errors in forecasting changes in relative prices versus those in the price
level. We can modify this analysis to forecasting errors in the rates of
increase in relative prices versus those in the general rate of inflation.
The LSR then implies that the deviations of output from the full-
employment rate are positively related to the unanticipated inflation rate.
Since changes in output and employment are positively related, changes
in the unemployment rate and output are negatively related (i.e., du/dy <
0), so that an increase in unanticipated inflation (π — πe) reduces
unemployment. Hence, similar to the arguments for going from the FSR
to the EAPC, the Lucas supply rule implies that

This equation has the same generic form as Friedman’s version of the
EAPC.

10.12 The Implications of the EAPC for Shifts in

the Phillips Curve and for Policy
The EAPC implies that there is a distinct Phillips curve for each expected
rate of inflation: the higher the expected inflation rate, the higher the
Phillips curve. Figure 10.5a shows three Phillips curves and the long- run
PC (i.e., for which Pe = P ) as PCLR.23 PC0 is the PC for πe = 0. It cuts the
PCLR at π = 0 and the (natural rate) of unemployment u0. If the expected
inflation rate rises to 5% (= 0.05), the Phillips curve shifts to PC1 which
cuts the PCLR at πe = π = 0.05. But if the expected inflation rate
accelerates to 10% (= 0.10), the Phillips curve shifts to PC2 which cuts the
PCLR at πe = π= 0.10.
Therefore, the PC is not stable if the monetary authority changes the rate
of inflation which induces the public to change its expected rate of
inflation. However, it will be stable if the expected inflation rate does not
change. Therefore, the PC does not shift during the period of unanticipated

493
inflation but does shift once its occurrence has been realized.

The relationship between the Phillips’ curve and the EAPC

The dependent variable of the Phillips curve is u . We can express u as

where u* is the short-run equilibrium unemployment rate specified by the

FSR and the LSR. These rules assert the EAPC, which is the statement that

Further, (u — u*) is involuntary unemployment uinv. Therefore,

If we were to draw the relationship between u and π in a PC type diagram,

the curve would shift with shifts in un and πe. The above equation is the
proper statement of the PC, with the EAPC being only one part of the PC.
As a corollary, the EAPC is not an accurate explanation of fluctuations in
u, unless it could be validly shown that the economy in question never has
uinv.
Since uinv varies over the business cycle, the impact of uinv occurs over
the business cycle between recessions and booms. Involuntary
unemployment uinv is positive in deficient-demand recessions. An
expansion of aggregate demand is likely to impact on both output and
prices,24 with the impact depending on the extent of the demand
deficiency: with a large demand deficiency and high uinv and excess
capital capacity, the impact is likely to be much greater on output than on
prices, as in a deep trough, while, as the economy nears full employment,
the impact is likely to be much greater on prices than on output. This
coincidental effect of changes in aggregate demand on both u, through
uinv, and n creates a negative statistical relationship between u and π. This
relationship is likely to be non-linear and convex.
Note that another part of the relationship between u and π occurs through

494
the EAPC component of the above equation. This part varies with
positively with (π — πe), as demonstrated by the FSR and the LSR.

10.13 The EAPC and the Non-Accelerating

Inflation Rate of Unemployment
The EAPC implies that if πe = π, the economy will be at the natural rate of
unemployment. Further, it implies that only PC is stable if πe = π. If π >
πe, the PC curve would shift up as expectations adjust to the actual
inflation rate. It would shift down if π πe.
However, in real-world economies, there is almost continuous
improvement of products and the creation of new ones, so that the public
might regard a small amount of inflation (say πnaim) as a reflection of such
improvements. In this case, workers would ask for compensation only for
(π—nnaim) but not for the full inflation rate. Therefore, the presence of
inflation less than or equal to πnairu will not produce nominal wage
increases or further inflation,25 so that the unemployment rate
corresponding to πnairu will not generate inflation and is called NAIRU —
translated as the non-accelerating inflation rate of unemployment. Given
that πnaim > 0, the NAIRU-modified PCLR will cut the PC at πnairu and not
at π = 0, as is shown in Figure 10.5b. The EAPC with this modification is

From the policy perspective, NAIRU implies that the central bank can
create or allow inflation equal to πnairu without causing a shift of the
Phillips curve. It can thereby lower unemployment on a long-term basis to
unaim, which is less than the natural rate, as shown in Figure 10.5b.

10.14 The Implications of the EAPC for the

Impact of Anticipated and Unanticipated
Demand Increases
Assume that the economy is initially in full employment and that
aggregate demand increases in an anticipated manner. This increase could
be due to an increase in investment or expansionary monetary and fiscal
policies. Since it has been assumed that this increase is to be anticipated by

495
the public, we will also assume that its impact on the rate of inflation is
also anticipated. Therefore, the increase in demand will not produce any
errors in the expected inflation rate, so that, according to the EAPC, the
economy will maintain the natural rate of unemployment. Hence,
anticipated changes in inflation make the economy move along PCLR: the
demand increase will not bring about a decrease in unemployment, but
merely cause a proportionate price increase.
In the case of unanticipated demand increases, the public would not be
able to anticipate their impact on the inflation rate, so that the
unemployment rate will fall. Under the Friedman version of the EAPC,
this would be due to the negotiated nominal wage being too low, so that
the real wage will fall and make labor cheaper. Under the Lucas version of
the EAPC, the increase in output will occur because some of the inflation
will be attributed by individual firms to an increase in their relative price,
which will induce the firms to increase their employment, thereby
decreasing unemployment. Hence, unanticipated changes in inflation
make the economy move along PC SR.
10.15 The Determinants of Inflation
As illustrated by the following Fact Sheet 10.7, inflation rates vary
considerably over time for a given country and across countries. Inflation
rates in very high double or triple digits are called hyperinflation.
Fact Sheet 10.7:
European and North American countries generally keep their inflation
rates within single digits. With this objective, when inflation rates
reached low double digits in the late 1980s, the central banks of
countries such as UK, Canada, New Zealand and many others pursued
contractionary monetary policies to bring inflation back to low single
digits. In the 1990s, many adopt inflation targeting, with inflation targets
as low as 2% or 3% on average.
Developing countries often have higher inflation rates than developed
ones. Several South American countries experience hyperinflation
during the 1970s and 1980s, as did Israel in 1980. Currently, Zimbabwe
has hyperinflation, with the inflation rate becoming over 24,000% in
2007. It increased further in 2008 and 2009.

496
 The AD-AS analyses of Chapters 5 to 9 provide the determinants of
inflation. Basically, these are:

(i) Demand factors

(ii) Supply factors
(iii) Equilibrium versus disequilibrium in the commodity market
The AD-AS analysis showed that increases in aggregate demand cause
inflation, while decreases in aggregate demand moderate inflation.
Decreases in aggregate supply cause inflation, while increases in aggregate
supply moderate inflation. Further, past inflation affects expectations of
the current inflation rate. The AD-AS model implies that:

n = f (change in AD, change in AS, a measure of disequilibrium, past

inflation rates).

The determinants of AD can be spelled out by several different proxies.

Among these are the target interest rate, the money supply, and fiscal
deficits, and, in the open economy, the level of net exports.
The determinants of AS can also be spelled out by several different
proxies. Among these are the average unit cost of the production of
commodities — which captures the improvements in productivity, changes
in wage rates, and changes in the cost of resources, especially oil and other
energy sources. Another determinant of AS in the open economy is the
domestic price of imported goods, which depends on changes in exchange
rates. Note that the shifts in aggregate supply will capture the shifts in the
natural rate of unemployment.
Deviations of the unemployment rate from the natural one can be used as
a proxy for the extent of disequilibrium (AD-AS) in the economy.
The economy usually shows inertia and persistence in the rate of

497
inflation, which is partly due to the dependence of expected inflation on
past inflation and partly due to the costs of adjusting prices rapidly.
Therefore, past rates of inflation are an additional element in explaining
the current inflation rate. This inertia means that inflation tends to lag
behind the growth of aggregate demand, including that of the money
supply. Over the business cycle, although inflation is procyclical, it picks
up relatively slowly at the start of upturns and falls relatively slowly as the
economy turns into a downturn.
Another way of stating the determinants of inflation
The types of inflation are sometimes classified as:
(1) ‘Demand pull ’ inflation, caused by increases in aggregate demand;
(2) ‘Cost push’ inflation, caused by decreases in aggregate supply. Among
the factors often listed in the cost push category is the demand by
workers, especially by unions, for higher wages, with firms unable or
unwilling to resist these demands. This pressure increases nominal wage
rates, which increase firms’ costs of production, so that firms react by
raising their prices, thus resulting in inflation. Cost push inflation can
also occur due to increases in the prices of inputs other than labor. For
example, increases in oil and gas prices increase the cost of production
and cause firms to raise their prices.

10.16 The Costs of Inflation

The costs of inflation depend upon whether it is anticipated or not, and
upon the relative economic and political power of the different economic
agents.

10.16.1 Inflation under perfect competition with fully

anticipated inflation for all future periods: the costs of
inflation in the analytical long-run case
Define fully anticipated inflation as one which is anticipated by all
economic agents on both sides of all ‘bargains’ (formal and informal
contracts) in the economy over all relevant future periods — a rather tall
assumption. Further, assume that:
(i) All markets are in perfect competition and are fully efficient, i.e., adjust
the prices immediately to equilibrium after any shock to demand and
supply functions.

498
(ii) No (private or public) agent has special power to take advantage of the
inflationary situation.
(iii) There are no costs of adjustment or ‘inflexibilities’ of any type which
prevent instantaneous adjustments in prices or quantities.
Under these assumptions, rational agents taking part in any negotiations
or decisions would agree to increase the relevant nominal variables
proportionately with the inflation rate. Therefore, all prices, nominal
wages, and interest rates — as well as the nominal values of other
variables such as mortgages, pensions, minimum wages set by the
government, etc. — would increase proportionately (and immediately)
with the inflation rate. Consequently, the economy will operate at its long-
run level, and employment and output would remain at their full-
employment levels. There would be no losers or gainers from such
inflation, so that its costs and benefits would not be significant. One could,
therefore, ignore such inflation as inconsequential from a purely economic
viewpoint.
From the perspective of realism, there are several objections to the
realism of the assumptions made in arriving at the above result. Among
these are:
(i) In the modern economy with explicit or implicit long-term contracts —
for example, on employment, loans/bonds, pensions, etc. — fully
anticipated inflation would require accurate anticipations over several
decades. This does not occur in real time.
(ii) The economy always has some deviations from perfect competition
and efficient markets.
(iii) Economic and political power is always unevenly distributed among
economic agents, of which the government is the most powerful one.
For example, the government exercises a preponderant power, e.g., in
the payments made out for various types of social benefits, and in
setting tax rates and tax brackets, etc. It often does not adjust these fully
to the anticipated inflation rate, or even ex post to the actual inflation
rate. Firms — especially if they are large and/or multinationals — also
often tend to have greater clout than workers — especially if the latter
are not unionized — so that nominal wages often do not increase
instantly and proportionately with the inflation rate, with the result that
real wages fall, usually when inflation accelerates.
(iv) There are always some adjustment costs or inflexibilities in the
economy, so that even fully anticipated inflation imposes some
economic costs. For example, interest is not paid on currency and some
other components of money, so that the purchasing power of such

499
 holdings decreases by the inflation rate and their holders lose. As
another example, firms have to post prices in price lists, which can only
be changed at some cost, labeled in Chapter 8 as ‘menu costs’.
(v) Further, even under the very restrictive and rather unrealistic
assumptions made for the long run, fully anticipated inflation always
imposes some non-economic costs, including psychological, social, and
political ones. For example, the continuous confrontation with the ever-
rising prices of products imposes a psychological cost/shock on
consumers and a ‘re-calculation cost’ when confronted with higher
prices at the point of purchase (e.g., in a supermarket). Furthermore,
practical experience indicates that very high inflation rates, even if they
are anticipated, tend to result in strikes, demonstrations, and public
dissatisfaction and increasing conflict among social classes, etc., thereby
tearing at the economic, social, and political fabric of the country.
In any case, fully anticipated inflation over all future periods never
occurs in real-time economies. The following makes a more limited
assumption on the extent to which inflation is anticipated.
An intermediate scenario: anticipated inflation in the commodity and
labor markets for some quarters ahead
A more realistic set of assumptions than made for the long run is that the
inflation rate is anticipated for the next few periods (a few quarters) only.
Also assume that this degree of anticipated inflation is all that is needed
for economic decisions in the commodity and labor markets. These include
those on labor supply and demand, on the duration of labor contracts and
the production decisions of firms. Further, assume that there is perfect
competition in these markets and they are fully efficient. Under these
assumptions, the FSR and the LSR showed that such anticipated inflation
will not produce short-run deviations from full employment or change the
real wage.
Further, according to the Fisher equation on interest rates, since the
inflation rate is anticipated for the next few periods, the short-term market
interest rates will rise by the rate of inflation, so that the short-term real
rates of interest will not be affected.
However, there are several costs of such limited anticipations of future
inflation. These include:
(i) Longer-term expectations may still prove to be erroneous, so that long-
term contracts would be based on incorrect expectations. Consequently,
long-term bond holders may gain or lose, as may pensioners, and others
locked into numerous long-term arrangements. In general, such limited

500
 anticipations would cause changes in the wealth distribution.
(ii) In addition, markets are not perfect and economic power is not evenly
balanced in the economy. Administratively set payments, e.g., in social
programs run by the government, may not be — and usually are not —
fully adjusted even to the current (known) inflation rate, let alone to
next period’s anticipated rate. Further, the salaries of government and
para-government employees may not adjust fully to inflation plus
average productivity growth in the economy. Any adjustments that
occur depend on political factors.
(iii) Under a progressive income tax system, the ‘bracket creep ’ of
nominal incomes caused by inflation puts some tax payers into higher
tax brackets — unless these are themselves indexed to the inflation rate
— and increases tax payments in real terms. It correspondingly reduces
the real disposable income of the public.
(iv) Since interest is not paid on currency and some other components of
M1, holders of these lose by the decrease in their purchasing power.

10.16.2 The costs of unanticipated inflation

Unanticipated inflation imposes costs and benefits additional to those of
anticipated inflation. These are:
• If we start with full employment in the economy with the expected
inflation rate equal to the actual one, an unanticipated inflation produces
a decrease in the real wage rate but a short-run increase in employment
and output along the SRAS curve. However, once the inflation rate
becomes anticipated, this gain disappears. Further, if the authorities
subsequently decide to lower the inflation rate and the public does not
immediately lower its expectations correspondingly, employment and
output would fall below the full-employment one. Hence, the initial gain
in employment and output from rising inflation has to be compared with a
subsequent loss in them when inflation is reduced.
• Asset prices, such as of equities, bonds, and houses, do not adjust
proportionately to the change in commodity prices, so that their owners
either lose or gain in real terms.
• As inflation accelerates, real wages fall for the duration of nominal wage
contracts, so that workers lose.
• Nominal interest rates do not incorporate the unanticipated inflation rate,
so that lenders lose and borrowers gain.
• Recipients of all kinds of fixed incomes, including pensions, social
benefits, etc., lose in the process.
• The tax brackets and collections of the government cannot be adjusted to

501
 neutralize the effects of unanticipated inflation on nominal incomes.
• Since interest is not paid on currency and some other components of M1,
holders of these lose by the decrease in their purchasing power during
unanticipated inflation.
• Once the past inflation rate is realized, various types of adjustment costs,
including the menu costs of changing price lists, have to be incurred.
A caveat
The assumption that the average inflation rate is anticipated hides the
following economic aspects usually experienced in the real world:
Inflation, irrespective of whether its average level is anticipated or not,
never occurs in the form of a simultaneous proportionate increase in all
prices. Some commodity prices increase earlier than others, and the wages
of some workers increase faster than those of others. This imposes all
kinds of adjustments on the economy, and some economic agents gain
while others lose.
Actual experience also indicates that inflation is never fully anticipated
over the next few years, even by what are supposed to be the most
knowledgeable economic agents in this respect. Even central banks and
financial institutions often make mistakes, sometimes with severe losses.
Inflation does not occur evenly over time. This increases uncertainty
about the pattern of future inflation rates. This has many consequences.
We illustrate two of these. For firms, uncertainty about the pattern of
future inflation rates increases the uncertainty of the future profitability of
investments and affects their investment plans. Lenders with an aversion to
the risk inherent in uncertain rates of inflation increase the required risk
premium in making loans, which increases the real interest rates.
Box 10.3: The Impact of Hyperinflation on Long-Term Output Growth
and the Standard of Living: Practical Experience in Contrast to Theory
Economic theory implies that anticipated inflation does not affect the
real variables. Further, under rational expectations, any unanticipated
future inflation is random and the effects of any unanticipated inflation
are eliminated as time passes and it becomes anticipated. Therefore, the
theoretical prediction is that increases in inflation rates would not do
much damage to the economy in terms of reducing the growth rates of
output, output per capita, or have a long-term impact on unemployment
rates. Many empirical studies confirm this finding.
However, studies of the economic history of hyperinflationary
economies and their development seem to indicate that such economies
usually do much more poorly than low inflation economies over long

502
 numbers of years. Comparison of this experience with the theoretical
implications seems to imply that economic theory does not encompass
all the effects of hyperinflation. Among the possible effects which are
ignored can be following.
(i) Hyperinflation increases uncertainty about the future returns to
physical and human investment, and destroys consumer and business
confidence in the future economic performance of the economy.
Investment in both physical and human capital decreases.
(ii) Invention and innovation suffer.
(iii) The increased conflict among the social classes, and between firms
and workers, is detrimental to economic performance.

10.17 The Sacrifice Ratio

Both unemployment and inflation impose costs on the economy. We can
use their sum (or weighted sum, with the weights chosen by society or the
policy makers) to provide an index of the sacrifice by the population over
time. This index is called the sacrifice rate or ratio, and has been used in
some studies to provide a rough measure of the failure of the economy to
perform relative to a full-employment, zero-inflation state.
Fact Sheet 10.8 : Sacrifice Ratio in the United States, 1950-2008
This Fact Sheet illustrates movements in the sacrifice ratio (defined as
the sum of the inflation rate and the unemployment rate) by looking at
that for the USA. While one could argue that its constancy over time
would be desirable as a social objective, in reality, the sacrifice ratio has
varied dramatically over the past 60 years. The relatively high sacrifice
ratio in the 1970s and the early 1980s was a result of stagflation, when
the central banks pursued expansionary monetary policies in an attempt
to offset the decline in GDP growth and the consequent rise in
unemployment due to increases in oil price increases during that period.
They failed to do so but instead resulted in higher inflation rates.
Note that the Phillips curve, even if it was stable, does not imply the
constancy of the sacrifice ratio.

503
10.17.1 Gradualist versus Cold Turkey Policies of
Disinflation
After a period of high and rising inflation rates, central banks often try to
reduce the inflation rate. However, the expected inflation rates do not fall
as fast. The Friedman and Lucas supply rules imply that disinflation in a
period in which expected inflation rates become above the actual ones
causes output to fall and unemployment to rise. Given this cost, central
banks could follow either a ‘cold turkey’ or a ‘gradualist’ policy of
disinflation. Under a cold turkey policy, the central bank pursues a
contractionary monetary policy — increase in the target interest rate,
supported by a decrease in the money supply — strong enough to
immediately lower the inflation rate to its desired level, even though this
causes a considerable gap between actual and expected inflation rates and
large losses of output. Under a gradualist policy, the central bank pursues
more moderate contractionary ypolicies but over a longer period of time so
as to decrease the actual inflation rate gradually to the desired one. This
pattern allows the expected inflation to drift along with the actual inflation
to lower levels. The gap between the two remains more limited and so do
the losses in output from disinflation.
However, expectations tend to adjust faster after the jolt of the cold
turkey policy, so that the recessionary period with output below its full-
employment level is shorter than under a gradualist policy. Therefore, the
cold turkey policy involves a bigger loss of output but over a shorter
period while a gradualist policy involves a smaller loss of output but over
a longer period. These two extreme patterns provide the central bank a
choice of the disinflation pattern that it would want to follow. The
preferred policy is likely to depend on the extent of the output losses

504
involved, the rapidity of adjustment of expectations, the central bank’s
preferences, and the reaction of the public to rising unemployment.

10.17.2 Indexation to the rate of inflation: should nominal

wages and interest rates be indexed?
Indexing nominal wages to the price level, which is a proxy for the cost of
living, prevents inflation from reducing real wages in an unanticipated
manner (see Chapter 8). Therefore, some economists favor such indexation
and support the idea of a cost-of-living clause (i.e., one which adjusts
nominal wages by the change in the price level) in nominal-wage
contracts. Similarly, indexation of the nominal interest rate to the rate of
inflation ensures that the real interest rate does not change because of
inflation. A similar argument applies to pensions, unemployment insurance
benefits, minimum wages, and other forms of payment. However, such
indexation raises several questions, some of which are:
• Should indexation be privately determined in contracts (e.g., in wage
contracts negotiated by firms and unions), or should it be legally
imposed?
• Should governments also be bound to index wages, welfare payments,
pensions, tax brackets, etc.?
• Should government bonds set a real interest rate, so that their nominal
coupon rate rises by the inflation rate?
• Should indexing be by the inflation rate or the inflation rate consistent
with NAIRU?
Eliminating the costs and the pain imposed by inflation reduces the
pressures to reduce it, so that inflation tends to continue much longer.
Should indexing not be done so as to retain this motivation for its
reduction?
If indexation is such a good thing, why does the central bank not avoid
the needed continual adjustments of nominal wages, etc., by just
maintaining stable prices?
These questions are deliberately not answered here but are raised in
order to encourage their discussion.

10.18 Conclusions
• The actual unemployment rate has three components. Of these, the
structural component of the natural unemployment rate un does not

505
 depend on aggregate demand, the price level, or the rate of inflation.
However, errors in price expectations and adjustment costs can cause
positive or negative values of the deviation of the SR equilibrium rate
from the natural rate, i.e. (u* — un). This component of unemployment
depends on the adjustment costs and errors in the expectations of the
price level and the rate of inflation. Disequilibrium in the labor market
can cause positive or negative values of the disequilibrium component (u
— u *). This component of unemployment depends on the extent of the
demand deficiency for commodities or a real wage higher than the full-
employment one.
• The PC is a real-world, real-time relationship between the actual
unemployment rate and the inflation rate in the economy. Changes in the
actual unemployment rate include variations in the natural rate, changes
in the short-run deviations of the unemployment rate from the natural
one, and changes in the disequilibrium level of unemployment.
• The EAPC only considers the short-run deviations of the unemployment
rate from the natural one due to errors in price expectations. The
disequilibrium level of unemployment is not encompassed in the EAPC.
The EAPC is an analytical relationship focusing exclusively on the
impact of errors in price expectations, while the PC curve also includes
other economic forces not captured in the EAPC. However, the EAPC
does validly show that the PC will shift with changes in the expected
inflation rate and that the latter will depend upon changes in the actual
inflation rate. Therefore, the PC cannot be taken to be stable. It cannot be
used by policy makers as a durable trade-off between inflation and
unemployment. The long-run PC is vertical under the definition of the
long run since this definition excludes adjustment costs or disequilibrium.
• The EAPC cannot be used as a durable trade-off or guide for the central
bank’s choice of the inflation rate and its accompanying unemployment
rate. Meaningful policy increases in demand and policy-induced inflation
cannot be random but must be systematic, which, under the EAPC,
cannot change the unemployment rate. Further, any systematic demand
increases or inflation, disguised or misinterpreted as being random ones,
will sooner or later lose their efficacy as their systematic nature becomes
understood. Hence, according to the EAPC, the policy makers cannot rely
upon systematic aggregate demand and inflation increases to reduce the
unemployment rate. Correctly anticipated inflation policies — such as
would be the case if they generated a constant or steadily increasing
inflation rate — generate a vertical EAPC, thereby not providing a
durable negative trade-off between unemployment and inflation. While
such a trade-off might be observed for short runs of data due to

506
 temporary misinterpretations of the nature of inflation, a durable trade-off
will not exist.
• On the impact of monetary and fiscal policies in changing the actual level
of unemployment in the economy, such policies change aggregate
demand which can change structural unemployment and frictional
unemployment, if they lead the economy out of long-run equilibrium.
However, their effect on involuntary unemployment is greater.26
• Empirical evidence has often rejected the stability of the Phillips curve
during inflationary periods. This curve shifts up with an increase in the
experienced rates of inflation, raising the presumption that this curve is
vertical in the limit, thereby providing justification for the concept of the
natural rate of unemployment. However, empirical evidence does not
support the constancy of the natural rate of unemployment. Nor does it
show the inability of expansionary monetary policies to reduce
unemployment under demand deficient conditions.
• These arguments lead to two competing paradigms — i.e., broad
approaches — to the study of unemployment and provide an introduction
to a topic not addressed so far in this book. These paradigms are:
1. The classical paradigm which focuses on fluctuations in the natural
rate itself both over the business cycle and the long run, and uses the
EAPC to explain the deviations of the actual rate of unemployment
from the natural rate. It rules out demand deficiency as a significant
source of the actual unemployment or changes in it. Involuntary
unemployment due to a demand deficiency is not a component of the
actual rate of unemployment in the classical paradigm.27
2. The Keynesian paradigm, which allows — but does not emphasize —
changes in the natural rate, but focuses on the deviations of the actual
rate of unemployment from the natural one due to the EAPC and also
due to fluctuations in the demand for commodities and labor.
According to the Keynesian paradigm, involuntary unemployment
due to a demand deficiency is considered to be a common and
significant component of the actual unemployment rate.
Therefore, the Keynesians argue that the authorities should keep a close
watch on the economy. When there is significant involuntary
unemployment due to a deficiency in aggregate demand, they should use
monetary and/or fiscal policies to increase demand by an appropriate
amount. If they succeed, the economy will eliminate such involuntary
unemployment and perform at its equilibrium level, which is the state
assumed by the classical paradigm. Chapter 11 discusses these paradigms
in detail.

507
 KEY CONCEPTS

Structural unemployment
Frictional unemployment
Natural rate of unemployment
Involuntary unemployment
Voluntary unemployment
Wage contracts
Hysteresis
The Phillips curve
Friedman’s expectations
augmented Phillips curve
Lucas’ expectations augmented
Phillips curve
Costs of unemployment
Costs of inflation
Benefits of inflation

SUMMARY OF CRITICAL CONCLUSIONS

• The Phillips curve plots the relationship between the actual rate of
unemployment and the actual rate of inflation.
• The EAPC shows that unemployment deviates from the natural rate if
there are errors in price expectations.
• Friedman’s version of the EAPC is based on the errors in the price
expectations — of households and firms — embodied in nominal wage
contracts.
• Another version of the EAPC is implied by the Lucas supply rule for
output. This is based on the errors made by firms in their
expectations/perceptions of the prices of their commodities relative to the
price level.
• Unanticipated inflation induced by unanticipated policies can bring about
changes in the deviation of the equilibrium unemployment rate from its
natural level.
• Both anticipated and unanticipated policies can bring about changes in
the deviation of the actual rate of unemployment from its equilibrium
level.
• Involuntary unemployment can be a significant component of the actual

508
 unemployment rate over the business cycle.
• Both anticipated and unanticipated inflation have costs and benefits.
Unanticipated inflation causes greater changes in output, employment,
the real interest rate, and the distribution of wealth.
• The major economic cost of unemployment is the loss of output that
would have been produced if these workers had been employed. There
may also be a long-term cost through the hysteresis effect.

REVIEW AND DISCUSSION QUESTIONS

1. Define the Phillips curve (PC). What is its justification?

2. What are the determinants of the natural rate of unemployment? Show
how changes in their values would change the natural rate.
3. Define the Expectations Augmented Phillips curve (EAPC). What
justification for it arises from Lucas’ contributions?
4. What is the difference between rationally expected inflation and fully
(i.e., accurately) anticipated inflation?
5. Is the EAPC under rational expectations vertical in the long run? Is the
EAPC under rational expectations vertical in the short run?
6. Define involuntary unemployment. What is the justification for it? Is its
existence likely in a recession?
7. What can the government (excluding the central bank) do to reduce
structural unemployment?
8. What can the government (excluding the central bank) do to reduce
frictional unemployment?
9. What can the government (excluding the central bank) do to reduce
involuntary unemployment?
10. What does NAIRU mean? What is its justification? What does it imply
for the position and use of the PC as a policy trade-off between
unemployment and inflation?
11. What costs occur, and to whom, if nominal wages do not increase by
the rate of inflation? Should nominal wages be indexed? Discuss.
12. What is a (a) cold-turkey policy, (b) gradualist policy, of reducing
inflation rates through a restrictive monetary policy? What are the
differences in their effects?
13. Why do governments rarely, if ever, index income tax brackets? What
are the consequences of not doing so (a) for the government budget, (b)
for the economy? What are the consequences of doing so?

509
 ADVANCED AND TECHNICAL QUESTIONS
T1. Suppose that the data on the economy shows that:
Population 22
Civilian population between ages 18 and 65 20
Armed forces 0.5
Unemployed but stopped looking for work 0.2
Employed, civilians, full-time 12
Unemployed 0.8
Employed part-time 0.9
Define and derive:
(i) The labor force.
(ii) The labor force participation rate.
(iii) The employment rate.
(iv) The (measured) unemployment rate; the discouraged worker rate.
(v) If the labor force and those employed both increase by 3%, recalculate
your answers to the preceding questions.
(vi) Starting with the initial data, if the unemployed increase by 5%, what
would become the new unemployment rate?
(vii) Because of the increase in unemployment in (f), some workers
become discouraged. If the discouraged workers increase by 10%, what
would become the labor force and the unemployment rate?
T2. [On involuntary unemployment due to deficient demand.] Suppose
that the economy was initially at full employment with a wage rate w* .
Now assume that aggregate demand falls, so that output produced and
its accompanying employment fall below the full-employment level, but
the wage rate (w 2) rises to equal the MPL (= MPL2). Show
diagrammatically the two ways of measuring ‘involuntary
unemployment’ at the wage rate w 2 under the two definitions of this
term? Explain why both of these can be considered to be appropriate
alternative measures of involuntary unemployment.
T3. [On unemployment due to a high real wage.] Suppose that the
economy was initially at full employment with a real wage rate w *.
Now assume that a shock to the real wage increases the wage rate to w
3, which is above w* . Show diagrammatically the amount of
involuntary unemployment at w3? Can monetary policy reduce such
involuntary unemployment?
T4. Suppose that the economy was initially at full employment with a real

510
 wage rate w*. Now assume that a shock to the real wage increases the
wage rate to w3 , which is above w*. Show diagrammatically the levels
of structural and frictional unemployment at w3? How do they differ
from those at full employment? Does the natural rate of unemployment
change?
T5. Assume that the Phillips curve has the relationship:

where πOil is the rate of increase of oil prices. Draw the two PC curves
(i) for πOil = 0.02 and (ii) for πoii = 0.10.
(a) What is the economy’s trade-off ( u/ n) between unemployment and
inflation?
(b) Calculate (and show diagrammatically) the shift in the Phillips curve
if oil’s inflation rate increases to 1.2, so that πOil1 = 1.20.
(c) Explain why the change in oil’s inflation rate shifts the Phillips curve?
T6. Suppose that the correct relationship is not the Phillips’ curve but the
expectations augmented Phillips’ curve of the form:

where π et is the expected inflation rate for period t . What is the natural
rate of unemployment? Draw the curves for (ut — un) against (πt — πte)
and for ut against (πt — πte).
T7. In the context of an EAPC such that

suppose the inflation rate had become 10% and was expected by the
public to remain at this level.
(a) What monetary policy is needed to decrease the inflation rate to zero?
(b) What would be the impact of this policy on the unemployment rate if
a cold-turkey policy (i.e., immediate reduction to a zero inflation rate,
without a change in the expected one) is followed?
(c) What would be the impact of this policy on the unemployment rate in
the long run once expectations adjust to a zero inflation rate?
(d) Explain, using the PC diagram, the impact of the reduction in the
inflation rate on the unemployment rate in the impact period and in the
long run.
T8. Suppose that in the context of an EAPC such that

with πt = πte = 10%, the central bank in period t + 1 brings the inflation
rate down from 10% to its target inflation rate of 3% by a cold-turkey

511
policy. What would happen to the unemployment rate (a) in the impact
period (specify your relevant assumptions for this period), (b) in the long
run (specify your relevant assumptions for this period)?

1For given levels of these, structural unemployment would be higher at

higher wages.
2The full employment state includes frictional unemployment, which
consists of workers who are qualified for the available jobs but are
unemployed for short periods while they get matched to jobs. This period
is likely to be fairly short so that such workers do not become discouraged
workers — and leave the labor force.
3Union pressure is often proxied by the percentage of workers who are
unionized. This number has been declining in recent decades in Western
countries.
4Some economists claim that one of the reasons for the higher
unemployment rates in Canada and Western Europe compared to those in
the United States are the more generous unemployment insurance benefits
and income support levels in the former.
5It is sometimes claimed that one of the reasons for the higher
unemployment rates in Europe than in the United States is the social
acceptance (or absence of social and family disapproval) of being without
a job.
6Seasonal unemployment affects workers who work in industries with
seasonal production, such as fisheries and construction.
7One part of frictional unemployment is search unemployment, which is
usually defined as the number of unemployed workers who have received
job offers but are choosing to stay unemployed while searching for a job
with a better offer. For such workers, economists use the theory of search
unemployment to explain why they chose to stay unemployed. The
remaining part of frictional unemployment consists of workers who have
not yet received job offers though there are enough jobs fo their skills and
other characteristics.
8As can be seen from Figure 10.1b, an increase in the wage rate (from w *
0
to w 1) can increase or lower structural and/or frictional unemployment
and, therefore, change their sum. However, at any time, the change in the
structural and frictional unemployment and their sum from their natural
levels will only be a small part of their total amounts. This variation is
being ignored as a simplification.

512
9Note that, under these conditions, the employment-wage relationship is
not given by the labor demand curve
10One way of measuring the rate of discouraged workers is given by D/L,
where D is the number of discouraged workers. It would equal [L(w) -
L(wf)/Lf ], where L is the actual labor force and Lf is the labor force that
would occur at the full-employment wage w f. For w <w f, L(w) < L(w f).
11It could also be represented by (ui + d) (i.e., the sum of the involuntary
unemployment rate and the discouraged worker rate). However, the data
on the rate of involuntary unemployment is usually not available or in
dispute.
12Note that involuntary unemployment — even when it exists — is not
included in the natural rate of unemployment.
13Some textbooks simplify by assuming that they always remain at these
levels, so that cyclical unemployment becomes synonymous with
involuntary unemployment.
14It can also change because of changes in the number of discouraged
workers. The discouraged worker rate changes over the business cycle but
is being omitted from the following discussion.
15Another procedure for estimating the natural rate regresses
unemployment on various macroeconomic variables, and proxies the
natural rate by the estimated one.
16This procedure gives erroneous estimates of the natural rate if the cycle
is not symmetrical or if full employment is not reached over the business
cycle, as occurred in the case of cycles within the Great Depression from
1930 to 1939.
17Disequilibrium in the economy due to a ‘high’ real wage or deficient
demand can also change structural unemployment since the change in
aggregate demand and employment is not proportionately shared among
all types of commodities and industries.
18Defining uinv as (n f—n d)/L, i.e., as the difference between full
employment and labor demand divided by the labor force, uinv has two
components, as in:

where (u*—u n) occurs because of the deviation of short-run equilibrium

unemployment along the SRAS curve from unemployment in LR
equilibrium along the LRAS curve. Though the economy is in SR
equilibrium at u *, (u *—u n) can nevertheless be considered to be a part of
involuntary unemployment since the individual worker who becomes

513
unemployed when u * > u n, does want a job at the existing wage rate but
does not have a job. (u—u *) is the disequilibrium element in the
involuntary unemployment rate.
19Both the average level and the distribution of unemployment among
industries together determine the Phillips curve relationship.
20The formal derivation of this curve was presented in Chapter 8 which
discusses expectations and wage contracts.
21This equals (∂ . ∂e).
22To convert Friedmans relationship between n * and P to that between u *
and,π* (u * —u n) is negatively related to (n *—n f ). Therefore, with (n *
—nf ) positively related to (P Pe), (u un) is negatively related to (P — Pe).
In a dynamic context, (P — Pe) is replaced by (π—πe), so that (u * —u n)
becomes negatively related to (π—πe), thereby giving the EAPC
relationship.
23In the long run, both the PC and the EAPC are identical — and vertical
at un —since there are no price misperceptions, adjustment costs, or
disequilibrium.
24This impact occurs because profit-maximizing competitive firms operate
along their marginal cost curves, which are upward sloping.
25In general, this non-accelerating inflation rate is taken to be 2% or less.
26Unanticipated policies may be able to change both (u *—u n) and (u — u
*) by causing unanticipated inflation. Anticipated policies can change (u
—u *) if there was an existing demand deficiency or by causing a demand
deficiency when the expected inflation rate does not adjust instantly.
27However, the classical paradigm does allow involuntary unemployment
due to a wage higher than the equilibrium one brought about by
administrative actions or market imperfections.

514
 CHAPTER 11
Paradigms in Macroeconomics

There are two major paradigms in macroeconomics, the classical

and Keynesian ones. The classical paradigm evolved over the past
two and a half centuries. Its theories focus on the long-run full-
employment) equilibrium of the economy. Its theories are
hierarchical in nature. The currently dominant formulation of this
paradigm implies that any deviations from full employment occur
due to expectational errors and are transitory or self-correcting.
In full-employment equilibrium, changes in aggregate demand,
brought about by systematic monetary and fiscal policies, can only
cause changes in the price level, but not in output or employment.
The Keynesian paradigm has developed since the 1930s. It
differs from the classical one in not focusing on the long-run
equilibrium, (full-employment) state of the economy, but rather on
its actual state, whether an equilibrium or disequilibrium one. Its
theories imply that any deviations from full employment in the
real-world, real-time economy are not always transient and self-
correcting, so that the economy could function at less than full
employment for significant periods. These deviations can arise
from a variety of quite different causes. Therefore, the Keynesian
paradigm has different theories corresponding to the different
causes of the deviations from full employment. These theories
usually imply that an increase in aggregate demand can increase
output and employment, and that appropriate monetary and fiscal
policies can improve on a macroeconomy working at less than full
employment.

There are several paradigms for studying the macroeconomy, as it exists in

most industrialized countries. The fundamental aspect of these economies
is a largely market-based system of production and allocation of factors of
production. Production in most of the sectors of these economies is by
private enterprises, often operating in some type of competition, though
governments have come to play significant roles in the production of

515
goods and services and as employers of labor and capital in the economy
as a whole. Among the existing paradigms for analysis of the
macroeconomy, the dominant ones are the classical and Keynesian ones.
This chapter focuses on these two paradigms and their various component
schools. It neither presents any material on less common schools of
thought, such as Marxism, post-Keynesian, Kaleckian, and Austrian for
studying industrialized economies nor presents any treatment of political
systems such as capitalism, socialism, and communism. A brief
specification of these was given at the end of Chapter 1.
The two major paradigms in economics are the classical and the
Keynesian ones. Their distinguishing elements are:
• The classical paradigm assumes that the market for each of the goods (so
far specified in this book as commodities, money, bonds, labor, and
foreign exchange) in the economy always clears, so that their demand and
the supply will be in equilibrium, which could be a long-run or short-run
one. Since one of the markets is the labor market, its clearance implies
that every worker who wishes to supply labor at the existing wage will
have a job and each firm will be able to employ all the workers that it
wants to at the existing wage. In the absence of rigidities (leading to
expectational errors and adjustment costs), this equilibrium state is that of
long-run equilibrium with full employment. Under uncertainty, errors in
expectations can occur, thereby producing a short-run equilibrium in
which output and employment can differ from the full- employment ones.
Such deviations from full employment are transitory and self-correcting
(see Chapter 8). Therefore, a hallmark of the classical paradigm is that it
tends to focus on the long-run (full-employment) equilibrium of the
economy, with any deviations from it being transitory and self-correcting.
In view of the classical paradigm’s emphasis on labor market clearance,
this implication of long-run equilibrium is often turned around and stated
as if it was an assumption. The literature, therefore, abounds with the
statement that ‘the classical models assume full employment’, which is
strictly not correct.1 If full employment were used as an assumption, it
would rule out the deviations from full employment, which occur under
the Friedman and Lucas supply rules (see Chapter 8), which are part of the
classical paradigm.
In the presence of uncertainty, short-run deviations from this full-
employment equilibrium are allowed due to expectational errors in prices
but the assumption of equilibrium — in this case, a short-run one — is
maintained. These short-run deviations are as specified by the Friedman
and Lucas supply rules, and are taken to be transitory and self-correcting

516
— since expectations will adjust to eliminate expectational errors as soon
as enough information becomes available and contracts are renegotiated.
• The Keynesian paradigm maintains that full employment may exist
sometimes but not always. When it does not, the economy could be in (a)
short-run equilibrium different from the long-run (full-employment) one
or (b) disequilibrium, for significant periods. Further, if it is away from
the full-employment equilibrium, the deviation is not necessarily self-
correcting and/or transient enough to be ignored. Among the major
reasons adduced for the deviation from full employment are the costs of
adjusting prices, employment, and production, errors in the expectations
of future demand and demand deficiency arising from the failure of
markets to reestablish market-clearing prices instantly after a demand
shock.
There are several schools or models within each of the two paradigms,
without a consensus on the proper delineation of either the paradigms or
their component models. This chapter sets out our classifications and
definitions. However, no claim is being made to our taxonomy being
acceptable to all economists.

Stylized Facts on Money, Prices, and Output

The usefulness of different theories can be judged by the stylized empirical
facts about the economy. Among these are:
• Over long periods, there is a roughly one-to-one relationship between the
money supply and inflation.
• Over long periods, there is no significant relationship between inflation
and output growth. This is also so for money growth and output growth,
though some studies show a positive correlation between these variables,
especially for low inflation countries (while others show a negative one).
• Over long periods, changes in interest rates tend to reflect changes in
inflation.
• In the short run, changes in money supply have a strong impact on both
aggregate demand and real output. Unanticipated money supply changes
affect output, as do anticipated ones. Negative shocks to money supply
have a stronger impact on output than positive ones.
• Increases in short-term interest rates lead to a decline in output.
• On monetary policy dynamics in the short run, monetary policy shocks
(i.e., changes in money supply or interest rates) build to a peak impact on
output and then gradually die out, so that there is a ‘hump-shaped pattern’
of the effect of monetary policy on output, with the peak effect occurring

517
 with a lag longer than one year, sometimes two to three years.
• As a corollary of the preceding point, the impact of monetary shocks on
prices and inflation occurs with a longer lag than on output.
• For the short run, since inflation responds more gradually than output to
monetary policy changes, expected inflation also responds more
gradually. Therefore, errors in price or inflation expectations do not
provide a satisfactory explanation of the response of output to monetary
policy shifts.
• Unanticipated price movements explain only a small part of output
variability. The impact of unanticipated monetary shifts on real output
does not occur through prior price increases.
• The responses of output and prices to monetary shocks differ over
different stages of the business cycle. They are also stronger for
contractionary than for expansionary monetary episodes.
• Contractionary monetary policies to reduce inflation do initially reduce
output significantly, often for more than a year.

11.1 An Analogy for the Two Main Paradigms in

Macroeconomics
We will use an analogy between the full-employment state of the economy
and the healthy state of the human body, and that between the deviations
from this full-employment state and the pathology (illnesses and
breakdowns) of the human body. This analogy provides insights into the
study of the economy by reference to personal experiences and knowledge
on the performance of the human body.
We know that the human body sometimes functions in perfect health and
sometimes suffers minor illnesses of a brief expected duration and without
any need for the help of a professional (physician) to get back speedily to
normal health. But it does sometimes suffer from serious illnesses and
breakdowns from which the recovery may occur but be slow and be
speeded up by the help of a physician, or even suffer ones from which
there is no recovery without the intervention of a specialist to return
speedily to normal health. There may also be illnesses from which there is
no cure and no recovery, but we do not include this limiting state within
our analogy. Among the serious illnesses, note there can be many
possibilities: infection by bacterium A rather than B, infection by a
bacterium versus a virus, infection versus collapse of lungs, collapse of
lungs rather than heart attack, etc. The list of the possible sources of the
deviations from the healthy state can be endless.
This analogy of the macroeconomy with the human body leads to the

518
following two fundamental — and highly plausible — axioms on the
performance of the macroeconomy:
a Axiom:
The economy, just as the human body, may sometimes function well and
sometimes not. Hence, it is essential to study both the healthy and the
‘sick’ states, with the former serving as the benchmark for the treatment of
the latter. This axiom implies that economists need to study the
determinants of both the full employment state of the economy and of
deviations (both small and large) from full employment.
β Axiom:
When the economy, just as the human body, is not functioning properly,
the causes, symptoms, and effective treatments of the malfunction can be
quite varied. The justification for the β axiom is that one cannot plausibly
attribute all possible departures from the healthy body to a single
underlying cause or attribute all potential causes to an overarching single
source. This axiom implies that economists cannot assume that each of the
deviations from full employment are due to a single cause and can be
captured by just one model. Different models need to be formulated to deal
with different potential causes of deviations from full employment, and the
model used needs to be tailored to the actual cause of the breakdown of the
economy.

11.1.1 The approach of the classical paradigm to the

pathology of the economy
The focus of the classical paradigm is on the study of the ‘healthy’ state of
the economy, meaning by this the state in which all available factors of
production are used efficiently and are fully employed, so that output in
this state is at its full-employment level. When the classical paradigm does
envisage deviations — e.g., along the SRAS curve with the (Friedman and
Lucas) EAPC due to expectational errors in prices2 — away from this
healthy state, the deviations are supposed to be minor, transitory, and self-
correcting. Under it, while the economic body may become ill, the
illnesses are never serious or long lasting, so that outside intervention (by
the government or the central bank) never becomes necessary or will not
really be worth the hassle and the cost — and/or its success will be
uncertain.
Since the classical paradigm focuses on the single state — the healthy
one — for the economy or self-correcting and transient deviations from it,

519
its various component models tend to have a close relationship among
them. Further, its models have almost a hierarchical nature, with
subsequent models continuing on from the earlier ones, with modifications
representing, in general, further refinements of the assumptions and
analyses.

11.1.2 The approach of the Keynesian paradigm to the

pathology of the economy
The Keynesian paradigm focuses on the pathology of the economy. An
implication of the β axiom is that this pathology cannot be properly
encapsulated within one model with one root pathogen. Hence, the
Keynesian paradigm, if it is to do its job properly, has to be a disparate
and, at best, a rather loose collection of models: the deviations from the
full-employment (healthy) state of the economy state can be due to
different pathogens or breakdowns of the different components of the
economy. Further, the Keynesian paradigm envisages the possibility of
serious departures from the full-employment state. Furthermore, it allows
for the possibility that the recovery may sometimes be slow and could be
speeded up with professional help (from the government and the central
bank). It also has to allow for the possibility that, in extreme and rare
instances, the recovery may never occur without such help.
To reiterate, by the nature of their attempts to deal with the pathology of
the economy, the Keynesian models have to be, and are, quite varied. If
they are to do their job properly of dealing with the different types of
deviations, such models need not — in fact, must not — all focus on the
same types of deviations from the overall equilibrium state or make the
same recommendations for policies to address the impact of very different
sources of deviations. Unfortunately, this aspect of the Keynesian
paradigm is often not recognized. Frequently, the presentations and
discussions of the Keynesian models miss this requirement for variety
within the Keynesian paradigm and seek to force the various Keynesian
models into a single format or view it as one unified model. The danger in
doing so is that a single policy prescription could be given as a cure-all for
very disparate causes and be inappropriate for many.3
Extended Analysis Box 11.1: The Fundamental Assumptions of the
Classical Paradigm
The classical paradigm is based on six fundamental assumptions. It
analyzes equilibrium levels of output and employment along the LRAS

520
curve for the long run (see Chapter 7) and along the SRAS curve, arising
from expectational errors (see Chapter 8), for the short run.
• No money illusion in demand and supply functions
Changes in the price level or the rate of inflation do not change the
demands and supplies of goods in real terms. In particular, the
demand and supply of labor depend on the real and not the nominal
wage.
• Flexible prices and wages
The prices of all the goods in the economy are assumed to be flexible
and adjust to equate demand and supply in the relevant market. They
increase if there is excess demand and decrease if there is excess
supply. These prices include wages, which is the price of labor.
• Continuous market clearance (market coordination)
Each market clears continuously, so that we can focus on the study of
the general equilibrium in the economy and its properties, while
largely ignoring the disequilibrium values of the variables.
Continuous market clearance requires some mechanism for bringing
about the continuous equality of demand and supply. One such
mechanism is that of centralized coordination in a bureaucratic
system;4 another is that of markets in a free enterprise system. The
mechanism for the latter is often referred to as the ‘invisible hand’ of
competition. [However, it is highly questionable whether either of
these mechanisms can, or ever delivered, the continuous equality of
demand and supply in all the markets of any real-world economy.]
• Transparency of equilibrium prices
All agents in making their demand and supply decisions assume that
such market clearance will occur instantly after any disturbance and
know/anticipate (or are informed by a ‘market coordinator’) the prices
at which it will occur.5
• Notional demand and supply functions
All economic agents assume that they will be able to buy or sell as
much as they want to at the equilibrium prices. They plan to produce,
consume, demand money, and supply labor at only these equilibrium
prices. The demand and supply functions derived under this
assumption are known as notional demand and supply functions. In
particular:
• Workers act on the assumption that they will be able to sell all the
labor that they want — that is, there will be full employment — so
that they can incur consumption expenditures based on full-

521
 employment income. [However, this is assumed to be so even if
actual employment falls below full employment and the unemployed
workers do not receive any income.]
• Firms must act as if there will be enough demand to sell all the output
that they want to sell, so that they produce at the full-employment
level. [However, this is assumed to be so even if there is not enough
demand for the full-employment output, though production by firms
at this level would mean an unplanned accumulation of inventories.]
• Under uncertainty and errors in price expectations, short-run
equilibrium analysis for the SRAS curve:
The short-run analysis of the classical paradigm is based on price
misperceptions under uncertainty, as presented in Chapter 8 for the
Friedman and Lucas supply curves. Otherwise, the preceding
assumptions are maintained. In particular, households maximize
utility and firms maximize profits, and markets clear on a continuous
basis.
Note that each of the above fundamental assumptions is related to the
others but is still distinct. Any one or more of these assumptions may
not be relevant to and valid for a particular stage of the business cycle or
at a particular time in a real-world economy.

11.2 Defining and Demarcating the Models of the Classical

Paradigm
All the models or schools within the classical paradigm share the common
beliefs that the real-world economy under consideration—and not just the
models—functions at full employment in the long run and that it does not
stay very long out of the full-employment state, so that it never gets stuck
in an under-full employment disequilibrium or equilibrium. Therefore,
states with less than full employment are either transitory states of
disequilibrium or short-run equilibrium during which the economy
continues to adjust fairly rapidly toward its full-employment equilibrium—
and not away from it. Further, one of the characteristics of the long-run
equilibrium for all the classical schools is the independence of the real
variables from the financial ones, so that money is neutral in the long-run
equilibrium.6 A major difference among the classical schools is whether
the real-world economy adjusts so fast as to maintain virtually continuous
equilibrium (so that it will not show any evidence of disequilibrium) or
whether there can exist transitory disequilibrium states for some duration.7

522
It refinements over time have, in general, tended to exclude disequilibrium
states, but allow short- run equilibrium states which deviate from the long-
run full-employment equilibrium state. However, these deviations, as
shown by the Friedman and Lucas supply rules in Chapter 8, arise not
from the failures of markets to clear but from the errors in the expectations
held by economic agents on market prices (not on sales and job prospects).
These errors in expectations are transitory and self-correcting, so that the
deviations of the short-run equilibrium from the long-run one are also
transitory and self-correcting.
The following classification of the models of the classical paradigm has
been chosen for reasons of clarity in separating each model from the
others, rather than leaving their differences ambiguous. Our classification
attempts to maintain consistency with the writings and folklore in the
history of economic thought.

11.2.1 The traditional classical approach

’The traditional classical approach’ consists of the somewhat disparate
ideas on the macro structure of the economy from the origin of
macroeconomics in the middle of the 18th century to the publication of
Keynes’ The General Theory in 1936. To quite a considerable extent, these
ideas were diffuse, varied among authors and changed over time. In any
case, there was no single compact version of the overall exposition, though
we do now treat them as if there had been a compact model.
The common themes of the traditional classical approach were:
• A microeconomic approach to economics, especially after the 1870s,
with market clearance and long-run equilibrium (see Chapter 7 for the
underlying assumptions of this approach).
• Loanable funds (or bond market) theory of the rate of interest, under
which the long-run rate of interest in a closed economy was determined
solely by full-employment national saving and investment (see Chapter
7).
• The quantity theory for the determination of prices (see Chapter 2). It
applied in the full-employment equilibrium of the economy. But outside
this equilibrium, during the adjustment phase, changes in the money
supply could change output and employment.
• Most of the traditional classical economists did not believe that the
economy functioned so well that it always maintained full employment or
that it did so most of the time. In fact, recessions and crises—any of them
originating in the banking sector or financial speculation or occurring due
to the response pattern of the financial sector to real shocks—were

523
 common during the 19th century. The widely held belief among
economists at that time was that such crises did affect output and
employment. Hence, the traditional classical school allowed for the
existence of disequilibrium to exist for significant periods. That is, they
did not assume continuous full employment. They also recognized that
changes in the money supply do cause changes in employment and output
over the short term and over the business cycle.
• Accompanying the economic analysis was a general conservative
approach to politics and public finances, which espoused (a) a small size
of government and (b) ‘fiscal responsibility’ by the government. Fiscal
responsibility meant that the government should not spend more than its
revenues, especially in peacetime, so that it should aim for a balanced
budget.8
The Great Depression of the 1930s with its massive unemployment over a
decade destroyed the public’s and economists’ faith in the traditional
classical approach and especially in its prediction that the economy will
normally be at or close to full employment. Its doctrines and predictions
were clearly remote from the public’s own economic experience. Further,
economic policies based on it tended to worsen the economy. Their failure
prepared the public and the economics profession for the acceptance of an
alternative paradigm that would explain the failures of the economy to
maintain full employment, especially in response to serious declines in
aggregate demand, and to provide more appropriate policy prescriptions
during these failures. The theory of why the economy can be away from
full employment, and the relevant policy prescriptions, were provided by
the Keynesian paradigm, which originated in 1936 in the midst of the
Great Depression.
Extended Analysis Box 11.2: The Founders of the Classical Tradition in
Macroeconomics

DAVID HUME (1711-1776)

Hume was a philosopher who made several contributions to economics.
These included the quantity theory of money, the loanable funds theory
of interest, and the benefits to the nation from international trade. On the
quantity theory, presented in Chapter 2, Hume argued that the long-run
effect of changes in the money supply was to proportionately increase
the price level. Gold coins were the main form of money in Hume’s
time, so that the changes in the amount of gold in the nation were
synonymous with changes in the money supply.

524
 On the loanable funds theory, Hume argued that the long-run rate of
interest was determined by the equality of national saving and
investment. Hume’s observation on the saving behavior of the three
main classes of society, the aristocrats, the merchants, and the peasants
(including the workers) was that the aristocrats and peasants spent all
their incomes on current consumption (including wasteful expenditures)
and did not save while the merchants tended to save any increases in
income and to invest them, often in their own small businesses. Since
increases in the money supply in the hands of the merchant classes
supplement saving and are lent, their short-run effect was to lower the
interest rate and increase national economic activity—a code phrase
meaning national output and employment in the modern terminology.
This short run could last several years since prices tended to change
rarely—be sticky. However, once the additional money supply was fully
integrated into the economy, both the interest rate and economic activity
would return to their long-run levels, but the prices would have
increased proportionately.
Therefore, national prosperity flowed from the amount of capital, the
workforce and its diligence, rather than the amount of gold and money
in the nation. This prosperity increased with trade with other nations.
Such trade helped poorer nations to export commodities, as well as learn
the technology of other nations, which promoted growth and
development. It also benefited richer nations by increasing their exports.
ADAM SMITH (1723–1790)
Although many economists, including David Hume, had earlier written
on economic matters, Adam Smith is widely called the ‘father’ of
economics because his analysis of the benefits from capitalism, greater
competition and free trade had very considerable impact on the
formulation of economic policy in Britain and other countries. His most
influential work was The Wealth of Nations (1776), which assumed that
people act according to their own self-interest—which is the
fundamental postulate of modern economic analysis. Competition
organized the individuals’ self-interest into exchanges that were of
benefit to both parties to the exchange and ensured the maximization of
national welfare. The pursuit of self-interest promoted national interest,
as long as competitive forces were allowed. As a corollary, state
interference in economic matters through regulation decreased national
welfare and impoverished the nation. The role of the government was to
be limited to providing public goods, such as defence, legal, and police
systems, and the provision of goods with large externalities—i.e., effects

525
 on other than the producer and/or the consumer.
To Smith, specialization in production and the division of labor
promoted greater production and growth. The larger the market, the
greater the scope for them. Therefore, the lesser the barriers to trade
within the country and among countries, the more prosperous the nation.
His famous example of this division of labor was that of production in a
pin-making factory. It showed how the division of the tasks among
workers permitted greater production than if each worker were to
produce the whole pin.
DAVID RICARDO (1772–1823)
Ricardo provided the first rigorous analysis of several topics in
economics. Among these were: the theory of distribution of income into
rents, wages, and profits; the theory of comparative advantage according
to which nations benefit by specializing in the production of items in
which they have a comparative advantage; the theory of value or relative
prices, according to which the relative prices of commodities would
depend on their relative cost of production in the long run and their
scarcity in the short run.
Ricardo was a supporter of the quantity theory of money, proposed by
Hume, under which changes in the money supply cause proportionate
changes in the price level.
At the level of the economy as a whole, Ricardo maintained that there
could not be general excess supply (i.e., output > aggregate demand) in
the economy.9 That is, the economy will always have enough aggregate
demand for the purchase of the economy’s output at full employment.
This proposition was later designated as Say’s Law, whose compact
statement is that ‘(full-employment) output creates its own demand’.
This proposition amounts to the denial of the empirical possibility of
deficient demand and involuntary unemployment, whose analysis was
presented in Chapter 9. While this proposition is consistent with the
long-run equilibrium of the economy, it does not hold when there is
disequilibrium, especially in the commodity and labor markets.

11.2.2 The neoclasical model

This name was given to the restatement of the traditional classical ideas
rebottled and reflavored in the post- General Theory period in a new
compact framework. The new bottle was the IS-LM technique or mode of
analysis, first proposed in 1937 by John Hicks to explain the arguments of

526
Keynes’ The General Theory. In some ways, this rebottling represented a
radical departure from the traditional presentation of the classical ideas, in
that the latter did not have an analysis of the commodity market
equilibrium while the IS-LM technique starts with a presentation of the
analysis of the commodity market. Note that the IS-LM and the IS-IRT
analyses of the commodity market embody the concept of the multiplier,
which was unknown to the traditional classical economists and was not
developed until the early 1930s. Chapters 4 to 6 incorporate the multiplier
analysis.
Further, certain elements of the traditional ideas such as the Quantity
Theory were discarded in the rebottling process and the analysis of the
money market was incorporated through the LM curve—again a concept
not familiar to the traditional classical school.
The neoclassical model continued to have both equilibrium and
disequilibrium aspects, and did not assume instant market clearance. In
this, it represents the ideas of the traditional (pre-Keynesian) classical
economists more faithfully than does the modern classical model.10
The classical paradigm—and its traditional classical and neoclassical
versions—was rejected by the majority in the economics profession from
the 1940s to the 1970s, though it continued to exist as an outcast.
However, refinements and additions to it continued to be made during
these decades. The dominant paradigm in these decades was the Keynesian
one. Following the Keynesian policy debacle of the mid-1970s that
erroneously addressed supply-side shocks (oil price increases) by
expansionary monetary and fiscal policies and led to stagflation, the
classical paradigm roared back in the 1970s and has since then taken
various forms. These are the 1970s Monetarism, the modern classical
model and the new classical model.
Box 11.1: Some Major Misconceptions about Traditional Classical and
Neoclassical Approaches in the Classical Paradigm
A common misconception nowadays is that the traditional classical and
neoclassical economists believed that the economy functioned well
enough to maintain full employment most of the time or that it had a fast
tendency to return to full employment following a disturbance and a
decline in employment. In fact, many traditional classical economists
believed that “the economic system is essentially unstable” (Patinkin,
1969, p. 50).11 Another misconception nowadays is that the traditional
classical and neoclassical economists believed that money was neutral in
practice and in theory for the real world over the short term.

527
 It was generally recognized that “cycles and depressions [are] an
inherent feature of ‘capitalism’. Such a system must use money, so that
the changes in the money supply and in its velocity were major sources
of business fluctuations.” Further, many traditional classical economists
believed that these fluctuations were heightened by the ‘perverse’
behavior of the banking system, which expands credit in booms and
contracts it in depressions. Among the reasons given for the real effects
of changes in the money supply and velocity changes was that firms’
costs have a tendency to move more slowly than do the more flexible
selling prices. In view of this, monetary policy was often envisaged and
recommended as a stabilization tool.
However, fiscal policy for the stabilization of aggregate demand was
usually not even considered a possibility nor was its analytical basis
known to the traditional classical (pre-Keynesian) economists. Its
consideration and recommendation as a major stabilization tool was due
to Keynes and the Keynesians, and needed as its basis the concept of the
investment multiplier, first derived in the 1930s. As a counter-
reformation, Barro’s Ricardian equivalence theorem (which implies that
aggregate demand cannot be changed by fiscal deficits and surpluses)
was proposed to again remove fiscal policy from the set of potential
demand management tools. This theorem is not presented in this book
but can be found in more advanced macroeconomics textbooks.

11.2.3 The 1970s monetarism

Monetarism and monetarists have been defined in a variety of ways. In a
very broad sense, monetarism is the proposition that money matters in the
economy.12 In a narrow sense, monetarism was associated with the St.
Louis School (named after the Federal Reserve Bank of St. Louis, whose
researchers popularized monetarism) in monetary and macroeconomics.
We will define monetarism in this narrow sense. This version of
monetarism is called the 1970s or the St. Louis School Monetarism. Its
ideas became quite popular in the late 1970s and early 1980s. It was
essentially a simplification of the neoclassical model for empirical tests of
the impact of money supply changes and fiscal deficits on aggregate
demand and on real output.13 Much of the work was statistical and
econometric in nature and provided a simple empirical procedure for
estimating the relationship between nominal income, money supply, and
fiscal variables.
The empirical and theoretical bases of the 1970s Monetarism was

528
provided by the economists at the Federal Reserve Bank of St. Louis. The
main tenets of this school were that:
• The short-run versions of their model neither assumed full employment
nor implied continuous full employment in the economy. Therefore, it
allowed for the possibility of involuntary unemployment in the short run
and supported the positive short-run impact of monetary policy on
nominal income, real output, and employment. In this, it was relatively
close to the then-Keynesian models. However, it denied on empirical
grounds the Keynesian claim of the efficacy of fiscal policy.
• In its long-run version, it belonged to the classical paradigm. In the long
run, changes in the money supply could not change full-employment
output but would cause proportionate changes in the price level.
• Among its empirical findings for the USA were:
• The lags in the impact of money supply changes on nominal income
were fairly short — about five quarters.
• Fiscal policy had a small positive impact on nominal expenditures
followed by a negative one, roughly canceling out over about five
quarters.
• This school also recommended that the central bank should aim at
controling and targeting one of the monetary aggregates rather than
interest rates.
• In general, the monetarists tended to be more conservative than the
Keynesians in their general political and economic philosophy. Along
with a recommendation that fiscal policy not be used as a stabilization
tool, they tended to favor lower taxes and smaller government.
• Another aspect of the general ideas of the Monetarist school was that the
central bank and the government neither possess an inherent superiority
of information on the economy nor do they always manage to improve on
the performance of the economy: their actions have worsened it at times.
In these beliefs, the monetarists agreed with Milton Friedman. Their
difference from Friedman was on an empirical issue: the duration of the
lag in the impact of money supply changes on nominal income.
Friedman’s estimates showed that this lag was long (6 quarters to 21
quarters) and variable, so that it was dangerous to pursue monetary
policy. The monetarists’ estimates showed that this lag was relatively
short (five quarters), so that monetary policy could be followed with
predictable effects.
In many ways, the 1970s monetarism was a hybrid between the
neoclassical model and the Keynesian paradigm, and made the switch
away from Keynesianism palatable for many economists. However, it had

529
a short life and was replaced in the early 1980s by ideas truer to the
classical paradigm, which eventually took the form of the modern classical
paradigm.

11.2.4 The modern classical model

This model is a statement of the classical paradigm under the assumptions
of continuous labor market clearance and rational expectations. Because of
the former assumption, it incorporates continuous full employment, except
in the short run when the economy operates along the SRAS curve—which
occurs due to errors in price expectations (including asymmetric
information among economic agents) in the Friedman and Lucas supply
rules. Given full employment as the long-run equilibrium state, rational
expectations imply that systematic monetary policy will not be able to
change real output and unemployment in the economy, though random
changes in the money supply can affect them. Further, for unanticipated
changes in the money supply and interest rates brought about by monetary
policy, the economy deviates from full employment but reverts to it soon.
The modern classical model differs from the neoclassical one since the
former assumes continuous labor market clearance, while the latter allows
the possibility of disequilibrium in labor markets. Further, the former
allows for uncertainty — addressed through the rational expectations
hypothesis — while the neoclassical one does not.
Hence, the constituents of the modern classical macroeconomic model
are:
• An emphasis on the microeconomic basis of macroeconomics,
maximization of profits by firms, maximization of utility by households,
and general equilibrium.
• The neoclassical model, modified by the additions of:
• Perfect competition and continuous market clearance (especially of the
labor market, so that there will not be any involuntary unemployment).
This requires that markets are efficient in establishing the price level
that equates aggregate demand and supply.
• Rational expectations hypothesis when dealing with uncertainty.
• Friedman and Lucas supply rules when there is uncertainty about
prices and inflation.
Intuitively, the modern classical model focuses on the economy in good
health (i.e., functioning at peak performance). Deviations from this state
can occur along the SRAS curve and are analyzed using the Friedman and
Lucas supply rules. These deviations are self-correcting and transient (i.e.,

530
short-lived). The full-employment state serves as the benchmark of what
the economy can do at its best. It implies that there is no need for
systematic governmental policies, either monetary or fiscal ones, to try to
improve on this level of performance. In fact, the economists of this school
point to the potential dangers of trying to do so.
Between about 1980 and 2006, the modern classical school seemed to be
the dominant one in macroeconomic theory but not necessarily among
policymakers.14

11.2.5 The new classical model

The new classical model imposes the assumption of Ricardian equivalence
on the modern classical model. Intuitively, Ricardian equivalence is an
aspect of intertemporal rationality and the extreme democratic notion that
the government is nothing more than a representative of its electorate and
is regarded as such by the public in making the decisions on its own
consumption. This government provides just the goods that the population
wants and its bonds, held by the public, are regarded by it (the public) as a
debt owed by the public to itself. For bond-financed deficits, the
implications of Ricardian equivalence are that, in order to provide for the
future tax liability15 imposed by a bond issue, the public increases private
savings by the amount of the deficit. In doing so, private consumption is
reduced by the amount of the deficit. Since the deficit increases aggregate
demand while the decrease in consumption cuts it, the two effects cancel
each other, so that the bond-financed deficit does not change aggregate
demand. Therefore, deficits do not even change nominal national income
and interest rates, let alone the real value of the economy’s output.
The constituents of the new classical model are:
• The modern classical model, modified by the addition of,
• Ricardian equivalence.
The only difference between the implications of the modern classical and
new classical models is that the former implies that fiscal deficits will
change aggregate demand (shift the IS curve), national saving, and interest
rate while the latter implies that they will not do so. Their other
implications are identical. Among these are:
• Both models do possess money supply changes as a policy tool for
changing aggregate demand in the economy.
• Both models imply the neutrality of systematic (which gets to be
anticipated) money supply changes in the full-employment state, so that

531
 the impact of the systematic money supply and velocity changes can only
be on the price level and not on real output and employment.
• Both models allow for the short-run deviations from full employment due
to expectational errors in prices when the money supply changes are
unanticipated.
• Both models imply that the deviations from full employment are at most
minor, transitory, and self-correcting, so that there is a strong tendency
for the economy to revert to full employment within a relatively short
period after any shock. Hence, there is neither need nor scope for
systematic monetary policy for changing the levels of output and
employment in the economy, so that such policies should not be pursued.
Extended Analysis Box 11.3: The Founders of the Classical Approach in
the Modern Period

MILTON FRIEDMAN (1912–2006)

We present the ideas of Milton Friedman in sufficient detail elsewhere,
so that we add little to them here. We have already studied the Friedman
supply rule (Chapter 8) and its associated expectations augmented
Phillips curve (Chapter 10).
Friedman was the foremost defender ofclassical ideas—at that time
called the ‘neoclassical approach’—during the Keynesian ascendancy
from the mid-1930s to the mid-1970s and contributed extensively to
updating the theories inherited from the traditional classical approach.
He espoused an emphasis on the long-run equilibrium analysis, which
implies the natural rate of unemployment and full-employment output.
However, he also accepted the possibility that unemployment could be
higher than its natural rate and that money supply increases could lower
it. Friedman tried to modernize the quantity theory.
Friedman was, however, sceptical about the discretionary use of
monetary policy since he believed that there were long and variables
lags in its impact, so that monetary policy could in effect, end up
worsening the economy as often, or more often, rather than improving it.
He recommended following a set policy rule that would increase the
money supply at a low, constant rate. He did not believe that fiscal
policy was an effective policy tool.
Friedman also argued for competition in the provision of services in
the various sectors of the economy, including education, health, etc, and
was opposed to government regulation.
ROBERT LUCAS (1937–)

532
Robert Lucas is the most prominent contributor to the modern classical
modeling of macroeconomics and is currently its best-known exponent.
His approach is to base macroeconomics on a microeconomic
foundation. Economic agents—households and firms—are rational
(utility and profit-maximizing subject to the relevant constraints) and
hold rational expectations. Markets clear. Since this tends to maintain
full employment and full-employment output, economic management of
the economy by the government and the central bank is not needed. We
have already studied some of Lucas’ ideas on the Lucas supply curve
(see Chapter 8). Lucas has also contributed extensively to other areas of
economics, including endogenous growth theory (presented in Chapter
15).
Note that the basing of macroeconomics purely on microeconomics
means that the former is just an extension of the latter. This approach
has led some economists to argue that macroeconomics is not a distinct
field within economics; some have even argued that there is no such
subject as macroeconomics. Milton Friedman occupies a special place in
the counter-reformation from Keynesian economics to the neoclassical
and modern classical ones. His major contributions from the 1940s to
the 1970s laid the basis for challenging the then current versions and
assumptions of the dominant Keynesian school. This school in the 1940s
and 1950s had downplayed the impact of money supply changes on the
economy and had a relative preference for fiscal over monetary policy.
In the 1950s, Friedman showed through his theoretical and empirical
contributions that ‘money matters’ — that is, changes in the money
supply change nominal national expenditures and income — as against
the then general view of the Keynesian school that changes in the money
supply brought about through monetary policy did not significantly
affect the economy or did so unpredictably.16 On the latter, Friedman
argued and tried to establish through empirical studies that the money-
income multiplier was more stable than the investment-income
multiplier, so that monetary policy was not less important or less
predictable than fiscal policy in its impact on nominal national income.
Other aspects of Friedman’s agenda to reestablish the doctrine that
money matters were to set out in the 1950s and 1960s the theory — and
to establish it empirically — that the demand function for money was
stable, with the result that the velocity of money also had a stable
function. Therefore, changes in the money supply had a strong impact
on aggregate demand. These arguments were accepted by the profession

Extended Analysis Box 11.4: The Economic Contributions of Milton

533
 Friedman
by the early 1960s, and contributed to the conversion of the Keynesian
macroeconomics to the Keynesian- neoclassical synthesis17 on
aggregate demand expressed by the IS-LM model for the determination
of aggregate demand.
On the relationship between the nominal variables and the real side of
the economy, Keynesians in the late 1950s and 1960s had relied on the
Phillips’ curve, which showed a negative trade-off between the rate of
inflation and the rate of unemployment. Friedman argued that the natural
rate of unemployment — and, therefore, full-employment output — was
independent of the anticipated rate of inflation, so that the fluctuations in
output and the rate of unemployment were related to the deviations in
the inflation rate from its anticipated level. This relationship came to be
known as the expectations augmented Phillips’ curve and incorporated
his contributions on the natural rate of unemployment.
While Friedman brought the role of anticipations on the rate of
inflation into discussions on the role and effectiveness of monetary
policy in the economy, he did not have the theory of rational
expectations. The rational expectations hypothesis had not yet entered
the literature and Friedman had relied on adaptive expectations (see
Chapter 8) in his empirical studies. Consequently, for Friedman, the
unanticipated rate of inflation — equal to the errors in inflationary
expectations — was not random with a zero expected value. Further,
since Barro’s contribution on the Ricardian equivalence theorem had not
yet been made, Friedman’s writings also did not incorporate this
hypothesis. Hence, Friedman was a precursor of the modern classical
school but not fully a member of it. Nor does this school follow all of
his ideas.18

11.2.6 Milton Friedman and the Keynesians

Milton Friedman was closer to the Keynesians in one important respect
than to the later modern classical school. He believed, as did the
Keynesians, that the economy does not always maintain full employment
and full-employment output — and does not always function at the natural
rate of unemployment, even though this concept was central to his
analysis. Hence, policy induced changes in aggregate demand could
induce short-term changes in output and employment. Therefore, money
mattered even to the extent that changes in it could induce changes in
employment and output, depending upon the particular stage of the

534
business cycle. While this view was shared with the Keynesians, Friedman
tilted against the Keynesians on the pursuit of discretionary monetary
policy as a stabilisation tool — especially for ‘fine tuning’ the economy —
because of his belief that the impact of money supply changes on nominal
income had a long and variable lag. He reported on the outside lag of
monetary policy that:

The rate of change of the money supply shows well-marked cycles

that match closely those in economic activity in general and
precede the latter by a long interval. On the average, the rate of
change of the money supply has reached its peak nearly 16 months
before the peak in general business and has reached its trough over
…12 months before the trough in general business …. Moreover,
the timing varies considerably from cycle to cycle — since 1907
the shortest time span by which the money peak preceded the
business peak was 13 months, the longest 24 months; the
corresponding range at troughs is 5 months to 21 months.
(Friedman, 1958)
With such a long and variable lag from changes in the money supply to
nominal national income, the monetary authorities cannot be sure when a
policy induced increase in the money supply would have its impact on the
economy. Such an increase in a recession may, in fact, increase aggregate
demand in a following boom, thereby only increasing the rate of inflation
at that time. Consequently, Friedman argued that discretionary monetary
policy, intended to stabilize the economy, could turn out to be
destabilizing. Friedman’s recommendation on monetary policy was,
therefore, that it should maintain a low constant rate of growth.

11.2.7 The relationship between the Monetarists and

Friedman
As we have explained earlier, the St. Louis Monetarism was based in large
part on Friedman’s ideas. The St. Louis Monetarists believed, as did
Friedman and the Keynesians, that changes in the money supply have a
significant and positive impact on real output, and prices, in conditions of
less than full employment. But, for the Monetarists, as with Friedman but
not the Keynesians, the economy tended to full employment in the long
run, so that the long-run impact of money supply changes would only be
on the price level.
However, the estimations of the St. Louis equation indicated a much
shorter and more reliable lag than Friedman had found. Therefore, contrary

535
to Friedman’s recommendations and consistent with Keynesian ones,
Monetarism was consistent with the stance that monetary policy could be
reliably useful for short-term stabilization of the economy.
On the transmission mechanism, the St. Louis Monetarists supported the
direct transmission mechanism—from the money supply to expenditures,
rather than indirectly from the money supply to interest rates, then to affect
investment, which changes aggregate expenditures—espoused by
Friedman. However, their estimation techniques were not fine enough to
separate the direct from the indirect channel of monetary policy effects.19
The St. Louis Monetarism represented a transitional stage in the
transition from Keynesian ascendancy in economics in the decades before
1970 to the ascendancy of the neoclassical and modern classical schools in
the 1980s and 1990s. In many ways, it was an amalgam of Keynesian and
Friedman’s ideas in macroeconomics, and led the way to the reemergence
of the classical doctrines.

11.2.8 Friedman and the modern classical school

While both Friedman and the modern classical economists are opposed to
the pursuit of discretionary monetary policy, they arrive at this position for
quite different reasons. For Friedman, money supply changes can change
output and employment but the long and variable lags in this impact make
a discretionary policy inadvisable. It could make the economy worse rather
than better. For the modern classical economists, the economy maintains
full employment, so that systematic policy changes in the money supply
cannot change output and employment, but only change the price level.
This joint stance against the pursuit of monetary policy hides a subtle
difference. For Friedman, money matters for the real sectors of the
economy, and changes in the money supply can alter output but are not
advisable because of long and variable lags. For the modern classical
economist, there are no lags in the impact of systematic money supply
changes, and there is no impact on output, so that from the perspective of
output and employment, it really does not matter what the money supply
is.
On the transmission mechanism from money supply changes to income
changes, Friedman supported Fisher’s direct transmission mechanism —
from money supply changes directly to expenditures changes—over the
indirect one, which is from the money supply to interest rates to
investment, as in Keynesian and IS-LM models. Friedman had, therefore,
recommended that the central bank should target monetary aggregates
rather than interest rates in its pursuit of monetary policy. However,

536
neoclassical and modern classical models espouse the indirect transmission
mechanism, just as the Keynesians do — rather Friedman’s direct one so,
than that they recommend, as do the Keynesians, that the central bank
should target interest rates rather than monetary aggregates.

11.3 The Keynesian Paradigm

The Keynesian paradigm studies the pathology of the macroeconomy. It
focuses on the causes, implications and policy prescriptions for the
potential deviations of the economy from a full-employment equilibrium.
Since there can be many causes of such deviations, their appropriate study
requires not one unified model but many, not all of which need be
variations on a theme or even compatible with one other. This chapter
provides a sample of their diversity. The original version of the Keynesian
paradigm was Keynes’ own ideas as set out in his book The General
Theory, followed by a number of evolving versions of the Keynesian
framework.20 Keynesianism has been evolving over time, so that there is
no single version of it.
The common strand in the Keynesian approaches is the anxiety over the
performance of the economy, and especially of the labor market. In
particular, the Keynesian models assume that:
• Even fairly or fully competitive markets in the real-world modern
economy are not fully efficient: they do not continuously bring about the
price level that would produce continuous market clearance.
Alternatively stated, even if there exists a high level of competition, the
‘invisible hand of competition’ is not efficient and fast enough to ensure
continuous and simultaneous equilibrium in all markets. There is also no
other coordinating mechanism for achieving this. This is often stated as
‘market failure’21 or ‘coordination failure’.22
• Given the sluggish response of markets relative to that of firms in
establishing the price that would equate supply and demand, firms make
and act on their profit-maximizing decisions on production and
employment in response to changes in the demand for their products.
These decisions include changes in their output, as well as changes in
their prices.
• The labor markets do not perform well enough to ensure full employment
as a continuous or almost continuous state for a variety of reasons. One of
the basic reasons for this is that the labor markets cannot be taken to
always clear at the current wage rate. They may do so, but not necessarily
always, nor is there always a strong and rapid tendency for them to move

537
 from a disequilibrium state to the equilibrium one. The latter is related to
the observation that the firms and the workers in the labor market do not
even negotiate a real wage but rather negotiate and set a nominal wage.
Further, there are very significant imperfections in the labor market for
various reasons, including the fixity and firm-specificity of acquired
skills, the differing geographical distributions of jobs and workers,
explicit and implicit contracts, etc.
• Information on the future is often incomplete and vague. This is a source
of considerable fluctuations in business and consumer confidence, whose
impact on investment, consumption, money demand, and bond purchases
and sales cause fluctuations in economic activity.
• There are various types of market imperfections and departures from
perfect competition in the economy.
Extended Analysis Box 11.5: The Founders of the Keynesian Paradigm

KNUT WICKSELL (1851–1926)

Wicksell made contributions in both microeconomics and monetary
economics. In monetary economics, Wicksell shifted the focus of
monetary theory from an emphasis on the quantity of money to that on
interest rates. His most famous analysis on this topic was based on a
pure credit economy, which does not use currency but only uses
checking accounts with the banks for all payments.
Wicksell argued that banks set the market rate of interest while saving
at full employment and the marginal productivity of capital determine
the natural rate of interest. If the market interest rate is lower than the
natural rate, firms can increase their profits by borrowing more from the
banks and investing these funds. This increase in investment increases
employment and income in the short run. The money supply also
increases in the process, but this increase is a consequence of the
divergence between the market and natural interest rates, not the cause
of it. This process continues until the banks raise their interest rate to
equal the natural rate.
Conversely, if the market interest rate set by banks is above the natural
rate, firms will reduce their borrowing, with the result that investment
and economic activity will fall. The money supply will also fall.
Economic activity thus depends on the market interest rate and
investment, so that unemployment will also depend on them. This
analysis was quite different from that of the quantity theory in several
ways: its primary emphasis was on interest rates and investment, and
output and employment could be altered by changes in the interest rate

538
 charged by banks. It is more in tune with the modern banking system
and macroeconomic analysis, and served as a prelude to the foundation
of modern macroeconomics outlined by Keynes.
JOHN MAYNARD KEYNES (1883–1946)
If Adam Smith is designated as the father of economics, Keynes
deserves to be regarded as the father of modern macroeconomic
analysis. There was no distinct field known as macroeconomics prior to
the publication of Keynes’ The General Theory of Employment, Interest
and Money (known as The General Theory) in 1936. Keynes was a well-
established economist and had published several books by that time.
These represented expositions of the traditional classical ideas. The
General Theory was a radical departure in many ways from these ideas
and represented a broad attack on the fundamental ideas of the
traditional classical school. It was written six years after the onset of the
Great Depression, with massive unemployment year after year, and is
based on the notion that while general equilibrium of the classical
framework shows full employment, the economy rarely shows evidence
of this state. The General Theory for the first time sets out a theory of
employment and output, under which these variables can deviate from
their full-employment levels under the impact of aggregate demand
changes. Keynes viewed investment fluctuations, acting through the
multiplier, as a major source of these deviations. Another source of
fluctuations was the speculative demand for money since it was based
on expected returns in the highly speculative bond markets. Keynes
dismissed the quantity theory of money, as well as the notion that the
labor market clears on a continuous basis to ensure full employment or
at least does so sufficiently fast not to leave excessive unemployment
over lengthy periods.
In the analogy with the human body used in this chapter, Keynes’ focus
corresponded to that on the pathology of the economy and its pathogens.
By contrast, the focus of the classical ideas corresponded to that on the
healthy state of the body and only transient deviations from it.
The ideas of The General Theory were sufficiently novel for various
expositions or ‘guides’ to explain it to the economics profession. One of
these, by John Hicks in 1937, became the source of the IS-LM analysis.
Out of The General Theory and its numerous expositions came the modern
definitions of many terms which are now basic to macroeconomics:
saving, investment, the multiplier, the speculative demand for money,
effective demand, failure of markets to clear and the dynamic analysis
relevant to this stage, the indirect transmission mechanism, etc.

539
11.3.1 Development of the Keynesian schools after Keynes
In the decades since 1936, variants of Keynes’ ideas have been
incorporated into a variety of quite different models. These models are
known as Keynesian models. In one version or another, they dominated
macroeconomics until the mid-1970s. Their major strength was in their
analysis of demand shocks and the policies for dealing with them. Their
major lacuna was their failure to consider supply shocks and the policies
relevant for economies reacting to such shocks. The resurgence of classical
economics in the mid-1970s, at a time of intense supply shocks from the
dramatic increase in oil prices, pushed the Keynesian paradigm into the
background. The classical paradigm still continues to be the dominant one.
The limitations of the classical paradigm were to be revealed by the
recessions of 2001 – 2002 and even more forcefully by that of 2008 – 10
in the world economy.
Both of these recessions experienced collapse of aggregate demand, both
in investment by firms and consumer demand by households. The
economy responded by decreasing output and employment, as predicted by
the Keynesian analysis of demand-deficiency (see Chapter 9), so that the
governments and the central banks were forced to pursue the Keynesian
recommendations of expansionary monetary and fiscal policies — by
cutting interest rates, increasing the money supply, and creating fiscal
deficits.

11.3.2 Frequent themes in the Keynesianparadigm

A common concern of the Keynesian models is with the potential for
involuntary unemployment, which is the deviation of actual employment
in the economy from its full-employment level. Consequently, these
models tend to pay special attention to the structure of the labor market: its
demand and supply functions and whether or not equilibrium holds
between them. Within this focus, some Keynesian models assume nominal
wage rigidity, often justified by theories of nominal wage contracts
between the workers and the firms. However, other Keynesian models do
not do so but consider the deviations from general equilibrium that could
occur even when the nominal wage is fully flexible.
The assumption of the rigidity or stickiness of prices in the economy is
often regarded as another common theme of Keynesian models. While this
assumption can impose deviations from a general equilibrium, it need not
be the only cause or reason for potential deviations. Therefore, models
within the Keynesian paradigm need not, and should not, all be based on

540
price rigidity. There is, consequently, also a place for Keynesian models
that consider the deviations from general equilibrium that could occur even
when the prices are fully flexible.
While some of the Keynesian models assume equilibrium in the
macroeconomic models, others do not do so. While some assume a special
form of the labor supply function, others assume the same general form of
the labor supply function — that is, labor supply depends on the real wage,
not the nominal one — as the classical paradigm. While some assume
nominal wage rigidity of some form, others do not do so. Similarly, while
some models assume price level stickiness or rigidity, others do not do so.
This variety in modeling within the Keynesian paradigm becomes even
more evident when the Keynesian and the neo-Keynesian models are
compared.
To reiterate, the variety ofmodeling—though perplexing and sometimes
seemingly contradictory, among different models—in the Keynesian
paradigm is essential to the proper study of the pathology of the economy.
As our comparison with the pathology of the human body illustrated, it
would be a mistake to force the Keynesian models into a single
straightjacket, even though this would impart an attractive uniformity and
a single model for the Keynesian paradigm as a whole.
New Keynesian economics
New Keynesian economics emerged as a distinct integrated model of the
economy in the 1990s and has become increasingly popular. It has IS
equations, a new Keynesian Phillips curve based on a price-adjustment
equation and a Taylor rule for monetary policy.

11.3.3 The variety of Keynesian models

Given the complexity of the functioning of the labor market, the
Keynesian models have resorted to a variety of simplifying assumptions
about it. At the risk of oversimplification, the differing critical assumptions
leading to different models of the Keynesian paradigm have been:
• The nominal wage is fixed (1940s and early 1950s). [Not presented in
this book.]
• The nominal wage is variable but the supply of labor depends on the
nominal and not the real wage (1950s and 1960s). [Not presented in this
book.]
• The structure of the labor market can be replaced by the Phillips curve
(late 1950s and early 1960s). [The Phillips curve was presented in
Chapter 10.]

541
• A fall in demand leads to a demand deficiency relative to full-
employment output because of the failure of commodity markets to
instantly reduce the price level to the appropriate extent. Firms react
faster and adjust output and prices. This leads to deficient demand
models, which assume that when the demand for commodities falls, firms
react faster than the economy and adjust the quantity produced (from
Keynes’ The General Theory in 1936 onwards). [Deficient demand
analysis was presented in Chapter 9.]
• In the labor market, given hiring and training costs, it is optimal for firms
and workers to enter into wage contracts, which tend to be staggered
among firms and industries. In the presence of such contracts, the
demand and supply of labor depend on the expected real (not nominal)
wage but that the expectations on prices, needed to derive the expected
real wage from the negotiated nominal one, are subject to errors and
asymmetric information between firms and workers (1970s and 1980s).
[The implicit contracts theory was presented in Chapter 8.]
• In the commodity markets, given the ‘menu costs’ of changing price lists
etc., it is optimal for firms to resort to sticky prices (neo-Keynesian
economics, 1980s and later). [The menu cost theory was presented in
Chapter 8.]
• Workers’ effort on the job is variable, leading to the efficiency wage
theory of wages and employment. (neo-Keynesian economics, 1980s and
later).
• An integrated macroeconomic model with sticky prices fixed by
monopolistically competitive firms resulting in slow staggered
adjustment of prices, Taylor rule for monetary policy and an IS equation
(new Keynesian model, since mid-1990s).
As discussed earlier, the focus of the Keynesian analyses is on states other
than full employment.
Monetary and fiscal policy effects in Keynesian models
The fundamental implication of the various forms of Keynesian models is
that in conditions of aggregate demand deficiency and less than full
employment, aggregate output depends upon aggregate demand and,
therefore, on the demand management policy variables of fiscal
expenditures, taxation, and the money supply. On the impact of monetary
policy on real output, the major implications of the Keynesian analyses
are:
• This impact will depend upon the existing demand deficiency in the
economy, so that a linear relationship between real output and the money

542
 supply, with constant coefficients, is not a proper representation of the
Keynesian implications. Notably, an increase in the money supply at full
employment will mainly or wholly produce inflation, without much
impact on output. But the same increase in the money supply when the
economy is below full employment with considerable excess capacity
will have a significant impact on output, though it may also produce
some inflation.
• At any given stage of the business cycle, both the unanticipated and the
anticipated values of the money supply—as also of the fiscal variables—
will affect output equally, as against the classical assertion that only the
unanticipated values do so. Therefore, the distinction between the effects
of anticipated and unanticipated changes in demand are not very relevant
to Keynesian models.
• The impact of money supply changes on employment and output does
not necessarily proceed through changes in the price level. In fact, the
impact of a money supply change on employment and output is greater,
the smaller is the concomitant price level change. By comparison, along
the expectations augmented Phillips curve associated with the classical
paradigm, the change in output is greater the greater the price level
change.

11.3.4 The critical role of dynamic analysis when aggregate

demand falls
The introduction of demand deficiency in a dynamic context does away
with the Keynesians’ need to assume the rigidity of prices and nominal
wages, or on the irrationality of the labor supply function based on
nominal rather than real wages. For such analysis, given a fall in aggregate
demand, the central issue is the nature of the individual firm’s response to
a fall in the demand for its product and the nature of the worker’s response
who is laid off or whose job no longer seems to be secure, in a context
where the numerous markets of the economy cannot realistically be
assumed to come into macroeconomic equilibrium instantly. This
represents a shift in the debate from comparative static to dynamic
analysis. There can be numerous plausible dynamic paths corresponding to
any one comparative static macroeconomic model, and not all necessarily
lead to full employment or do so within a short time. This implies a role
for Keynesian demand management policies depending upon the state of
the economy and the speed at which it is expected to redress deficient
demand or involuntary unemployment.

543
 Box 11.2: The Keynesian-Neoclassical Synthesis on Aggregate Demand
On the product, money, and bond markets, most Keynesians seem
willing to accept the IS-and LM relationships and the aggregate demand
analysis, as are those in the neoclassical model. This acceptance within
the main Keynesian approach is often labeled as the Keynesian-
neoclassical synthesis (of aggregate demand analysis). This synthesis
evolved in the 1960s through the acceptance by both the neoclassical
economists and the Keynesians of the IS and LM specifications for the
demand structure of the economy. This synthesis excluded both the
quantity theory range and the liquidity trap from the normal shape of the
LM curve. Consequently, both fiscal deficits and money supply
increases would be effective in increasing aggregate demand.
However, not all Keynesians go along with the acceptance of this
synthesis, with some Keynesians — especially post-Keynesians —
disputing the nature and/or the stability of the IS and LM curves and
their accuracy in representing Keynes’ own ideas in The General
Theory. At the other end of the spectrum, some classical economists
would wish to discard the IS-LM analysis while retaining the AD-AS
analysis.

11.4 The Reformulation of Keynesian

Approaches

11.4.1 NeoKeynesian building blocks for the reformulation

of Keynesian macroeconomics
With the resurgence of classical macroeconomics in the 1970s and 1980,
Keynesians sought to rebuild the foundations of their approach. This
rebuilding in its early stages (1970s to 1990s) consisted of contributions on
various aspects of macroeconomic modeling. These contributions were:
• The model based on menu costs, with a staggered price adjustment. This
was discussed in Chapter 8.
• Staggered nominal wage adjustments model, based on staggered wage
contracts. This was also discussed in Chapter 8.
• Implicit employment contracts based on hiring and training costs and/or
firm-specific skills. Their implications for labor hoarding were discussed
in Chapter 8. Such contracts can lead to staggered adjustment of
employment among firms.

544
• The efficiency wage model based on variable effort and the incentives to
increase it.
Each of the above implies that monetary policy can be used to change
output and employment in the economy for the short run but not for the
long run. This implication for the short-run effects of monetary policy is
shared — though not always the assumptions on which it is based — with
the earlier Keynesian models.
The models with one or more of these features were often labeled as neo-
Keynesian ones. Note that in these neoKeynesian approaches, as in some
of the earlier Keynesian ones, since the (sticky) price level does not rise in
response to an increase in aggregate demand, the short-run increase in
output occurs without a prior or accompanying increase in prices, so that
the impact of monetary policy on output is not a function of the change in
prices but rather of the change in the money supply itself. In fact, any
increase in prices reduces the impact of monetary expansion on output.
Therefore, in the neo-Keynesian approaches, the implication for monetary
policy is that it is neutral in the long run but not in the short run. Further,
higher rates of monetary expansion could evoke lower rather than higher
increases in output. This impact of monetary policy does not require a
reduction in real wages since the real efficiency wage — higher than the
market-clearing one — creates a long-run pool of involuntary
unemployment.
Many economists see the newer Keynesian approaches as replacing the
earlier Keynesian ones. However, there are several distinctive elements in
the earlier ones. Noteworthy among the elements of the earlier Keynesian
school were the lack of a macroeconomic mechanism for instantly
reaching equilibrium and the possibility of deficient demand. (see the
deficient demand analysis of Chapter 9), compare with their absence in the
neo-Keynesian approaches. Therefore, the newer Keynesian approaches
should be taken to be another addition — rather than complete
replacement of — to the stable of the models of the Keynesian paradigm.

11.4.2 Taylor rule and its incorporation into Keynesian

macroeconomics
The Taylor rule is that the central bank pursues monetary policy by
targeting the interest rate according to the equation:

In the Taylor rule, r is the real interest rate, (y – yf) is the real output gap

545
— i.e., the difference between actual and full-employment output — and
(π–πT) is the difference between the actual and target inflation rates. Such
a rule can be used in the context of any model that allows actual output to
differ from the full-employment one for significant periods and implies
that the economy can benefit from changes in aggregate demand brought
about by monetary policy. Both these propositions are not consistent with
the major proposition of the modern classical model that the deviations
from full employment are, at most, transient and self-correcting. Both are
consistent with the models of the Keynesian paradigm, so that the Taylor
rule, first proposed in the 1990s, came to be incorporated into some of the
subsequent formulations of Keynesian macroeconomics.

11.4.3 The new Keynesian model

Since the Taylor rule arose in the 1990s, at a time when the other
components of the neo-Keynesian ideas were being refined and organized
into a cohesive structure, some modern Keynesians have made it an
essential component of their models. The cohesive form of the latest type
of Keynesian models, developed since the mid-1990s, incorporates most of
the building blocks of neo-Keynesian economics as well as Taylor’s rule.
This model has been labeled as New Keynesian economics. Its main
components are:
• The IS curve analysis of the commodity market.23
• Staggered price changes (from the ‘sticky-price’ theory), based on the
assumption of monopolistic competition rather than perfect competition,
and staggered wage adjustments (from the wage-contract theory).
• Taylor rule for the specification of monetary policy on interest rate
targeting.
The above components specify the short-run model and determine the
deviation of output in the short run from the long-run one. For the long
run, the economy is taken to function at full employment, with all prices
and wage adjustments having been completed.

11.5 The Credit and Economic Crisis of 2007–2010: Which

Theory Can Explain It?
The credit and economic crisis of 2007–2010 originating in the USA and
spreading to many other countries, as well as Asian crises of the mid–
1990s (see Chapter 16), provide an acid test of the validity of the various
macroeconomic theories and schools. The stylized facts of these crises

546
include:
• The crisis in financial (money, credit, and stocks) markets caused a fall in
aggregate demand.
• The fall in aggregate demand caused a fall in output and rise in
unemployment. Output clearly fell below the full-employment level and
stayed below that level for several quarters and years.
• Money and credit were not neutral and a credit collapse was a major
determinant of the depth and duration of the recessions. The prices of
assets, such as houses, land, and stocks were also not neutral, and their
rapid declines were contributors to the crises.
• There occurred a reduction in the number of available jobs, so that many
workers who were otherwise qualified became involuntarily unemployed.
Hence, the economies clearly had involuntary unemployment.
• The recovery in aggregate demand required substantial monetary and
fiscal expansions; otherwise the economy would have had a worse and
longer recession, so that such policies speeded up the return toward full
employment.
• The collapse of credit played a key role in reducing aggregate demand
and aggregate supply, and delaying their recovery.
• The financial collapse and economic crisis spread rapidly from the USA,
the world’s largest economy, to many other countries.
Economics is a science and if its theories are to be valid and useful, they
must explain the preceding facts. The models of the classical paradigm,
especially the modern and new classical ones with their emphasis on the
full-employment general equilibrium state with minor, transitory, and self-
correcting deviations, does a fairly poor job of explaining the above facts.
Keynesian models as a whole do much better since they imply the non-
neutrality of money and credit, the impact of credit, the occurrence of
involuntary unemployment, etc. However, the new Keynesian model, with
its emphasis on general equilibrium, seems to do poorly. The
disequilibrium analysis of Chapter 9 is clearly very relevant to explaining
the above facts, so that an exclusive focus on long-run or short-run
equilibrium, as in many presentations of macroeconomic models, would be
inappropriate and invalid.
Does macroeconomics as a whole provide an adequate explanation of the
above stylized facts, and adequate guidance for policymakers? The answer
to this has to be mainly in the affirmative. Macroeconomics does provide a
fairly good understanding of the causes of the various crises and recessions
and what monetary and fiscal policies needed to be pursued.24 What it did
not provide is precise enough knowledge of the extent and timing of the

547
required policies. This extent and timing belong to the realm of dynamic
analysis, which is as yet imperfectly developed. In any case, the dynamic
patterns of the economy tend to differ from one recession to another. Their
recognition and prediction in any one recession at best remains an art
rather than a precise science, causing numerous differences and disputes
among economists in diagnoses and policy recommendations.

11.6 Which Macroeconomic Paradigm Should One Believe

In and Use?
While most textbooks and economists consider this to be a legitimate
question, our remarks above suggest that it is an improper — and quite
likely dangerous — one for the formulation of economic policies. The
proper study of the economy requires the study of both its healthy state
and its breakdowns. Since we cannot be sanguine that the economy will
always operate in general equilibrium, the models of the Keynesian
paradigm must not be neglected. Since we cannot be sure that the economy
will never be in general equilibrium, the models of the classical paradigm
must also not be neglected. Both paradigms have their relevance and
usefulness. Neglecting either of them can lead to erroneous policies that
impose high costs on the economy and its citizens.
For the practical formulation of monetary policy, the relevant and
‘interesting’ question is not the a priori choice between the classical and
the Keynesian models, but rather the perpetually topical one: what is the
current state of the economy like and which model is most applicable to it?
There is rarely a sure answer to this question. Consequently, the judgment
on this question and the formulation of the proper monetary policy are an
art, not a science — and very often rest on faith in one’s prior beliefs about
the nature of the economy.
While one cannot dispense with one’s beliefs and economists rarely give
up their conception of the nature of the economy, the fundamental role of
economics must be kept in mind. This is that economics is a positivist
science, with the objective of explaining the real world. This is done
through its theories, which, by their very nature, must be simplifications —
more like caricatures — of reality. As such, they may be valid or not, or
better for explaining some aspects of reality rather than others, or seem to
be valid for some periods rather than others. Intuition and econometrics are
both needed and useful in judging their validity and relative value.
A side implication of the positivist objective of economics is its
normative role — i.e., the ability to offer policy prescriptions to improve
on the performance of the economy, (hopefully) as a means of increasing

548
the welfare of its citizens. Both the Keynesian and the classical paradigms
are essential to these roles.
Box 11.3: The Anatomy of Two Quite Different Recessions: 1973–1975
and 2001 – 2002
The recession of 1973–1975 was mainly in response to increases in oil
prices, engineered by the OPEC oil cartel. During this period, oil prices
roughly quadrupled. This was a supply shock to the economy, which
seemed to be permanent. In any case, this rise in oil prices was not only
maintained through the 1970s but also further supplemented by another
doubling of oil prices in 1980–1981. The economy responded by severe
reductions in output and employment. The Keynesian economics of the
day recommended expansionary demand management policies to reduce
or eliminate these reductions. Such policies were attempted but — in
hindsight — were clearly an inappropriate demand-based response to
supply shocks. The results were recessions in 1973–1975 and 1980–
1981, along with a lengthy period of stagflation.
The recession of 2001–2002 was triggered by a collapse of stock
prices, starting with those of the Internet and computer stocks, which led
firms to cut back on their investment and employment. This was soon
followed by a collapse of consumer confidence and fall in consumer
demand, further supplemented by the economic pessimism following the
terrorist attacks on the World Trade Center in New York and the
Pentagon. Severe and successive cuts in interest rates seemed to be
ineffective in reviving investment since the firms had excess capacity
relative to the perceived demand for their products. The classical
paradigm would have predicted a severe fall in prices and wages to soon
restore full employment. These did not occur.
The economy went into a severe recession, and the policymakers were
forced to resort to expansionary fiscal and monetary policies.
The recession of 2008–2010 and the East Asian crises of the mid –
1990s have been discussed earlier in this chapter and are again discussed
in Chapter 16 on business cycles and crises.

11.7 Paradigms and Policies

11.7.1 Stabilization versus pro-active policies

Stabilization policies are ones that attempt to stabilize aggregate demand at

549
the level sufficient to buy the full-employment output of the economy at
the current price level. Proactive policies are those that seek to increase the
economy’s output above its full-employment level by increasing aggregate
demand.

11. 7.2 Rules versus discretion in the pursuit of policies

Both stabilization and proactive policies can be pursued according to a rule
(i.e., a set pattern) or in discretionary manner (i.e., not according to a set
pattern but as the central bank thinks fit under the emerging conditions in
the economy).
The Taylor rule, according to which the central bank’s target interest rate
depends on the output gap, has the objective of manipulating interest rates,
and through them aggregate demand, to produce full-employment output.
The Taylor’s rule is an example of a stabilization rule.

11.8 The Role of the Government in the

Macroeconomy
There has always been a role for the government in the economy. At the
minimal level, this meant the provisions of some of the public goods, such
as defence of the country, coinage, law and order, etc., provided by kings
in earlier times. At its most extreme, it can encompass total control over
the economy, including ownership of all production units and allocation of
factors of production to different uses and payments to them, as under a
communist system.
Modern industrialized economies tend to be largely, though not wholly,
capitalist ones, with production and income distribution determined by
markets and private economic agents (firms and individuals). The role of
the government in such an economy tends to vary among countries and
depends on their political system, philosophy, and history. Within the
context of a democratic country with a largely capitalist economy, the
‘conservative’ approach argues for a ‘small government’ which focuses
mainly on providing only public goods, with minimal regulation of the
non-government sectors of the economy. Those in favor of this philosophy
argue that governments, which are run by bureaucrats, are inefficient and
often impose regulations that make the economy worse rather than better.
Their arguments usually rely on the efficiency of production and factor
allocation in the economy with perfectly competitive markets and flexible
prices and wages. Such a system ensures that individuals and firms acting

550
in their own interests (i.e., maximizing utility or profits), and without
government interference and regulation, will maximize the economy’s
output and produce the socially optimal combination of commodities. This
argument had produced during the 19th century the economic philosophy
and approach to economic policies of laissez faire, discussed earlier in this
chapter. The economic policies of Reaganomics and Thatcherism followed
in the USA and UK during the 1980s and mentioned earlier in this chapter
followed this ‘small government’ philosophy. By comparison, the more
‘liberal’ approach focuses on the income inequalities produced by the
capitalist, free enterprise system, the failures of the market in the case of
externalities, and the departures of many sectors of the actual economies
from competition. The proponents of the liberal approach argue that
governments should interfere in the economy to provide the goods that can
be more efficiently provided by the government than by the existing
market structures, and the government should regulate in the national
interest firms and industries which tend to have monopolistic or
oligopolistic structures. Further, government should pursue social
objectives such as ensuring a minimum level of standard of living,
children’s welfare, access to medical care, and unemployment insurance.
Because of its usual inclusion of social objectives, this liberal approach is
now often associated with socialism.
There is no country in the world which has perfect competition and
perfectly efficient markets in all their sectors, so that the actual economic
structure does not provide the theoretical basis for the laissez faire
approach to economic policies. There is also now hardly any country in the
world which follows a pure capitalist philosophy. Further, the social and
political philosophy of most countries does ask for the pursuit of certain
social objectives. Therefore, nowadays, virtually all governments do
actively pursue economic and social policies, though some do so more
than others. Among the countries with somewhat lesser pursuit of
economic and social objectives through government policies than others is
the USA. European countries tend to be called socialist since their
governments are more active in pursuit of social objectives than the USA.
Most developing countries possess similar social objectives, but often lack
the fiscal means to pursue them effectively to any significant extent.

11.8.1 Evolution of ideas about the role of the government

On the whole, the success of the capitalist economies versus those of the
communist ones and of less regulated ones versus those of the more highly
regulated ones since World War II produced by the end of the 20th century

551
a general belief that controled and highly regulated economies led to lower
growth of output and standards of living. This practical demonstration led
to a movement in most countries during the last three decades of the 20th
century toward a smaller government size than had earlier existed, with
less ownership of production units and industries by the government and
less regulation of the economy in general. Conversely, there was a move
toward promoting competition both within the country and across
countries. The latter helped the process of globalization, the spread of
multinational corporations, and the expansion of international trade that
occurred in the past three decades. In these decades, the free markets
ideology seemed to have won the battle between market economies and
centralized ones, many of which had been communist ones. The starkest
example of this victory was the fall of state-dictated communist economies
in several countries, many of which were earlier in the Soviet Bloc, and
their transition to democratic socialism. In the case of China, while the
political system of communism was retained, economic reforms after 1980
opened up significant parts of the economy to private market forces,
thereby allowing the economic incentives of capitalism, which permit
income differences and private wealth accumulation, to enhance
production and improve economic efficiency. As a result, the output
growth rates of both Russia and China shot up. In the domestic context of
the USA, the supporters of capitalism and small government seemed to
have won against the liberal ideology, with the words ‘liberal’ and
‘socialist’ often becoming somewhat pejorative in the US politics during
the 1980s and 1990s.
The worldwide economic crisis of 2008–2010, initiated by problems in
the financial sectors of the US economy, were seen by many observers as
an outcome of a largely unregulated, free market, and capitalist financial
system. This belief led to demands for more regulation of financial firms
in terms of their capital requirements, the riskiness of their investments,
auditing of their accounts and in many other aspects.

11.8.2 Classical and Keynesian approaches and the debates

on the size of the government
How do the Keynesian and classical schools relate to the debates between
small versus large government, between liberalism or liberal socialism and
communism, or market versus centralized running of the economy? The
Keynesian and classical paradigms are modeling approaches on the
functioning of the macroe- conomy. Since they study the workings of
largely market-based economies, neither studies the communist,

552
bureaucratic economies. Beyond that, the analyses of the Keynesian and
classical schools can be applied to economies with small or large
government, and purely capitalist or liberal socialist economies. However,
since Keynesian economics focuses on the pathologies of the economy and
the need for governments to interfere through monetary and fiscal policies
in improving the performance of the macroeconomy, while classical
economists focus on the full-employment level of the economy and argue
against the need for monetary and fiscal policies, Keynesianism is often
loosely associated with a larger government size than recommended by the
classical economists. Further, Keynesianism is associated with greater
regulatory role of the government over the economy and for the active
pursuit of monetary and fiscal policies to stabilize the economy at full
employment.

11.9 Conclusions
• There are currently two dominant paradigms in macroeconomics. These
are the classical and Keynesian ones.
• The implications drawn from the classical paradigm are usually for its
full-employment equilibrium states. For such states, output is also at the
full-employment level. Since the objective of policy is normally specified
to be the maintenance of full employment, which occurs anyway in the
equilibrium of this model, there is no need or role for systematic demand
management policies in such a state. Further, since the classical
economists usually assume that full employment is restored within a
reasonable or acceptable period, the classical prescription even for the
disequilibrium states is that the economy should be left alone to return to
full employment.
• Not all economists believe that the economy always achieves full
employment or remains reasonably close to it. In particular, the
Keynesian paradigm believes that the economy does not always perform
at full employment for a variety of reasons.
• John Maynard Keynes was the founder of the Keynesian paradigm. It is
now accepted that Keynes did not assume that workers based their supply
behavior on nominal rather than real wages and did not assume that the
nominal wages were rigid. While the classical paradigm focuses on price
adjustments in response to demand or supply shocks, Keynes’ analysis
focused on quantity adjustments by firms in such situations to derive the
disequilibrium paths of output and employment. If the adjustment toward
equilibrium were a slow one, a shortfall in demand would result in a
reduction in output and employment for periods of significant duration. If

553
 the economy was beset by fresh disturbances arising frequently, such as
through bouts of pessimism or optimism about the future among firms’
managers, the disequilibrium state would be a persistent phenomenon,
with varying levels of employment or output.
• Neo-Keynesian economics came into being in an attempt to rebuild the
Keynesian framework after the decline of faith in the 1970s in the
Keynesian models and their policy prescriptions, and the resurgence of
classical economics in the 1980s and 1990s. The efficiency wage
hypothesis of neoKeynesian economics asserts the short-run rigidity of
real wages, in contrast to that of nominal wages, which had been a
component of some of the earlier Keynesian ideas. The neo-Keynesian
theory does provide a new basis for the short-run rigidity of prices
through its hypothesis of menu costs in monopolistic competition. Both
the Keynesian and the neo-Keynesian theories agree that monetary policy
need not be neutral in the short run.
• In the presence of uncertainty, expectations play an important role in
both paradigms. However, the classical paradigm assumes that economic
agents formulate their expectations on the market-clearing prices and
base their supplies of labor and commodities on these expected
equilibrium prices. The Keynesian paradigm assumes that firms
formulate their expectations in terms of the future demand for their
products and produce accordingly, while households formulate their
expectations on employment and determine their demand for
commodities based on their expected income. These differing approaches
to the formulation of expectations and subsequent responses by the two
paradigms produce very different dynamic paths of output and
employment in response to demand and supply shocks to the economy.
• Both the classical and Keynesian paradigms are consistent with the
rational expectations hypothesis and can incorporate it. This implies that,
in both cases, expectations are based on all available information.
However, the application of this idea—and, therefore, of rational
expectations—differs between the two paradigms. For the classical
paradigm, since full employment is the normal state of the economy and
any deviations from it are transient and self-correcting, the rational
expectation is always that of the full-employment values of the
economy’s variables. For example, if the available information indicates
that aggregate demand is falling, the rational expectation is that the
economy will adjust fast to restore it to its full-employment level, so that
at this rationally expected value of sales, the firms will continue to
produce at the full-employment level. For the Keynesian paradigm, since
the economy can stay away from full employment for considerable

554
 periods, rational expectations are those based on the expected actual state
of the economy for the period ahead. For example, if the available
information indicates that the economy is moving into a recession and
will have lower demand than the current one, the rationally expected
value of sales will be that they will fall — thereby leading firms to cut
production and employment, which would take the economy away from
full employment.
• In principle, interest rate targeting can be consistent with both the
Keynesian and classical paradigms. However, the output gap in the
Taylor rule implicitly incorporates the notion that the economy can stay
away from full employment for significant periods. This possibility is
more in line with the spirit of the Keynesian paradigm than of the modern
classical model.

Keynesian versus classical economics in relation to the real

world and real time
The classical and Keynesian paradigms represent different views of the
self-adjusting dynamic nature of the capitalist economy. The former views
it as being in full-employment equilibrium or close to it, with the dynamic
forces providing a strong tendency to return to full employment after any
deviation. The Keynesian paradigm allows the possible existence of full
employment but is concerned that the economy does not always, or most
of the time, perform at full employment. Further, a demand deficiency for
output can generate dynamic forces creating involuntary unemployment
that could subsist for extensive periods because there is no adequate and
rapid equilibrating mechanism, or the likely mechanisms are destabilizing.
From the perspective of monetary policy, it does not matter whether the
economy is away from full employment is described as being in
disequilibrium, in temporary equilibrium or in equilibrium, or even
whether there do exist mechanisms, such as the wage and price
adjustments which will eventually bring the economy to full employment.
If the economy left on its own can stay away from full employment for
some time, the appropriate question for monetary policy is whether it is
optimal for such policy to hasten the economy’s return to full employment.
On this issue, the neoclassical, and especially classical and new classical,
economists claim that it is better to leave the economy alone, no matter
what its state happens to be, while the Keynesians claim that monetary
policy can have a positive optimal role in certain states of the economy.
Historically, faith in these positions has tended to vary considerably. The
Great Depression of the 1930s in the industrial economies destroyed faith

555
in the traditional classical belief in a self-regulating economy. The fairly
stable macroeconomic performance of such economies in the 1950s and
1960s, though with active Keynesian demand management policies,
produced shallow, and short-lived recessions and led to a slow revival of
neoclassical economics — under the rubric of the Keynesian-neoclassical
synthesis. This was followed in the 1970s — a time of stagflation, partly
due to the Keynesian policy errors — which tended to restore faith in the
general classical beliefs. The dominant paradigm was the classical one
from the 1970s to the middle of the first decade of the 20th one.
The dragged out recession of the early 1990s and the increase in the rate
of unemployment to double digits in many industrialized countries in the
early 1990s was widely blamed on the restrictive monetary policies
pursued in these years to reduce the rates of inflation, followed or
accompanied by deficit-cutting strategies. The length of the recession and
the tempered recovery until the mid-1990s seemed to indicate that the
monetary and fiscal policies do affect real output and employment for
significant periods, as the Keynesians claim. Further, the recession of
2001–2002 was induced by the collapse of stock market prices following
the mania in Internet and computer stocks, leading to a collapse of
business and consumer confidence. These were heightened by the terrorist
attacks on the USA on September 11, 2001. The result was a severe fall in
aggregate demand, to which the economies did not respond by a quick and
sufficient fall in prices. Instead, firms cut back on output and employment,
with some moderation of prices. The financial and economic crises of the
world economy in 2007–2010 indicate the role of both financial factors
(bank solvency and reductions in the availability of credit) and aggregate
demand in causing severe declines in output. This causation strongly
supports the Keynesian paradigm.
To conclude, the financial and economic crises of 2007–2010 and the
collapse of credit markets (see Chapter 16), attributed by many to the
largely unregulated financial sector in the USA in the preceding decades,
brought serious doubts about the policies recommended by the classical
economists. In the ensuing backlash against such policies, aggressive
monetary and fiscal policies, as suggested by the Keynesian paradigm,
were pursued by many countries to prevent the crisis from turning into a
depression. There were also widespread calls for greater regulation of the
financial sector by government agencies, all of which are in line with
Keynesian ideas. Consequently, Keynesian doctrines, rejuvenated in newer
forms during the preceding three decades, were brought into popularity by
the recession of 2008–2010 and may become the dominant ones for the
next few decades.

556
 KEY CONCEPTS
Classical paradigm
Keynesian paradigm
Traditional classical approach
Neoclassical model
Modern classical model
New classical model
New Keynesian model
Reaganomics and Thatcherism
Comparison of the recessions
of 1973–1975, 2001–2002
and 2008–2010
Comparison of the booms of
the 1960s and 1990s, and
The differing interpretations
of rational expectations in the
classical and Keynesian paradigms.

SUMMARY OF CRITICAL CONCLUSIONS

• The main paradigms in economics are the classical and Keynesian ones.
Each consists of several different schools and theories.
• The classical paradigm focuses on the full-employment state of the
economy and transitory deviations from it due to price misperceptions.
• The main branches of the classical paradigm are the traditional classical
school, the neoclassical one, the modern classical one, and the new
classical one.
• The St. Louis Monetarist School represented a transition from the
Keynesian period to the modern classical one.
• An abiding theme in Keynesian models, originating with Keynes’ The
General Theory, is the failure of the economy to attain or maintain full
employment. This is attributed to the imperfect and slow functioning of
the competitive price mechanisms, and the likely responses of firms and
households in terms of quantity adjustments (production and
consumption) in disequilibrium. The Keynesian approach asserts that this
failure is especially symptomatic of the labor market, so that involuntary
unemployment is a common occurrence in the real-world economies.
• The Keynesian model based on this deficiency in the macroeconomic
environment is the demand-deficient model. This approach posits that the

557
 rational dynamic responses by firms and households to conditions of
inadequate demand and involuntary unemployment do not always take
the economy to full employment or do so within an acceptable period.
• The new Keynesian theories rely on rational long-run behavior resulting
in implicit contracts, staggered wage contracts, sticky prices due to menu
costs, etc.

REVIEW AND DISCUSSION QUESTIONS

1. What are the underlying themes (or theme, if only one) of the classical
paradigm?
2. What are the underlying themes of the Keynesian paradigm? Do they
justify the study of just one model, one variety of models, or several
different varieties of models? Why?
3. In order to explain the performance of the economy in economic
recessions and booms, would you rely on either the classical paradigm
or the Keynesian one, or sometimes on one and sometimes on the other?
Explain your answer with reference to the different phases of the
business cycle.
4. What are the distinguishing features of the modern classical model
relative to the traditional classical approach to the macroeconomy?
5. What explanations for a high actual rate of unemployment are consistent
with the modern classical model?
6. What explanation is offered by the Keynesians for unemployment when
it is above its natural rate?
7. Specify the main tenets of the St. Louis Monetarism.
8. Specify the major components of the new Keynesian approach.
9. Milton Friedman showed that changes in the money supply could have a
strong impact on national income and output. He also believed in the
possibility of long and variable lags in the impact of monetary policy.
What do these propositions taken together imply for the effective pursuit
of monetary policy? Discuss.
10. What are the major reasons for the 1973–1975, 1980–1982, 1990–
1992, and 2001–2003 recessions in the USA?
11. [Optional] What were the major reasons behind the strong expansion in
output and employment of the late 1990s?
12. In the USA, President George W. Bush in 2000 and President Barack
Obama in 2008–2009 proposed to Congress massive fiscal deficits,
which they justified as being needed to fight the shortfall in aggregate
demand at the time. In doing so, were they following the Keynesian or
classical policy prescriptions? What had caused the shortfall in
aggregate demand?

558
13. What is the Taylor rule? Explain why it is more in the spirit of the
Keynesian paradigm than in that of the modern classical school.

ADVANCED AND TECHNICAL QUESTIONS

T1. (a) Are there levels of real output, interest rates, and the price level
other than the full-employment ones in the neoclassical model? (b) How
would the economy move from a disequilibrium position resulting from
an unanticipated decline in investment to full employment? (c) Discuss
the likely responses of firms and households following such a shock.
T2. High real wages is sometimes the explanation offered by the classical
economists for unemployment above its natural rate in some countries
(e.g., in Europe). What is meant by high real wages in this explanation?
Show this diagrammatically. What can cause such high wages? Discuss
whether or not these provide a likely cause of the last recession in your
country?
T3. What were the main elements of Milton Friedman’s macroeconomics?
T4. During the 1980s and 1990s, many economists argued that Keynesian
economics give little guidance or even wrong prescriptions for dealing
with the current economic problems in Canada (or the USA or Britain or
any other country of your choice). What justified such comments? What
are your views on this issue and how would you justify them for the
country of your choice over booms and recessions?
T5. Sometimes, governments run fiscal surpluses and use these to reduce
the public debt. Discuss the effects of such surpluses and debt reduction
on aggregate demand, interest rates, and output. Very often, in running
fiscal surpluses, it is not the intention of the government to follow
Keynesian or classical policy prescriptions? What are some other
reasons for running a balanced budget or one with a surplus?
T6. Write an essay on the causes of the credit and economic crisis in USA
during 2007–2010. How did it spread to other countries? What policies
were used by the monetary and fiscal authorities to fight these crises?
What are the arguments for and against the decision of many
governments to bail out their banks and other financial institution?
T7. Write an essay on the likely impact of the worldwide financial and
economic crises of 2007–2010 on general views on the validity of the
Keynesian versus classical economics.

1The difference between an assumption of full employment and one which

is an implication of the equilibrium state is that the former rules out the
derivation of the properties of the system when it is in disequilibrium; the

559
latter does not necessarily do so.
2See Chapter 8 for this analysis.
3An example of this was economists’ inappropriate policy prescriptions,
based mainly on the traditional classical ideas, during the early stages of
the Great Depression in the 1930s. These worsened the depth of the fall in
GDP and lengthened the depression — and contributed to the demise of
faith in the traditional classical ideas. Another example of inappropriate
policies, based on the aggregate demand management approach in the
Keynesian paradigm, occurred in response to the supply shocks of 1973
and 1974. This led to stagflation and contributed to the demise of faith in
the Keynesian paradigm.
4For example, under the centralized economic system that used to exist in
the former Soviet Union and several other communist countries, the
bureaucrats made the decisions on production, employment, prices etc.
5In particular, if there are delays in establishing these prices or
communicating their knowledge, it is assumed that all production and
trade is put on hold during the delay and that any such delay does not
impose costs on economic agents. This is known as recontracting.
6The empirical evidence on this issue is provided by the stylized facts
presented at the beginning of this chapter. The evidence that money is not
neutral in the short run or during disequilibrium (in the economy in general
or following a change the money supply itself ) is very strong.
7Note that disequilibrium remains a hypothetical, analytical (but not
necessarily a real-time) state within all the models of the classical
paradigm. The difference among them is about its significant occurrence in
the real-world economies.
8It also advocated that the central bank should be ‘prudent’ in the way that
private banks have to be. This implied that it should not lend to
commercial banks which are illiquid and in danger of insolvency, so that
the central bank should allow the markets to determine whether private
banks fail or not.
9Ricardo’s contemporary, Thomas Malthus, disputed this proposition
though Malthus did not offer an acceptable theoretical basis for his
viewpoint. Another contribution of Malthus was to growth theory. This
contribution is presented in Chapter 14.
10This modern classical model is also different from the traditional
classical model in other ways. One of these relates to the existence of
speculative money demand in the modern classical model while the
quantity theory component of the traditional classical model did not have

560
such money demand.
11The various quotes are taken from Patinkin, a prominent neoclassical
economist, even though many of them are from passages quoted by him
from other writers. The relevant references are to: Patinkin, D. The
Chicago tradition, the quantity theory, and Friedman. Journal of Money,
Credit and Banking, 1, 1969, 46–70 and Friedman on the quantity theory
and Keynesian economics. Journal of Political Economy, 80, 1972, 883–
905.
12In this sense, John Maynard Keynes and the Keynesians — along with
Milton Friedman — were all monetarists, while the modern classical
school is somewhat less monetarist since it downplays the impact of
money supply changes on the real variables of the economy.
13Friedman’s ideas are presented later in this chapter.
14Example of policy makers’ refusal to follow the recommendations of the
modern classical approach occurred during the recessions of 2000–2002
and 2008–10 when central banks hastened to intervene in the economy by
lowering interest rates and increasing the money supply. These policies
constituted intervention in the economy and represented Keynesian
prescriptions for macroeconomic policy.
15This liability is to make the periodic interest payments and repay the
principal on maturity.
16Many Keynesians held this view in the 1950s and 1960s. It argued that
the money stock was only a small part of the liquidity of the economy,
which included short–term bonds, trade credit, overdrafts at financial
institutions, etc. Further, the degree of substitution between these
components to any contraction of the money supply was so high as to
render the impact of the contraction on aggregate demand negligible.
17See Chapter 5.
18For example, Friedman believed that the economy does not always
operate at full employment in which case changes in the money supply
would change real output, as well as prices.
19The modern classical school has abandoned the direct transmission
mechanism in favor of the indirect one.
20There is considerable dispute as to whether any of the Keynesian models
represents Keynes’ own work or not. A close reading of Chapters 2 and 3
of The General Theory shows that they do not. It is therefore appropriate
to make a distinction between the Keynesian models and Keyne’ own
analysis, though the former arose out of interpretations of the latter.
21However, this phraseology seems to indicate a rather harsh judgment on

561
a mechanism that works amazingly well for a complex and dynamic
economy — and definitely better than any system of administrative
coordination of demands, supplies, and prices, which are attempted in
centralized, communist countries. If the actual performance of
competitive, free economies is assessed against realistic alternatives, the
former comes out very well.
22The term ‘coordination failure’ is relative to the perfect coordination of
markets under the hypothetical and mythical construct of an all–knowing
’auctioneer’ or ’state planner’ who can instantly work out the changes in
prices following any shifts of demand and supply for individual
commodities and announce them to buyers and sellers. The assumption of
such an entity is unrealistic and can lead to erroneous conclusions relative
to how households, firms and markets actually function.
23The new Keynesian IS equation is somewhat different from the standard
one presented in Chapters 4 to 6 of this book, and incorporates
expectations about the future.
24However, it still seems to lack an adequate understanding of the role of
credit, which proved to be central to the cause and duration of the
economic crisis of 2007–2010, as of many other crises.

562
 PART III
Topics in Open Economy
Macroeconomics

563
 CHAPTER 12
The Foreign Exchange Market, IMF,
and Globalization

This chapter presents the analysis of the foreign exchange market

and the determination of the equilibrium value of the
exchangerate. It also discusses the role of the international
monetary fund (IMF) in assisting countries with balance of
payments problems. It also includes a presentation of the
economics aspects of globalization of commodity and capital
flows.

Chapter 3 had presented the basics of the open economy concepts. This
chapter starts with a brief review of the nominal and real exchange rates,
as well as of purchasing power parity (PPP) and interest rate parity (IRP).
Then, it proceeds to the analysis of the foreign exchange market and
movements in exchange rates. The domestic country is again designated as
‘our’ country, and the rest of the world is designated as ‘the world’,
‘foreign country’, or ‘foreign countries’.
Chapter 3 had defined the (nominal) exchange rate as the number of
units of a foreign currency required to purchase one unit of the domestic
currency. Using the dollar ($) as the unit of the domestic currency and the
£ (British pound) as the generic symbol for a unit of the foreign currency,
the nominal exchange rate is £s per $ (£/$). The Greek letter ρ
(pronounced ‘rho’) is our symbol for the nominal exchange rate. The
dimension of ρ for our definition of the nominal exchange rate is £/$. Note
that some other textbooks define the exchange rate as dollars per pound ($/
£).
The real exchange rate, designated as ρ r, is the amount of foreign
commodities required to purchase one unit of domestic commodities. Its
relationship with the nominal exchange rate was specified in Chapter 3 by:

564
where:
ρ = nominal exchange rate in units of the foreign currency per unit of the
domestic one (£/$),
ρr = real exchange rate,
P = domestic price level, and
P F = foreign price level.

Therefore, the real exchange rate is the nominal exchange rate adjusted for
the relative price ratio between the countries.
The nominal exchange rate may be left unregulated by the national
authorities to adjust, demand, or supply pressures in the foreign exchange
market or its movements may be controled rigidly by the central bank or
allowed to vary within certain specified limits. An uncontrolled exchange
rate is said to be a flexible (or floating) exchange rate. A rigidly specified
exchange rate is said to be a fixed (or pegged) exchange rate. There are
many combinations of these two basic practices. Thus, the authorities may
operate a system whereby the exchange rate is allowed to vary within wide
limits or is altered at pre-specified intervals.1 The exchange rate regime is
said to be a managed one if the exchange rate is flexible but the central
bank intervenes (i.e., buys or sells foreign exchange) at its discretion in the
foreign exchange market to ‘manage’ the magnitude of the movements in
it.

12.1 Review of PPP

Chapter 3 presented three versions of PPP. These are the absolute version,
the short-run relative version, and the long-run relative version. The
absolute and relative versions of PPP are:
Absolute PPP:

Relative PPP:

where k(k ≠ 1) stands for the degree of deviation from absolute PPP, and
depends upon: (i) barriers (tariffs, quotas, shipping, and other costs) to
trade in commodities, (ii) the proportion of non-tradable goods in the
calculation of the two price levels (P and P′′ ), (iii) significance of capital
flows, and (iv) imperfections in foreign exchange markets. k′′ is the rate of
change in k. In the short run, k may be taken as constant, which would

565
make k′′ = 0. In the long run, the determinants of k are definitely likely to
change, so that k′′ need not equal zero. Hence, the two versions of relative
PPP are:
Short-run relative PPP:

Long-run relative PPP:

Of the various forms, economists prefer the short-run relative PPP

equation over short periods. As argued in Chapter 3, under floating
exchange rates relative PPP explains changes in the exchange rate, while
under fixed exchange rates, it explains movements in the domestic
inflation rate. However, in neither of these roles does this theory do well
empirically. In modern economies, a major reason in addition to the
existence of non-traded goods, for the failure of PPP is the dominance of
capital flows in determining the demand and supply of foreign exchange
for the country. This is so for most modern economies with free and large
flows of foreign exchange on a daily basis. However, PPP and its relative
version can be quite validly taken to represent one of the forces operating
on the domestic inflation and exchange rates, but not the only one, and
quite possibly not the dominant one in the short term for economies with
large and sensitive capital flows.

12.2 Review of Interest Rate Parity (IRP)

Theory

12.2.1 IRP as a theory of the interest rate

While PPP deals with commodity flows, the theory of IRP deals with
capital flows. It was pointed out in Chapter 3 that the capital flows in
international trade signify financial capital flows — not physical capital
flows,2 which are included under commodity flows. These financial capital
flows are flows of money in exchange for financial assets (stocks and
bonds) and ownership of physical ones (as, for example, when a domestic
firm buys land and factories in other countries or a foreign firm buys land
and factories in our country). When we buy foreign assets, the amount paid
for them to foreigners represents an export (outflow) of the domestic
currency from our economy. Conversely, when foreigners buy our assets,

566
the amount paid to us represents an import (inflow) of foreign currencies
to our economy.
Chapter 3 presented the IRP theory. It asserts that, in perfectly
competitive markets,

where R is the domestic interest rate, R F is the foreign one, and ρ ′′e = (ρ e
– ρ )/ρ e. (1 + R) is the amount received from an investment of one dollar
in domestic bonds after one period and (1 + R F)(1 – ρ ′′e) is the amount
expected to be received from an investment of one dollar in foreign bonds
after one period. This equation is known as the uncovered interest rate
parity equation. By expanding this equation, we have:

For very small values of RF and p′′e, (RF • p′′e) is even smaller, so that the
IRP condition is often simplified to the benchmark uncovered IRP
condition, stated in Chapter 3, as:

where:
R = domestic nominal interest rate,
RF = foreign nominal interest rate, and
ρ′′e = expected appreciation of the domestic currency3 [= (ρ e – ρ )/ρ e].
If there are differences in transactions costs, in the risks of investing at
home or abroad4 or in the degrees of risk aversion applied by investors to
domestic and foreign investments, the benchmark form of the IRP
condition was modified to allow for a premium or discount (see Chapter 3
for this analysis). The modified form is:

whose simplified approximate version is:

Chapter 3 had shown in Fact Sheet 3.1 (reproduced below as Fact Sheet
12.1) for the USA–Canada differential in Treasury bill rates that α was
positive or negative but was rarely equal to zero, so that IRP does not hold
strictly even among developed economies with closely integrated
commodity and capital markets. The sign and value of α depend on many
factors, including the perceived riskiness of investing at home or abroad,
the future prospects of the economies in question, the relative degree of

567
knowledge of the relevant investment vehicles and economies, the degree
of risk aversion, whether the country is a capital importing or exporting
one,5 etc. Therefore, the sign and value of α cannot be assumed to be
constant over time. They tend to vary over time, as the determinants of α
shift.
Taking accounting only of the fact that the country in question is a
capital importing or exporting one. It is likely that α is negative for a
capital exporting country if the general preference of its investors is for
investing at home, in which case R < (R F – ρ ′′e), so that the domestic
return will be less than the foreign one. But α would be positive if the
general preference of its investors is for investing abroad,6 in which case R
< (R F – ρ ′′e).

The importance of stock market returns for capital flows: a problem for
the IRP theory
As Chapter 3 pointed out, the major (and possibly the dominant)
component of capital flows for certain countries are international
investments in stocks, rather than in bonds. The expected returns in stocks
are more speculative than those in bonds: the profits, dividends, and future
resale prices of stocks are all uncertain while the coupons and redemption
value on maturity are certain for bonds. Further, bubbles in stock prices
tend to be more common and more pronounced than bubbles in bond
prices, as discussed in Chapter 2. Therefore, the expected return on stocks
can differ quite significantly from bond yields. They also fluctuate in
different patterns.
Since international movements of capital due to investments in stocks
can nowadays be much larger than those in bonds, the former limit the
simple and straightforward application of IRP theory among countries, so
that, in practice, the interest rates on bonds are rarely identical among
countries.
Fact Sheet 12.1: Interest Rate Differentials between Countries
This fact sheet duplicates Fact Sheet 3.1 in Chapter 3. It shows the
interest rate differential (plotted as the USA rate minus the Canadian
Treasury bill rate) between the USA and Canada, which have closely
integrated financial markets, so that there should be perfect capital flows
between these countries. Interest Parity Theory implies that the interest
rates should be identical between them. However, as the following graph
shows, there is almost always a differential between them. The Canada

568
 T-bill rate was higher in Canada from the early 1970s to the mid-1990s;
outside this period, the Canadian rate was sometimes higher and
sometimes lower than in the USA. Therefore, the value of the interest
rate premium/discount α is usually not zero and also not constant over
time.

Fact Sheet 12.1: (Continued)

Simplified form of the benchmark IRP for short-run analysis

For simplification in comparing the rates of return among the developed,
politically stable economies and considering only riskless government
bonds, α is usually taken to be constant for purely short-run analysis.
Further, it is usually set at zero for analytical convenience. That is, the
short-run benchmark IRP condition is usually specified as:

Since our focus is on the short-run analysis7 for such countries, this
chapter will henceforth continue further with the use of the benchmark
form of IRP with α = 0, thereby implicitly making several assumptions.
These are that investors do not have a preference between investing at
home or abroad or that the risks are the same, the economies have similar
future prospects, there are no differences in the brokerage costs of buying
and selling domestic and foreign bonds, etc. Under the assumption of this
simplified short-run IRP, funds will flow to the country that has the
highest expected return.
Covered versus uncovered IRP
A forward contract in the foreign exchange market is a contract for the

569
purchase (or sale) of a designated amount of a foreign currency at a
specified future date at a specific exchange rate, with the contract being
agreed to at the present time. The market in which forward contracts are
traded is called the forward market (for foreign exchange). The financially
developed economies tend to have several such markets for buying and
selling foreign exchange at some of the future dates.
In the context of IRP, a dollar invested in foreign bonds will yield (1 +
RF) after one period in the foreign currency. To eliminate the risk of
exchange rate changes during the investment period, the investor can
contract now to sell the end-of-period receipt of (1 + R F) in the foreign
currency at the forward exchange rate. Designate the forward exchange
rate as ρ For and note that the current value ρ of the exchange rate is the
spot exchange rate.
For an investor to be indifferent between investing at home or abroad, and
under the assumption that α equals zero, we need:

whose approximation is:

where ρ′′ For = (ρ For – ρ)/ρ For. That if, for covered IRP, we replace in the
IRP condition the expected exchange rate ρ e by ρ For and ρ ′′e by (ρ For – ρ
)/ρ For. This gives the covered IRP condition as:

The difference between the original, uncovered IRP condition and the
covered one is that in the former, the expected exchange rate change (p′′e)
is not known, so that there is an exchange rate risk in making foreign
investments, while, in the covered IRP, the forward rate is known, so that
there is no exchange rate risk involved in covered foreign investments.
Effectively, there exists covered IRP if the investors protect themselves
against exchange rate fluctuations through forward contracts and
uncovered IRP if they do not do so. For covered IRP to exist in the foreign
exchange markets, there must be forward markets with zero or relatively
low transactions costs. While these normally exist to some extent in
economies with well-developed financial markets, they do not usually
exist for most of the LDCs, which do not have organized and competitive
forward markets for foreign currencies.

570
12.2.2 IRP as a theory of the exchange rate
To turn the benchmark IRP into a theory of the exchange rate, we used the
following procedure in Chapter 3. The simplified version of IRP (under the
continuing assumption that α = 0) can be written as:

Hence, R < R F implies that ρ ′′e > 0. That is, if the central bank lowers the
domestic interest rate below the foreign one, the public must expect an
appreciation of the domestic currency against the foreign one. This does
seem counterfactual, since everyday experience indicates that a decrease in
the domestic interest rate is usually followed by a depreciation of the
domestic currency. The reconciliation between the theoretical and
empirical observation is often explained in the following manner. Given
perfect capital flows and the cut in the domestic interest rate, foreign
bonds will pay a higher return than domestic ones. This will immediately
translate into net capital outflows sufficient to cause a big depreciation of
the country’s exchange rate. From this low-exchange rate, the currency is
expected to appreciate. For the expected appreciation to be validated, the
IRP condition requires that the initial exchange rate depreciation must
overshoot the eventual one.
Figure 12.1 illustrates the preceding scenario. It assumes that the
domestic and foreign interest rates are initially equal and the exchange rate
is stable at ρ 0, without a change in the demand and supply of foreign
exchange rate. At time t0 , the central bank lowers the domestic interest
rate below the foreign one. Investors react by sending abroad a sufficient
amount of capital to immediately to drop the exchange rate to ρ 1. From
this low level, the exchange rate gradually appreciates,8 as required by
IRP, so that it will eventually return to its initial value of ρ 0. This
appreciation would be expected by investors, so that the expected
appreciation implied by the IRP condition occurs, as if instantly. Note that
the drop in the exchange rate at t0 is excessive relative to its eventual level.
This excessive depreciation is called ‘overshooting ’ relative to the
eventual exchange rate.

571
The inconsistency between the IRP’s implications and common
observations
The scenario shown in Figure 12.1 asserts that the cut in the domestic
interest rate relative to foreign ones will immediately produce a sudden
depreciation from ρ 0 to ρ 1, followed by a gradual appreciation (i.e., ρ′′ >
0) for quite some time. This implication runs counter to the frequent real-
world observation that a cut in the domestic interest rate by the central
bank is not accompanied or followed by an appreciation but is followed by
depreciations (i.e., ρ′′ < 0) for quite some time (several quarters). While
this observation might be explained by assuming that the exchange
markets are imperfect and slow to adjust the exchange rate in response to
shifts in the demand and supply of foreign exchange, exchange rate
markets in the financially developed economies are known to be very
efficient and adjust the exchange rate on a continuous basis. Therefore, our
explanation, offered below, for the persistence in exchange rate
movements does not rely on the claim that the capital markets are not
efficient and are slow in adjusting the exchange rate to demand and supply
shifts.
Our preferred main explanation for the observed relationship (i.e.,
depreciations following an interest rate cut) relies on:
• Capital flows are not only for investment in short-term bonds but also for
investment in medium- and long-term bonds, as well as shares of firms.
The return on shares is never very certain. While the current interest rate
on bonds is known, future changes in interest rates are uncertain and can
cause capital losses or gains from investment in bonds. Further, since
future movements of exchange rates are uncertain, there is usually an
exchange rate risk in investments in foreign bonds and stocks.
• Foreign exchange markets are efficient in the sense that they equate
demand (orders to buy) and supply (orders to sell) instantly.
• Slow adjustments of portfolios by investors due to inertia, brokerage
costs, and the uncertainty of yields and future bond/stock values.

572
 Following a reduction in the domestic interest rate, investors do not
immediately make their long-run desired switch from domestic bonds and
stocks to foreign ones but do so gradually.9 Some investors adjust their
portfolios immediately, then others and then still more, and so on. Further,
even for a given investor, the uncertainty about the future course of
interest rates and exchange rates often causes the investor to shift his/her
funds gradually rather than in an abrupt manner to the new optimal long-
run composition of his/her portfolio, which is never definite anyway.
Overall, as the domestic interest rate is cut relative to the foreign ones, the
capital markets see a slow and cautious switch from domestic bonds to
foreign ones.10 The resulting gradual net outflow of funds will, under the
assumption of efficient capital markets, only gradually reduce the
exchange rate. Therefore, a cut in the domestic interest rate will usually be
followed by exchange rate depreciations for some time (i.e., ρ ′′ < 0).
Rational investors will anticipate this pattern of depreciations, so that they
will expect exchange rate depreciations (i.e., ρ′′e < 0), rather than the
appreciation (i.e., ρ′′e> 0) implied by the IRP theory. These expectations
of exchange rate changes will in fact be substantiated by the resulting
movement in the exchange rate.
• The dependence of the exchange rate on the expected one and on the
actual pattern of adjustments in the expected exchange rate value.
In the real world, expectations on exchange rates are not dictated by the
IRP theory but at least to some extent may be determined independently of
it. Among these determinants is the past experience on the non-random
part of exchange rate movements. This experience indicates persistence of
exchange rate movements (i.e., depreciations usually followed by
depreciations and by appreciations usually followed by appreciations) and
expectations on future capital flows.
The depreciations may accumulate to an amount such that the exchange
rate overshoots the long-run relationship asserted by IRP between the
domestic and foreign interest rates. However, given the slow adjustments
of portfolios by investors, any overshooting may occur not all at once but
over several quarters. Note that the IRP theory implies that an
overshooting depreciation phase will be followed by appreciations.
The different phases (depreciations, overshooting of the depreciations,
and eventual appreciations) do seem to occur in the real world, since it is
commonly observed that certain currencies do continue depreciating for
quite some time, that these depreciations continue to a level which seems

573
to be excessive/unjustified in terms of the performance of the domestic
economy relative to foreign ones, and that they eventually begin to
appreciate. An example of this pattern is provided by the Canadian dollar,
which fell against the US dollar from about US$0.80 in 2002 to as low as
US$0.62 in mid-2004, then climbed to about US$0.85 by the end of 2005
and to US$0.92 by mid-2006. It then fell to about US$0.85 in early 2007,
but then shot up in late 2007 and touched US$ 1.08, before falling to about
parity (i.e., one for one) in December 2007. The Canadian dollar fluctuated
between 0.80 US cents and 0.90 US cents during much of 2009. The Euro-
US$ exchange rate followed a somewhat similar pattern.

12.2.3 The impact of expectations on exchange rates

Rearrange the benchmark-covered IRP equation in the following manner:

where ρ′′e was our compact symbol for ( ρ e – ρ )/ ρ e, so that ρ′′e = ( ρ e –

ρ )/ ρ e = 1 – ρ / ρ e. Hence,

Hence, the current exchange rate not only depends on the interest rate
differential between the interest rates at home and abroad but also depends
positively on the expectations of the future exchange rate. One of the
determinants of ρ e is the interest rate differential (R – R F). If the rise in R
increases (R – R F), the expected exchange rate would increase, which
implies that ρ will rise, rather than fall, as implied by the IRP condition.
The expected exchange rate can clearly differ from the existing one. The
expectations of future exchange rates can be volatile and sometimes shift
rapidly. Suppose that investors suddenly form the expectation that the
exchange rate is going to depreciate. This raises the expected return on
foreign investments and sets up an outflow of funds, leading to the
potential for a balance of payments crisis and a decline in the actual
exchange rate to match the expected decline in it. If the central bank wants
to avert this, it will have to raise the domestic interest rate drastically to
compensate for the expected depreciation. However, sometimes, even such
a rise in interest rates does not avert the foreign exchange outflows, with
the consequence that the exchange rate would depreciate and continue to
depreciate for some time.

574
12.3 PPP and IRP Combined
PPP and IRP represent two equations for deriving the exchange rate. They
are:

Long run analysis

For the long run, including the assumption that the errors in expectations
are zero, we have ρ′′e = ρ ′′. This allows us to combine the IRP and PPP
conditions, which yield:

Note that this is a long-run condition. It shows that, ceteris paribus,

• An increase in the foreign interest rate will force the domestic one to rise.
• An increase in the inflation differential (π – π F) will raise the domestic
interest rate. The intuitive explanation for the effect of the inflation
differential on the domestic interest rate is that higher domestic inflation
will lead to a depreciation of the exchange rate,11 so that the domestic
interest rate has to be higher than the foreign one to compensate for this
depreciation.
Short run analysis
Short-run analysis takes k to be a constant, so that k′′ equals zero.
Therefore, assuming ρ′′e =ρ′′ , the short-run determination of the domestic
interest rate under flexible exchange rates is given by:

This equation can be rearranged as:

Hence, in the short run,

Further, in the short run:

• The domestic nominal return will rise relative to the foreign one only if
the domestic inflation rate rises above the foreign one.
• For a small open economy, since the foreign interest and inflation rates

575
 are taken as given, the interest rate differential can be changed by altering
the domestic inflation rate — for instance, by an expansionary or
deflationary monetary policy.
For most countries, the cross-country differentials in interest rates do not
always differ by a constant from the differentials in inflation rates for
daily, quarterly, or even annual data.12 There are several reasons for the
departures from the implications of the joint PPP and IRP theories: ρ′′e
does not always equal ρ′′, the indices for inflation include the prices of
non-traded goods and investors adjust their portfolios gradually because of
uncertainty. Further, for some countries, capital markets are not perfect for
international capital flows. While the above relationship does not apply in
a precise form in real-world economies, its following two major
implications are important and valid on average.
(i) An increase in the domestic inflation rate tends to raise the domestic
interest rate. Note that this conclusion also comes from the Fisher
equation on the nominal versus the real interest rate, so that there is a
very strong basis for it.
(ii) An increase in the foreign interest rate tends to raise the domestic
interest rate13 In practice, this conclusion is especially likely to apply to
small developed economies whose net inflows of capital are quite
sensitive to interest rate differentials.

12.4 The Market for Foreign Exchange, a

Review
Chapters 3 and 5 had presented the analysis of the foreign exchange
market. Both the demand and supply of foreign exchange are functions of
the exchange rate as well as of other variables. If the exchange rate is
permitted to be completely flexible and the foreign exchange market is
efficient and stable, the exchange rate adjusts to equate the demand and
supply of foreign exchange.
We supply our dollars in the foreign exchange market when we buy
foreign goods (our imports) or foreign financial assets (for which we
‘export capital’) and need to pay for them.14 The total payments for these
sum to the supply of our currency (S $) in the foreign exchange markets.
That is, the supply of our dollars (S $) arises when we import commodities
(Z c) or export capital (X k).
Correspondingly, foreigners demand our currency, so that they can buy

576
our commodities and financial assets and make unilateral payments to us
on a net basis. In other words, the demand for our dollars (D $) in the
foreign exchange market arises from foreigners’ purchases of our
commodities (our exports Xc ) or of our stocks and bonds (which constitute
our ‘capital imports’ Z k) or because of net unilateral inflows of funds from
abroad.15 The net unilateral flows consist of two items: net interest and
dividend payments NR to us (in excess of those we have to pay foreigners)
and net gifts and donations NT (in excess of those we give foreigners).
Therefore, the supply and demand of our dollars in the foreign exchange
markets are specified by:

S$ = Supply of our dollars in foreign exchange markets

= Nominal value in dollars of our purchases of foreign commodities
+ Nominal value in dollars of our purchases of foreign assets +
OR + OT (18)
= Zc (ρ r, y ) + X k(R , RF, ρ′′e) + OR + OT.
++-+-

D $ = Demand for our dollars in foreign exchange markets(19)

= Nominal value in dollars of foreigners’ purchases of our
commodities
+ Nominal value in dollars of foreigners’ purchases of our
assets + IR + IT
= X c(ρ r, y F) + Z k(R, R F, ρ′′e) + IR + IT,
–+ +–+

where:
X c = value in dollars of exports of commodities (goods and services),
X k = value in dollars of capital exports,
Z c = value in dollars of imports of commodities (goods and services),
Z k = value in dollars of capital imports,
IR = interest and dividend inflows in dollars,
OR = interest and dividend outflows in dollars,
IT = unilateral transfers (gifts and donations) to the domestic economy
from abroad, and
OT = unilateral transfers (gifts and donations) out of the domestic
economy.
A positive (negative) sign under a variable means that an increase in the

577
value of that variable increases (decreases) the value of the variable on the
left side of the equation. The determinants of imports and exports have
been inserted in the above equations.
An appreciation in the nominal exchange rate ρ means an increase in the
real exchange rate ρ r, which makes our goods more expensive relative to
foreign ones. Assuming that both exports and imports of commodities are
elastic16 with respect to price changes, increases in our prices cause X c
and D $ to decrease, while Z c and S $ increase. Capital flows depend on
the return R on domestic investments relative to the expected return (R F –
p′′e) on foreign investments, so that X K and Z K depend on R , R F and ρ′′e
. An increase in either the domestic interest rate R or an increase in the
expected rate of appreciation of the domestic currency increases the net
inflows of capital.
Fact Sheet 12.2: Interest Rates and Net Capital Flows in the USA, 1985–
2008
This Fact Sheet uses US data to provide some evidence on the
dependence of net capital flows on the interest rate. It shows that net
capital inflows (measured by the right-hand scale) are positively related
to domestic interest rate (measured by the left-hand scale) over some
periods but not over others.
There is a positive relationship between these variables since about
1997, but no clear-cut relationship before 1997.
Economic theory implies that capital flows would depend on the
interest rate differentials among countries and not merely on the
domestic interest rate. Since the USA is the world’s largest economy,
changes in its interest rate are often soon followed by those in other
countries, so that the assumption of ‘all other things being the same’
(meaning constant foreign rates) would not be valid in this case. Further,
capital flows depend on expectations of exchange rate depreciations and
appreciations, and a host of other factors, including the stage of the
business cycle among countries, riskiness of investment across
countries, political events, etc.

578
12.4.1 The relationship between the balance of payments
and (D $ – S $)
Chapters 3 and 5 had presented the definition of the balance of payments.
The balance of payments B is the amount of the net inflows (receipts less
outflow) of foreign exchange into the economy. These net inflows,
evaluated in dollars, equal the demand for our dollars by foreigners less
the supply of our dollars in foreign exchange markets. This amount also
equals the change in our foreign exchange reserves, which is measured in
dollars by R $. That is,

12.4.2 Diagrammatic analysis

If the nominal exchange rate rises (so that our currency appreciates),
foreign goods become relatively cheaper (and our goods become relatively
more expensive), so that our quantity of imported commodities increases.
With the elasticity of imports assumed to be greater than unity, the
(nominal) expenditures on our imports also increase, so that our supply of
dollars in the foreign exchange market increases. Hence, the supply curve
of our dollars in foreign exchange markets has a positive slope.
Conversely, as the nominal exchange rate declines (i.e., depreciates), our
commodities become relatively cheaper for foreigners, so that the quantity
of our exports increases. With the elasticity of exports assumed to be
greater than unity, foreigners’ expenditures on our exports also increase, so
that the demand for dollars in the foreign exchange market rises. Hence,
the (foreigners’) demand curve for dollars has a negative slope. Therefore,
the demand curve for our currency in the foreign exchange markets has a

579
negative slope while the supply curve has a positive one, as shown in
Figure 12.2a.
Figure 12.2a shows the demand curve D $ and the supply curve S $, with
the nominal exchange rate ρ on the vertical axis and the quantity Q $ of
dollars exchanged against foreign currencies on the horizontal axis. Net
capital inflows (Z k – X k) plus (NR + NT) are held constant in drawing the
curves in this diagram. The curves shown assume that the elasticities of
both imports and exports are greater than unity. The equilibrium value of
the nominal exchange rate is ρ* . In Figure 12.2a, if ρ > ρ *, as at ρ 2, there
exists an excess supply of dollars (the domestic currency) in the foreign
exchange market, which leads to the depreciation of its exchange rate. This
lowers the exchange rate to ρ *. But if ρ <*, as at ρ 1, there exists an excess
demand for our dollars in the foreign exchange market, which leads to
their appreciation, thereby pushing the exchange rate to ρ *. Hence, the
equilibrium exchange rate ρ * is stable. This stability is indicated by the
direction of arrows toward the equilibrium exchange rate.
Figure 12.2a has several limitations. These are:
• It only focuses on the impact of exchange rate variations on commodity
flows, while holding the interest rates and the expected future exchange
rate constant under the ceteris paribus clause since these variables are not
on one of the axes of the figure. As a consequence, this figure keeps
capital flows constant (or insignificant) and gives the erroneous
impression that the changes in the commodity flows alone determine the
movements in the exchange rate.
• The foreign exchange markets are financial markets that adjust the
exchange rate on a daily basis to shifts in demand and supply, so that the
shifts in capital flows — under interest rate parity — become the main
short-run determinant of exchange rate changes. This is particularly so in
the modern period in which capital markets have become increasingly
globalized.
• The responses of commodity imports and exports to exchange rate
changes are subject to significant lags.

580
• As pointed out earlier, the assumption underlying the curves in Figure
12.2a is that the quantities of both the exports and imports of the country
are elastic with respect to exchange rates and prices. Note that this
assumption does not always hold, as discussed in the next section.

12.5 Demand and Supply Elasticities

The slopes of the demand and supply curves shown in Figure 12.2a
assume that our imports and exports are elastic (i.e., with their elasticity
being greater than one) with respect to changes in the real and nominal
exchange rates. This means that, as our exchange rate depreciates and
foreign goods become more expensive relative to ours, the expenditures
(and not just the quantity) on our imports decrease and those on our
exports increase.17 However, this need not be so for all countries at all
times. Even if it applies in the long run for a given country, it need not be
the case in the short term while the buyers and suppliers take some time to
make adjustments to their purchasing and production patterns as a
consequence of a change in the exchange rate. Since our main concern
here is with the value of net exports rather than with the values of imports
and exports separately, we posit the following two scenarios.
1. An inelastic short-term pattern where the elasticities of imports and
exports are such that the value of net exports decreases in response to a
depreciation.
2. An elastic long-run pattern in which the elasticities of imports and
exports are such that the value of net exports increases in response to a
depreciation.
Figure 12.2a shows the long-run pattern. An extreme version of the
short-term pattern, in which both imports and exports are inelastic, would
have D$ sloping upwards and S $ sloping downwards. This case is shown
in Figure 12.2b.
To compare the stability of the exchange rate in the Figures 12.2a and

581
12.2b, suppose that a disturbance causes the exchange rate to fall to ρ 1
below ρ * in Figure 12.2a. Excess demand exists at ρ 1. The excess
demand for dollars below ρ * forces an increase in their value until
equilibrium is restored at ρ * through an exchange rate appreciation.
However, in Figure 12.2b, an accidental fall in the exchange rate to ρ 5
will cause an excess supply of dollars, which will further lower their value
— that is, cause a depreciation to ρ6 — so that the market adjustments will
lead to a movement away from equilibrium. Conversely, in Figure 12.2b,
an accidental exchange rate appreciation to ρ3 will cause a further
appreciation to ρ 4, rather than a return to the equilibrium rate ρ *. Hence,
while the equilibrium exchange rate is stable in Figure 12.2a, it is unstable
in Figure 12.2b. Therefore, depending upon the price elasticities of imports
and exports during the relevant period, a depreciation of the exchange rate
can cause either a decrease or increase in the net supply of foreign
exchange.

12.6 The J Curve for the Value of Net Exports

in Real Time
There are considerable disputes in the literature about the elasticity versus
inelasticity of net exports for the industrialized economies in the short
term. Some empirical studies do support such a possibility while others
deny it. Intuitively, the impact or short-term elasticities for many
commodities are often lower for some time, say for a year or more, than
their long-run ones. There usually exists, some household inertia and lags
in switching consumption between domestic and foreign goods, especially
when there is product differentiation. There are usually also lags in
switching production. In some, perhaps most cases, the short-term
elasticities of net exports can be less than unity for a given economy for
some quarters, so that a depreciation will worsen the balance of trade —
i.e., reduce the value of net exports — in the short term, even though this
balance eventually tends to improve in the long run.
Following a depreciation of the domestic currency, if the value of net
exports initially falls and then rises, the net nominal exports follow a
pattern resembling a J.18 This pattern is called a J curve and is applicable
to many countries. Such a curve is shown in Figure 12.3a. In this figure,
assume that the net exports were negative and there was a balance of
payments deficit at the pre-existing exchange rate ρ 0 at time t0. This

582
balance of payments deficit causes the exchange rate to depreciate (say to
ρ 1 at time t 0), which leads to the time-path of the (value of) net exports
shown by the curve marked net nominal exports: these fall for some time
and then increase. That is, the elasticity of net exports is initially less than
unity but increases over time to become greater than unity. Note that this
figure shows the time path of net exports following a single specific
depreciation. However, while net exports are falling, the balance of
payments worsens19 and causes further deprecations of the exchange
rate,20 so that the unstable market behavior induces a cumulative string of
depreciations, before net exports do pick up and reverse some of the
depreciations.
The corresponding figure for the exchange rate appreciations is Figure
12.3b, with an ‘inverted J curve’. For this case, start at time t0 with a
balance of payments surplus at the pre-existing exchange rate ρ 0. This
surplus causes an appreciation of the exchange rate (say to ρ2) at time t 0.
In the short term, this appreciation increases the value of net exports,
which leads to further appreciations. There occurs, then, a cumulative
string of appreciations, with successive shifts in the J curve. Note that
Figure 12.3b is drawn to show only the impact of the initial appreciation
on net exports. Eventually, net exports do fall and the overall appreciation
is partly reversed through depreciations.

Hence, the J curve analysis indicates the possibility that the exchange
rate depreciations and appreciations will exceed — overshoot — their
long-run amounts because of the short-term inelasticity of the net exports
of commodities.
Under flexible exchange rates, the existence of a J curve for a country
poses two types of problems in the short run:
1. A shock that appreciates the domestic currency will be followed for
some time by further appreciations, with the trade balance developing a

583
 larger surplus, i.e., improving rather than worsening under the impact of
the appreciation for some time. Conversely, a shock that depreciates the
domestic currency will be followed for some time by further
depreciations, with the trade balance worsening rather than improving
under the impact of the depreciation.
2. The cumulative movements in exchange rates and the trade balance are
likely to trigger supporting speculative flows of capital, thereby
increasing the former’s impact on the balance of payments and
intensifying the extent and duration of the exchange rate movements, as
well as surpluses or deficits in the balance of payments.
This analysis provides one of the reasons for:
• The observed persistence of trade deficits and surpluses even under
flexible exchange rates.
• The observed overshooting of the exchange rate, with depreciations
followed by further depreciations and appreciations followed by further
appreciations over the short-term, relative to its long-run value.

12. 6.1 Limitations of the commodity-based exchange

market analysis
Note that the preceding commodity flow analysis of the foreign exchange
market assumes that capital flows are either insignificant, which applies to
few economies nowadays, or that we can hold the interest rates, expected
exchange rate changes and hence capital flows constant under the ceteris
paribus assumption while the value of net commodity exports takes its
time to adjust in response to disturbances in the exchange rate. But the
faster pattern of dynamic capital market responses often dominate the
slower commodity flow responses and can turn a case of stability into an
actual case of instability in the overall foreign exchange market, at least
for some time — and vice versa. These capital flows are even more likely
to accentuate a case of the basic instability, due to the J curve arising out
of the commodity induced foreign exchange flows. In such a context, the
central banks are often tempted to intervene in the foreign exchange
market by buying or selling, as required, foreign currencies to try to
stabilize the exchange rate of their currency against what they perceive to
be short-term destabilizing influences.

12. 6.2 Policy implications of the J curve

If a country needs to depreciate its currency to correct a balance of

584
payments deficit and has a J curve, it needs sufficient foreign exchange
reserves to ride out the worsening of the balance of payments during the
early parts of the J curve. Alternatively, it could borrow foreign exchange
from other countries or international financial institutions. In addition, it
could raise its interest rate to boost capital inflows.
Extended Analysis Box 12.1: The General Experience of Industrialized
Countries and LDCs on the J Curve
The J curve analysis helps to explain the observed persistence and
swings of exchange rate changes among countries. These depend on the
duration of the declining part of the J curve, which is determined by the
composition and nature of the country’s exports and imports, their
substitutes in consumption and/or production, and upon the supply
conditions.
Even industrialized countries can possess a J curve for some time
following a depreciation of their currency. The reasons among these
include inertia in changing demand patterns between home and foreign
products, production and marketing delays, imperfect competition and
deviations from PPP, etc. Some empirical studies have supported the
existence of such a curve, even for such advanced economies as those of
the USA, Canada, Germany, and Japan.
Many developing countries export traditional types of products with
low elasticities of demand in foreign markets and low supply elasticities
at home. Their imports normally consist of industrial goods, which in
general have imperfect domestic substitutes with limited production, so
that the imports on the whole also have a low elasticity of demand. This
scenario corresponds to the unstable case shown in Figure 12.2b for the
foreign exchange market. Hence, these countries often face the
possibility of the trade balance worsening in response to a depreciation
of their currency and doing so for a longer period than for the
industrialized countries. For many of the developing economies,
attempts to eliminate the trade deficit and that in the balance of
payments through a devaluation/depreciation of the currency often
increase, rather than decrease, these deficits, at least for a significant
period. As pointed out above, this requires that
devaluations/depreciations have to be accompanied at least for some
time by selling foreign exchange from the country’s reserves or from
loans from other countries to support the balance of payments and the
exchange rate. They also often require other policy measures — such as
contractionary monetary and fiscal policies to decrease domestic
aggregate demand, import controls, etc. The long-run solution has to

585
rely more on increasing net exports and higher net capital inflows.
The declining part of the J curve also implies that any
devaluation/depreciation of the currency needs supportive monetary,
fiscal, and other policies to prevent cumulative
devaluations/depreciations. This need would depend upon the shape of
the J curve and differ among countries. The existence of the J curve also
points to the need of a country faced with a depreciation or devaluation
to have either adequate foreign exchange reserves or access to a line of
credit from the IMF and other countries, as a way of riding out the
declining part of the J curve.
Implications of the above analysis for movements of exchange rates
under flexible exchange rates
The preceding analyses of trade flows implied that, in the absence of
significant capital flows or constant ones,
• If the elasticity of net exports is greater than unity (as in the stable case
depicted in Figure 12.2a for the foreign exchange market), the
exchange rate moves toward its equilibrium value. If shifts in exports
and imports were to produce a balance of payments deficit (surplus),
the exchange rate would depreciate (appreciate) to restore equilibrium
in the balance of payments. If this return to equilibrium is sufficiently
rapid, the central bank need not interfere in the foreign exchange
market by buying or selling foreign exchange.
• If the elasticity of net exports is less than unity (as in the unstable case
depicted in Figure 12.2b for the foreign exchange market) the exchange
rate moves away from its equilibrium value. If shifts in exports and
imports were to produce a balance of payments deficit (surplus), the
exchange rate would still depreciate (appreciate) but this would further
worsen the deficit (increase the surplus) in the balance of payments. If
this tendency is sufficiently powerful, the central bank may have to
undertake offsetting sales (purchases) of foreign exchange to stop the
series of depreciations (appreciations) from continuing.
The standard assumption of macroeconomic analysis is that the foreign
exchange market is stable (first bulletted point) in the long run. Further,
many economists assume that this stability also applies in the short term.
If the market is unstable for the impact period of the
devaluation/depreciation, the market returns to the stable case within an
acceptable period and at an acceptable loss (if any) of foreign exchange
reserves. That is, the assumption of the elasticity of net exports being
greater than unity is reasonable for analytical and policy purposes. This

586
 is more likely to be valid for economies that mainly export homogenized
industrial and agricultural products than for ones exporting specialized
products in segmented markets.
The general view among economists is that it is acceptable for the
central bank to offset random and transitory variations in its exchange
rate. However, it is difficult to offset fundamental increases in the
demand for foreign exchange or decreases in its supply — since these
will cause continuing balance of trade deficits — by the (continuing)
sales of foreign exchange from the country’s foreign exchange reserves
or from funds borrowed abroad. In this case, it is better to allow market-
determined fluctuations in the exchange rate and, if needed, limit the
potentially cumulative depreciations imposed by the J curve through the
pursuit of various policies — including restrictive monetary policies —
at the command of the central bank. This policy pattern is that of
managed exchange rates within a flexible exchange rate regime.

12.7 Historical Experience with Flexible Versus

Fixed Exchange Rates
Whether a country has a stable or unstable foreign exchange market in real
time is an empirical question and generally one whose answer varies with
the circumstances. Up to the late 1960s, the common belief in economics
was that the foreign exchange markets tended to be unstable, so that it was
in general preferable to have fixed exchange rates. This belief was
embodied in the rules and recommendations of the IMF, with strong
support from the USA from the founding of the IMF in 1945 to about the
early 1970s. However, many of the major industrialized economies,
especially that of the USA, suffered systematic imbalances in their balance
of payments toward the end of the 1960s. Given these imbalances,
combined with the speculative runs on several currencies, it proved
impossible to maintain fixed exchange rates among currencies, and
between the US dollar and gold.21 In the early 1970s, the economics
profession generally began to recommend flexible exchange rates and the
IMF and the USA became strong advocates of flexible exchange rates.
In the years since the early 1970s, most of the world’s largest trading
nations have had flexible exchange rates22 and unhampered flows of
financial capital, without seemingly excessive exchange rate volatility,
though bouts ofsevere volatility have occurred for several emerging and
transitional — mainly East European countries in the transition from

587
communism to capitalism — economies. In general, this period has been
one of the very fast growths in the international flows of both commodities
and capital, accompanied by very rapid technical change (such as shifts in
labor productivity and innovation) and other shifts in the comparative
advantage of countries. These changes constitute potential pressures for
changes in the exchange rates. Such pressures have been accommodated
well by the flexible exchange rate system, while a fixed exchange rate
system would most likely have created severe imbalances in the balance of
payments of countries, which, reinforced by speculative and chaotic runs
on the exchange markets, would have forced periodic changes in the fixed
exchange rates.23Consistent with this experience, many economists
believe that the flexible exchange rates provide greater stability in practice,
especially in periods of significant shifts in the underlying determinants of
the demand for and supply of foreign exchange.
However, the Latin American and Asian exchange crises during the
1990s seem to indicate a high degree of volatility in the exchange rates of
individual countries. Part of the reason for this is that the extent of
currency speculation that can potentially occur into or out of a currency
can be a very significant proportion of the country’s trading flows of
foreign exchange and of its reserves. This is especially so for the smaller
developing economies. In such cases, a speculative run into or out of the
country’s currency has the potential of creating large fluctuations in its
exchange rate.
Economic theory and historical evidence do not make it unambiguously
clear whether a flexible or a fixed exchange rate regime acts as a greater
disincentive to such runs and/or contains them better. In general:
• Flexible exchange rates are better than fixed exchange rates at
accommodating long-run shifts in the foreign exchange markets related to
fundamental changes in commodity flows — e.g., from differential
developments of technologies, different inflation rates, etc. — and to
fundamental changes in the pattern of capital inflows resulting from shifts
in the relative profitability of investment among countries.
• But flexible exchange rates and high capital mobility open the door to
more frequent and potentially severe speculative runs into and out of
individual currencies — especially for countries with unstable political
systems and weak financial markets.

12.8 Measures to Reduce Balance of Payments

Deficits
588
Remember that:

There exists a balance of payments deficit — i.e., an outflow of foreign

exchange — if R $ < 0. The measures that can be undertaken to reduce a
deficit are:
• Assuming the elasticity of net exports to be greater than unity, devalue
the exchange rate (if it is a fixed one), or allow it to depreciate (if it is a
flexible one).
• One of the components of the balance of payments is net exports, (Xc – Z
c), which — assuming the elasticity of net exports to be greater than unity
— increase if domestic prices fall relative to foreign prices. Net exports
also increase if domestic income falls and reduces imports. These require
a fall in aggregate demand. Therefore, domestic deflationary policies,
such as an increase in the interest rate (accompanied by the appropriate
money supply decrease) or a reduction in the fiscal deficit (or increase in
the fiscal surplus), can be employed to reduce domestic aggregate
demand and domestic prices, which will increase net exports and reduce
the balance of payments deficit.
• Another component of the balance of payments is net capital exports, (X
k – Z k), which depend on domestic interest rates relative to foreign ones
and on the expected rate of change of the nominal exchange rate.
Therefore, raising the domestic interest rate, through monetary policy,
can increase domestic interest rates, which induce increases in capital
inflows and reduce balance of payments deficits.
• Other measures that can be used to reduce a balance of payments deficit
include increases in tariffs and/or reductions in import quotas,24 which
reduce imports and increase net exports. However, the country’s
maneuvrability to do so is usually subscribed by its international trade
agreements, such as the North American Free Trade Agreement
(NAFTA) and World Trade Organization (WTO) rules.
• Countries can also try to reduce their net interest payments to foreigners
(i.e., increase NR). These can occur through readjustments of the interest
rate or in the payment schedules related to foreign debts through
renegotiations with foreign creditors of the outstanding external foreign
debt. In a few cases, especially those relevant to the poorest countries, it
may occur through ‘debt forgiveness’ by the foreign creditors, especially
the IMF and foreign governments.
• Further, countries can try to increase their unilateral inflows of transfers
(NT), e.g., by pursuing policies that promote higher remittances by

589
 expatriates living abroad.
From a long-term perspective, what is needed to curb balance of payments
deficits is increases in the productive efficiency of the economy and an
improved ability to compete in foreign markets. These need to be
supplemented by measures to increase the attractiveness of the country for
capital inflows. This requires both higher domestic returns and greater
security of capital inflows, as well assurances to foreign investors that they
would be able to repatriate their profits and capital.

12.9 The International Monetary Fund (IMF)

National policies on exchange rates have to be formulated in the context of
the existing international financial system and arrangements. A core
element of this system is the IMF.
The IMF was set up under an international agreement reached at Bretton
Woods, New Hampshire, in 1945, and started functioning in 1947. Its
main mandate was to be an international financial institution to oversee the
member countries’ payments arrangements and exchange rate policies, and
to provide short- term financial assistance to members with balance of
payments problems. Member countries were required to maintain fixed
exchange rates and to devalue their currencies only in the case of a
fundamental disequilibrium in their balance of payments and only after
prior consultation in a specified manner with the IMF. The underlying
support to this system of fixed exchange rates was provided by the
decision of the USA to be on the gold exchange standard by maintaining a
fixed exchange rate for its dollar against gold (at US$35 to an ounce of
gold) and convertibility of the dollar against gold. This system came to be
known as ‘The Bretton Woods system’ .
The imbalance in trade balances, especially large US deficits leading to
turbulence in the foreign exchange markets, in particular the speculative
runs on the US dollar at the end of the 1960s and early 1970s, led the USA
to abandon the gold exchange standard in 1971 and let the US dollar float
against gold. Further, in 1973, the European countries allowed their
currencies to float against the US dollar. In 1974, the IMF adopted
guidelines promoting managed floating exchange rates among its member
countries. The encouragement and management of flexible exchange has
since then been a central tenet of IMF policies. These changes meant an
end to the Bretton Woods System.
Central to the IMF’s role with respect to exchange rates is its mandate to
exercise ‘surveillance’ over its members’ exchange rate policies and their

590
macroeconomic and related structural policies. It promotes policies aimed
at the elimination of balance of payments deficits, the creation of efficient
and sound financial institutions, and limits on fiscal deficits, etc.
In recent decades, the IMF has been a strong supporter of the
unrestricted international flows of both goods and capital, considering
these to be essential for continued economic growth and prosperity. It uses
its influence and powers to promote liberal trade practices and the free
movement of capital among countries. Its power to influence a country’s
policies becomes especially effective when the country runs into severe
balance of payments problems and needs large loans from the IMF.25

12.9.1 The IMF and disequilibrium in the balance of

payments of the member countries
A core function of the IMF is to provide short-term loans to its member
countries with balance of payments problems, whether due to negative net
exports or capital outflows. Such loans are often in the form of a ‘stand-by
arrangement’ — that is, a line of credit — and are usually accompanied or
followed by additional loans by other creditors. The stand-by loans are
made subject to certain conditions, which often include an explicit
commitment by the borrowing country to take specific and quantitative
remedial economic measures, including the reduction of fiscal deficits and
money growth rates. The objective of these is to reduce balance of
payments deficits and to make it more feasible to repay international loans
by specified dates. The policies and conditions accompanying the IMF
loans are known as ‘conditionality’ .
Extended Analysis Box 12.2: IMF Conditionality
In providing short-term loans to countries to cover their foreign
exchange shortages, the IMF performs a function similar to that of
national central banks when they act as a lender — at times, as that of
last resort — to their respective commercial banks. While such a role is
needed and of considerable benefit to member countries facing severe
shortages of foreign exchange, it also carries moral hazard. Moral
hazard arises because the availability of a loan or a line of credit reduces
the compulsions and incentives for the borrowing country to contain its
balance of payments deficits by appropriate policies.26 Conditionality is
a device for containing moral hazard, since the loan will be made
contingent on the pursuit of the economic adjustments required to enable
the borrowing country to repay the loan within a specified period —

591
 usually within three to five years. The conditions often take the form of:
• An (often drastic) exchange rate devaluation or the requirement of
floating the exchange rate of the domestic currency.
• Strict limits on fiscal deficits as a proportion of tax revenues or of
GDP.27
• Low monetary growth rates.
• Increased liberalization, internally and externally, of the economy in
the form of reductions in the tariffs and controls on foreign trade and
on domestic industries.
In recent years, conditionality has had an underlying philosophy
advocating free markets, competition, and democratic political
structures, which some countries have resisted. At the international
level, conditionality has consistently taken the form of supporting the
liberalization of import and export flows. The required pursuit of
policies under conditionality can have relatively minor consequences or,
as in the case where the exchange crisis needs for its resolution major
structural changes, have considerable economic, social, and political
repercussions — which the borrowing country may not wish to incur.
Many LDCs and economists have often contended that these
repercussions seem to be ignored by the IMF. Further, it is often argued
that the IMF policies on loans are really for the benefit of the richer
nations, which gain by more prompt payment of their loans, as well as
by the enforced import liberalization in the borrowing countries that
increase the lending countries’ exports.
For whose benefit are the IMF policies?
The IMF, the World Bank and the World Trade Organization (WTO) are
currently the three dominant international organizations of the world
economic system. Their policies are determined by their member
countries and are in practice dominated by their wealthier members,
especially the USA. They have often been accused of catering to the
economic interests of the richer, rather than the economic and social
interests of the poorer, nations. For the IMF, this accusation takes the
form that the conditions imposed on the borrowing countries are not
necessarily beneficial to their economies and often ignore their impact
on the poor in those countries.

12.9.2 The IMF and capital flows

The original Articles of Agreement of the IMF signed in 1945 had allowed
the use of controls by member countries on capital movements. Among the

592
reasons for this was the aim of preserving for the member countries a
degree of control over their monetary and fiscal policies. These controls
also enabled the European countries to decrease or limit their balance of
payments deficits and/or rebuild their reserves from extremely low levels
at the end ofWorld War II. Many countries, including most European ones,
did maintain such controls for several decades after 1945. But, in recent
decades, with the very considerable integration of the national markets for
goods and services into the global economy, and in the current context of
flexible exchange rates, the IMF has strongly supported the liberalization
of capital flows. Among the reasons for it is the belief that obstacles to
capital flows reduce growth.
However, unrestricted capital flows also pose a risk since a major part
ofthem is highly liquid and susceptible to sudden reversals — giving their
most liquid component the name of ‘hot money’. These reversals can be
due to shifts in the fundamental economic determinants of capital flows or
be merely due to non-fundamental factors, such as (unjustified) shifts in
the investors’ perceptions of the riskiness of investment in the country in
question. Reversal in capital flows can also occur because of international
developments, such as the decisions of other countries to raise their
interest rates or because other countries or regions come into ‘fashion’ as
more attractive investment havens. With the considerable increase in
international capital flows, the role of the IMF as a lender has become of
even greater importance than in earlier decades. A country faced with a
capital account crisis usually requires relatively large financial support at a
few days’ or few weeks’ notice. If this strategy or ‘cure’ works, the loan
can be repaid promptly; if it does not, there will be still greater demand for
loans.

12.9.3 Special drawing rights (SDRs)

Under the initial arrangements made in 1945, the IMF was not envisaged
as a central bank with the power to issue its own currency. This changed in
1969 with the creation of the Special Drawing Rights (SDRs) and their
allocation to member countries. The SDRs are an international financial
asset, which can be traded among member countries and used to settle
payments imbalances among them. The IMF has increased the outstanding
amount of SDRs at various times, allocating the increases among the
members in proportion to their quotas in the IMF, which differs from the
open market operations by which national central banks bring about
increases in the monetary base. The total outstanding amount of SDRs is
still a relatively small fraction (about 1.7% in 1997) of the non-gold

593
reserves of the member countries, so that increases in it do not
significantly affect the world money supply and the world price level.
With the replacement in 2001 of several European currencies by the
Euro, a unit of SDRs is now28 a weighted average of four currencies, the
US dollar, the European euro, the Japanese yen, and the British pound,
with its value calculated daily. As a weighted average, its stability is more
than that of any of the individual currencies. Besides serving as a reserve
asset for the member countries, it is also used as a unit of account in many
international transactions and even for denominating private financial
instruments. Countries can peg their currencies to the SDR, if they so
wish, though, so far, few have done so. The SDRs are essentially a unit of
account and their holdings are in the form of deposits. They do not have a
physical existence and are not traded in national markets.
The interest rate charged by the IMF on its SDR loans is also a weighted
average of the yields on specified short-term (mainly three month Treasury
Bills) instruments of the countries (USA, UK, the European Union, and
Japan) whose currencies are used for computing the value of the SDR.
Extended Analysis Box 12.3: The World’s Demand for Liquidity
The need for assets that can serve as an international medium of
payments arises from the world’s transactions need for liquidity to
finance international commodity and capital flows and as precautionary
reserves held by individual countries against unanticipated net outflows
of capital. This need for liquidity can be met by holding foreign
currencies and short-term foreign bonds. It is currently being met mainly
in this manner by the large outstanding supply of US currency and short-
term bonds to the world (non-US) economy. This gives the US Federal
Reserve System the role of providing liquidity to the world economy. A
somewhat similar, but more limited, role is also imparted to the central
banks of other countries, such as Britain, European Union, and Japan,
whose currencies and short-term bonds are widely held by other
countries. However, national central banks are likely to put the interests
of their own economies before the interests of the world economy. If
they are in conflict, the world economy’s interest is likely to
suffer.29Therefore, it is preferable to have a central bank that is
sufficiently remote from individual national interests and can focus more
directly on the interests of the world economy and more impartially
balance the conflicting interest of the individual countries.
Exchange and financial crises
At the international level, capital flows are now large, volatile, and

594
subject to contagious movements, thereby providing the potential for
exchange and financial (i.e., banking) panics. For a country with a
flexible exchange rate, a panic attack will force a sharp and severe
depreciation of the exchange rate. A country with a fixed exchange rate
will usually be forced to devalue. In both cases, the country’s financial
institutions faced with panic withdrawals of funds are likely to become
subject to exceptionally high failures, which can spread to firms in the
production and trading sectors. The resulting fall in domestic liquidity
will contract aggregate demand and output in the economy. All these
effects tend to be excessive and often cumulative if the panic is
unchecked. They can be moderated if the panic is managed through the
provision of emergency financial assistance and management of the
panic by the national central bank with adequate foreign exchange
support provided by a world lender of last resort. The IMF is at present
the only existing international financial institution that might evolve to
fill this role.
Is the IMF a central banker for the world?
The IMF was neither set up to function as a central bank for the world
nor yet fully evolved into this role. It is still much less than a world
central bank since it is not able to control or change the world’s
liquidity, which national central banks can do for their economies. It
does neither issue a widely traded currency nor have much influence on
the world money supply or the total reserves of foreign exchange held
by its member countries. The IMF’s main similarity with national
central banks is in its role as a short-term lender to member countries,
especially during exchange crisis and also to provide them with liquidity
while they tackle their balance of payments problems. This role is
sometimes viewed as that of a lender of last resort for member countries,
but is, in reality, somewhat different from and significantly less than the
extent to which the national central banks can perform this function for
their own commercial banks.
While the IMF has acquired considerable respect — mainly among the
lending countries, but not necessarily among the borrowing countries on
whom it has imposed severe and unpopular conditions — for its
recommendations and considerable leverage over the macroeconomic
policies and financial practices of the countries that seek assistance from
it, these still fall short of the powers exercised by national central banks
over their financial sectors. Further, some of the member countries,
especially the USA, exert greater influence on the world financial
system and on short-term lending to countries with exchange crises than

595
the IMF does.
Does the world economy need a central bank?
Two of the critical roles of a national central bank are the power to issue
a financial instrument that can serve as reserves for private financial
institutions and to be a lender of last resort. In the international context,
the former requires that the world central bank — if the IMF was to
become one — have the power to issue an international financial
instrument that can be part of the foreign exchange reserves of the
individual countries. While the SDRs do serve this role, the ability of the
IMF to create them is severely limited and their current total amount is
grossly inadequate as the main element of the foreign exchange reserves
of the member countries.
Some economists have argued that the limitations on the power of the
IMF to create an international currency or to do so to the required extent
need not prevent its acting, at least partially, as if it were central bank. It
can act to ensure liquidity for a country in crisis by being a ‘crisis
manager’ and standard bearer in approving loans and imposing
conditions, by the coordination of loans from national and private
lenders, and generally by the signals its loans and assessments provide
to the world markets.
The rising demand for the IMF to create a world currency
While the IMF has already created SDRs as an international reserve
currency, their amount as of 2009 is relatively insignificant as a
proportion of the foreign exchange reserves of countries or the world’s
need for an international medium of payments. However, the very
severe US financial crisis starting in 2007 (see Chapter 16) and the
continuing large outflow of dollars from the USA through the large and
continuing deficits in its trade balance, as well as its large fiscal deficits,
have created doubts in the world financial community whether the US
dollar will maintain its value in the coming years. Its depreciation will
mean that its holdings outside the USA will suffer corresponding losses,
so that foreign countries would prefer to hold a more stable unit of
international payments and foreign exchange reserves.30 This has led to
speculation among economic analysts about whether or not the IMF
should create enough SDRs to meet the world demand, or a sufficiently
large proportion of it, for increases in international reserves. Any such
step is likely to represent a depreciation of the US dollar, as well as
probably cause a global foreign exchange crisis. It would also mean very
significant losses on the large amounts of US dollar assets held outside

596
 the USA, so that such a move is likely to be resisted by the USA and
some of the countries with existing large holdings of US dollar assets. In
view of these, any decision to allow the IMF to create assets that can
displace the US dollar assets is likely to be a difficult political one, and
is unlikely to come about easily or soon.

12.10 The International Transmission of Crises

and Business Cycles
Modern economies are closely linked through exports and imports of
commodities and capital (see Chapter 16 on business cycles). Hence, a
recession or a macroeconomic crisis in one large economy, especially that
of the USA, which is the largest one, can affect other economies by
reducing commodity imports from them, as well as reducing capital
exports to them. This linkage is increased by the high degree of
international integration known as ‘globalization’. As an illustration, the
financial crisis that originated in the mortgage market of the USA in 2007
immediately affected financial firms in many countries and its
development into an economic recession in 2008 spread to other countries.

12.11 Economic Globalization

The effective globalization of the market for a single commodity would be
the integration of the national markets for that commodity into a single
one. However, the word ‘globalization’ means the integration of the
national and regional economies as a whole into the world economy. Full
globalization would require the effective unification of all markets on a
world scale, where the relevant markets are those of commodities,
financial capital, and factors of production, including labor. The world
markets and economies have not yet become integrated to this extent, so
that the current stage of globalization is ‘ truncated globalization’. The
past three decades saw a major push for globalization in commodity and
financial capital flows, though not very much in labor flows. In its current
phase, globalization has to be thought of as the process of increasing
integration of national commodity and capital markets into world ones.
The market system provides strong incentives in the form of larger
profits for firms to increase the size of the markets for their output and
inputs, first within the nation and then across nations. The latter requires
the reduction of national barriers (such as tariffs and quotas) to trade,

597
thereby implying liberalization of commodity flows. In the limit, the profit
incentive to expand the market size would result in a single world market
for the firm’s products — i.e., globalization of the market. This expansion
of the market requires: (a) a market-oriented system among countries, (b)
the dismantling of tariffs and quotas, and (c) the reduction of
communication and transport costs.
Historically, in the individual west European countries, the expansion of
the market had been more or less completed through effective political
unification and dismantling of regional barriers to trade by the beginning
of the Industrial Revolution in the late 19th century.31 The spread of
railways, steamships, and the telegraph during the late 19th century
deepened this process within nations by reducing communication and
transport costs. Further, during the 19th century, the incorporation of the
world into the empires of the European nations, promoted this process
across nations, so that some internationalization of trade was already in
place by the beginning of the 20th century. This process was interrupted
by World War I, the rise of protectionism during the Great Depression of
the 1930s during which nations tried to protect their national markets from
foreign competition, and World War II. It picked up momentum after
World War II under the aegis of the General Agreement on Tariffs and
Trade (GATT), which was the precursor of the World Trade Organization
(WTO). Both GATT and WTO set limits to the barriers that member
countries can place on imports and subsidies to exports. However, the
Communist countries that constituted a significant part of the globe were
not members of GATT. Several factors in the last quarter of the 20th
century produced a massive push toward internationalization of trade. One
of these was the collapse of the communist system in several countries,
accompanied by their adoption of market systems of production and
exchange. Countries with mixed economies — i.e., with a mix of socialist
and capitalist systems — decreased their regulations on production and
exchange, as well as privatized many formerly nationalized industries. The
information technology (IT) revolution reduced communication costs and
the reduction in airfreight charges very considerably reduced transport
costs and times. The net effect of these developments intensified the push
toward effective globalization. This was accompanied by the
unprecedented expansion of multinational corporations across the world.
For individual countries, the push for the globalization of commodity
markets means that their international trade (usually measured as the sum
of exports and imports)32 as a proportion of GDP increases substantially.
One estimate of the amount of this trade in 2000 was $6.6 billion per day.
For individual countries, the globalization of financial capital flows means

598
that the inflows plus outflows of financial capital also increase very
substantially as a proportion of their GDP. The spectacular increase in
foreign exchange transactions in the last few decades resulted in about
$1.5 trillion a day of such transactions at the end of the 20th century. In
addition, the transnationalization of production through direct foreign
investment and ownership of plants in different countries by multinational
corporations has increased very substantially. This trend toward
globalization is most evident among the countries of Europe, North
America, Japan, and some of the rapidly developing countries, such as
South Korea and Taiwan. But there are few countries in the world that
have not been affected by it.
The globalization of production and exchange is by no means complete
or even dominant relative to national markets. The high-income countries,
especially the USA, Europe, and Japan, have gone farthest along this route
and been the main beneficiaries of globalization. More recently,
developing countries, especially several Asian ones, have increasingly
joined this movement. Overall, the process of the integration of the
national economies into a global one is still in its early stages and has
much further to go.

12.11.1 Causes of the push for global markets

The forces that have brought about the increasing internationalization of
trade are:
• The reductions in transport and communication costs and delays. Among
the former are the very substantial reductions in airfreight charges.
Among the latter is the IT revolution.
• The self-interest of the world’s major economic and political powers,
which require ever-larger markets for their output and capital. Pre-
eminent among these countries is the USA.
• The strengthening of the IMF, the WTO and the World Bank and their
support for liberalization.
• The implementation of regional blocs and trade treaties, such as NAFTA.
• Political and economic unions, whose most successful instance is the
European Union.
Most countries seem to have benefited considerably by opening up
relatively closed economies to the international forces and the integration
of their economy into the world one. Globalization imparts its economic
benefits for several reasons.
• It increases competition from foreign firms, thereby forcing domestic

599
 firms to adopt the newer technologies (equipment, production techniques,
managerial methods, etc.). This usually results in rationalization of
domestic firms into larger and more efficient ones to compete with
foreign ones.
• It increases the size of markets and permits specialization based on
comparative advantage. These allow industries to benefit from economies
of scale. It also allows diffusion of the stages of production among
several countries (vertical diffusion) based on comparative advantage in
production.
• It forces a considerable shift from state regulation and/or domination of
national production and exchange to their determination by market
forces. The former tends to be relatively inefficient and promotes
corruption.
• It allows the inflow of knowledge on more advanced production and
management techniques. The impact of this factor on output growth will
be considered in Chapter 15.
• For firms, the integration of capital markets increases the availability of
the amount and sources of external capital and reduces its cost.
These factors force the modernization of production, increased
competitiveness, and specialization based on comparative advantage,
which lead to higher output growth rates. Many economists consider
globalization and market domination of economies, along with the IT
revolution and other technological innovations, to be the major forces
behind the rising growth rates of the 1990s in most countries and support
the dismantling of barriers to openness of economies. These benefits
become more visible once the institutions and technology of the domestic
economy have adapted to the new economic forces brought about by
globalization. However, it is likely that certain firms and groups that were
among the beneficiaries under the old regime could lose from the shift to
the new economic forces. It is also likely that certain firms and types of
labor would suffer losses during the adjustment phase to the new long run,
or from their failure to adapt to the new economic and political realities.
Overall, at the level of the country, the general assumption borne out by
the experience of most countries is that the losses will be less than the
benefits, as well being of shorter duration and mainly during the
adjustment phase, relative to the benefits, which are long run in nature, so
that the countries participating in the globalization process will experience
rising growth rates in their standards of living.

12.11.2 Measures of the extent of globalization

600
The extent of globalization is never absolute but one of degree, so that it
has to be measured by indices of internationalization. It can be measured
by various indices, among which the most important ones are:
• The level of trade flows relative to GDP.
Smaller and more industrialized open economies tend to have larger
ratios of trade flows relative to their GDP. Measuring trade flows by
exports, the ratios of exports to GDP at the end of the 20th century were
about: 61% for Holland, 39% for Germany, 28% for France, 25% for the
UK, 10% for the USA, 5% for China, and 2.5% for India. For the world
as a whole, exports as a percentage of GDP grew from about 5.5% in
1950 to about 10% in 1973 and then to 17% in 1998.
• The composition of trade.
As international trade intensifies and industrial production rises, the
composition of trade tends to shift from agricultural to manufactured
products, and then to service industries. This ratio is higher for the
industrialized economies than for others. For the world economy, more
than three-fourths of exports now consist of manufactured goods.
• Capital flows relative to GDP.
• Foreign direct investment flows and the transnationalization of
production.
The transnational expansion of production within a firm occurs
through foreign direct investment (FDI) in the countries in which the
firms’ plants are allocated. The amount ofFDI in 2000 was over US$1.25
trillion.
Profit-maximizing firms attempt to cut costs by dividing their
production into various components, which are produced in the firm’s
plants in different countries on a cost minimizing basis. This expansion
leads to the transformation of national firms to multinational or
transnational ones.33 A considerable part of international trade now
consists of the ‘within-firm’ trade of multinational corporations. One
estimate of the extent of this trade relative to total exports is about 40%.
Multinational firms have an especially strong incentive to push for free
trade and have become the driving force behind the acceleration of
globalization and its contemporary characteristics.
• The extent of foreign exchange trading relative to GDP.
The globalization of trade and capital flows has dramatically increased
the trade in foreign currencies. By 1998, the amount of daily foreign
exchange trades was over US$1.5 trillion, which was over 100 times the
daily world exports.

601
12. 11.3 Other aspects of globalization
Economic globalization often implies major changes not only in the
economic policy and structure of the economy but also in the political and
social spheres. Among the latter are:
• It has reduced the political and economic independence of governments
and increased the power of international trade organizations, such as the
WTO and NAFTA.
• It has reduced the power of governments relative to multinational
corporations. It has favored a more laissez faire stance by governments
toward the activities of multinational corporations.
• It has increased the impact of external shocks on the domestic economy.
• It has led to ‘cultural internationalization’, with their threat to national
cultural patterns and industries. It has also diluted national control over
the environment and social programs.
• The market orientation of economies seems to have increased income
inequalities, at least in the short term, while lessening the ability of
governments to redress this through progressive taxation and social
welfare programs. As against this increase in economic inequality, the
growth rate of GDP and wealth per capita have increased very
significantly in most countries that liberalized foreign trade. This has
reduced poverty and produced higher standards of living, even if income
inequality did increase.
Most less developed nations that opened their economies seem to have
benefited in economic terms. However, some have suffered, at least in the
short term, on a net basis from some of its effects. In some countries,
globalization has meant de-industrialization from the displacement of
domestic production by cheaper or better quality imports and the take-over
of domestic firms by foreign ones. Many have also experienced rising
inequality between the capitalist class and the working one, and between
the working classes. It has also heightened exposure to external shocks,
which the relatively small and backward economies are ill equipped to
handle.

12. 11.4 Globalization and modernization

Although modernization and globalization are in principle different from
each other, it is difficult to separate their effects in practice. Economic
modernization is the process of the adoption of up-to-date production
techniques and the appropriate education and training of the labor force.
Inventions and innovations change these over time, sometimes drastically

602
as in the Industrial Revolution or, more recently, by the technological
revolution in data processing and communications. The adoption of newer
techniques raises productivity, wages, and profits. Since technology keeps
improving, modernization is a continuing process. Competition forces
firms to adopt the newer technologies earlier than they would otherwise
do. The experience with modernization over several centuries has
established that over the long-term modernization brings considerable and
continuing increases in productivity and standards of living for the
population as a whole, so that it is highly desirable for the country.
Globalization increases both competition and the exchange of
information on new processes and products, so that it is a major
contributor to the modernization process. Since much of the improvement
in technology may first originate in other countries, interaction with other
countries is essential for continuation of the modernization process.
Another part of this interaction occurs through foreign trade, which tends
to boost modernization by forcing domestic firms to become more efficient
in their attempt to remain competitive with foreign firms. Hence,
globalization of the economy boosts its modernization. In turn, continuing
modernization boosts the ability to compete internationally and therefore
tends to increase exports. Conversely, withdrawing from the process of
globalization often means falling behind in the continual modernization
process. This reduces productivity growth and the growth in the standards
of living.
Note that the modernization process often involves restructuring of the
economy in which some sectors expand and some firms and workers
benefit (in terms of job opportunities and wages), while other sectors,
firms and workers lose. These losses can be quite significant, sometimes
dramatic, over the short term for the affected sectors, firms and groups of
workers. Firms and workers which cannot make a successful shift to the
new technology and the sectoral shifts lose on a long-term basis while
those which do so may only suffer short-term losses. Long-run analysis
assumes that such adjustments have already been made, so that the losses
during the adjustment process are ignored in discussions of the long-run
consequences of globalization.

12.11.5 The persistence of nationalistic mercantilism —

globalization limited to truncated globalization
Mercantilism is the economic form of nationalism and consists of the
pursuit by countries of their own national advantage in the international
sphere. The economic tools of mercantilist are tariffs, subsidies, quotas,

603
and other devices to prevent competition from foreign countries. The latter
include administrative delays at the borders, ban on imports based on
(sometimes unjustified) charges of lack of safe production techniques
abroad and unfair trading practices (dumping), etc. Since countries try to
pursue their own national advantage, the international free trade
agreements and globalization tend to be truncated (i.e., limited), so that the
actual form of globalization in practice can be labeled as truncated (or
tilted ) globalization. The particular form of the truncated globalization put
into practice by the international or regional trade agreements/agencies
tends to favor countries with greater political and economic power.
Therefore, many economists and countries claim that the rich
industrialized nations only pursue international liberalization, through their
policies or those imposed through international institutions such as the
WTO, for those industrial goods in which they have comparative
advantage. In general, the industrialized countries collectively seek to do
so for high-tech industrial goods in which they possess a comparative
advantage over the poorer, less technologically developed countries, while
blocking the imports of low-tech goods from the latter. The textile and
apparel industries are relatively low-tech and are among the first industries
to develop en route to industrialization. They are also among the ones to
which the economically richer and more powerful countries have been
least willing to allow unhindered imports. Another such industry is
agriculture. The industrialized countries have protected such industries
through tariffs, subsidies, administrative rules, etc., and continue to do so
extensively. While the barriers to textile and apparel imports from the
developing countries have been gradually reduced, those to agricultural
imports, especially though subsidies to domestic farmers, remain high.
Such charges are often leveled at the USA, the European Union, and
Japan.

12. 11.6 The possibility of net losses from globalization

While, as pointed out above, the general experience has been one of
significant increases in standards of living and average wealth levels in
countries that joined the globalization bandwagon, some economists and
organizations believe that under the existing form of truncated
globalization, some countries, especially the poorer ones at lower
technological levels, experience net negative effects from globalization.
One reason for this net effect is that while the rich industrialized nations
pursue the liberalization for industrial goods, their agriculture remains
protected through tariffs, subsidies, administrative rules, etc., in which the

604
agriculture- based poor countries might have a comparative advantage.34
Another reason for this net effect arises from the restrictions on the
diffusion of new technologies through the long-duration patents by the rich
industrialized nations. The result of the tilted pattern of liberalization and
globalization in practice becomes lop-sided in terms of net benefits across
the two types (i.e., predominantly agricultural versus industrial) of
economies.
However, as economic internationalization has progressed, the economic
cost of maintaining closure to market forces and international competition
has increased, so that few countries can now afford to keep their
economies closed or isolated. There is considerable empirical evidence
that the refusal to adopt the market system and to participate effectively in
the international exchange of commodities and capital results in economies
which become increasingly inefficient and technologically backward, with
low per capita output and low growth rates. As a consequence, there are
few countries that choose to remain closed to the international trade in
commodities and/or continue to have state domination of production.35
The preceding focus on the net costs and benefits to the country as a
whole hides the distribution of these net benefits across the different
income and social groups within each country. Increasing the openness of
the economy benefits those sectors (owners and workers) that are able to
compete in the international sphere and expand; it has net costs for sectors
that are not able to compete and decline. The more skilled workers who
can adapt36 to the skill requirements of the expanding industries benefit
through greater demand for their skills see increases in their wages, while
less skilled and less adaptable workers tend to lose.

12. 11.7 Effects of globalization on commodity and

capital flows
The globalization of commodity and capital flows has had both positive
and negative effects. Among these are:
• They have substantially increased world trade, capital flows, and world
economic growth, which have raised living standards significantly in all
countries.
• They have increased the chances and scale of exchange crises and
heightened the adverse impact of such crises on individual economies,
especially smaller ones.
• They have also increased the possibility of the contagion of crises, with
an exchange rate crisis for one country spreading to other countries, often

605
 in the same region. This is a worrisome externality of the closer
integration of economies and world trade.
• The spread and size of multinational corporations have magnified
dramatically. This has increased the economic power of multinational
corporations relative to that of national governments, especially in the
case of some of the smaller and poorer countries. Some experts view this
as a disturbing erosion of national sovereignty.
• The increase in the size, competitiveness, and power of markets over that
of national governments has increased substantially, raising the levels of
competitiveness and efficiency in production and exchange. The power of
governments to control and regulate trade and production has
correspondingly decreased. Given the past record of many governments
in this respect and the corruption that it fostered, the relative increase in
the power of markets over governments is generally viewed as a very
positive consequence of the liberalization process.
The responsibility of the IMF to act as a lender at short notice and of last
resort — that is, when no other lenders/countries may be willing to lend to
the country in question — to the national central banks and governments
is, consequently, now of vital importance to the stability of exchange rates
and the performance of the national and world economies. The IMF
performed this role to a significant extent during the 2007–2009
worldwide banking and economic crises when commercial banks in the
USA and Europe drastically reduced their lending not only to other
domestic banks but also to foreign governments and banks.

12.12 Conclusions
• The world economy has experimented extensively with fixed, flexible,
and managed exchanged rates. The post-World War II period started with
a strong belief in the greater benefits of fixed exchange rates over those
of flexible ones, and over the belief that devaluations other than those for
really fundamental imbalances in the balance of payments were to be
discouraged. The modalities of this determination were embodied in the
rules established in 1946 for the IMF whose permission was needed
before member countries could devalue their currencies relative to others.
• Another strong belief in 1946 was in the desire to maintain some form of
control over the world’s money supply through some form of the Gold
Standard or the Gold Exchange Standard.37 After 1946, this was done
through setting a fixed exchange rate for the world’s strongest currency,
the US dollar, against gold at the rate of US$35 to an ounce of gold. The

606
 exchange rates for other currencies were set against the US dollar or
against one of the world’s other major currencies, such as the British
pound or the French franc.
• As the economies of the world recovered from their war-torn states and
expanded extensively during the 1940s and 1950s, the relative strength of
the economies and the relative competitiveness of their export sectors had
changed markedly by the later 1960s from their status in 1945. In
particular, by the end of the 1960s, there were persistent deficits in the
US balance of payments, which caused persistent speculative runs on the
US dollar, as well as on many other major currencies, including the
British pound (£). The fixed exchange rate system came to be viewed as
an undesirable and unmanageable constraint on needed adjustments and
on world trade. Eventually, in a series of steps taken during the late 1960s
and early 1970s, the twin pillars of the fixed price of gold in terms of US
dollars and the fixed exchange rates among currencies were abandoned.
The general pattern of exchange rates among countries since that period
has been one of flexible, though managed, exchange rates, though with
relatively fixed ones by a few countries, of which (as of 2009) China was
the one with the largest economy.
• Even though the trade in commodities grew enormously in the second
half of the 20th century, that in capital flows came to dominate the
foreign exchange markets on a short-term basis. Consequently, the short-
term behavior of exchange rates, especially for countries with developed
financial systems, is now explained by the interest rate parity condition.
The application of purchasing power parity, especially in its relative
form, for the equality of inflation rates is reserved for the long-run
movements in exchange rates. The actual foreign exchange markets also
exhibit a great deal of randomness, so that the deviations from the
exchange rates implied by the interest rate parity occur on a daily basis.
• International flows of capital for certain countries are dominated by
investments in stocks rather than in bonds, and the expected yield in
stocks can have a different pattern than the yield on bonds. The capital
flows for investments in stocks limit a straightforward application of the
interest rate parity doctrine. Hence, for this and other reasons such as a
home or foreign country preference for investment and the differences in
risk, the interest rates among countries are rarely, if ever, equal.
• National economies have become more complex and more closely
integrated into the world economy. Economists’ opinions and national
policies have gone from favoring fixed exchange rates among currencies
— as in the years from the end of World War II to the early 1970s — to a
strong recommendation for flexible exchange rates. The latter are now

607
 regarded as more suited to ensure balance of payments equilibrium in the
face of day-to-day and longer-term shifts in the commodity and capital
flows among countries. In particular, the IMF now advocates flexible
exchange rates. However, there continues to be concern that a flexible
exchange rate regime can expose the national currency to destabilizing
speculative runs and thereby contribute to the volatility of its exchange
rate. This is especially likely to happen in smaller or poorer countries if
their exchange reserves are small relative to their potential inflows and
outflows of financial capital.
• The general adoption of flexible exchange rates since the 1970s seems to
have mitigated the potential disruption from fundamental shifts in the
competitive positions of countries and protected the domestic economies
from the differences in the inflation rates among countries. However,
flexible exchange rates do hold the potential of large speculative runs on
the currencies of selected countries and of the depreciation of their
currencies in excess of what is implied by the fundamental economic
forces. Combined with the short-term inelasticity of net exports as
encompassed in the J curve, these speculative episodes can become self-
reinforcing and damaging for the affected domestic economy, as well as
for its trading partners.
• Expectations play a very significant role in speculation in the foreign
exchange markets. To illustrate, the interest rate parity condition
incorporates the difference between the expected rate and the spot foreign
exchange rate. Speculation arising from differences in expectations and
shifts in them can buffet currencies through speculative runs.38 This is
especially so since the amounts of liquid assets at the disposal of the
world’s major financial institutions, private firms, and even some
individuals are extremely large, and shifts in them of even small
percentages from a given currency to others can cause major changes in
the demand and supply of foreign exchange for the given currency, and
thereby in its exchange rates against other currencies.
• The integration of national economies into a world economy is known as
globalization. Its actual form is still a truncated one.
In terms of the short-run macroeconomic models, taking account of the
existence of foreign trade and capital flows modifies the assumptions of
our earlier closed economy models, so that the specifications of the
expenditure and monetary sectors for the open economy change. These are
dealt with in the next chapter.

KEY CONCEPTS

608
Purchasing power parity
Interest rate parity
The demand for the domestic
currency (dollars) in the foreign
exchange markets
The supply of the domestic
currency (dollars) in the foreign
exchange markets
Elasticity of net exports
The J curve
Fixed versus flexible exchange rates
Hot money
International Monetary
Fund (IMF)
The Bretton Woods System
Special Drawing Rights (SDRs)
IMF conditionality
Globalization
Mercantilism
Truncated globalization, and
The IMF as a central bank.

SUMMARY OF CRITICAL CONCLUSIONS

• Purchasing power parity occurs under perfect commodity flows without
transactions costs and is the doctrine that the real exchange rate between
countries will be unity. Relative purchasing power parity is the doctrine
that the depreciation of the domestic currency in terms of another will
equal the difference between the inflation rates of the relevant countries.
• Interest rate parity occurs under perfect capital flows and is the statement
that, under fixed exchange rates, the domestic and foreign rate of return
on riskless assets will be identical. Under flexible exchange rate, the
domestic interest rate will (approximately) equal the foreign one less the
expected rate of appreciation of the domestic currency (alternated stated
as: plus the expected rate of appreciation of the foreign ones).
• Investors in secure economies seem to possess strong home country
preference, with the result that the portfolios held are overwhelming
composed of domestic stocks and bonds. Consequently, the correlation
between domestic saving and investment is extremely high, contrary to
the implications of perfect capital flows resulting in IRP.
• Long-run analysis is based on the assumption that net exports are elastic.

609
 However, they may be inelastic in the impact period and the short term.
Therefore, many economies experience a J curve following a depreciation
of their currency. This results in overshooting of the depreciation from
that required for the long run.
• While the original mandate of the IMF had been the promotion of fixed
exchange rates except in cases of long-run imbalances in the balance of
payments, its policies shifted in 1971 to the promotion of flexible
exchange rates.
• In making loans to the member countries, the IMF often imposes
conditions on the pursuit of monetary (low money supply growth rates)
and fiscal policies (smaller fiscal deficits). In addition, there is also
pressure toward increased liberalization of the economy of the borrowing
country – that is, the reduction of restrictions (tariffs, quotas, and
regulations) on commodity and capital flows.

REVIEW AND DISCUSSION QUESTIONS

1. [This question is partly based on material in Chapter 3.] Using your own
experiences, provide some examples of the differences in the prices of
seemingly identical goods among stores (a) in the city centre and the
suburbs, (b) in different parts of the country, and (c) between countries.
What factors account for these ostensible deviations from purchasing
power parity? Are these factors likely to be stronger for (c) than for (a)
and (b)?
2. [This question is partly based on material in Chapter 3.] How does
relative purchasing power parity differ from (absolute) purchasing
power parity between countries? Why might neither do well over the
long term for a given pair of countries (say USA and Mexico or India)?
3. What are the determinants of the supply of the domestic currency
(dollars) in the foreign exchange markets? Explain.
4. Identify two variables changes in which shift the demand curve in the
foreign exchange market for our/domestic currency (dollars) to the right.
Identify two variables changes in which shift the supply curve in the
foreign exchange market for the domestic currency to the right.
5. What is the J curve? Why might countries experience a J curve
following a depreciation of their currency?
6. Using the IRP hypothesis, discuss whether Canada or another country of
your choice normally offers a premium over the corresponding interest
rate in the USA, or vice versa. Discuss the reasons for this premium.
7. Explain the notion of the stability versus instability of the foreign
exchange market. What does each imply for the movement in the
exchange rate following an accidental decline?

610
8. Why are balances of payments deficits not always corrected or at least
reduced by depreciation, and surpluses not corrected or at least reduced
by an appreciation of the exchange rate?
9. In a two-country model, how would changes in the domestic economy
be transmitted to the foreign one?
10. What is the IMF? What are its major functions nowadays?
11. What was the Bretton Woods system? Does this system still apply? If
not, how has it been changed?
12. [Optional] What is meant by moral hazard as it applies to loans to
countries (see Extended Analysis Box 12.2)? What implications does it
have for risk taking by borrowing countries and the subsequent
repayment of loans? What mechanism has been created by the IMF to
limit moral hazard?
13. Why does globalization promote increases in the standard of living of
countries that liberalize their trade in commodities and capital?
14. What is mercantilism? How does it modify globalization to its
truncated form? Give at least two examples of this truncation.

ADVANCED AND TECHNICAL QUESTIONS

T1. You are given the following information:
P = 3, P F = 2.

(a) If the central bank fixes the nominal exchange rate at 0.5, what is the
real exchange rate?
(b) What does the absolute PPP imply for the nominal exchange rate?
T2. If r F = 0.25, P = 2, P F = 4, = 4 (that is, the exchange rate has been
fixed by the central bank at this level), and the public expects it to
remain at this level, what does (the benchmark) interest rate parity imply
for the domestic interest rate?
T3. You are given the following information:
P = 110, P F = 220.

(a) What does absolute PPP imply for the nominal exchange rate?
(b) If the domestic rate of inflation is 5% and the foreign one is 20%,
what does absolute PPP imply for the appreciation/depreciation of our
(domestic currency’s) nominal exchange rate?
(c) What does PPP imply for the change in the real exchange rate?

611
T4. Let the domestic currency be the dollar and the foreign one be the
Japanese yen. Suppose that you are given the following information and
questions:
Interest rate on dollar assets (i.e., assets denominated in dollars): 8% per
annum; interest rate on yen assets (i.e., assets denominated in yen):
10%; current exchange rate: 100 yen per dollar; and actual exchange
rate after one year: 120 yen.
(a) For a domestic resident, what is the actual rate of return on investing
$5000 in yen assets?
(b) For a domestic resident, what is the actual rate of return on holding
$5000 in dollar assets per year?
(c) A priori, what does (the benchmark) interest rate parity theory imply
for the expected rate of change in the yen–dollar exchange rate? Is
this different from the actual rate of change. If so, why?
1For example, under the Smithsonian Agreement reached in August 1972
between the USA and 19 other major trading nations, which are members
of the IMF, the exchange rates between the US$ and the currencies of
certain specified countries were allowed to vary within ±2.25% of the
designated exchange rates. It had fluctuated within a band of ±1% of the
designated rates earlier. In March 1973, the US$ was set free to float
against other currencies.
2Such physical capital flows occur if a firm exports the components
(which are in our category of commodities) of a factory and sets up the
factory abroad. This action does not generate purchases or sale of foreign
currency.
3Note that in the following expression, ρ e, not ρ , is in the denominator. If
period analysis is used, ρ e will be represented by the expected future
exchange rate ρ et+1 while ρ will be the actual current exchange rate ρ t.
4These risks depend on the likely fluctuations in the exchange rate, the
extent of political and economic uncertainty, home country advantage in
information, etc.
5Given a general preference for investing at home, capital exporting
countries usually have lower domestic interest rates and capital importing
countries have higher ones.
6LDCs and countries with political uncertainty usually fall in this
category.
7Note that the determinants of α can vary for long-run analysis, since, as
normally assumed in the definition of the long run, the factors determining
its sign and value would not be a priori held constant .

612
8The overshooting hypothesis does not explain why this adjustment is
gradual and why the exchange rate does not immediately adjust to its
eventual value.
9The reasons for this cautious switch of portfolios lie in the uncertainty of
future domestic and foreign bond returns and prices, and in the uncertainty
of the future course of domestic and foreign interest rate policies, so that
there is uncertainty about the optimal composition of the portfolio over the
future.
10Remember that there is always uncertainty about whether foreign
countries will soon follow by corresponding cuts in their interest rates.
Further, a considerable part of capital flows are for investment in foreign
equities, whose returns are never known with any degree of precision, so
that small cuts in the domestic interest rate on bonds rarely lead to
immediate massive outflows of capital.
11Remember that domestic inflation higher than the foreign one will make
our commodities more expensive relative to foreign ones and decrease net
exports. This will cause a balance of payments deficit, so that the exchange
rate will be expected to depreciate. To offset this expected depreciation,
investors will seek a higher interest rate from domestic bonds than from
foreign bonds.
12Note that this observation is about the short-term — not the short-run or
the long-run — behavior of interest rates.
13An exception to this occurs if the central bank targets the domestic
interest rate and chooses not to raise it. However, in this case, the central
bank will have to balance the net capital outflows by selling foreign
exchange out of its foreign exchange reserves.
14This supply of our dollars (S $) in the foreign exchange markets can also
be looked at as the demand for foreign currencies (D £)since we need the
foreign currencies to pay foreigners for their goods and financial assets.
15From another perspective, the supply of foreign currencies (S £) to us
arises when foreigners buy our commodities (our exports)or buy our
financial assets (which becomes our imports of capital) or make unilateral
payments to us, and pay for them in foreign currencies.
16That is, the quantities imported (or exported) have elasticity greater than
unity with respect to changes in their prices, so that the nominal
expenditure on them rises as their price decreases.
17If the supplies of both imported and exported goods are infinitely
elastic, the demand elasticities required for stability are known as the
Marshall–Lerner condition. This is the requirement that the absolute values
of the import and export demand elasticities sum to more than unity. The

613
Marshall–Lerner condition ensures that the trade balance improves in
response to a devaluation or depreciation of the domestic currency
(assuming that the supply elasticities of commodities are infinite).
18Along the J curve, the elasticity of net exports is less than one for some
time after a depreciation or devaluation but increases over time until it
becomes greater than one.
19Capital flows are being held constant under the ceteris paribus
assumption.
20Each succeeding depreciation would create a new J curve emanating off
the preceding one. Figure 12.3a illustrates this by showing three J curves
for depreciations at times t 0, t 1, and t 2.
21The USA is the world’s largest single trading nation and has enormous
amounts of capital inflows and outflows. The currencies of most of the
other major trading nations of the world had their exchange rates fixed
directly or indirectly in terms of the dollar before March 1973. However,
at that time, some of the currencies, such as the Canadian dollar, had a
floating exchange rate against the dollar. Therefore, the US dollar, faced a
mixed case of fixed and flexible exchange rates. The US dollar was set
free in March 1973, to float against gold. Most other industrialized
countries also moved to flexible exchange rate regimes during the 1970s.
22Over the last three decades, the European Union as a regional bloc
including several major trading countries moved increasingly closer to a
system of fixed exchange rates among its members, with flexible rates vis-
à-vis other countries not in the union. In the earlier part of this period, the
national currencies were permitted a limited float in a pre-designated
narrow band, known as a ‘snake’.
23The substantial fall in the exchange rates of Thailand, Indonesia, Korea,
and some other Asian countries in the late 1990s is at first sight indicative
of the volatility and instability of floating exchange rates. However, many
of these exchange rates were not fully flexible but tied to the US dollar or
the Japanese yen, and had became overvalued in effective terms against
other currencies. The final assessment of whether this experience is
indicative of the instability or not of flexible exchange rates is still
pending.
24The implications of such policies for the open economy require a
respecification of the IS-LM model and are dealt with in the next chapter.
25An example of the need for such loans occurred for Argentina in late
2001 and early 2002 to offset speculation against the devaluation of its
currency, the peso. Argentina was not willing to meet the conditions
required by the IMF, so that it was refused the required loans. Argentina
defaulted on its interest payments on its bonds held by foreigners and,

614
among other steps taken in January 2002, devalued the peso by 40%.
26Therefore, the existence of moral hazard implies that the borrower is
more likely to resort to more risky choices if the loan is made than if it is
refused. The loan transfers some of the consequences of the risky choices
from the borrower to the lender.
27These usually imply reductions in social welfare and anti-poverty
programs. Such policies are unpopular with the public since they tend to
lower incomes and increase poverty.
28Initially, an SDR was a weighted average of four currencies, which are
the US dollar, the euro, the Japanese yen, and the British pound.
29The Great Depression was intensified by the absence of a world lender
of last resort and the failure of any of the major national banks to address
the international dimensions of their policies.
30During 2009, there were rumors that China and some other foreign
holders of substantial US dollars wanted to decrease their foreign
exchange holdings of US dollars and shift to other currencies. There was
even discussion of the need for the IMF to create large amounts of SDRs,
and the replacement of the US dollar by SDRs as the key currency for
holding foreign exchange reserves.
31This is never completely so. For instance, several provinces in Canada
still maintain barriers to the flows of commodities and labor from other
provinces.
32The ‘openness’ of an economy is usually measured by the sum of
exports and imports, not by net exports.
33One estimate of the number of the parent multinational firms is over
64,000 in 2000, with over 850,000 foreign affiliates.
34Of course, the rich industrialized nations also try to protect those of their
industries for which they do not have a comparative advantage.
35An excellent example of this is provided by China with a communist
system, which is in principle opposed to the capitalist basis of
globalization. Over the last two decades, it has chosen to accommodate the
capitalist forces in many spheres, and benefited from very much higher
growth rates than if had remained a closed economy.
36These are often the recent graduates and the young while the older
workers tend to be less adaptable.
37These are explained in Chapter 13. The Gold Standard was a system of
fixed exchange rates in which the national currencies were fixed in value
against gold. The Gold Exchange Standard was a system under which the
national currencies had a fixed exchange rate against a key currency
(mainly the US dollar), which had a fixed exchange rate against gold.
38A speculative ‘run’ (i.e., a movement by many in the same direction)

615
occurs if a lot of speculators simultaneously buy (or sell) the domestic
currency. The cause of the run is the occurrence of a general belief among
investors that the domestic currency is going to appreciate (depreciate).
The result of the run under flexible exchange rates is to cause the currency
to appreciate (depreciate). Under a fixed exchange rate, such a run requires
the belief that the exchange rate can be made to change, and its purchases
(sales) are called a speculative ‘attack’.

616
 CHAPTER 13
The Open Economy Under a Fixed
Exchange Rate Regime

This chapter extends the macroeconomic analysis of the open

economy to the fixed exchange rate case .
For a country with a fixed exchange rate, fiscal policies are
likely to be more effective than monetary policies in changing
aggregate demand in the economy .
Under a fixed exchange rate, since the exchange rate does not
change to maintain equilibrium in the foreign exchange market,
the pursuit of monetary and fiscal policies is more complicated
than if the fixed exchange rate is floating .

This chapter presents the IS-LM model to derive aggregate demand for the
small open economy that maintains a fixed exchange rate. As discussed in
Chapter 5, the real interest rates are more relevant for the commodity
sector and its IS curve, while the nominal rates are more relevant for
international capital flows and the BP curve, which specifies the
equilibrium points of the balance of payments. However, our use of the
Fisher equation on interest rates and the assumption that the expected
inflation rate πe equals zero allows the simplification that R = r, so that R
and r are used interchangeably for the interest rate in this chapter.
Chapter 6 with a flexible exchange rate derived the IS-LM framework
for the open economy. Our arguments in this chapter show that for an
economy that has a fixed exchange rate,
• We need to supplement the IS and LM curves with a balance of
payments equilibrium curve (called the BP curve). The BP curve is
upward sloping in the IS-LM diagram. The BP curve may be flatter or
steeper than the LM curve.
• The IS curves between the fixed and flexible exchange rate cases have
somewhat different properties.

617
• Balance of payments surpluses and deficits are likely under fixed
exchange rates. These have the potential for changing the monetary base
and the money supply. The central bank may allow this impact to occur
or sterilize the balance of payments surpluses and deficits. The money
market analysis and the LM curve are different under sterilization from
that without sterilization.

13.1 The Balance of Payments (BP) Equilibrium

Curve Under Fixed Exchange Rates
The equation for equilibrium in the balance of payments was derived in
Chapters 3 and 12 as:

where:
B = balance of payments,
Xc = value of exports of commodities (goods and services),
X k = value of capital exports,
Zc = value of imports of commodities (goods and services),
Zk = value of capital imports,
NR = net interest and dividend inflows,
NT = net transfers from abroad,
ρ = nominal exchange rate (£/$),
ρr = real exchange rate (= ρ P/PF),
P = domestic price level,
P F = foreign price level,
r = domestic real interest rate,
R = domestic nominal interest rate,
rF = foreign real interest rate,
RF = foreign real interest rate, and
πe = expected inflation rate.
To reiterate, an assumption made earlier in this chapter was that πe = 0 for
both the domestic and foreign economies, so that we have R = r for both
economies.
There are three endogenous variables (ρ r, r, and y) and two exogenous

618
variables (yF and rF) in the preceding equation. We can use this equation
to plot a curve, called the BP curve, whose points indicate equilibrium in
the balance of payments. The three potential forms of the BP curve
relevant to the IS-LM framework are shown in Figures 13.1a, 13.1b, and
13.1c. Equilibrium in the balance of payments exists at each point of each
of the BP curves in these figures.
In Figure 13.1a, start with the equilibrium point (r0, y0 ). Given the
interest rate r0 , an increase in domestic income y beyond y 0 increases
imports, leaving exports unchanged. For equilibrium in the balance of
payments, this increase in imports has to be offset by an additional net
capital inflow, which requires a rise in domestic interest rates. That is, an
increase in national income y requires a rise in the interest rate r (with ρr
held constant) to maintain equilibrium in the balance of payments, so that
the BP curve between y and r has a positive slope.
There are also two extreme cases of the BP curve. One of these,
illustrated in Figure 13.1b, occurs if there exists perfect capital mobility,
implying interest rate parity. The second case, illustrated in Figure 13.1c,
occurs if the capital flows are totally insensitive to interest rate changes.
These are explained in the next section.

Note that the BP curve shifts if ρ r changes. To illustrate, in Figure 13.1a,

start with (r0 , y0) on the curve BP0. If P rises, ρr would increase, causing
net exports at y0 to decline, so that a larger net capital inflow induced
through a higher interest rate than r 0 will be required to maintain
equilibrium. Let this interest rate be r 1 and the BP curve through (r1, y0)
be BP1. Hence, the increase in the real exchange rate ρ r shifts the BP
curve upward from BP to BP1. In Figure 13.1b, if ρr rises, the BP curve
will also have to shift up. This cannot happen in Figure 13.1c: the capital
flows have zero interest sensitivity.

13.2 The IS-LM Model for the Open Economy

619
 Under Fixed Exchange Rates
The IS-LM diagram for the fixed exchange rate case has four curves in the
(y, r) space: a negatively sloping IS curve, positively sloping BP and LM
curves and a vertical LRAS curve. Since both the LM and the BP curves
have positive slopes, we need a hypothesis to determine their relative
slopes.

13.2.1 The slope of the BP curve

On the slope of the BP curve, it was mentioned above that there are two
extreme cases. These are:
1. Perfectly elastic capital flows. From Chapters 3 and 12, the IRP
hypothesis under perfectly elastic capitalflows is:
1

where ρ ′′e is the expected change in the exchange rate. Assuming that
the exchange rate is not only fixed but is also expected by the public to
remain unchanged, ρ′′e = 0, so that IRP becomes:

Hence, for a small open economy with a fixed exchange rate, the
domestic interest rate is set by the foreign interest rate. In this case, the
BP curve for any given exchange rate is horizontal, as shown in Figure
13.1b, at the interest rate r0.
2. Capital flows are zero or completely insensitive to the domestic interest
rate. In this case, the BP curve for any given exchange rate would be
vertical, as shown in Figure 13.1c.
Economies with the financial markets well integrated into the global ones
would have a BP curve closer to the horizontal form, while developing
economies with virtually non-existing or isolated financial markets would
tend toward the vertical BP curve.

13.2.2 The relative slopes of the LM and BP curves

We can, therefore, base further analysis on one of the extreme cases or on
two slightly less extreme ones, one of which assumes that the BP curve is
flatter than the LM curve, while the other assumes that it is steeper. The
former is likely to be more applicable to the developed economies and is
treated as the normal case for them.It will be called the interest-sensitive

620
capital flows case. The latter is likely to be more applicable to some of the
developing economies and will be called the interest-insensitive capital
inflows case.
Extended Analysis Box 13.1: Differences between the IS-LM Diagrams
for the Flexible and Fixed Exchange Rates
There are two major differences between the IS-LM diagrams for the
flexible and the fixed exchange cases.
First, if the nominal exchange rate is fixed, there would exist a BP
curve showing the (y, r) combinations for equilibrium in the balance of
payments. But if it is flexible and adjusts to yield continuous
equilibrium in the balance of payments, the balance of payments
equilibrium can exist at any (y, r) combination, so that there will not be
a separate BP curve. Therefore, we have two quite different versions of
the IS-LM diagram, one with a BP curve for the fixed exchange rate
case and the other without a BP curve for the flexible one.
Second, any exogenous shift in the fixed exchange rate will shift the
IS curve. But, for the flexible exchange rate case, a change in the
exchange rate due to a change in the domestic interest rate will not shift
the IS curve: the slope of the IS curve will incorporate within it any
impact of this on net exports and thereby on income.2
Note also that the LM curve is identical between the flexible and fixed
exchange analyses if the balance of payments flows are not allowed to
influence the money supply under an active monetary policy. The short-
run equilibrium analysis of the flexible exchange rate case assumes
balance of payments equilibrium, so that there are no net inflows or
outflows of foreign exchange. Therefore, the assumption of the
sterilization of balance of payments surpluses and deficits is not needed
for it. It is needed for the fixed exchange rate case.

13.2.3 General equilibrium under fixed exchange rates, a

diagrammatic treatment
Figure 13.2a shows general equilibrium for the developed economy in the
fixed exchange rate case. All the curves intersect at the general equilibrium
point a. To reiterate, the steeper slope of the LM curve compared with the
BP ones assumes that the international capital flows are more sensitive to
interest rate changes than is the domestic money market,3 so that an
increase in y requires a larger increase in the interest rate to keep the
monetary sector in equilibrium than that required to keep the balance of

621
payments in equilibrium. Figure 13.2b shows what we referred to as the
interest insensitive capital-inflows case. All the ‘curves’ are shown as
straight lines.

In order to limit the number of cases that are investigated, we will

henceforth mainly analyze the developed economy case, mostly without
even referring to it as such. That is, the default assumption in the rest of
this chapter is that the BP curve for any given exchange rate has a flatter
slope than the LM curve. However, some analysis of the interest
insensitive capital-inflows case — relevant to many LDCs — is also
presented in the Extended Analysis Boxes 13.2 and 13.3.
To analyze the roles of monetary and fiscal policies in the open
economy, remember that these are demand management policies. As such,
the analysis of their role can be simplified by first focusing only on
aggregate demand (AD) in the economy and only later studying the
interaction between aggregate demand changes and supply responses.

13.3 Managing Aggregate Demand in the Open

Economy: Monetary and Fiscal Policies Under
Fixed Exchange Rates
The basic difference between the fiscal and the monetary policies in the
macroeconomic static models arises because an expansionary fiscal policy
raises the interest rate, while an expansionary monetary policy lowers
them. A rise in the domestic interest rate increases net capital inflows,
while a fall in this interest rate decreases such capital inflows. Therefore, if
the balance of payments is initially in equilibrium, an expansionary fiscal
policy creates a surplus ( RF > 0)4 by raising interest rates. But an
expansionary monetary policy creates a deficit ( RF < 0) in the balance of
payments by lowering the rate of interest.
Assume that the economy has a fixed exchange rate, has a flatter BP

622
curve relative to the LM one, and is initially in overall equilibrium, as
shown in Figure 13.3a.

13.3.1 The impact of an expansionary fiscal policy on

aggregate demand
In Figure 13.3a, an expansionary fiscal policy will shift the IS curve from
ISo to IS1. This would increase aggregate demand to d’, at the intersection
of the IS and LM curves. As shown in Figure 13.3a, this policy would
raise the domestic interest rate and create a balance of payments surplus
(since the interest rate at d’ is higher than required for equilibrium in the
balance of payments) while increasing domestic demand. The further
effects are:
• If the balance of payments is sterilized under an active monetary policy,
the money supply will not expand, but the domestic interest rate will stay
above the foreign ones. The balance of payments surplus will continue,
with a resultant continuing increase in foreign exchange reserves. Many
countries consider this outcome to be quite desirable.
• If the balance of payments is not sterilized, as under a passive monetary
policy, it will cause an expansion of the monetary base and hence of the
money supply, thereby shifting the LM curve to the right from LM0 to
LM′. This will place downward pressure on interest rates, causing interest
rates to fall toward their former level, while causing a further expansion
of demand. Aggregate demand will eventually settle at d″ at the
intersection of IS′, BP, and LM′, so that the fiscal expansion is supported
by an induced expansion in the money supply.

13.3.2 The impact of an expansionary monetary policy on

aggregate demand
Now consider the effects of an expansionary monetary policy. The
increase in the money supply shifts the LM curve from LM0 to LM1 in
Figure 13.3b, thereby increasing aggregate demand in the economy to d′

623
while lowering the interest rate. The latter reduces net capital inflows and
causes a deficit in the balance of payments. This deficit causes the money
supply to fall unless this effect was offset by the monetary authorities
through an active monetary policy. Hence, the two further scenarios are:
1. If the balance of payments is sterilized, the money supply would remain
at its new enlarged level and its expansionary impact upon the economy
would continue. But so would the external deficit, with the result that
the country will run down its foreign exchange reserves by a
corresponding amount. Such a policy clearly has a limit and can be
pursued only up to the point where the reserves begin to fall to
unacceptably low levels. The money supply will then have to be cut
back to its initial level so that aggregate demand will also return to its
initial level at d.
2. If the balance of payments deficit is not sterilized, the money supply
will fall, shifting the LM curve to the left and thereby decreasing
aggregate demand. This counteracts the initial monetary expansion. The
balance of payments deficit will continue until the money supply has
fallen to its initial amount and the LM curve has returned to its original
position. Therefore, the restoration of general equilibrium will require a
return to the point d in Figure 13.3b. Hence, the expansion in the money
supply would be offset through outflows of funds from the economy.
Therefore, for the developed economies’ case with fixed exchange rates,
an expansionary monetary policy is clearly circumscribed by the
availability of foreign reserves to finance deficits, while an expansionary
fiscal policy is not thus circumscribed and can be pursued indefinitely. The
latter is also reinforced by an induced expansion in the money supply in
the case of a passive monetary policy. Hence, the pursuit of fiscal policy is
preferable to that of monetary policy in the fixed exchange rate case.
Extended Analysis Box 13.2: Monetary and Fiscal Policies in the Case of
Interest—Insensitive Capital Flows and Fixed Exchange Rates

If the net capital inflows are relatively insensitive to domestic interest

624
rates, as in many LDCs, the BP curve could be steeper than the LM one.
This case is shown in Figures 13.4a and 13.4b.
For the expansionary fiscal policy analysis in Figure 13.4a, the IS
curve shifts from IS0 to IS1. At the new level of interest rates given by
the point d′, there would be a balance of payments deficit, which causes
a loss of foreign reserves. If it is not sterilized, the money supply will
fall, with a contractionary effect running counter to the expansionary
objective of fiscal policy, so that the desired expansion of aggregate
demand will not be achieved. But if it is sterilized, the balance of
payments deficit will continue and foreign reserves will continue to fall.
For the expansionary monetary policy analysis in Figure 13.4b, the
increase in the money supply shifts the LM curve from LM0 to LM1,
increasing aggregate demand from d to d′. The point d ′ is associated
with a balance of payments deficit, so that the earlier analysis based on
Figure 13.3b continues to apply, though the deficit will be even greater
in Figure 13.4b.5 As before, this balance of payments deficit, in the case
when it is not sterilized, will result in a continuing reduction in the
money supply until the original money supply is restored. If it is
sterilized, there will be a continuing loss of reserves.
Hence, with fixed exchange rates in the interest-insensitive capital
flows case, both expansionary fiscal and monetary policies will cause
balance of payments deficits and loss of foreign exchange reserves.
Further, in the passive money supply case, the expansionary fiscal
policy will be accompanied by an induced contractionary rather than
expansionary money supply, so that the desired expansion will not
occur. An expansionary monetary policy will also cause balance of
payments deficits and loss of foreign exchange reserves. Neither type of
policy can be pursued for long unless the country has foreign exchange
reserves whose depletion through deficits can be tolerated by the
authorities. Since LDCs tend to have interest- insensitive capital flows,
their pursuit of expansionary monetary and fiscal policies tends to pose
relatively greater problems in terms of balance of payments deficits and
loss of foreign exchange reserves, leading to an eventual reversal of
such policies.

13.4 Bringing Aggregate Supply into the IS-LM

Analysis with a Fixed Exchange Rate
For an economy with developed financial markets, we have shown that

625
an expansionary fiscal policy is more effective in increasing aggregate
demand than monetary policy. However, the benefits from this increase in
demand depend upon the supply response of the economy. As in the closed
economy analysis, there are three different types of supply responses. They
are:
1. Long-run analysis : the economy maintains full employment. In this
case, the increase in demand will merely increase the price level,
without an increase in output. It will also decrease net exports and
worsen the balance of payments, as we show in the next section.
Therefore, the relative efficacy of fiscal policy in increasing demand
becomes a disadvantage.
2. Short-run analysis: the increase in aggregate demand induces an
increase in the equilibrium employment and output beyond their full-
employment levels. Therefore, there is some advantage from pursuing
an expansionary fiscal policy rather than an expansionary monetary one.
However, this advantage is temporary since the economy will eventually
move to its full-employment state.
3. Disequilibrium analysis : if the economy was experiencing deficient
demand, the relative efficacy of fiscal policy in increasing aggregate
demand becomes a major consideration in pursuing it rather than
monetary policy.

13.5 Macroeconomic Equilibrium in the Small

Open Economy with Full Employment and
IRP
The modern classical approach assumes continuous full employment and
perfect capital markets. The following specifies the open economy
macroeconomic model for this approach.
Perfect capital mobility between the domestic and foreign capital
markets implies interest rate parity (IRP), which is the condition that:

IRP should properly be stated in terms of nominal interest rates R rather

than real interest rates r. However, we have used the Fisher equation, R = r
+ πe, and the assumption that the expected inflation rate πe is zero, to state
IRP in real interest rates r. Further, assume that the exchange rates are
fixed and that the public also expects them to remain unchanged.6 In this
case, ρ′′e = 0, so that IRP becomes:

626
Hence, r is exogenously determined by rF . In Figure 13.5, this condition
is shown by the horizontal IRP curve, which is the form of the BP curve
under the interest rate parity assumption.
Further, given continuous labor market clearance, there would be
continuous full employment and output would remain at yF. This is shown
by the LRAS curve in Figure 13.5. Therefore, under the assumptions of
IRP and full employment, the IS-LM figure would become that shown in
Figure 13.5. The equilibrium interest rate would be at rF and the
equilibrium output will be at y f.

13.5.1 Fiscal policies under IRP and full employment

Fiscal policy cannot alter the equilibrium shown in Figure 13.5 since it
cannot affect the exogenously determined interest rate or the full-
employment output, and must accommodate itself to (r F, y f). Hence,
expansionary fiscal policy cannot be used to increase domestic output
and/or change the interest rate. The only impact of fiscal policy will be on
net exports. This impact occurs through the IS equilibrium condition
derived in its three-gaps version in Chapter 5. Under the assumptions of
this section, this condition, with r = rF and y = y f, can be written as:

where the left side is net exports, so that:

In this equation, investment i and saving s are determined by the interest

rate given by rF and national income at its full-employment level yf, so
that the only variable that can be affected by a fiscal deficit is net exports.
Hence, from this equation, an increase in the fiscal deficit, by decreasing
the right side, will decrease the equilibrium level of net exports and
worsen the balance of payments on current account. Consequently, it will

627
worsen the balance of payments and cause a fall in the country’s foreign
exchange reserves. Conversely, a contractionary fiscal policy will improve
the balance of payments and build up foreign exchange reserves, without
causing a fall in output or rise in the interest rate.

13.5.2 Monetary policy under IRP and full employment

Under the assumptions behind Figure 13.5, an expansion in the money
supply also cannot change output or the interest rate: the interest rate is set
by IRP at r F and output is set at its full-employment level y f by the
assumption of long-run equilibrium. Hence, the position of the LM curve
cannot change either rF or yf, so that we did not even bother to draw it in
Figure 13.5.
In this context, the demand for real balances is specified by:

With the demand for real balances determined by yf and rF, the
equilibrium domestic supply of real balances must adjust to this amount.
The process by which this result is achieved is as follows. An attempted
expansion of the nominal money supply would lower the interest rate,
which leads to an outflow of capital, a balance of payments deficit and an
outflow of the domestic currency by enough to maintain the interest rate at
rF. Therefore, the equilibrium amount of the domestic money supply will
adjust to the unchanged domestic demand for real balances, making
domestic monetary policy completely ineffective. Hence, if the money
supply is decreased, the interest rate will rise, leading to a balance of
payments surplus, which will increase the money supply sufficiently, so
that it equals the demand for money at (rF, yf). That is, the domestic
money supply becomes endogenous to money demand at the fixed
exchange rate, world interest rates, and full-employment output, only
changing if these variables change.
Therefore, with a fixed exchange rate, IRP and full employment, there is
no role in equilibrium for monetary or fiscal policies for changing output.
In fact, with output already at its full-employment level, there is no scope
or need for such policies. The only role of these policies is in inducing
inflows or outflows of foreign exchange: contractionary monetary policies
and expansionary fiscal ones will induce an inflow.

13.6 PPP, the World Rate of Inflation and the

628
 Convergence in National Inflation Rates
This section investigates the implications of PPP in the context of the
quantity equation, which was presented in Chapter 2 and is an identity.

13.6.1 The world rate of inflation in the fixed exchange rate

case
Assume that all the countries of the world have fixed exchange rates, so
that we can treat the world as a single economy, with a world money
supply defined by the weighted7 sum of the money supplies of the
individual countries.
From the quantity equation, we have:

where the superscript w stands for the world, V is the velocity of

circulation and ′′ again indicates the rate of growth. This equation implies
that a general expansion of the money supplies among the countries will be
a major source of world inflation. This occurred during the 1960s and
1970s when many countries simultaneously followed expansionary
monetary policies in an attempt to increase their output and employment.
The result was worldwide inflation. By the mid-1980s, most countries had
abandoned such policies and were exercising much stricter controls on
their money growth. As a consequence, the world rate of inflation in the
1990s fell drastically from its levels in the 1970s and 1980s.
If we also assume relative PPP, then the fixed exchange rate for the j th
country implies that:

where πj is the rate of inflation in country j and k′′ is the change in its
relative competitiveness. Assuming k′′ = 0, this condition implies that the
individual country inflation rates will be identical with the world one — or
at least converge to it. Since the domestic inflation rate is determined by
the exogenously specified world one, the equilibrium money supply of
each country must adjust to accommodate this inflation rate and the money
supplies of the individual countries become endogenous. While money
supply growth and the price level in each country would then be closely
related, note that causation runs from the world inflation rate to the small
economy’s inflation rate, and then to its money supply, not from its money
supply to its price level.8

629
 Box 13.1: The impact of a large economy on smaller economies
The USA has the largest single economy in the world. Its economy
cannot be assumed to be a ‘small’ one since its economic variables
influence those in other countries. This influence on other countries
occurs through commodity and capital flows. This influence can be seen
in various ways:
• An increase in the interest rate in the USA puts upward pressure on the
interest rates in other countries, since a reduction is the US interest
rates leads to capital flows to other countries. These are significant
enough for their smaller economies to cause a reduction in their interest
rates. The reverse occurs if the USA raises its interest rates.
• A recession, with a decrease in the GDP, in the USA decreases other
countries’ exports to the USA. These are significant enough for their
smaller economies to cause a reduction in their aggregate demand and
output. The reverse occurs if the USA has a boom.
• A fall in the US price level (or a lower inflation rate) increases the US
exports and the decreases the US imports. These reduce the net exports
of the smaller economies and lower their aggregate demand and price
level (or inflation rate).
Consequently, in empirical studies, since the US price levels and interest
rates cause similar movements in those of the small open economies
trading with the USA, the US price levels and the interest rates are often
used as proxies for those of the world economy in the PPP and IRP
equations for those economies. This causation is only one-way from the
US economy to the smaller economies.
However, the emergence of the European Union as a single country
and the rapid economic growth of Russia, China, and India in this
century imply that the rest of the world as a whole, but not any single
country, can be treated as a large economy for the study of the USA
economy. That is, similar exogenous (to the USA) economic
developments in these economies collectively can exert a powerful
influence on the US economy.

13.7 Economic Blocs, Fixed Exchange Rates and

the Convergence in Inflation Rates
The above analysis can be adapted to the case of a bloc of countries
committed to a fixed exchange rate among its members. In the case of

630
regional blocs such as the European Union with fixed exchange rates
among its members, the member countries’ inflation rates will converge to
those of the bloc as a whole. This has been the experience of the European
Union countries since the 1980s, as the European Union has increasingly
limited the permissible range of exchange rate variations among its
member countries. Its long-term objective of a European Monetary Union
with a European currency and fixed exchange rates among its members’
national currencies and the European one would effectively impose a
common inflation rate on the member countries and limit their monetary
independence.

13.8 Disequilibrium in the Domestic Economy

and Stabilization through Monetary and Fiscal
Policies: The Fixed Exchange Rate Case with
Interest Sensitive Capital Flows
Figures 13.2a and 13.2b had shown the case of general equilibrium in the
developed economy, with all the curves, including the long-run supply
LRAS curve, intersecting at the same point a. Figure 13.6a shows the case
corresponding to Figure 13.2a, but for an economy with disequilibrium
and therefore with several intersections. Assuming as before for analytical
purposes that the commodity and the money markets adjust faster than the
labor market and the balance of payments, the economy will be at the point
d (intersection of the IS and LM curves) in Figure 13.6a. At this point, the
economy has deficient demand, relative to output yf. There is also a
balance of payments deficit, measured from the point d to the BP curve:
the interest rate at d is below that necessary to restore equilibrium in the
balance of payments, so that the capital inflows are inadequate for
equilibrium. The deficient demand at d forces prices to fall, so that the LM
curve would shift down from LMo to LM′ to go through the point b .
While this movement will lower interest rates and worsen the balance of
payments on capital account, the fall in prices will increase net exports.
However, this increase need not be enough to ensure balance of payments
equilibrium at full employment and is in any case going to be a long run
rather than a short-term solution. In the meantime, the country will lose
foreign exchange reserves. Since the exchange rate is fixed, appropriate
monetary and fiscal policies will be needed to bring the economy into
overall equilibrium. This could, for instance, be done in Figure 13.6a by an

631
expansionary fiscal policy and a contractionary monetary policy [not
shown in Figure 13.6a], so that the IS and LM curves shift to go through
the point a. Such a procedure will also avoid the need for price
adjustments, in addition to avoiding balance of payments problems.

A caveat
The preceding arguments have made the implicit assumption that the BP,
IS, and LRAS curves do not shift as prices change. Our earlier arguments
show that they do so. Thus, if we start again from the point d in Figures
13.6a and 13.6b, as prices fall, the IS curve shifts to the right, the BP curve
shifts down and the LRAS curve shifts left (as in some of the Keynesian
models). It is unlikely that there would exist any fall in prices so fortuitous
as to make all the curves intersect at the same point, so that one or more
policy tools will still be necessary to restore equilibrium.
Extended Analysis Box 13.3: Disequilibrium in the Domestic Economy
and Stabilization through Monetary and Fiscal Policies: The Fixed
Exchange Rate Case with Interest Insensitive Capital Flows
In the interest-insensitive capital-flows case depicted in Figure 13.6b,
the initial equilibrium at the point d has deficient demand and balance of
payments deficit. Let an expansionary fiscal policy shift the IS curve
from IS0 to IS1 so as to pass through the point c. The excess demand at c
would raise prices and shift the LM curve up (not shown) toward the
point a. While this price level increase might fortuitously reduce the real
money supply by just enough to take the LM curve through the point a,
its likelihood is not encouraging. Alternatively, an appropriate
expansionary fiscal policy, combined with an appropriate contractionary
monetary policy could (hopefully) ensure that all the curves intersect at
the point a, thereby obviating the need for domestic price level changes
and also eliminating the balance of payments deficit.

13.9 The Need for Monetary and FiscalPolicies

632
 under Fixed Exchange Rates
Suppose that the authorities are willing to pursue both monetary and fiscal
policies to restore equilibrium in Figures 13.6a and 13.6b. An
expansionary fiscal policy could shift the IS curve to pass through the
point a; a deflationary monetary policy could shift the LM curve to also
pass through this point. An expansionary fiscal policy and a deflationary
monetary policy could, therefore, cure a deficient demand and a balance of
payments deficit.
Now consider the case shown in Figure 13.6c. At the point d , there is
excess aggregate demand relative to output yf. There is also a balance of
payments surplus. A contractionary fiscal policy could shift the IS curve to
pass through the point a. A contractionary monetary policy could also shift
the LM curve to pass through a. A contractionary fiscal policy and/or an
expansionary monetary policy could thus cure a balance of payments
surplus and excess demand in the context of Figure 13.6c.9
The analysis of the above cases shows that the achievement of overall
equilibrium in the context of fixed exchange rates is likely to require the
use of fiscal and/or monetary policies. Alternatively stated, the economy
with fixed exchange rates does not possess enough mechanisms that would
automatically bring the economy into equilibrium so that appropriate
policy actions are essential for maintaining equilibrium. Such a situation
arises because one of the equilibrating’ price variables, the exchange rate,
has been fixed, so that a policy adjustment has to compensate for its failure
to adjust.

13.9.1 Internal versus external balance

Chapter 12 had defined the internal balance of an economy as the equality
of its domestic aggregate demand and domestic full-employment output,
while external balance was defined as the equality of the demand and
supply of foreign exchange. Fixed exchange rates prevent changes in the
exchange rate from maintaining the latter. This imbalance does not pose a
serious problem if the external balance is such that there is a balance of
payments surplus and the country is willing to let its foreign exchange rate
reserves increase, while sterilizing their effect on the money supply.
However, a balance of payments deficit implies an outflow of foreign
exchange. The country could tackle this outflow through the sale of
foreign exchange from its foreign exchange reserves. Once these threaten
to fall below their desired level, continuing balance of payments deficits
will have to be eliminated. Fixed exchange rates prevent the use of the

633
exchange rate from eliminating them, forcing the use of at least one policy
tool to ensure the external balance of the economy. In principle, this tool
could be either monetary or fiscal policy. In practice, fiscal policy is too
inflexible a tool relative to the speed of movement in trade and capital
flows, so that monetary policy has to be diverted to maintaining the
external balance of the economy. If this is done, monetary policy cannot be
employed for maintaining internal balance.

13.9.2 A note of caution on the use of monetary and fiscal

policies
A mix of monetary and fiscal policies can restore equilibrium in the open
economy with fixed exchange rates and obviates any need for adjustments
in prices. However, a strong cautionary note on the pursuit of such policies
must be added here. To pursue such a policy efficiently and with the
confidence that the wrong policies are not pursued requires a great deal of
knowledge about the inter-relationships among the sectors. This
knowledge may not exist or there may be shifts in these relationships
because of the public’s response to the policies pursued, so that the pursuit
of discretionary stabilization policies can be risky: their outcome may
make the economy worse rather than better. The impact of the attempted
policies on the expectations of — and, therefore, speculation by — both
domestic and foreign investors about the ability of the government to
maintain its exchange rate must also be taken into account, and adds to the
difficulty in deciding on the appropriate policies.

13.10 The Effect of Relatively Different

Improvements in Productivity at Home and
Abroad under Fixed Exchange Rates
One of the big problems with fixed exchange rates occurs when the rate of
productivity growth is lower in the domestic economy than abroad. This
reduces the international competitiveness of domestic industries relative to
foreign ones — unless the lower level of domestic productivity is
accompanied by a proportionately lower real wage — and leads to a fall in
exports and a rise in imports. This fall in net exports has to be financed by
a capital inflow, which increases the indebtedness of the domestic
economy to foreigners and raises interest payments for the future. If the
domestic productivity continues to fall relative to the foreign one, the

634
process of falling net exports accumulates foreign debt and increases
interest payments to foreigners until the existing fixed exchange rate
becomes unsustainable. The country is then forced to devalue or let its
currency float, which immediately leads to its depreciation. If the
accumulation of foreign debt and interest outflows has become quite large,
the devaluation/depreciation may have to be by a fairly large percentage. It
may also be accompanied by large capital outflows and have severe
consequences on the economy in the form of a decline in output and
employment and a rise in inflation.
In order to keep exports competitive, the alternative to these adjustments
is for the domestic real wage to decline relative to the foreign one to match
the relatively slower domestic productivity growth. This is likely to lead to
migration to foreign countries, with warnings in the media about a brain
drain.
Therefore, in the long run, a fixed exchange rate in a context of relatively
lower productivity growth at home than abroad can pose severe problems.
Conversely, in the long run, the domestic economy is likely to do quite
well if it has higher productivity growth than abroad.

13.11 The Political Economy of Exchange Rates:

The Choice between Fixed and Flexible
Exchange Rate Regimes
This chapter and preceding one have considered some of the ways in
which equilibrium in the balance of payments can be brought about or
maintained in the face of disturbances. It could be done by market-
determined changes in the exchange rate in a system of flexible exchange
rates. But in a system of fixed exchange rates, equilibrium could only be
maintained either by allowing induced changes in domestic income, prices,
interest rates, etc. or by the appropriate use of the monetary and fiscal
tools. It would, therefore, seem preferable for each country to allow its
exchange rate to float rather than fix it, with all the constraints that the
latter implies. However, there are several reasons why a country
sometimes fixes its exchange rate against some or all foreign countries.
One reason for using a fixed exchange rate system is to avoid the
possibility of instability in the foreign exchange markets, under the belief
that flexible exchange rates promote such instability. The latter remains a
moot point. Its likelihood is now minimized by many economists,
especially those in the modern classical approach. Another reason for

635
keeping a fixed exchange rate is that exporters and importers of both
commodities and capital can make contracts without facing the uncertainty
of exchange rate changes, which hinders international flows of both
commodities and capital.
During the period from 1945 to the early 1970s, an often compelling
reason for a country to have a fixed exchange rate was its membership of
the International Monetary Fund (IMF), which had been set up in 1944.
Most of the nations of the world are such members. A basic rule of the
IMF from its inception to the early 1970s was that the member countries
would fix their exchange rates and change them only in cases of a
‘fundamental disequilibrium’ in the balance of payments.10 This rule was
adopted to limit the uncertainties affecting international trade and thus to
promote such trade. It was also designed as a way of preventing
manipulations (devaluations) in the exchange rate by any one country as a
way of improving its own balance of trade with others, with the possibility
of retaliatory changes in the exchange rates by other countries whose
balance of trade worsens. Such a retaliatory war of devaluations creates a
great deal of uncertainty in international trade and is usually accompanied
by other attempts by countries to curb their imports and promote their
exports by other means such as tariffs, quotas, etc. The resulting reductions
in exports and imports tend to substantially lower incomes and
employment in the countries involved in such a ‘war’.
A disadvantage of a country’s adoption of a fixed exchange rate is that
there sets in a great deal of reluctance for the country to change that rate
even if a long-term disequilibrium emerges in the balance of payments. If
the country has a surplus, which is sterilized and thus does not increase the
money supply, it has no incentive to appreciate its currency since it could
merely allow its reserves of foreign exchange to increase indefinitely. It
might, however, face the jealousy of other nations and some international
disapproval of its accumulation of reserves. In rare cases, it might even be
faced with economic and political pressures from other countries to
revalue. Thus, Germany and Japan, which had sizeable surpluses
throughout the 1960s, came under strong pressures from 1971 to 1973
from the USA to revalue. Part of such pressure was a 10% (surcharge)
duty imposed on imports into the USA with a promise that it would be
abolished after the appropriate revaluations had been made. Both Japan
and Germany were forced to revalue in 1971 and again in 1973 under such
pressure from the USA. But such explicit pressure tactics tend to be rare
are not always effective, and can be pursued by only a few powerful
nations.
A country with a fixed exchange rate and a persistent deficit in its

636
balance of payments could eliminate this deficit by devaluation. But
devaluations tend to be unpopular with the electorate. They are often
interpreted as a ‘cheapening of our money’ and ‘a loss of national face’
and prestige. Further, devaluations represent a worsening of the terms of
trade and increase the prices of imported goods. In some countries, often
low-income ones, these price increases are resented by the public, which
blames the government for them. Therefore, deficit countries, such as
Britain and the USA in the 1960s, often let their reserves fall or used
restrictive monetary or fiscal policies at home, rather than devalue. Such
policies usually prove to be palliative rather than cures of the underlying
disequilibrium. Thus, the USA, after resisting devaluation through the
1960s, finally succumbed to the market pressures, reinforced by
speculation, in March 1973, and allowed its dollar to float.

13.12 Other Tools for Handling Balance of

Payments Deficits
Countries also have at their disposal other tools to reduce deficits. These
consist of direct measures to change exports and imports and capital flows.
Imports can be hindered by the imposition of tariffs on their value or
quantity, or by quota limitations on the quantity that can be imported,
while exports can be encouraged by subsidies. These measures were
formerly ruled out by the General Agreement on Tariffs and Trade
(GATT) and now by the World Trade Organization (WTO), of which most
nations are members. This is because such actions by one country hinder
the exports of other countries and invite retaliation. Overall, such actions
reduce the gains in efficiency and welfare that come from free trade.
However, countries do often try to get around such rules, often in the form
of explicit or hidden subsidies to exporters and home producers.
Controls on the flows of capital can similarly consist of a ‘tax’ on capital
inflows or outflows, administrative barriers, or quota type limitations. A
country trying to reduce its capital outflow could impose a ‘tax’ on such
outflows.11 A country trying to reduce its capital inflows could similarly
impose a tax on the capital inflows or on the subsequent returns to them.12
Quotas on the inflows of capital are commonly imposed by countries that
are concerned about the foreign ownership of their real and financial
assets.13 Most LDCs have experimented with such controls. However,
there was a marked worldwide trend during the decade of the 1990s
toward the dismantling of such controls.

637
Extended Analysis Box 13.4: The Gold Standard and the Gold Exchange
Standard
The Gold Standard and its variation, the Gold Exchange Standard,14 is a
system under which countries maintain fixed rates of exchange of the
national currencies against an ounce of gold. The Gold Standard is a
system under which one or a few countries maintain fixed exchange
rates of their national currencies — called ‘key currencies’ — against an
ounce of gold, while other countries maintain fixed rates of exchange of
their national currencies against a key currency. Both systems have very
similar economic effects. Most countries of the world operated on the
Gold Exchange Standard under the Bretton Woods System of the IMF
from 1946 to about 1970, with the US dollar being the main key
currency.
The Gold Standard was a major tenet of exchange rates regimes prior
to World War II. Britain had set a fixed conversion rate between its
currency — the pound — and gold, so that its currency was on a ‘Gold
Standard’. British colonies and dominions had fixed their exchange rates
against the British pound, so that their currencies could also be
converted into gold, if not directly, then through their conversion into
pounds and then into gold. Canada was one of these countries. France
and its colonies were also on the Gold Standard. These empires
constituted trading blocs akin to customs unions, with trade among the
member countries facilitated by fixed exchange rates and with gold
flows serving to maintain these exchange rates.15 While it was not
obligatory for a country to be on the Gold Standard, most of the world’s
major trading nations were directly or indirectly on it.
From 1946 to the early 1970s, the US dollar was fixed against gold at
$35 per ounce of gold. Most of the world’s currencies maintained —
directly or indirectly through other currencies — a fixed exchange rate
against the US dollar, so that the world was said to be on the Gold
Exchange Standard.16
Being on the Gold Standard limits the country’s independence to
create money, which limits its ability to pursue an independent
discretionary monetary policy. When it is pursued fully, a balance of
payments surplus for a country causes the inflow of gold to increase the
money supply through exchanges of the gold inflow with the central
bank for the local currency, while an outflow decreases the money
supply. Consequently, aggregate demand and prices rise in the countries
with balance of payments surpluses, thereby increasing their imports and
decreasing their exports, so that their balance of payments would tend to

638
equilibrium. Correspondingly, demand and prices fall in the countries
with balance of payments deficits, thereby decreasing their imports and
increasing their exports, so that their balance of payments also tends to
equilibrium. When these changes are allowed by the individual countries
to occur, the Gold Standard serves as a mechanism for equilibrating the
balance of payments of the individual countries. It also prevents
excessive creations of new money by national central banks.
The 19th and early 20th centuries were a period in which few
countries had central banks and the monetary theories of inflation and
output were poorly developed. The creation of currency was usually in
the hands of private individuals and firms and subject to their private
interests, so that there was little national control over the money supply
in the economy. In these conditions, the Gold Standard provided
countries with external restraints on their private institutions (usually
banks) in creating currency, and in the aggregate over their national
money supplies.
The Gold Standard has several defects. Among these are:
• The fixed exchange rate may not be the equilibrium one for the foreign
exchange market. If the exchange rate for a national currency is set
higher than the equilibrium, it will suffer a persistent balance of
payments deficit. This would produce an outflow of gold and a
decrease in its money supply. The consequent fall in aggregate demand
would result in a fall in employment and output. This is likely until the
exchange rate is lowered.
• Even if the exchange rate is initially set at the equilibrium level, there
exists the possibility that economic changes will render the national
currency over- or under-valued. This can cause speculation about
adjustments in the exchange rate and panic in the foreign exchange
markets. Such bouts of intense speculation are especially troublesome
in the modern period with the increasing international flows of capital.
• Countries may choose to adopt an independent monetary policy and try
to sterilize the inflows and outflows of gold. This is especially likely
for inflows of gold, since this would mean an increase in foreign
exchange reserves, usually considered a desirable objective. This
policy would hinder the automatic mechanism of the Gold Standard for
correcting balance of payments disequilibria.
• The world supply of gold limits the world money supply, so that if it
fails to grow proportionately with world trade and output, its lower
growth rate will produce a persistent tendency toward inadequate world
demand and recessions — which will hamper output and employment
in national economies.

639
 Starting in 1971, the Gold Standard was effectively abandoned by
virtually all nations during the third and fourth quarters of the 20th
century because of several developments. One of these was the setting
up of the central banks, which wanted to maintain control over their
money supply, as they thought appropriate for the national interest,
rather than let it be determined by balance of payments flows.
Accompanying this was the elimination of private note issues. Another
was the increasing magnitude of capital flows, which can cause intense
speculation against a currency leading to a crisis in its balance of
payments. A third was the break-up of the large empires that had
imposed fixed exchange rates on their colonies, dominions, and
affiliated countries. While there are still some calls for the restoration of
the Gold Standard, it is highly inadvisable in the modern age.

13.13 Summary of the Costs and Benefits of a

Fixed Exchange Rate
The major benefit of a fixed exchange rate accrues from its elimination of
exchange rate fluctuations. This reduces the risk from such fluctuations for
importers and exporters and for investors, so that the volume of trade and
investment would be larger. This allows the country to benefit more from
specialization in the products in which it has a comparative advantage, and
also to attract foreign investment.
Another advantage arises from the constraint that a fixed exchange rate
imposes on the ability of the national central bank to expand the domestic
money supply and cause inflation rates higher than in other countries. This
becomes valuable if the central bank has a record of mismanagement in
this respect. While this can provide the needed discipline for an otherwise
irresponsible central bank, it does not necessarily do so if the central bank
resorts to periodic devaluations to solve the problems caused by an
excessive creation of the money supply.
The disadvantage of a fixed exchange rate is that monetary policy
becomes constrained by the need to divert monetary policy (money supply
and interest rates) to support of the level at which the exchange rate has
been fixed. For example, if the interest rate is too low relative to ones
abroad, the resulting net outflow of capital will soon exhaust the country’s
foreign exchange reserves, to be followed by either abandonment of the
fixed exchange rate regime or tightening of monetary policy by raising
interest rates. Further, under fixed exchange rates, monetary or confidence

640
shocks are transmitted to the domestic economy unless the central bank
rapidly moves domestic interest and inflation rates to compensate for such
shocks.
By comparison, a floating exchange rate maintains equilibrium in the
foreign exchange markets (i.e., external balance ) so that monetary policy
can concentrate on the objective of maintaining aggregate demand equal to
the full-employment output (i.e., internal balance ).

13.14 Dollarization
‘Dollarization’ means the adoption, by a country other than the USA, of
the US dollar as its domestic currency, without a distinct national currency
in existence at the same time. An example of this occurred in 1999 when
Ecuador announced that it would adopt the US currency as its national
currency.
Under dollarization, any fundamental deficit (surplus) in the balance of
payments will have to be met by a fall (increase) in the domestic money
supply, with a consequent fall (increase) in aggregate demand — ending
up with a fall (increase) in domestic prices and output. For monetary
purposes, the country adopting dollarization becomes in some ways an
extension of the US economy. This promotes trade with the USA, and
capital and technology flows. However, it does not necessarily attain the
perfect flows of commodities, labor, capital, and technology that occur
within the USA.
An increase in the monetary base represents a net increase in government
revenues equal to the amount of the increase. This increase in revenues is
called ‘seigniorage’ . The retirement of a national currency in favor of the
US dollar means that the seigniorage inherent in the existing monetary
base and its future increases is lost by the domestic country while the USA
gets this seigniorage. This is a very significant amount, so that the return
by the USA of the seigniorage to the dollarizing country, or at least its
division in some proportion, has to be negotiated and settled between the
two countries.
Dollarization of a national currency has three consequences:
1. It represents the adoption of a fixed exchange rate (at one to one)
between the national currency and the US dollar. This promotes
international trade in commodities and increases capital flows.
2. It represents a commitment that this exchange rate will not, and cannot,
be changed either by the policymakers (which they can do even under
fixed exchange rates if there was a separate national currency) or even

641
 by the markets. This eliminates the possibility of speculative attacks on
the currency in circulation, which has become the US dollar one.
3. The dollarizing country loses seigniorage 17 from the issue of the
currency circulating within it. Unless there is an agreement to sharing
this seigniorage, the USA will get the seigniorage from the adoption of
its currency and its increased circulation in the economy.
Dollarization usually is an attempt to stabilize the value of the circulating
medium of payments and to prevent speculation against its depreciation.
As shown in the analysis of monetary policy under a fixed exchange rate,
the adoption of a fixed exchange rate through dollarization implies loss of
control over the domestic money supply and, therefore, reduces the scope
for monetary policy to address the ills (e.g., deficient aggregate demand)
of the domestic economy. This is a potential cost of dollarization, as of a
fixed exchange rate, which can become significant in demand-deficient
recessions. But this limitation can also be of benefit to the economy if the
national central bank has shown a lack of internal discipline on monetary
policy and has itself been a significant source of inflation in the economy.
This would have happened if the central bank itself had on its own volition
created high money supply growth rates or had done so at the behest of the
government. Such behavior has sometimes caused high rates of inflation,
even hyperinflation in some countries in some periods, which significantly
lowered output and employment. Therefore, whether the loss of national
control over the money supply — and through it on domestic aggregate
demand — is a cost or benefit depends on the public’s trust in — and past
experience with — the central bank’s handling of its monetary policy
instruments.
Having one’s own currency with a flexible exchange rate allows an
independent monetary policy, which can be used for counter-cyclical
stabilization policies, as well as for reducing unemployment and increasing
the output growth rate. However, the adoption by many central banks of
price stability as the preferable goal itself imposes limits on the use of
monetary policy, provided this goal is pursued with determination.
The importance of point (2) lies in dollarization preventing exchange
rate speculation. Dollarization will prevent speculation not only against the
local currency when there are (limited and transient) random fluctuations
in the balance of payments — which occur under both fixed and flexible
exchange rates — but also when there are fundamental shifts in the
balance of payments. The latter scenario is the significant one for
evaluating the merits of dollarization.
Why should one country adopt the currency of another country as its

642
own, especially when it is likely to lose at least some of the seigniorage
from issuing a national currency? Clearly, this should only be done when
the net benefits from dollarization are positive. Often the expected benefit
arises because the national central bank through substantial money supply
increases had created very substantial inflation rates in the past and these
have contributed to the collapse of production and employment in the
economy in the recent past. Dollarization is meant to ensure a more stable
currency and price level — and thereby to restore production and
employment to their full employment levels. Another argument for
dollarization arises if it is part of a package that includes free access to the
US markets or a customs union (which has free flows of commodities
among its member countries) with the USA.
Should the USA welcome dollarization by other countries? It earns
seigniorage from the dollars floating in other countries, which benefits the
US government and economy. However, it is likely that the country
adopting dollarization would want to share in this seigniorage, so that a
deal has to be struck on this division. Further, if the dollarizing economy is
quite significant in size relative to the US one, the strength and stability of
the US dollar in world markets will become partially dependent on the
performance of the dollarizing economy. This represents a loss of control
of the Fed over the internal and external value of the US dollar, which the
Fed may be reluctant to accept.
For some countries, especially in Asia and Africa, the trade flows with
Europe are more significant than with the USA. For them, appropriate
dollarization would mean the adoption of the euro as their national
currency.18
The forces in favor of dollarization recently have been:
• Globalization has meant lowering tariffs and quotas and increasing the
levels of imports and exports relative to GDP, as well as increasing
capital flows across countries (this topic was covered in Chapter 12).
• The implementation of regional economic blocs or unification of several
nations into one. The European Union provides a good example of the
latter. Its currency, the euro, has replaced the national currencies of most
of the countries in the European Union.

13.14.1 Dollarization and different productivity growth

rates
The problems posed by a lower productivity growth at home than abroad
were discussed above in the context of fixed exchange rates. Dollarization
intensifies these problems. If the dollarizing economy has lower

643
productivity growth than the USA, then its real wage must decline relative
to that in the USA. This will cause migration — often a ‘brain drain’ of the
best workers — from the domestic economy to the USA. If the free flow
of labor is very limited or is not permitted at all, then the domestic
economy is likely to maintain relatively high unemployment rates, as well
as lower real wages and standard of living. It would become what is
termed an ‘economically depressed area’.
The Atlantic Provinces in Canada provide an example of this long-run
phenomenon: these provinces share the Canadian dollar with the other
Canadian provinces and so have ‘Canadian dollar dollarization’. Canada’s
Atlantic Provinces have had slower productivity growth than some of the
other Canadian provinces, such as Ontario, Alberta, and British Columbia,
on a long-run basis. Consequently, over the 20th century, the former lost
quite a bit of their population to these other provinces, but still had higher
unemployment rates and lower real wages. By extension, if Canada was to
adopt the US dollar as its currency but had slower productivity growth
than the USA, Canada is likely to suffer a significant out-migration of its
population to the USA, as well as experiencing higher unemployment rates
and lower real wages.
Conversely, dollarization should be great for Canada if its productivity
growth was higher than in the USA. It is then likely to experience
increasing exports and favorable balance of payments, an inflow of
population from the USA, lower unemployment rates and a higher
standard of living.
Extended Analysis Box 13.5: A Dual Currency System: Dollarization
Along with a Separate National Currency
Instead of complete dollarization with only the US dollar circulating as
the sole medium of payments, a country may have two currencies
functioning equally or almost equally efficiently as media of payments.
Such cases are likely to involve a foreign currency, often the US dollar,
in addition to a distinct domestic one. Such as case occurs in Lebanon,
in which the US dollar and the Lebanese currency circulate
interchangeably and payments, even in shops, are acceptable in either
currency.
A country with two circulating media of payments can maintain a
fixed exchange rate or a flexible one between the two currencies. A
simple example of such a fixed rate case is the circulation of the cent,
the five- and ten-cent coins, along with the dollar notes, within the USA.
However, under a fixed exchange rate between the national currency and
the US dollar, the central bank has to offset any shifts in their relative

644
 demands and supplies which threaten to change the pre-set exchange
rate between them. This means that the central bank must have enough
reserves of US dollars to meet any excess demand for them at the fixed
exchange rate.
Alternatively, the country could allow a floating exchange rate
between the currencies. In this case, changes in the demands and
supplies of the two currencies will change the equilibrium/market
exchange rate, so that the central bank does not need to buy or sell US
dollars. However, the public must acquire continuous information even
throughout the day on fluctuations in the exchange rate. This occurs on a
continuous basis in Lebanon.

13.15 Currency Boards

A country, more so in the past but rarely nowadays, may have a currency
board instead of an independent central bank.19 With a currency board, the
country maintains a fixed exchange rate against a designated foreign
currency (often the US dollar), and the monetary base — a liability of the
currency board — is backed fully by its foreign exchange reserves. As
these reserves increase — e.g., through a balance of payments surplus —
the currency board increases the monetary base and the money supply in
the economy increases. Conversely, as foreign exchange reserves fall, the
monetary base and the money supply are decreased. Other than this, the
currency board does not have discretion to change the money supply or
manage interest rates and, therefore, cannot pursue independent monetary
policies for achieving domestic goals.
Currency boards were common in the colonies of imperial countries —
e.g., the UK— during the first half of the 20th century. They were a means
of linking the currency and the economies of the colonies to those of the
imperial country. Further, if the imperial currency was under the gold
standard — i.e., with its value fixed in terms of gold — the colonies also
indirectly adhered to the gold standard. Such currency boards were usually
replaced by central banks on independence. In other cases, countries,
though independent, maintained currency boards with a strict adherence to
the gold standard, implying a fixed value of the domestic currency in terms
of gold.
The main reason for having a currency board, as under dollarization, is
to impose discipline on the central bank, so that it cannot increase (or
decrease) the domestic monetary base in an ‘irresponsible’ manner (e.g.,
such as to cause very high inflation rates). Under a currency board,

645
changes in the domestic money supply are strictly determined by inflows
and outflows of foreign exchange, just as under dollarization.
The disadvantage of having a currency board or dollarization is that the
central bank does not have the discretion to pursue an independent
monetary policy in the national interest, e.g., to stabilize aggregate demand
at a level such as to ensure full employment in the economy or to reduce
the impact of foreign shocks from on the domestic economy. In the
modern context, countries prefer to have a central bank that can perform
these roles, so that, with rare exceptions, they do not have a currency
board.

13.16 Conclusions
• Fixed exchange rates continue to dominate at the intra-national and intra-
bloc levels, with flexible exchange rates being the common rule between
countries and between blocs of countries.
• One of the most remarkable changes in monetary arrangements in this
respect in the world economy in recent decades has been the creation of a
new currency called the euro, and the conversion of the national
currencies of the European Union to a single one.
• Open economies impose considerable limitations on the successful
pursuit of monetary and fiscal policies. The complexity of the
interactions among the various sectors is greater for the open than for the
closed economy, and greater under a fixed exchange rate than under
floating ones.
• The requirements of the purchasing power parity and the interest rate
parity operate as constraints on the policies that can be successfully
pursued under fixed exchange rates. In the limiting case of the continuous
clearing of the domestic labor markets at full employment and perfect
capital flows, while monetary and fiscal policies could change aggregate
demand in the domestic economy, they could have no impact on domestic
real employment, output, and interest rates.
• A major problem with maintaining fixed exchange rates arises when
there are shifts in productivity among countries — especially if the
domestic economy has a relatively low growth rate of productivity
leading to a fall in net exports.
• If a country is unable to control its money supply growth and inflation
rates, adoption of dollarization imposes a regimen for this control and
tends to lower growth rates — at least in the early years. However, a
relatively lower domestic growth rate of productivity or political
instability will lead to a balance of payments crisis and either increased

646
 unemployment or the abandonment of dollarization.
• A dual currency system with dollarization and a national currency is
usually not sustainable over time because of differences in relative
productivity growth rates, etc.
• A given nation does use a single currency and, therefore, embodies
dollarization among its various regions. However, this is supported by
unhindered capital and labor flows, so that workers who become
unemployed in regions with lower productivity growth tend to move to
regions with higher productivity growth, thereby keeping unemployment
lower and raising output. Such flows do not occur under the adoption of
another country’s currency, unless the latter allows free labor and capital
mobility between the countries — which usually does not happen. A
precondition or corequisite for successful dollarization over time is the
free flow of labor, capital, and technology.

KEY CONCEPTS
Balance of payments curve
World inflation rate
Convergence of inflation rates under fixed exchange rates
Convergence ofinterest rates under fixed exchange rates
The Gold Standard and the Gold Exchange Standard
Seigniorage
Dollarization
A dual currency system with dollarization and a national currency, and
Currency board .

SUMMARY OF CRITICAL CONCLUSIONS

• Fiscal policy is more effective than monetary policy in changing
aggregate demand if the exchange rate is fixed.
• Under a fixed exchange rate, with interest rate parity and continuous
labor market clearance at full employment, there is no role for monetary
and fiscal policies in changing output or employment or interest rates.
• Under a fixed exchange rate, with interest rate parity and continuous
labor market clearance at full employment, an expansionary fiscal policy
reduces net exports and worsens the trade balance. A contractionary
fiscal policy increases net exports and improves the trade balance.
• Under purchasing power parity and a fixed exchange rate, the domestic
price level is fully determined by the foreign one. The domestic inflation
rates among countries will converge to the world inflation rate.

647
• Fixed exchange rates reduce the flexibility of the economy in adjusting to
shocks and increase the difficulties, the uncertainty and the limitations on
the successful pursuit of monetary and fiscal policies.
• Dollarization, in addition to the disadvantages of fixed exchange rates,
involves loss of seigniorage and a risky economic strategy over time
unless it is accompanied by unhindered flows of labor, capital and
technology.

REVIEW AND DISCUSSION QUESTIONS

1. What are the main benefits to a country of maintaining a fixed exchange
rate against other currencies?
2. Define the BP curve. Why is it essential to incorporate it into the
analysis of the open economy with a fixed exchange rate?
3. Given a fixed exchange rate, what is the effect of an increase in (a) the
foreign inflation rate and (b) exports, on the domestic money supply and
price level? Base your analysis on the IS-LM framework for the open
economy, making explicit your assumptions.
4. Given a small economy with a fixed exchange rate, why might the
effects of fiscal policies differ for the developed economies with
developed financial markets from those for the developing economies
with under-developed financial markets? Base your analysis on the IS-
LM-BP framework for the open economy.
5. Given a small open economy with a fixed exchange rate, what does
interest rate parity imply for the determination of the domestic interest
rate?
6. Under a fixed exchange rate regime, should the monetary authority try
to manipulate the flows of commodities and capital to its own country?
If so, what policies are appropriate for this purpose?
7. What are likely to be the advantages and disadvantages to the Canadian
economy and the US economy of dollarization (replacement of the
Canadian dollar by the US one)?
8. Why can dollarization be a threat to the maintenance of full employment
and living standards in the dollarized country? What other policies are
necessary to make it successful over time?
9. What is meant by the term “the Gold Standard”? Under the Gold
Standard, what happened to an economy’s money supply when it had a
deficit or surplus? How did this restore equilibrium in the economy’s
balance of payments?
10. Why have most economies abandoned the Gold Standard? Should it be
reinstituted for the modern economies? Discuss.
11. Discuss, from the perspective of (a) a small country and (b) the USA,

648
 the pros and cons of a free trade area with fixed exchange rates
encompassing the Americas?
12. What is an exchange-rate crisis? What can cause it? What can a
country undergoing an exchange rate crisis do to moderate it in (a) the
short term and (b) the long run?

ADVANCED AND TECHNICAL QUESTIONS

T1. Can the same policies prescribed by the IMF for a developed economy
and a developing economy, with both facing the same problem, benefit
one economy but be detrimental to the other one. Discuss the relevant
features of the two economies that bring about these results.
T2. Suppose controls on capital flows were imposed during an exchange
rate crisis. What is likely to be the response of the investors to the
imposition of such controls? Would this response improve or worsen the
balance of payments situation?
T3. In the light of the European experience over the past decades on the
evolution of the European Union and its single currency, the euro,
discuss the following. What are the costs and benefits to Canada (or any
other country of your choice) and the USA if they were to not become a
single country but jointly (a) set their mutual exchange rate in a narrow
band, (b) adopt a fixed exchange rate between their currencies, or (c)
adopt a single currency (with Canada adopting the US dollar)? What are
the pros and cons of these options for (i) the country that you chose and
(ii) the USA?
The Basic AD-AS Model for this
Chapter

The commodity sector:

Additional information: P = 2,P F = 4, = 4 (i.e., the exchange rate has

been fixed by the central bank at this level). Further, assume that for the

649
given price level, output is determined by the level of aggregate demand.
T4. Calculate the following.
(a)Derive the IS equation.
(b)Assuming that the central banks sets r = 2, what is the value of
aggregate demand y d [y 0d]?
(c)Assuming r = 2, if the government increases its expenditures to 1000,
what is the value of yd [yd1]?
(d)What is the fiscal multiplier for aggregate demand?
T5. Given the basic model for the commodity market but with the interest
rate set by the central bank at 0.25, what is the level of income? What
becomes the level of income if autonomous investment i becomes 800?
What is the investment multiplier in this case?
T6. Given the basic model for the commodity market, a perfect capital
market and interest rate parity, as well as r F = 0.25, P = 2, P F = 4, = 4
(and with no change expected in this exchange rate):
(a)What is the domestic interest rate?
(b)Given this interest rate, what is the level of output (hint: use the
commodity market equation for this derivation)?
(c)Suppose the money demand function is m d = 0.25y — 60r . Given
the interest rate determined by the IRP condition, what is the money
demanded equal to? What is the level of the money supply consistent
with the above (i.e., derived in (a) and (b) above) levels of the interest
rate and output?
(d)Given the money demand function as m d = 0.25y – 60r and that the
central bank maintains a constant money supply at 1,000, what will be
the domestic interest rate (Hint: this is determined by the domestic
commodity and money markets, i.e., IS and LM curves)?
(e)If r F = 0.25, P = 2, P F = 4, and IRP holds, what will be the implied
devaluation or revaluation of the exchange rate?This page
intentionally left blankPART IV Growth Economics This page
intentionally left blank
1As the preceding chapter pointed out, this is an approximation.
2However, as discussed earlier, changes in the domestic or foreign prices
or foreign interest rates will shift the IS curve.
3The international and domestic bond markets are different in nature: the

650
former represents an exchange of domestic for foreign bonds while the
latter represents exchanges between money and bonds and requires
equilibrium in the market for money.
4Note that RF refers to foreign exchange reserves.
5The size of the balance of payments deficit can be related to the interest
gap between the one given by the intersection of the IS and LM curves
and, for this level of aggregate demand, that given by the BP curve.
6The second assumption is sometimes violated when the authorities have
fixed the exchange rate but the public believes that this exchange rate
cannot be maintained. Suppose that the public expects a devaluation. In
such a case, there are usually speculative outflows of foreign exchange,
which reduce the country's foreign exchange reserves and worsen the
balance of payments, thereby putting pressure on the authorities to devalue
in an attempt to stem the outflow of reserves.
7The weights would be the national exchange rates.
8However, at the world level, causation does run from the world money
supply to the world price level.
9These arguments involving shifts in several curves become quite complex
when price changes are considered. The complexity arises basically from
the dependence of the various sectors on domestic prices. It is, therefore,
preferable to work directly with the equations derived earlier. If one
wishes to stay with the diagrammatic form, it would be better to adapt the
arguments to the (y, P) space while eliminating r as a variable through a
combination of the expenditure and the monetary sectors and a
combination of the foreign and the monetary sectors.
10A fundamental disequilibrium is to be thought of as a long-run
disequilibrium rather than short-run — monthly or annual —
disequilibrium.
11The USA, for example, did so in 1964 when it imposed a 15% ‘interest
equalization tax’ on interest and dividend payments from foreign stocks
and bonds bought after 1964.
12Several European countries did so in 1973. An example of a quota on
outflows of capital was the ‘foreign credit restraint’ program instituted by
the USA in 1965 and designed to limit the growth of US investments
abroad.
13Japan has had some such controls through most of the post-1945 period.
14The Gold Exchange Standard differs from the Gold Standard in that
gold coins circulate in the latter but not in the former. In both, the central
bank is willing to exchange its paper currency for gold bullion and vice
versa.
15In the modern period, the European Union has instituted fixed exchange

651
rates among its member countries, though without having a fixed exchange
rate between the euro and gold, so that it is not on the Gold Standard.
16Under the Gold Exchange Standard, gold coins did not circulate and
private individuals in many countries were not permitted to hold or trade in
gold (other than for jewelry) but countries could trade their currencies
against gold.
17Seigniorage is the gain in revenue to the agency/country which issues
the currency, since this currency costs very little to produce while it is used
to buy bonds which pay interest or to buy commodities.
18See A. Alesina and R. J. Barro, Dollarization, American Economic
Review, 91, May 2001, 381–385, for a compact discussion of this topic.
19As of 1998, currency boards exist in Hong Kong, Argentina, Estonia,
Lithuania, and a few other countries.

652
 PART IV
Growth Economics

653
 CHAPTER 14
Classical Growth Theory

The underlying model of growth analysis is known as the

neoclassical, classical, or Solow growth model. It shows the
impact of increases in physical capital and the labor force on the
long-run growth rate of output and the standards of living. Among
its distinguishing features are its assumptions of exogenous
technical change and the absence of money. This chapter presents
this analysis.
The next chapter extends the growth analysis of this chapter to
encompass endogenous technical change and the relationship
between the economy’s output growth and its financial
development.

This chapter studies the economic forces that increase the standard of
living by examining the long-run determinants of the output growth rate.
Improvements in living standards come about because of increases in
output per worker. These increases are cumulative over time, so that even
quite small (for instance, 1 to 4%) rates of increase, when compounded,
can produce very large improvements in living standards over periods of
50 years or a century. For instance, over a century, a 4% growth rate leads
to an output per capita that is seven times greater than what a 2% growth
rate will produce. Since growth rates differ across countries, the
underlying reason for the prosperity of some nations, while others are
relatively poor, lie in the former having had higher growth rates over long
periods in the past. Therefore, the study of the prosperity of nations is a
study into the long-run growth of their output.
This chapter and the following one present the economic theories on the
determination of the growth rate. The starting point of these theories is the
‘classical or Solow growth model’. We have used this term for it because
this model represents the basic growth theory element of the classical
paradigm, as defined in Chapter 11. It was originally called the
‘neoclassical growth model’ and many expositions of it still continue to
use this name. It is also known as the Solow model, since Robert Solow

654
was the economist who proposed it in 1956.1
Fact Sheet 14.1: Long-Term Real GDP Growth in Selected Countries,
1960–2003
This Fact Sheet illustrates the variations in annual real GDP growth rates
across countries and, for a given country, over time. For any given
county, growth rates vary over the business cycle. 1980 and 1980 were
recession years in most economies, so that the growth rates in these
years were lower than in other years. Growth rates have been
consistently lower in west European countries than in the north
American ones. While India and China had lower growth rates than the
USA in earlier years, China’s and India’s growth rates rose dramatically
after 1990 when they started limited liberalization in production, foreign
direct investment, foreign trade, etc.

The very considerable variation in the growth rate of real GDP for a
given country is illustrated in Fact Sheet 14.2 from data for the USA.
Fact Sheet 14.2: Real GDP Growth Rates in the USA
This Fact Sheet illustrates from the US data the very considerable
variation in real GDP growth. A major reason for the variation in this
growth rate is fluctuations in the rate of technical change, which is a
major focus of this chapter. Other reasons for variation in this growth
rate are fluctuations in aggregate demand and economic crises.
Consequently, the variation in real GDP growth rates partly causes
business cycles and is party caused by the other determinants of business

655
 cycles (see Chapter 16).
Note that if the growth rate is negative, real GDP falls. A period of
negative growth is a symptom of a recession.

The assumptions of classical growth theory

Several fundamental assumptions distinguish the neoclassical/classical
growth theory from the short-run macroeconomic models of the classical
paradigm. For its growth model:
• The emphasis is on the evolution of the long-run equilibrium of the
economy in response to changes in its three determinants of growth,
which are the labor force, capital, and technology.
• All markets are assumed to be continuously in equilibrium. There is no
uncertainty, so that the impact of errors in expectations is omitted. There
are also no adjustment costs, whether of prices, employment, capital
stock, or technology. Hence, the economy is always in long-run
equilibrium at full employment, and the deviations from long-run growth
path, along an SAS curve in the short-run analysis or due to
disequilibrium, are omitted from the analysis.
• The economy is a closed one. It is also without a government sector.
Further, the money and bond markets are omitted from the analysis.
• The physical capital stock is variable.
• The labor force is variable.
• Technology can change, so that the shifts in the production function are
allowed. However, when technological change is incorporated in the
standard classical growth models, the change in technology is assumed to
be exogenously given.
These assumptions delineate the separate spheres of short-run models
and growth theory. The raison d’être of short-run macroeconomic models
is to examine the nature of the equilibrium in the different macroeconomic
markets and the consequences of deviations from these. Growth theory, by
assuming equilibrium in all markets, takes the full-employment state of the

656
short-run models for granted. As a consequence, the main items of interest
in the short-run macroeconomics models — such as the determination of
prices, unemployment, deviation of output from its full-employment level,
monetary and fiscal policy effects, etc., which are related to aggregate
demand in the economy — are normally not considered in growth theory.
The focus of the growth theories is on the impact of variations in capital,
labor force, and technology on the evolution of full-employment output.
Classical growth theory studies the long-run equilibrium paths of the
economy. A special state of such long-run growth paths is steady state
(SS) paths along which the growth rates of designated variables,
particularly the capital–labor ratio, do not change. Quite clearly, the actual
economy of any country at any particular time may not even be in long-run
equilibrium, let alone in an SS. However, most economists believe that,
over long periods, economies tend toward the SS and that its study reveals
the long-run tendencies of the economy. Given this belief, the usual focus
of classical growth theory is on examining the existence and stability of SS
output. But in practice, the public and the policymakers maintain a lively
interest in the level of per capita output and its pre-SS (i.e., before the SS
is reached) growth rates, so that we will also examine the long-run
determinants of these variables.
Our interest in the growth theories will focus on three issues:
1. The SS level of output per capita, since this reflects the standard of
living.
2. The SS growth rate of output per capita, since this reflects the growth of
the standard of living.
3. The growth rate of the economy in the long-run equilibrium prior to or
on the way to the SS one, i.e., the ‘pre-SS growth rate’. This is an
important and topical state to study since virtually no country at any
point in time can be said to have reached the SS, but can be said to be on
the way to one.
Exogenous shifts and policies that only change the level of per capita
output are said to have level effects, while those which change its growth
rate are said to have growth effects.
Basic mathematics of growth theory
For a variable Y (output), the change in Y over time has been compactly
designated in this book as Y or Y ′ and its growth rate has been specified
as ( Y/ Y ) or Y ′′. These are calculated for period t as:

657
Since Y = y . L, where y is output per worker and L is the labor force, we
have Y′′ = y′′ + L′′ .2 Therefore, for y = Y/L, we have y′′ = Y′′ – L′′.
Similarly, for k = K/L, we have k′′ = K′′ – L′′.
If the growth rate of output Y is constant at the rate 4%, we will often
write it as 0.04. In the latter version, Y″ = 0.04 (= 4/100). If the growth
rate is γ,we write Y′′ = y .3 If Y does not change, its growth rate will be
zero.
Extended Analysis Box 14.1: Setting the Boundaries of Macroeconomic
Growth Theory
The boundaries of the standard treatments of classical growth theory
analysis are set by:
• Classical growth theory focuses on output growth resulting from labor
force growth, growth in the capital stock and change in technology. It
ignores the growth in the stocks of money and bonds (including
equities). In fact, many economists believe that the long-run growth of
the economy is independent of the money stock and even of its growth
rate. Consequently, the inclusion of money in growth theory is left to
the special topic of monetary growth theory.
• Most treatments of growth theory ignore the government sector —
thereby ignoring government expenditures, taxation, and deficits.
• Most presentations of growth theory also ignore the external sector—
thereby ignoring imports, exports, the terms of trade, the balance of
trade, and the balance of payments.
The economy is always at full employment. Therefore, classical growth
theory does not examine the determination of unemployment or changes
in its rate in the short or even long run. Therefore, it does not see a need
to differentiate between employment and the labor force. Further, it
usually assumes that the labor force grows at the same rate as the
population.
A critical assumption of short-run macroeconomic theory is that the
saving and investment functions are distinct from each other. In reality,
saving is done by households under utility maximization (subject to a
budget constraint) while investment is done by firms based on profit
maximization. Given a closed economy without a fiscal sector, short-run
equilibrium requires the equality of saving and investment. Since growth
theory starts by assuming such equilibrium, saving and investment have
to be identical (always and not merely equal in equilibrium) in growth
theory. While this identity could be defined in terms of an exogenous
investment function and accommodative saving, classical growth theory

658
 assumes an exogenous saving function and accommodative investment.4
That is, classical growth theory specifies the saving function from
household behavior and assumes that investment always equals saving
— which is determined independently from investment. Hence, an
investment function independent of the saving one is not specified in
classical growth theory. Further, since this theory assumes continuous
full employment, saving is at its full-employment level, while
investment is determined by — and always equal to — the full-
employment saving.

14.1 The Classical (Solow’s) Growth Model’s

Assumptions
This section sets out the benchmark classical growth theory. It was
formulated by Solow in 1956 and is also known as the Solow model or the
benchmark neoclassical growth model.5
Contrary to our earlier usage (in the short-run models) of capital symbols
to designate the nominal values of the variables in the short-run models,
we will use for growth theory capital symbols to designate the real values
of the variables and the corresponding lower case symbols to designate
their per capita values.

14.1.1 The technology of production

Assume a production function of the form:

where:
Y = real output,
K = physical capital stock,
L = labor force,
Yk = marginal product of capital (MPK) (= ∂Y/∂K),
Ykk < 0 = decreasing marginal product of capital (decreasing MPK),
YL = marginal product of labor (MPL) (= ∂Y/∂L),
YLL < 0 = decreasing marginal product of labor (decreasing MPL), and
YLK > increase in the marginal product of labor as capital is
=
0 increased.

659
This production function is assumed to have constant returns to scale. This
means that if both labor and capital are increased in the same proportion
(say α), output will also increase in the same proportion (α). Hence:

where α can be any positive constant. If we set a equal to 1/L, this

equation implies that:

which can be rewritten as:

where:
y = output per worker6 (= Y/L ),
k = capital per worker (= K/L), )
yk = marginal increase in output per capita as capital per worker
increases (= MPk = MPK ),
ykk = change in MPk as capital per worker increases,
ykk <
= diminishing MPk, and
0
ykk >
= increasing MPk.
0
This form of the production function asserts that, for the given technology,
output per worker depends only on capital per worker.

14.1.2 Saving, investment and the change in the capital

stock
The Solow model assumes that in the aggregate the average propensity to
save (APS) is constant and always equal to the marginal propensity to save
(MPS). Designate the value of the APS byσ (which is a Greek symbol,
pronounced as ‘sigma’). Since APS equals S/Y , we have S/Y = σ .
Therefore, the economy’s total saving S is given by:

S = σY.

Designate the per capita saving (= S/L) by s. Therefore:

where:

660
 S = aggregate saving in the economy,
s = saving per worker (per capita saving), and
σ = average propensity to save (APS) out of output.7
The capital stock K changes by the amount of real investment I . Since
there is continuous equilibrium in the commodity market in this closed
economy without a government sector, investment (I) and the change in
the capital stock (K′) must equal saving S .8 That is, designating the change
in K as K′(= ∂K/∂t = change in K per period), we have:

K′ = I = S,

so that,

14.1.3 Labor force growth

The labor force growth rate is assumed to be constant at n. Using the
convention that the growth rate of the labor force L is designated by L′′,
this assumption is:

where L′(= ∂L/∂t) is the change in L during the period and L′/L gives the
growth rate of L.

14.2 The Analysis of the Solow Model

Equations (1) to (6) constitute the basic assumptions of the benchmark
classical growth model. The analysis based on them is as follows.
First note that k = K/L, which implies that:

k′′ = K′′ – L′′.

Since k′′ = k′/k, K′′ = K′/K, and L′′ = n, we have:

k′/k = K′/K – n.

Multiplying each term in this equation by k gives:

where K′ = S and, so that {K′. (k/K)} on the right-hand side of Equation (7)

661
equals S/L. Hence:

Since S = σY, S/L equals σY/L. Therefore:

k′ = σY/L – nk.

Further, since Y/L = f(k) , we get:

The last equation has two components, σf(k) and nk, on the right-hand side.
Of these, σf(k) is the increase in the capital stock per worker through
saving and can be interpreted as ‘the availability of new capital per
worker’ (i.e., on average over all workers). Examining the second
component, n is the number of new workers through population growth
and k is the existing capital intensity (i.e., the capital per worker available
to the existing workers), so that nk can be interpreted as ‘the capital
requirements of the new workers (at the existing capital intensity)’.
Therefore, we have the intuitive interpretations:

σf(k) = (= S/L ) the availability of new capital per worker and

nk = the (per–capita) capital requirements of the new workers.

The value of k (designate it as k*) at which the equality of these terms

occurs is sometimes also referred to as the break-even capital-labor ratio,
since the new capital then becoming available is just enough to equip new
workers with the existing level of equipment per worker. Clearly, if σf{k)
equals nk, the capital-labor ratio will not change, so that k* is the SS value
of k.
The SS of the Solow model without technical change is defined as the
long-run equilibrium with a constant capital/labor ratio. That is, in the SS:

k′ = 0,

so that, from Equation (9):

This is the fundamental SS equation of the benchmark Solow model

(without technical change). It determines the SS by the condition:

Availability of full-employment saving per worker = the capital

requirements of new workers. (11)

662
14.3 Diagrammatic Analysis of the Solow Model
Figure 14.1 graphs the fundamental equation of the SS for the Solow
model with k on the horizontal axis. In this and similar figures later, we
have designated the units of the vertical axis as output units. This axis9
will be used to represent two different variables (σf(k) and nk) whose
measurement units are output units. The concave curve representing σf(k)
measures the per-capita availability of new capital (through saving), while
the straight line marked nk measures the capital requirements of new
workers at the existing captial-labor ratio.
In Figure 14.1, the curve marked f(k) is output per capita. Since the MPL
is positive but diminishing, this curve is concave. The curve marked σf(k)
represents saving per capita. This curve is concave because f(k) has a
concave curve because of the diminishing marginal productivity of labor
per capita. Since n is a constant, nk is represented by a straight line from
the origin. The SS, with σf(k) = nk, occurs at kj. To the left of kj, at k1,
saving (i.e., new capital) is greater than required to equip new workers
with capital at the existing capital/labor ratio. Since all workers have to
have the identical amount of capital per worker, the capital provided to all
workers will increase, thereby causing a rightward movement toward k*0.
To the right of the point k*0, say at k2, saving (i.e., new capital) is less
than required to equip new workers with capital at the existing
capital/labor ratio. Since all workers have to have the identical amount of
capital per worker, the capital provided to all workers will decrease,
prompting a leftward movement to k*0.10 Therefore, if the economy is
away from k*0, it will move back to k*0. That is, k*0 is a stable SS
captial-labor ratio.

663
14.3.1 The SS growth rate of output
In Figure 14.l, in the SS at k0*,k (which equals K/L) is a constant at k*.
Substituting the SS value of k, designated as k*0, in the production
function gives the SS level of y* as:

Since k*0 is constant, y must also be a constant, say at y*0. Hence, in the
steady state, y has a zero growth rate. That is:

Hence, the SS growth rate of output per capita is zero. Therefore, the
standard of living becomes constant in the SS.
However, since y = Y/L and y′′ = Y′′ – L′′, in the SS:

Hence,

That is, the SS growth rate of output is determined by, and equals, the
growth rate>That is, the SS growth rate of output is determined by, and
equals, the growth rate of the labor force.
We now derive the SS growth rate of capital. Since k = K/L, k′′ = K′′ – L′
′. Hence, in the SS, k*′′ = K*′′ – L′′ = 0, so that K*′′ = n. That is, the SS
growth rate of capital, just as that of output, is determined by, and equals,
the growth rate of the labor force.
Our conclusions for the SS of the Solow model without technical change
are: re:
• The SS values of y and k are constant. Therefore, the standards of living
and capital intensity (i.e., capital per worker) are constant in the SS.
• The SS growth rates of k and y are zero. That is, the standards of living
do not change in the SS.
• The SS values of Y and K grow at the labor force growth rate. The lower
the labor force growth rate, the lower the growth rate of output and
capital in the SS.
• In the pre-SS stage, the growth rates of k and y are positive. Therefore,
capital intensity and the standards of living increase as the economy goes
to the SS.
Extended Analysis Box 14.2: Variability of the Saving Rate
The preceding model assumes that the saving rate will remain constant

664
 as the standards of living improve and the country approaches SS. In
practice, poor countries with a static standard of living have a low
marginal propensity to save (MPS), countries in the early stages of
development and high growth rates often have high MPS but the MPS
declines as the standard of living becomes quite high. The hypothesis of
a constant saving rate in the preceding growth theory was, therefore, a
simplifying device for the exposition. However, note that while this
assumption may be suitable for static economies, it is not very realistic
for economies during the process of development from the pre-industrial
state to the industrial one.
Now suppose that the saving rate is endogenous and is a function of
output per capita. Since the latter depends on capital intensity, the saving
rate becomes a function of k. A plausible further assumption would be
that the saving rate increases as the standard of living begins to rise
above the low/subsistence levels (i.e., at low capital intensities) but
declines at higher ones, eventually becoming a constant.11 This
hypothesis can be expressed as σ = σ(y) or as:

σ = σ(k),

so that the SS condition becomes:

σ(k)f(k) = nk.

With this assumption for the saving propensity σ, σ(k) first has a convex
(increasing rate) segment, followed by a concave (decreasing rate) one.
Such a curve is likely to cut the ray nk at three points, so that there
would be three SS positions (not shown diagrammatically) for this
curve.

14.3.2 The impact of shifts in the saving rate

Our conclusion from Solow’s fundamental growth equation above was that
the SS growth rates Y *′ and K*′ depend only on the growth rate of labor
and not on the saving rate. Hence, a change in the saving propensity σ
cannot change the growth rates of capital and output. This result can also
be established diagrammatically, as shown in the following analysis.
An increase in the marginal propensity to save (MPS) from σ0 to σ1
shifts the σf(k) curve in Figure 14.2a upwards from σ0f(k) to σ1f(k) and
establishes a new SS at k*1, with k*1 > k*0. Since the capital per worker

665
has increased to k*0, the output per worker — and therefore the standard
of living — will be higher. However, the fundamental growth equation
implies that the growth rate of output per worker in the new SS, as in the
old one, will be zero, so that the growth rate of output per worker does not
change from one SS to a new one. Therefore, in both the old and new SS,
output will grow at the labor force growth rate n, so that the SS output
growth rate is not changed by the increase in the saving rate. However,
note that the output growth rate will rise above n during the transition from
k*0 to k*1: in Figure 14.2a, under the higher saving rate, the output growth
rate is higher at k2 (i.e., in the pre-SS) than at k*0 and k*1.
Conversely, if there is a decrease in the saving rate, the Solow model
implies a fall in investment and a reduction in both the growth rate and
living standards in both the transition (which can take decades) to the new
SS and the new SS itself — even though it does not change the SS growth
rate of output per capita.

The Solow model’s conclusions for the comparisons among countries

that have different saving rates but are otherwise identical (the ceteris
paribus assumption) are that:
• In the pre-SS stage, ceteris paribus, countries with higher saving rates
will have higher growth rates of output and of output per worker, so that
their standards of living will improve at a faster rate. However, their
growth rate will fall over time to the SS one.
• In the SS, ceteris paribus, countries with higher saving rates will have
the same growth rate (zero) of output per worker as those with lower
saving rates.
• In the SS, ceteris paribus, countries with higher saving rates will have
higher levels of output per worker and, therefore, higher standards of
living.
To illustrate, many countries, including Canada and the USA, have low

666
saving rates between 0% and 10%. These low saving rates are a cause for
concern about their implications for the pre-SS growth rates of these
countries. Conversely, the relatively high growth rates of certain
developing countries, such as Japan during the 1960s and 1970s or Korea
in the 1980s, were often attributed to their high saving rates. Note also that
their high growth rates did eventually fall to the levels of the American
and West European countries, as predicted by the Solow model.
Mathematical Box 14.1: Numerical Example A
Assume that a country has the production function:

y = 20k – 0.4k2.

Also assume that the country also has the marginal and average
propensity to consume (APC) identical at 0.9 and a labor force growth
rate of 4%. That is, σ = 0.1 and L′′ = 0.04.
Q: What are its SS captial-labor ratio and its output per worker?
The SS condition is: σf(k) = nk. Therefore:

0.1 (20k – 0.4k2) = 0.04k.

Opening the bracket on the left side of the preceding equation, we can
cancel k from both sides and solve the equation for the SS value of k.
This yields k* = 49. Substituting this value in the production function,
we have:

y* = 20k – 0.4k2 = 20(49) – 0.4(49)2 = 19.6.

For use in further numerical examples in Mathematical Boxes 14.1 and

14.2, designate these solutions as:

k* = 49 and y* = 19.6.

Q: What are the growth rates of output per worker and of output?
In the SS, since the SS captial-labor ratio k* (= 49) does not change,
the SS output per worker y* (= 19.6) also does not change. Hence, y*′′ =
0 and the SS output per worker remains at 19.6.
In the SS, y′′ = (Y/L)′′ = Y′′ – L′′ = 0 implies that Y′′ = L′′ = 0.04.
Therefore, Y′′ = 0.04. Hence, while output per worker remains constant
at 19.6, output grows at 4%.

667
14.3.3 The impact of shifts in the labor force growth rate
Many countries of the world are currently experiencing decreases in their
labor force growth rates. Therefore, the following analysis derives the
effects of a decrease, rather than an increase, in the labor force growth rate.
A decrease in the labor force growth rate from n0 to n1 will lower the nk
curve in Figure 14.2b from n0k to n1k. This increases the SS values of k
from k*0 to k*1 — that is, there is more capital to distribute among the
smaller labor force. This higher value of the SS captial-labor ratio
increases the SS output per worker, thereby improving the SS standard of
living.
However, the growth rate of output per worker is zero in the SS.
Therefore, the growth rate of output is n0 at k*0 while it is n1 at k*1, with
n1 < n0, so that the SS output growth rate falls as the labor force growth
rate falls — though the output per worker has increased.
The Solow model’s conclusions for comparisons among countries that
have different labor force growth rates but are otherwise identical (the
ceteris paribus assumption) are that:
• In the pre-SS stage, countries with lower labor force growth rates will
have lower output growth rate but higher growth rate of output per
worker, so that their standards of living will improve over time.
• In the SS, countries with lower labor force growth rates will have the
same (zero) growth rate of output per worker but lower growth rates of
output.
• In the SS, countries with lower labor force growth rates will have higher
levels of both capital and output per worker and, therefore, higher
standards of living.
Comparing the effects of increases in the APS σ and of the labor force
growth rate n on output Y and output per worker y, the Solow model’s
conclusions are that:
• SS effects on y: increases in APS σ increase SS output per capita (y*) but
do not change its growth rate (y′′*,which remains at zero), while
decreases in the labor force growth rate n increase the SS output per
capita (y) but do not change its growth rate (y′′*, which remains at zero).
• SS effects on Y : increases in the APS σ raise the SS output level y* but
do not change its growth rate Y ′*. Decreases in the labor force growth
rate n lower the growth rate Y ′′*.
• In the pre-SS stage, the higher the saving rate and the lower the labor

668
force growth rate, the faster the growth in capital and output per worker.
Mathematical Box 14.2: Numerical Example B
In Numerical Example A in Mathematical Box 14.1, the economy had
the production function y = 20k – 0.4k2 and L′′ = 0.04. Now assume that
the saving rate rises from 0.1 (in Example A) to 0.2. Hence, our
information becomes:

y = 20k – 0.4k2, σ = 0.2 and L′′ = 0.04.

Q: What are the SS values of k and y?

The SS value of k is calculated from the SS fundamental equation,
which gives:

0.2(20k – 0.4k2) = 0.04k.

We can simplify this equation by canceling k in all the terms. This

yields:

4 – 0.08k = 0.04,

so that k* = 49.5.
Substituting this value of k in the production function gives:

y* = 20k* – 0.4k*2 = 20(49.5) – 0.4(49.5)2 = 9.9.

Designating its solutions as k*1 and y*1 , we have k*1 = 49.5 and y*1 =
9.9. Comparing Examples A (in Mathematical Box 14.1) and B, the
increase in the saving rate increased the SS captial-labor ratio (but
lowered output per capita).
Note that while the captial–labor ratio increased, output per worker
fell, which is a very unrealistic result.12 This result is a consequence of
the assumed quadratic production function, which is easy to work with
but unrealistic. The more realistic result would be that the greater capital
per worker increases output per worker. For such a result, a more
appropriate production function would be one with an exponent less
than two on the second term, but this would make the calculations much
more difficult. Such a function is the Cobb–Douglas production function
used in the Appendix [optional] to this chapter.
Q: What are the SS growth rates of k, y, K, and Y after the increase in the
saving rate?

669
In the SS, since the SS captial-labor ratio k*(= 49.5) does not change,
the SS output per worker y*(= 9.9) also does not change. Hence, y*′′ =
0.
In the SS, y′′ = (Y/L)′′ = Y′′ – L′′ = 0 implies that Y′′ = L′′. Therefore, Y
′ = 0.04. Hence, while output per worker remains constant at 9.9, output
grows at 4%. Note that this SS growth rate is the same as in Numerical
Example A in Mathematical Box 14.1. Therefore, the increase in the
saving rate in B did not change the SS growth rate of output per capita
(which was 0%) and the growth rate of output (which was 4%).
Numerical Example C
Now assume that the labor force growth rate decreases to 0.02. Hence,
our information becomes:

y = 20k – 0.4k2, σ = 0.2 and L′′ = 0.02.

Its SS values are to be calculated from the SS fundamental equation,

which gives:

0.2(20k – 0.4k2) = 0.02k.

We can simplify this equation by canceling k in all the terms. This

yields:

k* = 49.75.

Substituting this value of k in the production function gives:

y* = 20k* – 0.4k*2 = 20(49.75) – 0.4(49.75)2 = 4.975.

Designating these solutions as k*2 and y*2, we have k*2 = 49.75 and
y*2 = 4.975. Comparing Examples B and C, the decrease in the labor
force growth rate increased the SS captial-labor ratio.13
Q: What are the SS growth rates of output per capita and of output?
In the SS, since the SS captial-labor ratio k*(= 49.75) does not change,
the SS output per worker y*(= 4.975) also does not change. Hence, y*′′
= 0.
In the SS, y′#x2032; = (Y/L)′′ = Y′′ – L′′ = 0 implies that Y′′ = L′′ .
Therefore, Y′′ = 0.02. Hence, comparing Examples B and C, the
decrease in the labor force growth rate lowered the output growth rate
from 4% to 2%.

670
14.4 The Growth Implications of a More General
Production Function
A common form of the production function used in microeconomic
analysis has a region of increasing MPK followed by one with diminishing
MPK. We illustrate this case with an example. Suppose that the existing
technology dictates a minimum functional type of a machine (computer)
that can be useful. Below this size, if there are more workers than
machines, there is forced and inefficient sharing of the machine among the
available workers or some workers have to be left waiting or unemployed.
Increases in capital are then likely to result in increasing marginal
productivity. Once all the workers are equipped with the machines, further
increases in the number of machines for the given number of workers
results in decreasing marginal productivity. The analysis of this range
yields the Solow diagram in Figure 14.3. In this figure, σf(k) is, as before,
the saving done in the economy at the constant saving rate a . The growth
rate of the population is still assumed to be constant at n.
Figure 14.3 shows that the use of the more general production pattern
generates two SS points k*1 and k*2. Of these, if the economy is to the left
of k*, it does not generate enough new capital through its own savings to
meet the capital requirement of new workers, so that it gradually decreases
its capital per worker and therefore moves to lower standards of living.
This movement continues until a new SS is reached. If the economy is to
the right of k*1, it generates more new capital through its own saving than
needed to meet the capital requirement of new workers, so that the
economy gradually increases its capital per worker and therefore moves to
higher standards of living. Hence, k*1 represents an unstable SS: the
economy moves either to the left or to the right of k*1 , but not toward it.
k*1 is called the take–off captial-labor ratio: to the right of it, the economy
‘takes off’ to self-sustaining growth leading to increasing higher standards
of living.

671
 k*2 represents a stable SS point: once the economy is to the right of k*1 ,
it moves towards k*2 .
Hence, the more general production function produces an unstable SS at
k*1 , but a stable one at k*2. Economies which are to the left of the take-
off ratiok*1 will move to lower SS captial-labor ratios and will become
increasingly poorer. Economies to the right of the take-off ratio k*1 will
progress to higher captial-labor ratios and will become increasingly richer.
Over time, there occurs convergence in the living standards among the
better-off countries (i.e., those to the right of k*1) while the standards of
living of poorer countries (i.e., those to the left of k*1 ) diverge from those
of the richer countries.
It is quite an oversimplification to fit the actual experience ofcountries to
a single difference in the determinants of growth. Keeping this in mind,
consider the growth experience of China in the context of a production
function of the type shown in Figure 14.3. In recent decades, China
drastically reduced its birth rate through a state-instituted one-child-per-
family policy. This would have lowered its nk curve and reduced its take-
off captial-labor ratio, as well as raised its pre-SS growth rate — thereby
facilitating its entry onto a self-sustaining high growth path. Many sub-
Saharan African countries have very low captial-labor ratios and high
labor force growth rates, raising the possibility that they are below their
take-off captial-labor ratios. Countries in this situation tend to remain poor
and possess low growth rates unless some shift or shifts reduce their take-
off capital ratios. Among these shifts would be increases in their capital
growth rates (financed by increases in their own saving rates or through
continuous inflows of foreign aid over time), or decreases in their labor
force growth rates. A third possibility is a shift in technology. Certain
forms of technical change can shift the σf(k) curve in such a way as to
lower the take-off captial-labor ratio. However, other kinds of shifts could
raise it. The role of technical change in growth theory is discussed later in

672
this chapter and in Chapter 15.

14.5 Convergence Versus Divergence in Output

per capita among Countries
We now return to the benchmark Solow model with diminishing MPK.
This model implies that there is a tendency for all countries to move
toward and converge to the same SS k and y ratios if they have:
• the same production function with a single technology possessing
diminishing marginal productivity of capital,
• the same saving propensity, and
• the same population growth rate.
Even if countries have different production functions, saving propensities,
and labor force growth rates, the Solow model implies that their capital per
worker and output per worker would become constant in their respective
SS, though at different levels.
The Solow model, assuming a given technology and saving rate, has
several implications that are contrary to the general experience of growth.
Among these are:
• All the countries converge to the same SS output per capita.14 However,
empirical studies do not show a general worldwide tendency toward the
convergence of living standards among all countries of the world. In fact,
the gap in the standard of living between the richest countries and the
poorest ones seems to increase rather than decrease.
• The SS growth rate of output per worker for all economies will converge
to the same (zero) growth rate. This is also contrary to the real-world
experience since most countries continue to experience improving
standards of living over long periods.15
• Countries with lower labor force growth will have lower output growth
rates. This is often contradicted by experience, for example, between
European and North American growth rates with those in Africa.
• For countries tending to the same SS but still in the pre-SS stage,
countries with lower values of y and k will have higher growth rates of y
than richer countries with higher values. While this is consistent with the
experience of some countries, it is not always so.
Given these discrepancies between experience and the implications of the
benchmark Solow model, we need to modify some of its assumptions. The
most important modification to it is the introduction of technical change.

673
This modification of the benchmark Solow growth model is based on
another contribution of Solow (1957) to growth economics. This
modification is presented in the next section.

14.6 Assessing the Importance of Technical

Change in the Solow Model16
This section provides an empirical assessment of the importance of
technical change. The following procedure is based on the contributions of
Robert Solow.17 For this purpose, modify the production function to:

where, taking technology to be a measurable variable, T is the ‘technology

index’. Solow (1957) offered a procedure for the estimation of technical
change under the assumption that the production function takes the
specific form:

For this production function, output results from the joint contribution
through g(K, L) of labor and capital and the technology level, whose effect
on production is captured in a multiplicative form by A(T). The
contribution of technology to output is assumed to be independent of the
levels of the labor force and capital stock. T is measured by chronological
years. Since the total contribution of the two factors of production, labor,
and capital, to output/product is captured by g(K, L), g(K, L) is called the
total factor product. Note that since labor is in man-hour units and
physical capital is in terms of commodity units, they cannot just be added
to arrive at an index of all the inputs, so that an index of inputs needs to be
constructed.18 This index can be constructed by using the function g(K, L),
which is the output that is produced at the technology level of a designated
base year, with T normalized at unity for the base year. A(T) is
correspondingly called the total factor productivity (TFP) or multifactor
productivity. The index for A(T) can be constructed from:

A(T) = Y/[g(L, K)].

Intuitively, A(T) is the ratio of real GDP to the (index of) total inputs and
shows their productivity.
A major emphasis of growth analysis is on the ‘growth rate of
technology’, which is measured by A′′(T). Intuitively, A′′(T) represents the

674
growth rate of output that results from improvements in technology, with
all inputs held constant. This assumed pattern of technical change shifts
the production function and increases output year by year even if there are
no changes in labor and capital inputs. Further, A′′(T) increases the
marginal productivities of labor and capital in the same proportion, so that
this form of technical change called neutral technical change.
Solow’s contribution in 1957 — in addition to his contribution of 1956
on the growth theory model specified earlier — provided the following
method of measuring this increase. The contribution of each unit of labor
to output equals its marginal product.19 Therefore, the contribution of
labor as a whole to output equals MPL . L (note: not APL . L), so that its
contribution per unit of output Y is given by:

αL = (MPL . L)/Y,

where MPL is the marginal productivity of labor. αL is the elasticity of

output with respect to labor, i.e., the percentage increase in output for an
increase in labor. Under perfect competition, the profit maximizing firm
employs labor up to the point where the MPL equals its real wage rate
(i.e., MPL = w ), so that:

α L = (w . L)/Y,

where w is the real wage rate and wL is the total payment out of output to
labor. This method of measuring αL makes it the income share of labor
(i.e., the share of labor in income) and provides a way of measuringαL.
Similarly, the contribution of capital as a whole to output is captured by
(MPK. K) and its contribution per unit of output Y is given by:

α K = (MPK. K)/Y,

where αK is the elasticity of output with respect to capital.

The wage rate can be thought of as the rental or user cost per unit of
labor per unit of time. Similarly, designating the rental or user cost of
capital20 under perfect competition by ρK, profit maximizing firms will
use capital up to the point at which the MPK equals the rental cost ρK of
capital, so that:

αK = (ρK K)/Y.

675
This provides the interpretation of aK as the income share of capital and
provides a way of measuring ρK.
Adding the contributions of labor and capital to output gives the total
factor product (i.e., the weighted sum of the contributions of the factors of
production to output) g(K, L) as:

g(K, L) = αL . L + αK . K.

Measuring αL and αK at an initial level (i.e., in a given base year), the

growth rate of total factor product g(K, L) is specified by:

g′′ = αL L′′ + αK . K′′.

Therefore, the output growth rate can be decomposed into three

components: the contribution of labor, the contribution of capital and a
remainder, which is called the contribution of technical change to output.
Since Y = A(T) . g(L, K), we have:

Y′′ = A′′(T) + g′′(L, K),

where g′′(L, K) = αKK′′ + αLL′′. Therefore:

To reiterate:
A′′(T) = growth rate of total factor productivity (TFP),
αK = share of capital in aggregate income/output, and
αL = share of labor in aggregate income/output.21
For many industrialized countries, the estimated value of αL (labor’s share
of national income) is about 0.7, while αK (capital’s share of national
income) is about 0.3.
In the preceding equation, (αKK′′ + αLL′′) is the share-weighted growth
rate of output due to the growth of labor and capital. A′(T) measures the
growth rate of output that cannot be accounted for by factor growth and is
the growth rate of TFP. It is usually attributed to the change in technology.
The preceding equation is known as the growth accounting equation
since it ‘accounts for’ (separates) the output growth rate by the
contributions of technology, capital and labor. Since A′′(T) is unobservable
while all the other parameters and variables in the preceding equation are
observable, rewrite the above equation as:

676
 A′′(T) = Y′′ – αKK′′ – αLL′′ . (18)

All the terms on the right side of this equation are observable, so that the
data on them can be used to calculate the technology growth rate.22

14.6.1 Solow’s estimates of the contribution of technical

change to the improvement in living standards
Solow used the growth accounting formula to calculate A′′(T) and found
that historically the main increase in per-capita output in the USA had
been due to technical change rather than to increases in capital per worker,
though there was considerable variation from year to year in the former.
His conclusion for the period 1909–1949 for the USA was that while
output per capita almost doubled, about 87.5% (i.e., 7/8) of this increase
was due to technical change and only about 12.5% (i.e., 1/8) was due to
more capital per worker. Another recent study for the USA indicates that
between 1995 and 2000, output per hour grew annually by 2.79%.23Of this
amount, the growth of multifactor productivity (i.e., A′′(T′′)) accounted for
about 1.31% or 47% of the annual growth of output per hour of labor
time.24 Somewhat similar results indicating the importance of technical
change in the growth of output per capita have been confirmed by many
studies on other countries. Given this importance of technical change, we
need to incorporate it explicitly into the formal growth models and study
its determinants. The former is done later in this chapter. The latter is done
in the next chapter.
Mathematical Box 14.3: Numerical Examples on Technical Change
To illustrate the calculation of the technology growth rate, suppose that
the annual growth rate of labor is 2% and of capital is 3%, while output
grows annually at 4%. With αL = 0.7 and αK = 0.3, we would have:
growth of output due to capital growth = αKK′′ = 0.3(0.03) = 0.009 =
0.9%,
growth of output due to labor force growth = αLL′′ = 0.7(0.02) = 0.014
= 1.4%, and
growth of output due to technical change = A′′α(T) = 0.04 – 0.3(0.03)
– 0.7(0.02) = 0.017 = 1.7%.
Hence, with total output growth at 4%, output grows by 1.7% each year
as a result of the improvement in technology, while it grows by 0.9%
because of the growth of capital and by 1.4% because of the increase in

677
 labor.
If we convert these results to the growth rate of output per capita (i.e.,
y′′ = Y′′ – L′′), which represents the growth of the standard of living, we
have:
growth of output per capita = y′′ = Y′′ – L′′ = 0.04 – 0.02 = 0.02 = 2%,
growth of output due to growth of K and L = αLL′′ + αKK′′ = 0.7(0.02)
+ 0.3(0.03) = 0.023,
growth of output per capita due to capital growth per capita = {(aLL′′ +
aKK′′) – L′′}25
= {0.7(0.02) + 0.3(0.03)} – 0.02 = {0.014 + 0.009} – 0.02 = 0.023 –
0.02 = 0.003 = 0.3%, and growth of output per capita due to technical
change = 0.02 – 0.003 = 0.017 = 1.7%.
Therefore, the total growth in living standards because of the growth of
both labor and capital is only 0.3% per year. This figure of only 0.3% is
very much smaller than the contribution to the growth of living
standards due to technological change, which is 1.7%. While we can use
other plausible numbers as calibration exercises for the growth
accounting equation, they tend to show a similar point: the growth of
total factor productivity (TFP) is the main cause of the improvements in
living standards.
Now take the example of rapidly industrializing/growing economies,
sometimes referred to as ‘economic tigers’ — a designation for a
country with exceptionally high growth rates for many years. They tend
to have a high output growth rate, a relatively high labor force growth
rate and a relatively high capital growth rate. Suppose that their output
growth rate is 10%, labor force growth is 3%, and capital growth is 8%.
With αL = 0.7 and αK = 1 – αL = 0.3, these figures imply that:

A′′(T) = 0.10 – 0.7(0.03) – 0.3(0.08) = 0.055 = 5.5%.

Hence, the relative importance of technical change to the growth of

output in this example increased dramatically over that in the first
example: 5.5% per year as against only 1.7% per year. Therefore, the
high growth rate of the economic tigers is likely to be largely due to
their very rapid technical change.

14.6.2 The residual as the unexplained component of

growth
The derived value of A′′(T) from the growth accounting equation is the

678
residual (of the output growth rate) after deducting the contributions of
labor and capital. It has been nicknamed the “Residual”. Since Solow’s
growth theory does not explain the determinants of this residual, it
represents a lack of knowledge, so that another nickname given to it has
been ‘The Measure of our Ignorance’. A continuing objective of growth
studies has been to explain ever-larger percentages of the growth rate, so
that the residual — or the unexplained measure of output growth — is
gradually whittled down. One method of doing so is explained in
Mathematical Box 14.4.
Mathematical Box 14.4: Detailed Estimation of the Contributions to
Economic Growth
Estimating the returns to different educational levels
The growth accounting equation measures the return per period to
labor by the wage rate. But labor is of many different kinds, usually
differing in the amounts of human capital, and being paid different wage
rates. We need a procedure for calculating the returns to different
amounts of human capital. The following provides an illustration of the
procedure used in growth accounting.
Assume that we have some workers (numbering XHS) who have only
a high school education while there are others (numbering XBA) who
have a BA degree. The economy provides data on the wages of these
two types of workers, designated respectively as wHS and wBA .
Assuming that the only difference between these two types is in their
educational levels,26 the difference (w BA – w HS) would measure the
market return to the college education of the workers with the BA, while
w HS will be the return/wage of the workers with the high school
education.27 The income shares of all workers up to the completion of
high school would be [wHS(XHS + XBA)] while that of college
education would be [(w BA – w HS)XBA ].
This procedure can be usually used to calculate the returns to a wide
variety of inputs.
A general procedure for assessing the contribution of each input to
output
For the general case of J inputs in the production function, we can
calculate the contribution to output of the jth input as:

Income of jth input = (MPj. Xj) j = 1, 2, …,J

679
where MPj is the marginal product of the jth input and Xj is the quantity
of the jth input. Under the assumption that each unit of each input is paid
the marginal product of the input, we can write the income share of the
jth input (i.e., the share in national income Y ) as:

αj = (ρ.Xj)/Y j = 1, 2,…,J

where:
αj = income share of j th input,
MPj = marginal product of the j th input,
ρj = payment (rental/user cost) per period to the j th input, and
Xj = quantity of j th input used in production.
The general growth accounting equation
We can now generalize the growth accounting equation28 to:

where Xj′′ is the growth rate of the jth input. This equation provides a
growth-accounting procedure according to which the contributions to
the growth of output by technical change and the various inputs can be
measured. Its empirical application requires the calculation of the
income shares α j, which equal ρjXj/Y, so that we only need data on ρj
(the rental cost per period), Xj (the input quantity), and total income Y.
The above procedure can be used to calculate the relative contribution
to output for a wide variety of inputs. Edward Denison in an influential
study in the early 1960s used this growth accounting procedure to derive
the contribution of numerous factors to output growth in the USA. This
procedure was adopted in many studies for many countries and
continues to be used for evaluating the importance of the various inputs
to output growth. Among the factors included in such growth accounting
measures are the physical capital stock, number of hours worked,
various educational levels, sex composition, and age, as well as factors
that are more difficult to quantify. Among these are the workers’ health,
firms’ competitiveness, the economy’s openness to foreign trade, etc.
The residual of the growth rate from this procedure encompasses the
contribution of all factors that are not explicitly incorporated in the
growth accounting calculations. In general, the larger the number of
factors taken into account, the smaller becomes the residual that needs to
be attributed to the diffuse term ‘technical change’.

Box 14.1: Deficiencies of the Growth Accounting Equation

680
The growth accounting procedure has several deficiencies, so that it can
yield erroneous measurements of the contributions of the various inputs
and the growth rate of technical change. These deficiencies/errors are
related to:
(i)The measurement of the contributions of physical and human capitals.
These are not only the inputs into production but also the carriers of
technical change: without them, many forms of technical change will
not occur. The above growth-accounting procedure ignores this
carrier role of labor and capital, so that it underestimates their
contribution to growth. Conversely, it overestimates the residual
attributed to pure technical change.
(ii)Variations in work effort — without corresponding changes in wages
— and in capital utilization over the business cycle: As Chapter 8
argued, the work effort of employees changes over the business cycle.
In recessions, firms tend to hoard labor, so that its productivity falls.
Further, the rate of capital utilization is decreased, so that even in the
absence of technical change, output falls more than employment.
Hence, total factor productivity falls. This fall is not a reflection of a
negative growth rate of technology but only in the rates of utilization
of labor and capital. However, if A′′(T) was actually equal to zero but
labor productivity fell relative to wages in the recession, the growth
accounting calculation would erroneously yield a negative value for
it. As demand and output rise in the upturn, work effort and capital
utilization increase, so that productivity rises more than wages.
Hence, if A′′(T) was actually equal to zero, the growth accounting
calculation would erroneously show a positive value for it in the
upturns. Therefore, the calculated value of A′′(T) would be too low
during recessions and too high during upturns.
(iii)The growth of intangible capital such as organizational and
production skills and other skills acquired on the job. This type of
capital is difficult to measure accurately. Further, the expenditures on
such skills are usually treated as a current expense of firms rather than
as an investment. Failure to include an accurate value of intangible
human capital understates the amount of human capital in the
economy, and so reduces the human capital’s contribution in the
growth accounting calculations.
(iv)Discrimination in the payments to factors of production. If the labour
markets do not have perfect competition but are segmented in some
way — e.g., by gender, race, color, etc. — the wages paid to workers
need not equal their marginal productivity. In the presence of

681
 discrimination, wages would be greater than marginal productivity for
the favored workers and less than marginal productivity for the
discriminated-against workers. Further, some of the benefits of
discrimination are likely to increase, as they normally do, the profits
paid to capital beyond its marginal productivity. Therefore, measuring
the contribution of the different types of inputs to the nation’s output
by their payments will overestimate the contribution of the favored
workers, as well as of capital, and underestimate the contribution of
the discriminated-against workers. It will also give erroneous
estimates of technical change.
Other reasons for the deviation of the payments to the factors of
production from their marginal productivity include monopoly,
monopsony power, etc.

Table 14.1 Sources of output and productivity growth in the USA, %

annual growth rates,30 1959–1998.

14.7 Some Empirical Findings on the

Contributions to Growth
In Canada, total productivity growth was about 2% annually29 over 1962–
1973, after which it declined to about 0.45% over 1974–1992, but picked
up sufficiently in the 1990s to become about 0.8% over 1974–1999. That
is, there was a marked slowdown in technological growth after 1973.
Similar patterns were observed for most western nations and Japan. Table
14.1 provides the findings of one study on US growth rates.
There are several points worth noting about the information in Table
14.1. The growth rates both of output and of output per hour worked
decreased from 1959 to 1995 and then increased very significantly. This
was also the case for the growth rates of capital per worker and of total
factor product (TFP). This upsurge is a reflection of the benefits from the
very extensive revolution in information technology (IT), based on the
innovations of computers and the Internet, which started much earlier than
1995 but needed to reach sufficient maturity for their economic benefits to
become apparent at the macroeconomic level.

682
 The growth of TFP was a very significant reason for the growth of
output per worker in every period. To illustrate, during 1995–1998, TFP
growth (0.99) accounted for about 42% of the growth in output per worker
(2.37). The growth of capital per worker (also called capital deepening) at
1.13 accounted for an even larger percentage (about 48%). Hence, the
major source of increases in the standard of living came from more hours
worked per worker, more capital per worker and higher TFP.

14.8 The Solow Growth Model with Exogenous

Changes in the Quality of Labor
In view of the above results on the importance of technical change, we
need to introduce technical change into the Solow growth model. While
we can introduce neutral technical change of the type specified above into
the growth model, the analysis is easier if technical change was assumed to
be such that it increases the efficiency of each worker (i.e., the quality of
labor) over time at a constant rate. To incorporate worker efficiency into
the production function, write it as:

Y = F(K, E), (19)31

where E is the labor force measured in efficiency units.32 Set kE = K/E,

and assuming constant returns to scale to proportionate increases in K and
E, we can write:

Applying the Solow procedures presented above for this form of the
production function modifies the equation for the change in the amount of
capital per efficiency worker, i.e., in kE, to:

In the steady state, kE = 0, so that the fundamental SS equation with

technical change is:

We now assume that the change in the efficiency of labor is caused by

‘technical change’33 T , so that we can specify E as:

where B is a function of technology T . Therefore:

683
where B′′ is the rate of growth of the average level of efficiency per
worker, while L′′ is the labor force growth in terms of the number of
workers.34 Assuming that B′′ is a constant equal to n (which is a Greek
letter pronounced ‘eta’), we get:

Substituting this into the above fundamental SS growth equation with

technical change, we have:

where n is the growth in labor efficiency. Equation (26) is Solow’s

fundamental SS growth equation when technical change increases the
efficiency of labor.35 This labor efficiency can be increased by increases
in the average level of skills (e.g., by better or more education on average,
more experience, etc.). The present model assumes that this increase is
exogenous.
The SS of the current model occurs when kE′ = 0 — i.e., when there is
no change in kE, which is the capital per efficiency worker. Since this
amount is constant in the SS, the SS output per efficiency worker is also
constant, so that there is no change in yE and its growth rate is zero. That
is, in the SS:

yE′′=0.

Note that yE = Y/E, so that yE′′ = Y′′ – E′′ = Y′′ – (n + n). This equals zero
in SS. That is, in the SS:

Hence, in the SS:

Hence, SS output grows at the rate (n + n ). The growth rate of output per
(base–level productivity) worker/capita, which specifies the standard of
living, is given by:

Similar reasoning shows that k′′ = n. Hence:

• Output per worker and capital per worker will continue to increase even
in the SS at the growth rate of efficiency per worker.

684
• Countries that innovate and increase labor efficiency at a faster rate
would have a higher rate of growth of output and output per worker.
• Birth rates per couple in Western countries have fallen in recent decades
to less than 2, which is required for a constant population (i.e., L′′ = 0). In
some countries, it has become as low as 1.5. Barring an adequate
immigration level, this would eventually lead to continuous declines in
the labor force, so that L′′ would become negative. However, what is
required for the determination of the SS is the sum of L′′ and the growth
rate n of labor efficiency. Hence, for the SS, it is not necessary for L′′ to
be positive, as long as (L′′ + n) is positive.

14.8.1 Diagrammatic analysis

To incorporate changes in labor efficiency through labor-augmenting
technical change into the diagrammatic analysis, Figure 14.1 is modified to
Figure 14.4. The horizontal axis now measures kE. The output growth rate
equals (n + n). Figure 14.4 assumes that n and n are both constant so that
their sum is also a constant. Assume that this sum initially equals (n0 +
n0), so that (n0 + n0)kE specifies the capital requirements of new
efficiency-workers (at the existing captial-labor ratio) and is represented
by a straight line from the origin. The SS K/E ratio is k0E*. At this K/E
ratio, Y/E becomes static but Y/L continues to increase at the rate n0.
An increase in the technical change rate n from n0 to n1 raises the growth
rate of labor in efficiency units to (n0 + n1), which lowers the SS K/E ratio
to k1E*.36 Though this ratio falls, as does the Y/E ratio, there occurs an
increase in the SS Y/L ratio, so that the workers’ SS standards of living
improve — and continue to improve over time.
Our conclusions on the effects of the increase in n (the efficiency level of
workers) are that:
• It reduces the SS K/ E ratio, so that less capital is used per unit of
efficiency-labor. But the SS K/L ratio could rise (so that more capital is
used per worker) or fall (so that less capital is used per worker). This
would depend on the increase in the efficiency of labor and the fall in the
K/E ratio.

685
• Since the SS K / E ratio falls, so does the Y / E ratio. That is, output per
unit of efficiency worker falls. However, our concern is with the growth
rate of the Y/L ratio, which is the index for the standard of living. The SS
standard of living increases over time at the efficiency growth rate n .

14.8.2 The role of human capital in changing the quality of

labor
Labor-augmenting technical change in the above analysis can be viewed as
a result of the increase in human capital, which represents the acquisition
and use of skills relevant to production. The stock of human capital in the
economy can increase in a variety of ways. Among these are:
• The allocation of labor and capital to impart the existing stock of
knowledge in educational institutions. This is labeled as ‘schooling’.
• The allocation of labor and capital to impart the existing stock of
knowledge in firms through on-the-job training.
• Changes in the stock of knowledge.
The following distinctive aspects of this classification are worth noting.
• The mere duplication of human capital through schooling and on-the-job-
training — as in the first two points of the above list — of new entrants to
the labor force leaves the level of knowledge per capita (and E/L)
unchanged.37 Therefore, it does not change the average efficiency level
of the workers so that the value of n does not change.
• An increase in human capital through more years of schooling on
average for the population or a better quality of education increases the
average efficiency of workers, which increases the value of n . This shifts
the (n + n )kE curve upwards.
• New techniques of production — from innovations and inventions —
constitute new knowledge, which raises the productivity of capital per

686
 efficiency worker and shifts the production function upwards. Therefore,
they shift up the of(kE) curve.
The first two aspects of human capital have already been discussed in this
chapter. The nature of the third element and how it comes about is
presented in the next chapter.
Extended Analysis Box 14.3: The Variability of the Labor Force Growth
Rate
The historical experience of growth suggests that the labor growth rate is
not a constant. It increases as standards of living rise from low levels,
since this allows better food, greater access to medicine, better hygiene,
etc. These reduce mortality rates, especially among infants, so that the
population growth rate rises. At higher living standards, such effects on
mortality become relatively insignificant, while control by families —
given the availability and affordability of birth-control devices — on the
number of children decreases the birth rate per woman. Consequently,
the population growth rate increases at low values of y, but decreases at
higher values. This has been the experience of many countries, including
Canada, the USA, and Europe over several centuries. In these countries,
during the early stages of industrialization in the 19th century,
population growth rates rose dramatically as a result of the rise in their
standards of living. They fell in the 20th century, especially after 1960
following the invention of the birth control pill, the increasing
participation of women in the labor force and the cumulative rise in the
standard of living. Other factors adduced for this decline are the feminist
movement and the increasing costs of raising children, which includes
the costs of providing them with an appropriate — as well as longer
years of — education, the loss of income while the mother is out of the
labor force, etc.
Some economists have interpreted the above historical evidence to
suggest that the population growth rate is a function of the standard of
living, i.e., of output per capita. Since the latter depends on capital per
worker, this argument implies that:

L′′ = L′′ (k).

The suggested form of this function implies that, over very long periods,
L′′ would be first positive and rising, then falling and becoming
negative, so that population will first increase (at low captial-labor
ratios), but eventually decline (at higher captial-labor ratios). It may or

687
 may not eventually stabilize. If it does so, its growth rate will eventually
become zero.
In Canada, the fertility rate (number of children per woman) was about
3.7 in 1926, fell to about 2.7 in the 1930s — the period of the Great
Depression, with exceptionally high unemployment — and then climbed
to about 4.7 in the late 1950s. It started declining rapidly in the 1960s
and continued declining until 1988 when it reached 1.4 children per
woman. It recovered marginally in the 1990s to about 1.5 in 1997. The
fertility rate for the USA is still above two, though marginally so. It is
well below two for many west European countries, Japan, and Canada.
Family sizes in many of the richer countries of the world are now quite
small, with an average of less than two children per couple. With births
less than deaths on an annual basis, their population would decline
unless there is adequate immigration. The Solow model without
technical change poses problems for the analysis of a negative growth
rate of the labor force.38 Its exposition is more appropriate in the Solow
model with technical change.

14.8.3 Technical change with falling or negative labor force

growth rates
In the richer countries, the growth rate of the labor force has not been
constant but has declined in recent decades. Many economists believe that
as standards ofliving (output per worker) increase, L′′ becomes smaller.
Since output per worker is positively related to the increase in capital per
worker, the hypothesis being advanced is that the labor force growth rate L
′′ is negatively related to k (i.e., capital per worker, not its growth rate). L′′
may even become negative in some cases. Assume that it stabilizes at
some value no, which could be negative.39 Further, assume that the labor-
efficiency growth rate n0 is exogenous and constant. Under these
assumptions, (L′′ + n0 ) declines as k increases, eventually becoming
constant at (n0 + n0 ). Making the assumption that growth in labor
efficiency more than offsets the decline in population, (n0 + n0) > 0. Under
these assumptions, the Solow model implies that the economy’s SS growth
rates of capital and output would be (n0 + n0 ). The SS standard of living
will rise at the rate n0, though population will be falling at a constant rate.

14.8.4 Shifts in participation rates

688
The participation rate is defined as the rate of participation of the
population in the labor force. That is:

The labor force consists of workers who want a job for pay. The labor
force participation rate of women — i.e., the ratio of women in the labor
force to all women in the country — has increased very substantially since
the 1960s. Some of the effects of this increase in the female participation
rate have been offset by a decrease in the participation rate of males. Fact
Sheet 14.3, also presented as Fact Sheet 10.1 in Chapter 10, presents the
shifts in the female, male, and total participation rates for the USA since
1960.
Fact Sheet 14.3 : Participation Rates in Canada and the USA, Male and
Female, 1980–2008
The following table classifies participation rates in Canada and the USA
as the fraction of the total working age population (over 15 years old)
currently in the workforce.

Source: For 1960–1970 data, Sorrentino, C. International comparisons

of labour force participation, 1960–81. Monthly Labor Review,
February 1983, 23–36. For 1980–2008 Canada data, CANSIM
V2062816. For 1980–2008 US data, US Bureau of Labour Statistics.

Box 14.2: The Implications of Increases in the Labor Force Participation

of Women
What would be the effects of the increasing labour force participation of
women in the context of the Solow model?
To answer this question, we need to distinguish between the labor
force and population, and between the output per worker (y) and the
standard of living. The relevant proxy for the latter is output per person
(i.e., per unit of the population, not per unit of the labor force). Assume
that there is no technical change and that, in the initial state, the
participation rate for the population was constant (so that L = β .
population, where f is the population’s labor-force participation rate) and

689
that the growth rate of the population — and, therefore, of the labor
force — was initially n0. Let the increase in the participation rate occur
in one step, so that the labor force growth rate jumps to (n0 + α). After
this step increase, the participation rate stabilizes at its new long-run
level and the labor force growth returns to n0. Further, assume that the
skill levels of the new entrants are identical with those of the initial
workers.
The Solow growth diagram implies that the increase in the
participation rate will reduce the K/L ratio. If the economy was initially
in the SS at k*0, it will move below this SS level, say to k1,k1 < k*0 (not
shown diagrammatically, though the earlier Figure 14.1 can be used as a
guide for the arguments here). Note that at k1, as compared with that at
k*, output per worker (y) and the real wage rate (w) are lower, but the
growth rate of y is higher. Since k* is a stable SS, the economy will
eventually return to k*, at which time the economy will revert to its SS
levels of y and w.
Since the participation rate has risen, the number of jobs (and pay-
checks) per representative family would rise correspondingly. During
the adjustment phase, family incomes would reflect two opposing
changes: decrease in wages due to the lower k ratio and increase in the
number of jobs per family. The net effect is likely to be an increase in
family income. Since the rise in the number of jobs per family has its
converse in the time left for non-job activities, i.e., for ‘leisure and rest’,
the increase in family incomes comes at the cost of the decrease in its
leisure and rest time.
Now add the assumption that the new entrants to the labor force have
lower skills — due to lower educational levels and years of experience
— than do the initial labor market participants.40 This implies that their
marginal productivity and wages will be lower than of the initial
participants. Hence, the increase in the family income will be
correspondingly less. As the new entrants accumulate experience and
skills, their marginal product and wages would rise.
Hence, there are several stages in the likely effects.
1. The first stage incorporates the initial impact of an increased
participation rate for the population and a lowering ofthe average skill
level. These reduce the average real wage for two reasons: capital per
worker falls and the average skill level falls. [However, note that it is
likely that the income of the representative family (which now has a
higher participation rate) rises. The increase in hours spent by the

690
 family in market jobs also reduces the average hours spent in home
activities and time spent in the home on child rearing.]
2. The second stage occurs after this initial impact. In this stage, the
economy starts recovering and moves toward its SS position. During
this movement, both capital per worker and skills accumulate, which
produces increases in real wages from the lower level of the first
stage. A positive growth rate of output per worker occurs at the lower
(pre-SS) level of the captial-labor ratio.
3. In the third stage, the economy has returned to the initial SS values of
k, y, and wages, all of which have again become constant. Family
incomes will have risen to their new SS level in proportion to the
increase in the participation rate.
4. Note that the above findings were derived under the ceteris paribus
assumptions that there was no technical change and no population
growth. On the latter, we have already discussed the increase in the
labor force participation rate of women in the last three decades of the
20th century. Associated with this trend was a decrease in the number
of children per woman, which resulted in a decrease in the labor force
growth rate some 20 years later.

14.9 SS Growth versus Level Effects of

Exogenous Shifts
Comparing the impact of shifts in the various exogenous variables and
parameters, the Solow growth model makes a distinction between those
shifts that alter the SS growth rate (of output per capita) and those that
only change its SS level. Exogenous shifts in the production function, the
stock of schooling, the saving propensity, and population growth induce
changes in the SS level, but not the SS growth rate of output per capita,
while technical change also alters its SS growth rate.
Therefore, economies that want to grow faster in the SS must find ways
of increasing the rate of technical change. However, this rate is exogenous
to the Solow growth model and is not affected by the accumulation of
capital, labor and other aspects of the model itself.41 In recent decades,
economists have become convinced that the rate of technical change is not
invariant with respect to these variables. Allowing for such effects makes
the rate of technical change endogenous to the model. The next chapter
will examine some of the sources of endogenous growth and the
possibility of increasing growth rates by policy measures.

691
14.10 The Implications of the Solow Analyses for
Policy
The above exposition of the Solow model has the following implications
for policy.
• In terms of the standard of living, countries are better off with a higher
saving rate and a lower population growth rate. Therefore, policymakers
desirous of a higher standard of living for the population should aim at a
higher saving rate and a lower growth rate of the labor force.
• In the pre-SS stage, an increase in the saving rate will increase the
growth rate of output per capita. This offers another justification for
governmental policies that can raise the saving rate. Similarly, a decrease
in the population growth rate will increase the growth rate of output per
capita. However, a declining population is also an ageing population,
with an increase in health costs and a decline in the working-age
population. The adverse consequences of these are considered to be quite
serious for economists not to generally advocate the use of governmental
policies to reduce the country’s population below its existing level.42
• However, variations in the saving rate and the labor force growth rate do
not alter the SS growth rates of output per worker or capital per worker,
which are both zero.43 Therefore, if we were to focus only on the SS
output per capita, governmental policies to change the saving rate or the
population growth rate would not have a significant impact.
• In the SS, the growth in living standards occurs only due to technical
change. Therefore, governments should try to promote R&D (research
and development) expenditures in firms and universities, etc., and also
promote entrepreneurship and investments to incorporate new
technologies in the physical capital and the labor force.
Fact Sheet 14.4 : Real GDP Per Capita, 2003 Population Growth Rates
This Fact Sheet provides a scatter graph of the relationship between real
GDP per capita and population growth. The graph for the year 2003
shows a strong negative relationship between GDP per capita and
population growth: countries with high GDP per capita tend to have
lower population growth rates. This relationship is negative because, in
the modern period, high-income families tend to have fewer children per
couple than low-income families. However, Box 14.3 points out that
earlier periods in history (especially the Malthusian stage and even the
post-Malthusian (early industrial) one had a positive relationship
between the standard ofliving and population growth because higher

692
living standards increased the survival rates without a corresponding
decrease in birth rates. Many developing countries still follow this
pattern, especially in their poorer rural areas, which have a higher
number of children per family than the middle class in the cities.

Box 14.3: Historical Growth Patterns, Labor force Growth, and

Technology Shifts
To analyze the historical growth pattern of societies, we need to
incorporate both technical change and changes in population/labor force
growth rates into the Solow model. Doing so allows us to use growth
theory to divide the historical economic experience of societies into the
following stages/ epochs:44
I.The Malthusian stage/’trap’
In this stage, technology does not change significantly and the long-run
standard of living is virtually constant at or close to a (physical)
subsistence level. Further, if increases in the standard of living do occur,
population increases sufficiently in response to the higher living
standards such as to cancel any improvement in them beyond the
subsistence level.
II.The early industrial (post-Malthusian) stage
In this stage, technical change occurs and is greater than the population
growth rate, so that output per capita rises. While population growth rate
increases significantly through declines in the mortality rates, which
occur in response to the improvements in the standard of living, output
per worker, and living standards rise on a long-run basis.
III.The modern growth stage
In this stage, industrialization deepens and technology continues to
improve, so that the standards of living continue to rise. A critical part of
the technical change in this stage is the invention and widespread use of
birth control technology. This reduces the number of births per family,

693
even though mortality rates fall and longevity increases because of
higher incomes and improvements in medical technology, sufficiently
for the population growth rate to fall. It may eventually even become
negative, in which case the average number of children per couple will
become less than two, so that the country’s population (ignoring
immigration) will fall. The modern growth stage existed in most of the
richer countries at the end of the 20th century. It has a falling, rather
than constant, population growth rate and rising standards of living.
In the above classification of growth stages, the critical difference
between the Malthusian stage and the post-Malthusian one is the
acceleration in technical change while that between the post-Malthusian
stage and the modern growth one is the demographic one: in these
stages, population is endogenous and depends on the standard of living.
Starting close to the physical subsistence level at the end of the
Malthusian stage, population first expands in response to improvements
in the standard of living. At high living standards and availability of
birth control devices, it declines in response to further improvements in
the standard of living.
IV.Globalization and the modern growth stage
Globalization (see Chapter 12) not only increases trade and capital flows
but also accelerates the spread of technology, including the birth control
one. The less-developed economies benefit from all of these, so that
their time frame for moving through the various stages above is
compressed from centuries to decades. Therefore, in recent decades, the
developing economies tend to go from the historical Malthusian stage to
the modern growth stage in a matter of decades rather than centuries.
V. The post–modern growth stage
This stage is still somewhat speculative. It is based on the premise that
while the average number of children per couple in the country will have
fallen during the modern growth stage to less than two, (which is
necessary to maintain population at a constant level, i.e., zero growth
rate), this number will eventually stabilize by the end of the modern
growth stage, so that the growth rate of population (ignoring
immigration) will become constant, though with a negative value. This
premise separates the post–modern growth stage from the modern
growth stage (which had a falling, rather than constant, population
growth rate). We will also assume that the rate of technical change will
continue to be significant and have a positive value. For simplification
so as to be able to use the Solow model, we will assume that the sum of
the rate of technical change and of population growth rate will be
positive. Hence, we deduce from the Solow model that, in the post–

694
 modern growth stage, SS output will growth at a rate equal to the sum of
the rate of technical change and of population growth rate, while the
standards of living will rise at the rate of technical change. Note that in
this stage, the population will keep declining.

14.10.1 Malthus’s theory of economic growth and the

Malthusian stage
This theory was presented by Thomas R. Malthus in 1798. Its main
assumptions were:
• The factors of production are labor and land. Of these, land is fixed in
supply, so that the marginal productivity of labor, the only variable input
in this theory, diminishes. Further, any expansion of the population
reduces the land/population ratio.
• There is no effective birth control technology. Population is positively
related to living standards and grows faster than productivity per worker
in the economy. This occurs through rising live birth rates45 and/or
falling mortality rates due to improvements in living standards.
• Technology does not change. Innovations and inventions, when they do
occur, are sporadic and with limited impact on the macroeconomy.
Malthus’s theory was quite appropriate for the pre-industrial societies
dominated by agriculture. In these societies, the capital/labor ratio was
quite low and did not change much over long periods, so that the explicit
treatment of capital was not necessary. The emphasis on land was justified
since agriculture was the dominant economic activity. Standards of living
were close to the physical subsistence levels (in terms of calories per day)
for most of the population. Population increased rapidly in response to any
improvements in the standard of living, through some increase in live birth
rates, but more significantly through a fall in the mortality rates, especially
among the very young.46 In the absence of improvements in technology,
population growth and the diminishing marginal productivity of labor
caused the living standards to fall, which induced malnutrition, starvation,
famines, and disease, as well as wars, so that births fell and deaths rose.
The SS of this model was at the physical subsistence level.
Malthus concluded for the pre-industrial societies that the long-run
standard of living fluctuated around the subsistence level, with population
growth rates adjusting to drive the economies to this SS.

14. 10.2The post-Malthusian stage

695
This stage differs from the Malthusian one because of the addition of the
following assumptions:
• Technology improves very significantly over time and makes output per
capita grow faster than the population, so that the standards of living rise.
• Population growth remains positively related to the improvement in the
standard of living but is less than the technology-induced growth rate of
output.
• A massive shift to industrial production occurs, which reduces the
relative significance of agriculture and land while enhancing the
importance of physical capital for the future growth of the economy.
The increase in the population growth rate occurs through a falling
mortality rate as nutrition and private hygiene (for instance, by regular
bathing and the use of soap) improve and public health measures (such as
putting in flushing toilets, sewers, vaccinations, etc.) are undertaken. The
number of births (especially live- births) per woman may also increase.
The population growth rate offsets some of the increase in output per
worker that would have occurred due to technical change but not fully, so
that living standards rise. This is the SS of the Solow model modified to
incorporate technical change and population growth positively related to
standards of living.

14. 10.3The modern growth stage

This stage opens with living standards that are already substantially above
the subsistence level. The analysis of this stage can also be based on the
Solow model, with or without endogenous technical change, but with the
addition of the assumption that the population growth rate is now
negatively related to the (further) improvements in the standard of living.
Mortality has already declined considerably prior to this stage and does not
significantly fall further. But birth rates fall due to a decrease in the desired
number of children per woman,47 which can be implemented by an easily
available birth control technology. The desired number of children per
family falls partly because of the advancing technology, which raises the
return to human capital and induces increased investment in the education
of (fewer) children — which in turn stimulates the speed of technical
change. Technological advances also reduce the labor time required for the
home production of goods — such cooking, laundry, etc. — and the time
required for ensuring the good health of children and other family
members, over those in earlier times. These promote the increasing
participation of women in the labor force, whose incomes and rising skill

696
levels substantially increase the opportunity cost of having and raising
children.48 Other causes for the fall in the birth rate include changes in the
social and cultural patterns affecting the socially desirable family size,
which impacts on personal decisions on the desired number of children per
family.
In the modern growth stage, the improvement in the standard of living
occurs due to technical change and falling population growth rates — and
the increase in the labor force participation rate.
For Europe and North America, this broad and rough classification
places the centuries prior to the Industrial Revolution49 in the Malthusian
stage, the period from the Industrial Revolution to the 1960s in the post-
Malthusian stage and the period after the 1960s in the modern growth
stage.
Galor and Weil (2000)50 summarize the historical evidence on standards
of living as:
For thousands of years, the standard of living was roughly constant and
did not differ greatly across countries … the growth rate of GDP per
capita between 500 and 1500 was zero … the real wage in England
roughly the same in 1800 as it had been in 1300 … real wages in China
were lower at the end of the eighteenth century than they had been at the
beginning of the first century … even in the richest countries, the
phenomenon of sustained growth in living standards is only a few
centuries old. (p. 807)
The growth rate of total output in Europe was 0 percent per year
between 1500 and 1700, and 0.6 percent per year between 1700 and
1820 … the growth of income per capita was only 0.1 percent per year
in the earlier period and 0.2 percent per year in the later one …. Thus
the initial effect of faster income growth in Europe was to increase
population. (p. 808).
Their conclusion on population growth prior to the Industrial Revolution
was:
The rate of population growth in Europe between the years 500 and
1500 was 0.1 percent per year … the growth rate of world population
from the year 1 to 1750 was at 0.064 percent per year. (p. 807)
Galor and Weil also pointed out that:
The prediction of the Malthusian model that differences in technology
should be reflected in population density but not in standards of living is
also borne out … prior to 1800 differences in standards of living across
countries were quite small by today’s standards; yet there did exist wide

697
 differences in technology. China’s sophisticated agricultural
technologies, for example, allowed high per acre yields, but failed to
raise the standard of living above subsistence. Similarly, in Ireland a
new productive technology — the potato — allowed a large increase in
population over the century prior to the Great Famine without any
improvement in standards of living. (p. 807).

14.11 Conclusions
• The Solow growth model assumes constant returns to scale in labor and
capital, diminishing returns to capital per worker, and exogenous
technical change, as well as the absence of the government and financial
markets in the economy. These assumptions imply that the growth rates
of countries tend to the SS rate, in which per capita incomes grow at the
rate of technical change.
• According to the Solow model, per capita incomes do not grow endlessly
unless there is technical change.
• Assuming identical and static production technology across countries, the
Solow model implies that the per capita growth rates of countries will
converge to the same (zero) SS rate.
• The Solow model further implies that both the growth rate of the
economy and the return to capital decrease as capital intensity increases.
That is, the growth rates and the return to capital are higher in countries
in which it is scarcer. In this model, while there is some impact of
macroeconomic demand policies on growth, this impact is severely
limited because the determinants of growth are exogenous. In the limiting
case of the SS, there is no impact of policy on the per capita growth rate,
which is determined by the exogenous rate of technical change.
Some of these implications of the classical model seem to be contrary to
the factual evidence. Among these are:
For example, Romer (1986) analysed the per capita growth rates for the
USA and showed that these had a positive trend over the period 1800 to
1978, rather than the decline in the growth rate predicted by the traditional
classical theory.
For at least some of the middle-income countries in the world in the last
couple of decades, the rates of growth have been increasing rather than
decreasing over some periods. Among such countries have been Korea,
Taiwan, Singapore, Malaysia, Brazil, China and India.
Comparing the capital-intensive developed economies of North America
and Europe with the least capital– intensive ones, mainly in Africa, the

698
former economies have higher growth rates than the latter.
For pre-industrial societies, the Malthusian growth model is more
appropriate than the Solow one.

KEY CONCEPTS
Constant returns to scale in
production
Diminishing returns to capital per worker
Steady state output per capita
Steady state growth rate of output
Pre–steady state growth rate
Convergence and divergence of
growth rates among countries
Exogenous technical change
Labor in efficiency units
Growth accounting
Total factor productivity
The Solow model
Malthus’growth model
The post-Malthusian growth stage, and
The modern growth stage.

SUMMARY OF CRITICAL CONCLUSIONS

• The Solow model implies that in the steady state without technical
change, output per capita will be constant, so that the standards of living
will become constant.
• In the Solow model, an increase in the saving rate does not alter the SS
growth rate but will increase the SS output per capita.
• In the Solow model, countries with identical technology, saving rate and
labor force growth rate, will converge to the same SS output per capita.
• The Solow model with exogenous technical change produces the SS
growth rate of output per capita equal to the rate of technical change in
labor productivity.
• Labor force growth rates have been falling in recent decades in most
industrialized countries.
• Malthus’ analysis of pre-industrial economies implied that the long-run
standards of living would remain close to the subsistence level, as they
did in many pre-industrial societies over centuries.
• Most economies have historically gone through several distinctive

699
 growth stages.

REVIEW AND DISCUSSION QUESTIONS

1. Present the benchmark Solow growth model (including its assumptions)
without technical change and show diagrammatically its steady state
position. Is this position stable? Explain your answer, using the
appropriate diagram.
2. Is the economy always in the steady state? If not, assuming that there is
no technical change, what are the differences in the growth rates of (a)
output and (b) the standard of living, between the steady state position
of the economy and the positions prior to reaching it?
3. It is often claimed that (a) the growth rate of output, (b) the level of the
standard of living, and (c) the growth rate of the standard of living, are
negatively related to the population growth rate. Using two different
labor force growth rates for illustration in your analysis, show if each of
the above relationships is implied by the Solow growth model without
technical change or not. Give your answers for (i) the steady state and
(ii) the pre–steady state.
4. It is often claimed that the economy’s growth rate is independent of the
saving rate of the economy. To investigate its implications in the Solow
growth model without technical change, show the impact of increases
and decreases in the saving rate on (i) the SS growth rate and (ii) the
pre-SS growth rate.
5. It is sometimes claimed that while a war (on one’s own land) destroys a
great deal of the economy’s capital, it also leads to very significant
inventions and innovations that are incorporated into civilian production
after the war. Assuming that the economy was in the steady state at the
start of the war, present its diagrammatic analysis for the Solow-type
economy, comparing the pre- and post-war SS output per capita. What
happens in the ‘recovery’ period during which the capital stock is still
below the pre–war SS level?
6. ‘Countries (including the LDCs) in which governments pursued plans to
increase their saving rates as a means of boosting their growth rates
were misguided.’ Discuss.
7. What is meant by ‘convergence’? How does the Solow model explain
convergence?
8. According to the Solow model, what factors determine the SS growth of
output per capita? Can the long-run growth rate of output differ from the
SS one? Can the short-run growth rate of output differ from the long-run
and the SS ones? Explain your answers.

700
9. How does the growth accounting method calculate the contribution (to
the host economy) ofimmigrants? What does the Solow model imply for
the impact of a very significant rate of increase in immigration on the
steady state?
10. Present the Malthusian theory of growth. What are its critical
assumptions that cause long-run living standards to remain at the
subsistence level? Are there countries for which this analysis is still
applicable or at least indicative of their growth pattern?
11. Discuss the relationship (or different types of relationships) that seems
to have occurred historically between population growth rates and living
standards.

ADVANCED AND TECHNICAL QUESTIONS

T1. Assume that labor’s and capital’s shares of output are respectively 0.7
and 0.3, and that output is growing at 3% annually. There is no technical
change. What is the growth rate of total factor productivity if:
(a)Both labor and capital are growing at 2%?
(b)Labor is stationary while capital is growing at 3%?
T2. Suppose that increases in the labor force participation rate occur over a
particular decade, after which the participation rate stabilizes. Discuss
its effects on the growth rate of output per capita over the decade
relative to those before and after it. Specify any assumptions that you
consider to be relevant to your analysis.
T3. What explanations have been offered for the falling population growth
rates in the modern period? Can it be offset and how? Is there a limit to
the fall in the population of (a) individual countries and (b) the world?
Discuss.
T4. Assuming that all countries in the world have entered the ‘modern
growth stage’ (as defined in this chapter), present the analysis of the
long-run steady state for the world population and its standards of living
under the modern growth theory with exogenous labor-saving technical
change exogenous technical change.
Basic growth model for this chapter:

Y/L = 100(K/L) – 0.5 (K/L)2

APC = MPC = 0.8

L′′ = 0.04.

701
T5. For the basic growth model of this chapter, calculate the marginal
product of capital per worker (MPk). Is MPk constant in the steady state
of the basic Solow model without technical change? Explain your
answer.
T6. Using the basic growth model of this chapter:
(a)What is the SS value of k?
(b)What is the SS value of y?
(c)What are the SS levels of (a) consumption per capita and (b) saving
per capita?
(d)If the MPC changes to 0.7, recalculate your answers to the previous
questions.
(e)What was the change in the SS growth rates of output and capital for
the two values of MPC?
Revised growth model A for this chapter, with labor–saving
technical change:

yE = 1000kE – 2kE2

APC = MPC = 0.8

L′ = n = 0.02 and n = 0.02,

where n is the growth rate of workers and n is the rate of growth of

efficiency per worker.
T7. Using the Revised growth model A of this chapter:
(a)What are the SS growth rates of kE and k?
(b)What are the SS growth rates of yE and of y?
Revised growth model B for this chapter, with labor-saving
technical change:

yE = 2000kE – 4k E2

APC = MPC = 0.9

E′′ = n + n

n = 0.005, n = 0.015.

702
T8. Using the Revised growth model B of this chapter:
(a)What is the SS value of kE?
(b)What is the SS value of yE?
(c)What is the growth rate of consumption per capita in the steady state?
Suppose that because of a low birth rate, aging population with rising
numbers of retirees and inadequate immigration levels, the labor force
growth rate falls to –0.01 while n remains at 0.015. Recalculate your
answers to (a) to (c). Use a diagram to illustrate the general nature of your
results for this part of the question.
[ General note: in the presence of technical change, the information
provided by models A and B (with technical change) above is inadequate
to allow the calculation of the SS levels of variables such as y, k, c, and 5,
but allows the calculation of their SS growth rates.]
Mathematical Appendix
[OPTIONAL. For students who know differential calculus.]
A Mathematical Example of the Solow Model, Using the Cobb–
Douglas Production Function
Assume a production function of the Cobb–Douglas form:

so that

The return to capital

The return to capital equals its marginal product, which can be derived as
YK or as yk since they are identical. The MPK is given by:

Since (a’ – 1) < 0, ykk = a’(1 – a’)Aka –2 < 0, so that the return to capital
falls as k increases. In perfectly competitive markets, this return specifies
the real interest rate in the economy, so that, as argued earlier, this model
predicts a decline in the real rate of interest as k increases in the pre-SS
stage.
The SS growth rate
The labor force growth rate is still assumed to be n. That is, L′ = n. As in
the earlier arguments, the average propensity to save is assumed to be
constant at a , so that K’ = aY , and:

703
 K′ = aY/K = ay/k.

From Equation (15):

Y’ = YK K + YL L’,

so that:

which implies that:

Hence:

dy′/dk = a’ (a’ – 1)aAk

Since (α ’ – 1) < 0, the growth rate of per capita output declines — and
goes to zero51 — as k increases. To derive the SS growth rate, rewrite
Equation (33) as:

In steady state, σy = kn, so that the SS value y′′* = 0, where * indicates the
SS value.
Growth in the pre-SS stage
But, in the pre-SS stage, σy > kn, so that y′′ > 0. Further, with α’ = yk .
(K/Y):

so that, in the pre-SS stage, y′′ depends positively on the saving rate and
the marginal productivity of capital, but negatively on the labor force
growth rate and the capital–output ratio.52 Hence, countries with higher
saving propensities (or better access to the savings of other countries) and
a higher return to capital will have higher pre-SS growth rates of output
and output per capita.
The impact on growth of a tax on the return to capital
Now, suppose that, for some reason, the rate of return on capital, YK, is

704
reduced by T K. Such a reduction could be due to the imposition of tax on
dividends and capital gains or due to the inefficiencies resulting from
regulations imposed on the capital or financial markets. Let this reduction
lower the return on capital from Y K to (1 – T)YK. From Equation (34),
this lowers the pre-SS growth rate of output per capita to:

Therefore, taxing the return to capital is inimical to the growth in the

standards ofliving and in the captial-labor ratio. Further, the slower growth
of the latter will increase the time taken to reach the steady state.
Diagrammatic Appendices: Sources of Economic Growth

Calculation of Capital Input

705
 1Solow, R. M. A contribution to the theory of economic growth.
Quarterly Journal of Economics, 70, February 1956, 65–94.
2Y = y.L, Y′ = (y . L′) + L . y′, Y′′ = Y′/Y = y . L′/Y + L . y′/Y = y . L
′/yL + L . y′/yL = L′/L + y′/y = L′′ + y′′.
3If Y′′ = ã, then the mathematical form for Y will be Y = Y0eãt, where
Y0 is the base period output and e is a mathematical symbol with the value
2.718. The natural log of e is 1. Therefore, we can also write output in logs
as ln Y = lnY0 + ãt.
4Some Keynesian theories assume exogenous investment and
accommodative saving through changes in output.
5Solow, R. M. A contribution to the theory of economic growth.
Quarterly Journal of Economics, 70, February 1956, 65–94.
6We have made the assumption here and throughout our analysis of
growth theory that the population and labor force can be taken to be
identical for analytical purposes.
7σ is a Greek symbol and is being used as a parameter. It is pronounced
as “sigma”.
8An independent investment function is not being specified in this
model. It is assumed that the interest rate adjusts instantly to determine the
equilibrium level of investment by the amount of saving. In the short-run
models, an independent investment function was specified so that the
determination of the interest rate could be explicitly studied.
9We will follow this pattern of not labeling the vertical axis for similar
figures later in this and the next chapter.

706
 10This incorporates an implicit assumption that capital is fungible and
like putty, so that the adjustments to the implied capital/worker ratio can
occur each period.
11The saving rate in normal times always tends to be positive because of
the need to save for one’s retirement age, and also because of
precautionary saving against unexpected declines in income or increases in
consumption needs.
12This occurs in both Examples B and C when compared with Example
A.
13It lowered output per capita but this is a result of the unrealism of the
assumed production function.
14Note that the classical model does not necessarily imply convergence
in the level of output per capita: it does not do so if the production
functions are different, the saving propensities are different, or if the
production function has segments with increasing MPK.
15Even if different countries have different production functions–though
with diminishing MPK–and different saving rates, they would still
converge to the same (zero) SS growth rate of output per worker. Such
convergence has also not been observed in the real world.
16Solow, R. M. Technical change and the aggregate production function.
Review of Economics and Statistics, 39, August 1957,312–320.
17Ibid.
18This index can also be viewed as a composite index of all the inputs
(factors of production) used in production, where each input is weighted
by its income share in the base period and the weighted amounts of the
inputs are summed.
19The assumptions for this procedure are constant returns to scale in
production, perfect competition, profit maximization, and that technical
change is independent of the quantities used of labor and capital. The
procedure is explained in the appendix to this chapter.
20The rental cost of a machine is what it would cost to rent it for one
period. In perfect capital markets for both new and used machines, it
equals the interest cost on the funds to buy it, plus depreciation and the
decline in its market value (other than due to depreciation) during the
period. This is often simplified to the interest cost alone by ignoring the
last two components.
21The assumption of constant returns to scale implies that (àL + àK) = 1.
That is, the shares of labor and capital in national income sum to unity.
22Equation (17) can be simplified further under the assumption of

707
constant returns to scale, in which case àL + àK = 1, so that àL = 1.àK.
Substituting this in Equation (17) gives A′′(T ) = Y′′ .L′′ .àK(K′′ .L′′).
Since y = Y/L and k = K/L, y′′ = Y′′ .L′′ and k′′ = K′′ . L′′,A′′(T ) = y′′ .
àKk′′. That is, the growth of output due to technical change is the
difference between the growth of output per worker and the weighted
value of the growth of capital per worker.
23Martin N. B. Distinguished lecture on economics in government: the
new economy: post mortem or second wind? Journal of Economic
Perspectives, 16, Spring 2003, 2–22.
24This estimate is down from Solow’s corresponding estimate of 87.5%.
Most studies indicate that the rate of technical change decreased from
1973 to the mid-1990s, when it again picked up.
25This calculation subtracts 2% because we are dealing with output per
worker and the labor force grows at 2%.
26This implicitly assumes that the two types of workers are otherwise
identical, e.g., in their abilities, and there is perfect competition in the
labor market.
27Under a perfectly competitive system, this provides the marginal
productivities of the two inputs, high school graduates and college
education resulting in the BA, respectively as ñHS(= wHS) and ñBA(=
wBA . wHS).
28Remember its assumptions: constant returns to scale, perfect
competition, and profit maximization.
29This rate of productivity growth is quite large. It was partly due to a
substantial expansion of the university system and the consequent increase
in human capital per worker.
30Jorgensen, D. W. and Stiroh, K. J. Raising the speed limit: economic
growth in the information age. Brookings Papers on Economic Activity,
2000 (1), 125–211; Stiroh, K. J. What drives productivity growth? Federal
Reserve Bank of New York Policy Review, March 2001, 37–59.
31This form of the production can also be written as Y = F(K, E(L, T )).
32For example, suppose we have two workers, one of whom produces 20
units per hour and the other produces 40 units per hour. Using the less
productive worker as the base worker, the two workers in terms of
efficiency together represent three (base level) efficiency workers. In this
example, L = 2 and E = 3. Therefore, E would increase if L increases
and/or its average efficiency level rises.
33Such ‘technical change’ could take the form of higher levels of
education or more skills on average for the workers.

708
 34That is, labor measured in constant efficiency units grows for two
reasons: the growth of the number of ‘raw’ (i.e., at ‘base level’
productivity when T = 1) workers and the growth of efficiency per worker.
The total growth rate of labor in constant efficiency units is n + n.
35Such technical change is of a different type from the neutral one used
in the growth-accounting equation. This form of technical change is said to
be labor-augmenting technical change —as against being capital-
augmenting or being neutral between the capital and labor inputs —since it
acts in the same way as an increase in the quantity of base-level workers.
36In this SS with technical change over time, each unit of output is
produced by a smaller amount of raw labor, as well as a smaller amount of
capital.
37Using the analogy with physical capital, this process corresponds to
equipping the workers with capital at the existing K/L ratio.
38This can be seen by its diagrammatic exposition for n < 0 and à > 0,
since the only intersection between the nk and the àf(k) curves would then
be at zero. This SS will be unstable and there will be continuous growth of
capita and output per worker.
39Note that a negative value implies that population will fall.
40The average wage of women in the last four decades of the 20th
century was about 70% of male wages, so that their marginal product was
assumed to be correspondingly lower. This was taken to reflect their lower
skill levels.
41Further, while it represents the minimum to which the economy’s
growth rate converges under a stable SS, actual economies often
experience increasing rates of growth during some stages. The standard
classical growth model does not cover this possibility.
42In conditions where the country’s existing population has a negative
growth rate, many economists tend to favor immigration to prevent the
population from declining.
43The former also does not alter the SS growth rate of output while a
decrease in the population growth rate decreases the SS output growth rate.
44Our classification and its discussion are partly based on Galor, O. and
Weil, D. N. Population, technology, and growth: From Malthusian
stagnation to the demographic transition and beyond. American Economic
Review, 90, September 2000, 806–828.
45This occurs because improvements in living standards provide better
nutrition to mothers and infants, thereby reducing their mortality.
Conversely, a fall in living standards raises mortality rates, especially for

709
the very young.
46This often happened after a major epidemic such as the Black Death in
13th century Europe, which reduced the population by about a third and
led to a substantial increase in the living standards of the remaining
population. The settlement of new lands (mainly Americas, Australia, and
New Zealand) from the 17th to the 20th century by European settlers also
raised living standards and led to a rise in labor force growth rates in West
European countries until the twentieth century.
47This stage requires the ability to control the number of births.
48The desire to build a career first tends to raise the mother’s average
desired age for the birth of the first child into the 30s (often the late 30s)
and reduces the number of childbearing years.
49Britain was the first country to go through the Industrial Revolution in
late 18th and early 19th centuries.
50Galor, O. and Weil, D. N. Population, technology, and growth: From
Malthusian stagnation to the demographic transition and beyond.
American Economic Review, 90, September 2000, 806–828.
51As k approaches its SS value, our earlier analysis implied that the SS
growth rate y′′ will go to zero, so that in the steady state, output will grow
at the same rate as the labor force.
52However, note that Y′′ = àyK + (1 . à′)n, so that the growth rate of
output will also increase with n.

710
 CHAPTER 15
Advanced Topics in Growth Theory

This chapter extends the theory of growth to include endogenous

technical change, which is the outcome of decisions made by
economic agents to change knowledge on the production
technology. This knowledge can then be embodied in human or
physical capital. Examples of such endogenous technical change
are those resulting from the acquisition of new skills, research and
development, investment in new or improved equipment, etc.
The second topic explored in this chapter is the relationship
between inflation and growth, and that between the growth of the
money supply and the growth of output. A somewhat distinct
relationship from these is that between the development of the
financial sector and output growth.

The classical growth model presented in the last chapter has several
implications that run contrary to empirical evidence, at least for some
countries. Among these were:
(i) If we focus only on the steady state, the Solow growth model without
technical change implies that the standards of living would be constant.
Common experience at least during the last two centuries of the Western
economies suggests that the standards of living continue to increase over
time, so that either the countries never achieve steady state, or it is
continually shifting, or there is an important missing element from the
theory.
We have already shown that the Solow model with exogenous
technical change does imply rising living standards even in the steady
state.
(ii) If we assume that all countries have the same production function, but
different saving propensities and labor force growth rates, their SS
output growth rate in the absence of technical change would be set by
their labor force growth rate, which is exogenously given to the model.
Hence, countries with higher population – and therefore higher labor
force – growth rate should be observed to have higher output growth.

711
 This runs contrary to the general cross-country experience, since the
industrialized countries with low population growth rates often have
higher growth rates than developing countries with higher population
growth rates.
One possible explanation that might be adduced for this contrary
observation would be that the countries are not in steady state, but are
scattered along the k (capital per worker) axis, with the industrialised
countries having higher values of k and being closer to the steady state.
However, starting from k = 0, as we move to higher values of k, the
growth rate of output per capita (y″) should decline as the economy
moves toward its steady state. This again does not appear to be
consistent with the general growth experience across countries:
countries with higher k ratios do not always have lower growth rates of
output per worker.
A more satisfactory explanation for the higher growth rates of the
richer and more industrialized economies seems to be that their
technology and knowledge tend to increase continually while many of
the poorer countries stay with a relatively static technology. This
possibility is the focus of the first part of this chapter.
(iii) The real interest rate represents the real cost of borrowing funds with
which firms buy physical capital. The marginal product of capital
(MPK) is the return that they get from this capital. Therefore, profit
maximizing firms will borrow and invest until their MPK equals the real
interest rate. For the assumed production function, the MPK is
diminishing. Since the real interest rate (excluding a risk premium)
equals the MPK, the model implies that the economy’s MPK and the
real interest rate decline as it moves to higher capital/labor ratios and,
therefore, higher output per worker. That is, countries with higher
capital per worker and output per capita would have lower real rates of
interest. Since investors prefer investing in countries with higher
returns, capital would flow from the developed to the less developed
countries. But, in reality, the poorest economies do not usually have
higher real rates of return to capital than more developed economies and
capital sometimes flows on a net basis from the former to the latter.
This suggests that we need to modify the theory to somehow eliminate
the assumed decrease in the MPK as capital per worker rises, and
replace it with either constant or increasing MPK at the economy’s
level. This is done by some of the endogenous technical change theories
presented in this chapter.

712
15.1 Technology, Knowledge and Externalities
Technical change is a change in the methods of production and manifests
itself as shifts in the production function, as well as shifts in the marginal
productivities of labor and capital. Technical change is partly exogenous
and partly endogenous. Of these, endogenous technical change is that part
of technical change that arises from the decisions of economic agents,
often in response to economic incentives, to acquire new knowledge and
embody it in workers or physical capital or both. Examples of these
decisions are research and development (R&D) expenditures, research in
universities, etc. These improvements in technology may not require
investment in human and/or physical capital, but usually do so.
Endogenous technical change is often of a type that benefits its innovator
in terms of higher profits and incomes. In many instances, it also benefits
other economic agents — in the case of universities, the adoption by firms
of the results of research conducted in the universities, and, in the case of
firms, by the shifting of other firms’ production functions and marginal
productivities. The effects on economic agents other than the innovating
one are known as externalities. Endogenous technical change may or may
not have externalities.

15.1.1 The distinction between endogenous and exogenous

technical change
The distinguishing features of endogenous technical change versus the
exogenous technical change in the Solow model are:
(i) Endogenous technical change is the result of economic decisions to
bring about a change in the technology of production and usually
involves some cost. The theories that model these decisions are known
as endogenous growth theories. By comparison, exogenous technical
change does not require decisions by the economic agents and is usually
postulated as occurring gratuitously; as such, economic reasoning is not
used to explain it or derive its determinants.
(ii) Endogenous growth theories model technical change in such a way
that, at the macroeconomic level, the economy’s ( also known as the
social) MPK becomes constant or even incremental, even though the
private MPK to the firm in question is diminishing. This implies that the
economy does not reach a stable steady state but can continue to
increase its output per worker (y) and capital per worker (k) beyond the
steady state.

713
 The modeling of endogenous technical change is still relatively new and
imperfectly understood, so that there is no consensus around a single
model. There are different ways of modeling endogenous technical
change, leading to different versions of growth theory. The main rationale
for the macroeconomic study of endogenous growth in recent years has
been to use this concept to produce improvements in the standard of living
even in the steady state. To achieve this, the endogenous growth theories
have emphasized the externalities of the new knowledge. When this new
knowledge occurs in a firm, it may arise costlessly from ‘learning-by-
doing’ by workers or be a good produced through costly R&D
expenditures. Its externality to other firms occurs because the knowledge
thus created is not only utilized in the originating firm, but also becomes
available to other firms and increases their productivity. This externality
makes the social return to the creation of new knowledge greater than the
private one.
Knowledge can be introduced into the analysis in two different ways:
(i) As a distinct input, which is itself produced, e.g., by firms through
research.
(ii) Embodied in physical capital (such as in new types of machines)
and/or human capital (such as in workers with more education and
better skills).

15.1.2 The definition of capital in endogenous growth

theories
Innovation and new knowledge often enter production through new forms
of equipment or the knowledge and skills of the workers, so that physical
capital and/or human capital often serve as the vehicles for endogenous
technical change. The acquisition of both of these in endogenous growth
theory occurs through economic decisions involving (physical and human)
investments and both affect the economy’s output and its growth rate.
Given this similarity between the accumulation of physical and human
capital, the endogenous technical change approach usually combines the
two and defines the term ‘capital’ in the production function to consist not
only of physical capital, but also to include human capital. For this
combined capital, the definitions of investment and saving are suitably
modified to encompass those occurring in both physical and human
capital. Among the latter are the amounts spent on formal education or
otherwise on the acquisition of skills. Also note that the investment by
governments, as in roads and other infrastructure, is part of this capital.

714
15.1.3 The externalities of new knowledge
The generation of new knowledge can be separated from its dissemination
— i.e. ‘spillover ’ — to other firms. The former usually occurs within
firms, universities, and research institutions. The latter is the process of its
copying and adoption by other firms or by workers in other firms, and is
an externality from the perspective of the firms and other units originating
the new knowledge. The creation of new knowledge can therefore possess
an externality. These externalities often occur because of the ability of
firms to observe the types and quality of products, the methods of
organisation and the technology employed by other firms and, if these are
superior ones, to be able to replicate or adapt them to their own
production.
Externalities in the economy can also arise in other ways, such as from
the expansion of the industry and the economy. For example, as the output
of some firms in an industry expands, the facilities for the transport of its
raw materials, the training of workers appropriate to the industry and for
the marketing of output, etc., tend to improve and lower the relevant costs
for all firms in the industry.
Externalities of knowledge represent a public good, 1 which is available
to all firms, even to those that do not contribute to its creation. Access to it
does, however, usually require some action and investment in new
physical capital by the firm. For example, new investment may be required
to buy the machines embodying the economy’s new knowledge of
production techniques, so that new investment acts as the gateway to the
firm’s use of the increased knowledge. In some cases, the new machines
may not require an increase in the firm’s capital stock but rather require
replacement, which could be covered through depreciation of the old
machines.
New knowledge can have not only positive externalities but also
negative ones for the economy. The incorporation of endogenous technical
change into growth theory, in general, assumes that the net benefit from
such change is positive; that is, there is an increase in output for given
levels of labor and capital in the economy. We will continue with this
assumption, but will later include some mention of the negative
externalities.
Box 15.1: Inventions, Innovations, and Patents
Innovations versus inventions as forms of technical change
Looking from the perspective of economic history, we can take technical
change to be characterized by

715
(a)‘Macro inventions’ . Inventions of this nature not only affect the
particular industry in which they occur or first occur but sooner or
later affect a wide range of industries in the economy — so that their
affects are macroeconomic rather than being confined to a single
industry. They are often technological breakthroughs. Illustrative of
these are the inventions of the steam engine, blast furnace, cotton
ginning, etc. during the Industrial Revolution in Great Britain during
the 17th and 18th centuries, and the invention of the transistor, of the
computer chip in the third quarter of the twentieth century, and of the
Internet in the last quarter of the 20th century. Such breakthroughs are
essentially unpredictable and occur unevenly over time. Since their
nature and time of occurrence cannot be anticipated in advance,
inventions are treated as being exogenous.
(b)‘Innovations ’. They consist of refinements of the existing technology
and are largely confined in their effects to a single industry.2
Therefore, innovations represent a cumulative process that builds on
previous discoveries and previous innovations. This process can occur
over very long periods following the macroinventions or other basic
inventions. Among the hindrances particular to a country to the pace
and extent of such innovations are the smallness of markets, the
weakness of scientific education, the lack of economic incentives,3 as
well as the unwillingness of labor to adapt its ideas and skills, work
habits, etc.
Since macroinventions are rarely continuous, their impact on the
economy will eventually peter out so that they cannot be made the
main vehicle in a theory of continuing technical change even in the
steady state. Therefore, the main emphasis of endogenous growth
theories for growth is usually placed on innovations.4 With the
emphasis on innovations rather than inventions as the engine of
continuous technical change, the ability of a nation to generate or
improve on imported ideas, its propagation mechanisms for
knowledge spillovers among its industries, and the learning
capabilities of its workers become the main elements of its
endogenous growth. Since government policies can affect these, the
endogenous technical change theories imply that the long-run growth
rate of the economy can be changed by such policies.
Further, it is not the occurrence of inventions that changes production
and causes growth; rather, it is their adaptation and incorporation in
production processes that does so. This process is essentially one of
innovation and takes time, sometimes many years and even decades.

716
 Therefore, we should not expect an important invention and its early
limited knowledge and implementation to have a significant impact on
productivity at the macroeconomic level for quite some time. A potent
example of this is the invention of the microchip and the Internet. While
they were invented and in use long before 1990, their impact on
productivity was not felt until after the mid-1990s: as Table 14.1 showed
for US data, the slowdown in productivity growth continued through the
1980s and early 1990s, before picking up dramatically after 1995 (see
also Table 15.1).

15.1.4 Patents and the protection of copyright, intellectual

property and trade secrets
Intellectual property protection grants legal rights to the originator for
written as well as oral inventions and innovations. Among such rights,
patents grant legal rights to the innovator for the use of a new physical
product or process. Copyright grants legal rights to the author for a new
written or oral work, such as a book, or a song or software code.
Most countries have patent laws that confer on the inventor/innovator the
exclusive ownership of the item or process that has been patented. To be
granted a patent, the item or process must be original — and therefore, not
known prior to the application for the patent — so that it represents new
knowledge in the context of our earlier discussion. Copyright is the
corresponding term for oral and written works, such as books. Both are
encompassed in the concept of intellectual property rights. Owners of
intellectual property rights get the exclusive right to market the goods and
services arising from their creation of new knowledge for a limited period
and within a specific jurisdiction.
The patented or copyright item or process can only be used with the
owner’s permission, which is usually not given without some financial
compensation such as a fixed payment or through royalties, etc. Therefore,
patents and copyright represent a way in which their owner can capture
some or all of the positive externalities associated with new knowledge, so
that the divergence between the private and social returns to new
knowledge is reduced. Within a closed economy, ceteris paribus, patenting
involves a redistribution of income to the owner of the patent from others
who use it. Growth theory is not concerned with the redistribution of
incomes but only with whether production increases or does not increase
because of patenting.
Patents and copyright increase the returns and incentives to inventions

717
and innovations, and therefore, provide incentives to the creation of new
knowledge. But, during the life of the patent, they can also hinder or slow
the dissemination of the patented knowledge, as well as hinder or reduce
further innovation emanating from the patented knowledge.
At the international level, patenting confers a net economic advantage on
the country in which the patent owners reside (since they get the profits
from it) and in which the patent is put into production. But what of other
countries, which get neither the profits from a patent nor have local
production based on it? Their consumers can benefit from the imports of
products based on the foreign patent. However, their production of the
products displaced by the foreign patent will cease. Given the very limited
immigration permitted by/among countries, workers thus displaced in a
given country by foreign patents may be unable to flow to the jobs created
in the foreign industries expanding on the basis of the patent. There could
then be a net increase in unemployment and decrease in the standard of
living in the former set of countries. The externalities of new knowledge
controlled by patents can therefore have different effects from those when
they are not thus controlled, and need not necessarily make countries better
off when the patents and production are confined to other countries.
Patents create monopoly power for a specified duration. The period for
which the patent is granted is relevant for endogenous growth.
Lengthening the period increases the returns and incentives for the owner
of the patent, but it also restricts usage of the patented product or process
by others within the country and in other countries. Further, stronger
patents discourage sequential research and follow-on inventions, especially
by others. These constitute opposing effects of patents on the creation and
use of new knowledge and imply that while some patent length is likely to
be desirable, extremely long patent protection can stifle the growth of
knowledge and reduce the long-term growth of the economy.
At the international level, longer patent protection periods work to the
advantage of nations — such as the United States and the European Union
— that generate more patents and to the disadvantage of nations that
generate less.
Recent changes in patent protection
The United States in recent decades has strengthened patent protection.
This has been done by extending patent protection to areas not previously
covered — such as genetically engineered bacteria, plants and software —
and by lengthening the life of the patents. Further, it has pushed for their
stricter and more vigorous enforcement in other countries by using its
influence in world bodies such as the World Trade Organisation (WTO).

718
For example, countries such as China that did not at one time subscribe to
the international patent and copyright protection laws, were persuaded to
sign them and gradually to enforce them effectively.

15.2 The Production of New Knowledge

To evaluate the role of knowledge in production, assume that the firm (or
different firms) produces two types of goods: ‘commodities ’ and ‘new
knowledge ’. Their respective production functions include four inputs:
• labor,
• capital,
• the firm’s (existing) knowledge, and
• the economy’s (existing) aggregate knowledge.

The production functions for both commodities and new knowledge have
the usual assumptions made in the Solow growth model: constant returns
to scale and the diminishing marginal productivity of each input, with all
other inputs held constant. In deciding on the production levels of
commodities and new knowledge, the firm maximizes profits by choosing
its levels of capital and labor, while taking its existing knowledge and the
economy’s existing knowledge as given.
The firm’s cost of production of the new knowledge can be labeled as its
R&D expenditures. For the firm doing it, the production of knowledge
through such expenditures is subject to diminishing returns to the variable
inputs, which are labor and capital. Consequently, each profit-maximizing
R&D firm (or the research done by the production firm) would produce
only a limited amount of it since it5 will only get its private (and not the
social/macroeconomic) return to the new knowledge created by its R&D
expenditures. However, this knowledge can be only partially kept secret or
patented, so that the new knowledge, once created, also benefits other
firms in the economy.6 The output of new knowledge enters as an
intermediate good in the production of (final) commodities.
From the economy’s perspective, two distinct cases emerge for the
production of final commodities. These are
a. The new knowledge created in the economy and its externalities are not
significant enough to create constant or increasing returns to the
capital-labor7 ratio at the macroeconomic level. In this case, even after
taking account of the externalities of new knowledge, the production
function for the final commodities has diminishing returns to capital per

719
 worker, as in the preceding chapter, and the Solow growth analysis
applies.
b. The new knowledge and its externalities are strong enough to create
constant or increasing returns to capital per worker in the production of
final commodities for the macroeconomy .8 Assuming the case of
increasing marginal productivity of capital for worker for the economy
as a whole, the economy’s production function has the convex
(increasing MPK) shape shown in Figure 15.1. Note that this case
requires not only a significant enough creation of new knowledge each
year, but also its sufficiently extensive dissemination to the production
of final commodities.

15.2.1 Diagrammatic analysis

The diagrammatic analysis for countries with increasing returns to capital
intensity (i.e. capital per worker) was presented in Chapter 14. The
increasing marginal productivity of capital per worker implies that the
σf(k) curve would be convex rather than concave. Its diagram is repeated
as Figure 15.1. This diagram shows two intersections between the σf(k)
and nk curves and therefore, two SS values of k, k 0* and k 1*, at each of
which σf(k) = nk. Of these, k 1* is stable but k 0* is unstable. This is shown
by the following arguments:
a. For k > k 0*, saving per capita exceeds the capital requirement for new
workers so that the capital-labor ratio k rises over time for all workers.
The economy gradually moves further to the right of k 0*.
b. For k 1* < k < k0*, saving per capita is less than the capital requirement

720
 for new workers so that k decreases over time towards k 1*.
c. For k < k* , saving per capita exceeds the capital requirement for new
workers so that k increases towards k 1*.

Since any deviation from k i* brings the economy back to it, it is a stable
SS value. But since a deviation of k from k 0* leads the economy away
from it, k 0* is an unstable SS value. Even though k 0* is unstable, it has
the appealing empirical implication for economies with k > k 0* that the
return to capital does not decrease as capital per worker increases nor do
the growth rates of capital and output per capita go to zero, which seem to
correspond to the experience of the developed economies. An
objectionable implication of this case from the perspective of empirical
validity is that the growth rate of the economy will continue to increase
indefinitely, which is highly doubtful.
Consequently, the case of the increasing MPK was discounted by the
growth theorists, at least for the developed economies, so that, for several
decades following Solow’s (1956) contribution, the profession mainly
opted for the standard classical production function having diminishing
MPK and implying a stable steady state. The endogenous growth theories
allow only diminishing MPK at the firm (or individual’s) level but, if all
the firms collectively innovate enough, allow increasing MPK to physical
and human capital — since they embody new knowledge with its
externalities — for the macroeconomy.
In Figure 15.1, as shown above, the SS point at k 0* is unstable, so that
countries to the right of it (at k 2) will be on an ever-expanding path of
output per worker. Countries (say, at k1) with k less than k 0* will be on a
decreasing path of output per worker. Therefore, k 0* represents a critical
capital/labor ratio for the country and policy makers should ensure through
their policies that their economy has a larger capital-labor ratio than this
critical one. Such a critical ratio is called a takeoff or threshold capital
intensity, and a ‘growth trap ’ occurs from k 1* to k 0*.

15.3 The Dissemination of Knowledge Across

Countries
The preceding analysis can be extended for comparisons of growth

721
across countries. As a simplification for the following discussion, we will
adopt the country’s existing capital-labor (K/L = k) ratio as the proxy for
its existing knowledge as well as for its capacity to adopt new knowledge
from other countries. For this purpose, ‘capital’ is defined as the sum of
both physical and human capital.
For a country with the most advanced technology, there is no benefit
from copying the inferior technology of other countries (with lower capital
per worker).9 But countries with less advanced technology have access to
a large pool of knowledge, hitherto untapped by them, but available in
other countries. But to take advantage of this pool requires:
(a) Willingness to adapt and change in an appropriate manner.
This is likely to depend on social and cultural patterns and familiarity
with foreign knowledge and practices. It may require cultural and social
changes.
(b) Adequate absorptive economic capacity.
This is likely to depend on the gap in the existing knowledge and skills
and on the difference in the country’s industrial structure from those of
foreign countries.
(c) Investment in human and physical capital.
Such investment is usually needed to access, adapt, and utilize the
foreign knowledge hitherto not used in the domestic economy. In free
enterprise economies, the amount of such investment is likely to depend
on the incentives to the acquisition of new knowledge and the adoption
of new technologies. In centralized economies, it will depend on the
resources committed to this process.
Some countries may fulfil these conditions or do so at certain times or
for certain types of changes. Others may not do so at all. It is often
contended that open, free enterprise and capitalist economies give more
scope to the creation and economy-wide use of new ‘intellectual capital’
by workers and firms than closed, traditional, or centralized and
bureaucratic ones.10 Further, countries with too low a capital (physical
and/or human) intensity may not be able to effectively imitate the
advanced technologies and, therefore, not be able to fully benefit from the
knowledge already existing in developed countries. Still other countries
may have an adequate base for understanding the new knowledge but not
be sufficiently open to its acquisition or adaptation or be unwilling to
invest sufficiently in its adoption. The reasons for this could lie in the
nature of society and polity as well as in its economic structure.
The above arguments imply that:

722
(i) The prosperity and economic growth of countries depend in very
important ways on the social, cultural, and political environment. These
factors include the political system, the rule of law and protection of
property and person, the capitalist ethic, ethnic and religious diversity,
openness to new ideas and institutions, lack of corruption, level and
distribution of income and of education, etc.
(ii) The growth rate is endogenous, so that it depends on the economy’s
capital intensity and its increase through investment and innovation.
(iii) The most technically advanced country, designated for simplification
as the one with the highest value of k (with k = kmax), is already at the
frontier of the world technology and benefits mainly from its own
inventions and innovations. These increase its own — as well as
worldwide — knowledge.
(iv) Other countries (with k < kmax) not only benefit from their own
inventions and innovations, but also by copying techniques known to
the more advanced countries. Hence, over a certain range, some
intermediate set of developing economies (with lower existing values of
k than kmax) can grow more rapidly through a faster change in their
technology than more developed ones. This usually requires a faster
accumulation of both physical and human capital.
(v) A country with extremely low technology and capital-labor ratio —
and/or a rigid social and economic structure — may not be able to adopt
other countries’ techniques. If its own ability to innovate is also absent,
it could continue with a relatively static technology, diminishing MPK,
and low standard of living.
If there exist extensive flows and adoption of technical knowledge, we
should expect standards of living among countries to converge. The
European Union provides an example of such a tendency among its
member countries. However, not all countries in the world have similarly
high flows of technical knowledge, the capacity for the effective
acquisition and use of others’ knowledge, and the ability and willingness
to invest in its adoption. Hence, some countries may continue to diverge
over long periods from others in their standard of living: while the per
capita incomes and growth rates of countries in the European Union in the
coming decades are expected to converge, significant convergence need
not occur between the European and some African countries during the
next decade.

15.4 Diagrammatic Analysis

723
Figure 15.2 illustrates the general scenario. This diagram uses k as a proxy
for the level of technology. Hence, k on the horizontal axis represents not
only physical capital per worker but also the technology level, and
movements along f(k) incorporate technology shifts as k increases. The
most advanced countries (with very high values of k) would have
diminishing returns to capital intensity so that they would be on a concave
segment of the curve σf(k) in the neighborhood of the stable SS ratio k 2*.
By comparison, countries in the intermediate range of the capital-labor
ratio will benefit from the externalities of the knowledge of the more
advanced economies. Some of these will have increasing growth rates and
increasing returns to capital. Such countries will be in the region k 1* to k4.
However, these higher-growth countries cannot increase or even maintain
their higher growth rates indefinitely: their higher growth rates of output
and saving per worker take them towardk 2* and lower their growth rate.
Eventually, their growth rates converge to those at k 2*.

The poorest countries (with low capital-labor ratios to the left of k 0*)
can stagnate at low values of capital intensity and low growth rates. These
countries are in a growth trap in the sense that even if they were somehow
to advance above k 0*, they will fall back to it — unless they manage to
get to values of k higher than k 1*, which will set them on a growth path of
improvements in their standard of living. Further, being quite poor with
limited savings, they have a limited capacity to invest in new knowledge,
even in just taking advantage of the knowledge that already exists in the
rest of the world. For them, the point k 1* in Figure 15.2. represents a
threshold or take-off capital-labor ratio, which they need to exceed if they

724
are going to enter a path of increasing standards of living. Shocks
(famines, civil, and other wars, diversion of the economy’s output to
consumption or wasteful expenditures rather than capital accumulation,
etc.) and policies (e.g., those discriminating against capital accumulation)
which shift the economy from the right of this threshold point to the left
are highly undesirable, while those which do the reverse put the economy
onto a path of ever-increasing standards of living. This danger is acute for
an economy close to the point k 1 *.11 Hence, it is very important for
policy makers to push and position their economies appropriately along the
k axis, and if they are LDCs with very low capital intensities, they must
somehow exceed the threshold capital intensity.
Therefore, the countries with lower capital-labor ratios do not always
have higher growth rates, though some of the countries with (intermediate)
capital-labor ratios in the range k 1*to k 2* will have the highest growth
rates and the highest return to capital. Furthermore, this igure implies both
a divergence and a convergence: divergence between the countries moving
toward k 0* and those tending to k 2*, while there is convergence of the
poorest countries to k 0* and the richest ones to k 2*.

15.4.1 Growth rates and capital flows among countries

The preceding analysis implies that economies can be divided into three
categories:
(a) The ‘poorest’ converging to k 0* with very low values of k and y.
(b) The fast-growing (intermediate) countries (‘the economic tigers’)
moving rapidly through high growth rates toward the k and y values of
the most developed economies.
(c) The ‘richest’ industrialized countries converging to k 2* with still
higher values of k and y.
In this scenario, the growth rates are higher for the economic tigers than
for the other economies. Further, since the MPK is highest for the
economic tigers, capital flows from the poorest and the richest countries to
the economic tigers. There is divergence of living standards between the
poorest countries and those over the takeoff point k 1*— that is, the
economic tigers and the richest countries — while there is bipolar
convergence, with that of the poorest countries to k 0* and its implied low
output per capita, while the economic tigers and the developed economies

725
converge to k 2* and its implied high output per capita.

15.4.2 Convergence and divergence in living standards

between countries
In the context of endogenous technical change, the convergence in living
standards between countries occurs for two reasons: the movement toward
roughly the same stable steady state as in the Solow model and the
convergence of technologies through adoption of the more advanced
technologies of other countries. Divergence can also occur for one or both
of two reasons: an unstable steady state as in the Solow model and the
divergence of technologies when there is an effective failure to adopt the
more advanced technologies of other countries.

15.4.3 Continuous growth in the advanced economies

Some studies in the endogenous growth literature suggest that advanced
economies, including the country with the highest capital intensity (i.e.,
closest to k 2*) and the most advanced technology, may escape for some
time, diminishing returns to capital per worker if their creation of new
knowledge is strong enough to produce constant returns or even increasing
ones. If this were so, they would continue to have increasing output per
worker and standards of living over time: the steady state capital/labor
ratio rises continually rises because of endogenous technical change.
However, this cannot be taken for granted: it may not happen and, when it
does, it may cease after some time.
Box 15.2: Foreign Aid and Investment — and Capital Drain
Foreign aid is usually given to provide immediate relief from the misery
caused by disasters such as droughts and earthquakes. It is often not
enough to restore the prior standards of living of the recipients. It also
often does not restore their prior levels of capital or saving. These
disasters still reduce the amount of capital and investment — and
consequently the growth rate.
Figure 15.2. provides a justification for aid and foreign investment
from the context of growth theory. In Figure 15.2, increasing MPL
occurs from k3 to k4 . Countries that are between points k3 to k4
experience a rising growth rate — providing they can take advantage of
it through adequate investment. Countries between k 1* and k 4*do have

726
 adequate savings to raise their capital per worker and are on the path of
rising growth rates. However, countries between k 0* and k 1* do not
have adequate saving of their own to do so, so that over time their
capital per worker and their growth rate falls. This provides one
rationale for foreign investment and aid: if these, plus domestic saving,
increase the capital/labor ratio, the growth rate rises. If these flows are
maintained long enough to take the country beyond k 1* — the self-
sustaining take-off point — the country enters a high growth path
sustained by its own saving.
Conversely, an outflow of domestic saving or human capital — a
‘capital drain’ — to foreign countries will lower the domestic growth
rate. In an extreme scenario, such a capital drain could push the country
below its take-off capital/labor ratio and send it on the path toward
decreasing capital per worker and lower standards of living.

15.5 The Historical Experience of Growth

Romer (1986, p. 1008) pointed out that “Growth for a country that is not a
leader will reflect at least in part the process of imitation and transmission
of existing knowledge, where the growth rate of the leader gives some
indication of growth at the frontier of knowledge.”12 He argued that the
leader was Netherlands during 1700–1785 (with a growth rate of GDP per
man-hour of –G.G7%), the United Kingdom (UK) during 1785 to 1890
(with growth rates of 0.5% during 1785–1820 and 1.4% during 1820–
1890) and the United States (USA) since 1890 (with a growth rate of
2.3%). This evidence suggests that the leader can change over time and
that the growth rates of the leading country have been increasing over
time.
Looking at the more recent experience with growth, the USA, Japan,
Canada, and many western European countries are at or close to the
frontiers of world knowledge. Their growth rates were fairly stable or
declining over several decades until the mid–1990s, when the benefits of
the Information Technology (IT, including computers and the Internet)
revolution eventually began to be realized.
The growth rates of Japan increased during its rapid development in the
1950s and 1960s, but decreased in the 1980s and 1990s. The growth rates
of Korea, Taiwan, Thailand, Malaysia, and some other Asian countries
increased in the 1980s and early 1990s as their pace of development
picked up, though it remains to be seen if they will eventually also revert

727
to slower growth rates.13 This slowdown and convergence to a common
growth rate are implied by the above model for countries as they approach
the frontiers of world knowledge. Currently, China, India, Russia, and
Brazil, along with some other countries, are going through a growth spurt.
The years since the mid–1990s have been especially exciting ones for the
growth experience. Growth rates in many advanced economies increased
rather than declined, contrary to the predictions of the Solow growth
model. Proponents of endogenous growth theories have attributed this
increase to the IT revolution, which had its beginnings in the 1980s but
only seemed to make its macroeconomic impact in the late 1990s. The
process consisted of the original major breakthroughs in knowledge,
sporadic inventions, followed by decades of innovations and
dissemination. Heavy investments in labor and capital had to be devoted to
these. Countries that had the initial capacity to innovate and were willing
to make the investments benefited the most in terms of higher growth and
increases in the standard of living. Among these were Canada, the USA,
and western European countries.
Box 15.3: Economic Globalization and Endogenous Technical Change
As discussed in Chapter 12 on globalization, the market system provides
strong incentives in the form of larger profits for firms to increase the
size of their market first within the nation and then across the nations. In
the limiting case, this process leads to a single world market for
individual products — i.e. globalization of the market. This expansion of
the market requires a market-oriented system among countries, the
dismantling of tariffs and quotas and reduction of communication and
transport costs.
Economic liberalization:
• Increases the degree of competition, which acts as a stick for domestic
industries to innovate so as to be able to compete with foreign firms.
• Increases the inflow of knowledge on more advanced production and
management techniques.
Liberalization has, therefore, intensified the forces that promote
endogenous growth among countries, especially among countries whose
technology was not at the frontiers of knowledge and who are not
successful in the creation of new knowledge.
Technology Clusters
In practice, it is not just one country that is at the frontiers of knowledge.
Even for a given industry, many countries are often clustered at this
level. There is both similarity and diversity of knowledge and

728
 production among these countries. Each is able to adapt the inventions
and innovations originating in other countries. Consequently, there is
considerable similarity in their standard of living and their growth rates.
Over time, some other countries initially outside the cluster at the
frontiers of knowledge, begin to grow even more rapidly. Part of this
growth comes from being able to share in the knowledge of the
advanced cluster. Often, the countries that are best able to do so lie on
the periphery — geographical, cultural, linguistic, etc. — of the
advanced countries, so that development often proceeds in ever-
expanding circles around an advanced country or countries. The formers
eventually catch up with the cluster and join it. Subsequently, their
growth tends to slow down to match those of the cluster. The invention
of the Internet and the decrease in communication and transport costs
have considerably reduced the effects of physical distance, so that
physical distance has become relatively less important to the spread of
innovations.

Globalization and Patents

The pressure for liberalization of the trade in commodities has often
come from countries with an overall competitive edge over others. Since
these are often the countries with greater innovations and innovations,
they have also tended to espouse the strengthening and broadening of
patent and copyright protections (in laws and implementations at the
national level, at the regional bloc level and through international
organizations such as the World Trade Organization) and lengthening
the duration of the monopoly rights conferred by them. This increases
the economic benefits from globalization to the countries whose firms
obtain this power — and increases the national income and wealth of
their countries — while reducing the benefits of globalization and
liberalisation of external trade for other countries.

15.6 Human Capital

While human capital is often measured by the amount of schooling, the
decision on the amount of schooling is like those on the amount of
investment in physical capital. Increases in human capital improve output
per capita, as we saw from the Solow growth model for increases in the
saving and investment rate. For continuous growth in the SS standard of
living, what is critical is not the schooling of children up to the existing
levels of knowledge, but a continuous increase in the average knowledge

729
of the population. One component of the latter is the generation of new
knowledge. This depends on the ability and willingness to create new ideas
or borrow and adapt them on an economy-wide basis, and on the
incentives for doing so. It will ultimately depend on the economic
decisions taken by the individuals, society, and government on the
proportion of the human capital to be allocated to the creation of new
human capital. Research-oriented universities and institutes are an
important element in this creation, as are the R&D expenditures by firms.
Box 15.4: The Innovative Process and Its Creative Destruction
The pedagogical nature of classical, and even Keynesian, economics
does not focus on the institutions and processes that lead to technical
change in the economy. As against these schools, the Austrian School in
economics, whose most famous exponent was Joseph Schumpeter,
emphasizes the role of the institutional structure in the evolution of
technology. According to it, self-interest in competitive markets
constantly drives firms or entrepreneurs to improve their products and
their technology in order to gain a competitive edge over others. This
can be through invention or innovation. The successful firms create
monopolistic rents by being the first to produce a new product or use a
new process, so that such rents (which exceed normal profits under
perfect competition) are the driving force for invention and innovation.
Other firms try to reduce their losses by copying successful innovations
and innovating on their own. Firms which do neither lose and are often
forced out of production. Therefore, the production structure of the
economy is in a continuing state of flux, with new firms coming into
existence, some existing firms growing, while others become bankrupt
or are taken over. This Austrian notion that the firms and the technology
of an economy are constantly in flux clearly differs from the reliance on
the mainly static (or with exogenous technical change) production
function of the Solow growth model. The change in the economy is
often compared to that in a forest, in which new trees are always coming
into being and some are growing larger while others collapse. The
introduction of endogenous technical change in growth theory in recent
years is in some ways a movement toward modeling the Austrian view.
The above arguments suggest that the competitive capitalist economies
possess an advantage in bringing about endogenous technical change
since their firms are continually attempting to gain — by creating
monopoly profits — from changing their products and/or techniques of
production. By comparison, the centrally planned or highly regulated
economies do not provide sufficient market incentives for their workers

730
and production units to change and penalties for not keeping pace with
competitors, with the result that such economies do not innovate as fast
as the capitalist ones.
An aspect of the struggle among firms to gain an advantage over other
firms through innovation and invention is that some of these will not
prove successful in the market, so that the investment in them will be
wasted. Another is that technical change not only has positive
externalities, but also causes negative externalities. The process of
innovation usually leads to what Joseph Schumpeter called ‘creative
destruction ’, in which some firms lose their market share and may even
close. Further, some of the existing machines embodying the older
technology will become obsolete while some will need investments to
update them. The benefits from the new technology in terms of output
have to be offset by the losses in production resulting from the
obsolescence of existing capital and the failure of firms which can no
longer compete. The net social benefit is therefore reduced by these
costs and may even become negative.
The innovative process also affects labor. The employees of firms that
can no longer compete will be laid off and will need to find other jobs.
Further, some of the workers even in the successful firms may be left
with out-of-date skills and become unemployed. Such creative
destruction by innovations thereby decreases some types of employment
and by some firms, even while the innovations increase the average
productivity of the employed workers and could increase employment in
the economy.
Another source of the impact on employment arises from the
substitution of capital for labor, or vice versa, in the shift from the old to
the new technology. Whether this increases or decreases employment in
the economy depends on the nature of technical change as being capital
or labor intensive, the adaptability of labor to the new technology, and
the increase in labor demand due to the shift in the production function.
It is often suggested that the overall effect of innovations is to raise the
unemployment rate in the short run but not necessarily in the long run.
However, the effects are likely to differ for different types of technical
change.
In the national context, the innovative process will lead to some
industries expanding while others contract, and sometimes even
disappear. Among the regions of a country, this process will cause
output and employment in some regions to increase, while those in other
regions decline. This usually translates into labor migration from the
declining to the expanding regions and population shifts among the

731
 regions of the country.

The negative impact of globalization and endogenous technical

change
In the international context with free trade in commodities, if the
innovating and expanding firms are abroad and the declining ones are
domestic ones, the domestic economy would suffer a net decrease in
employment and output from endogenous technical change. Therefore,
worldwide increases in knowledge do not necessarily benefit all
countries equally and may be detrimental on a net basis for some
countries, especially those lagging in innovations. Since poorer
countries have limited resources to allocate to the creation of new
knowledge and less absorptive capacity for innovations occurring in
other countries, they are likely to be the laggards — and, therefore,
likely to be relative losers from the globalization of the trade in
commodities and knowledge. Consider the computer and Internet
revolution. The main innovating firms such as Microsoft were in the
USA and earned monopoly profits on a worldwide basis. These
benefited the Microsoft shareholders and workers, and the USA
benefited from the inflow of royalties, etc. Other countries also benefited
from the new technology but less so than the USA.
Since population flows across countries are controlled, population
usually does not flow in adequate numbers from the losing countries to
the expanding ones. Therefore, for the losing countries, the free flow of
commodities but not of labor can produce declining standards of living
and higher unemployment over long periods.

15.7 The Implications of Endogenous Growth

Theories for Macroeconomic Policies
The Solow model assumes exogenous growth rates of labor and technical
change, so that there is no scope for policy in influencing the SS output per
capita and the SS growth rate. However, note that even in such a model, an
increase in the saving rate would have increased the pre-SS growth rate, so
that there was some scope for policy to increase the growth by increasing
the saving rate in the economy. Since a particular economy may not reach
its steady state for long numbers of years, the impact of increased saving
rates on the growth rate would be observable over long periods even in the
context of a strictly classical economy without endogenous technical

732
change. Further, policy-induced changes in labor force growth rates will
change the capital-labor and output-labor ratios.
The endogenous growth theories provide for effects of government
policies on per capita output growth rate if they increase (or decrease) the
growth rates of knowledge. Among such policies are those that promote
research in universities, R&D expenditures by firms, and other creative
processes. Any policies which lower the return to human or physical
capital or reduce saving — and thereby reduce the growth rates of physical
and human capital — reduce the growth rate of the standard of living.
Among these policies are the imposition of taxes on profits and interest
earnings, and policies detrimental to research, R&D, and innovation in
general. Among other policies that could do so are excessive regulations,
rampant corruption, hindrances to financial development, etc.
There can also be effects of short-run fiscal and monetary policies on the
long-run growth rates. We illustrate these by considering the possible
effects of recessions on growth. The Austrian school’s notion of creative
destruction from technical change implies that recessions weed out firms
that are not sufficiently innovative in finding new markets and/or new
processes. The ones that survive are more successful at doing at least one
of these. Recessions, therefore, serve as a stick for staying competitive and
for innovation and change. Stabilization policies that eliminate or
moderate recessions could, therefore, lower the rate of endogenous
technical change. So could rescuing (‘bailing out’) failing firms by
providing huge government subsidies, as occurred in the USA and the UK
in the financial and economic crisis of 2007–2010. Conversely, sharp and
unexpected recessions which reduce demand beyond the capacity of firms
to protect themselves by innovating would also eliminate some firms
which, had they survived, would have innovated. However, the workers
laid off in recessions do not acquire skills that occur by learning on the job
and their previous skills may atrophy during the unemployment period.
The resulting decrease in skills could, depending on the severity and
duration of the recession, have a long-run detrimental impact on output.
This is one type of hysteresis effects.14
To conclude, the endogenous technical change theories imply that
governmental policies that affect the rate of technical change will affect
output per capita over the long run, and not merely over the short run.
Further, the former could prove to be much more significant for the long-
run prosperity of the nation. Government policies are even more critical
for countries close to an unstable SS equilibrium, where the policy effects
can push the economy to either side of the threshold.

733
 Extended Analysis Box 15.1: International Linkages and Growth
The recognition of the role of knowledge as a very important ingredient
for the growth of economies enhances the role which international trade
and capital flows play. The benefits of trade can be classified as:
(i) static, and
(ii) dynamic.
The static benefits can be defined as arising from the ability of the
country to benefit from comparative advantage in relative factor
intensities at its existing knowledge of production techniques. The
dynamic ones can be defined as arising from the new knowledge
acquired through trade and capital flows and other contacts. New
knowledge can include knowledge of the goods produced — and
processes used — in other countries but not yet in one’s own, with some
of this knowledge leading to innovations in the country’s production
techniques. The importation or imitation of existing techniques and
products promotes — or reduces the cost of— the search for new
ones.15
Education abroad, personal travel, and business contacts can
contribute to increases in the country’s knowledge of products and
techniques. Direct investment by foreigners in the home country and
agreements on the transfer of technology also contribute to the country’s
knowledge. Conversely, barriers to foreign education, travel, trade, and
investment stem this flow of knowledge.
Therefore, from the perspective of economic growth, international
flows, whether of products, capital or people, carry externalities in the
form of knowledge flows. These knowledge flows are likely to be of
greater benefit to those countries that lack the requisite knowledge than
to those which already have it. They will also be of greater potential
benefit for those that have the ability and production systems to use the
knowledge of other countries than for those that cannot do so. Hence,
developing economies are expected to be the greater beneficiaries of the
international flows of knowledge, provided that they possess the
openness, the absorptive capacity, and the environment — consisting of
the political, social, and economic systems — to utilize them. The extent
to which they actually benefit depends on their investment in this
enterprise.
Historically, the openness of the economy and extensive international
trade have been among the characteristics of rapidly growing
economies. Also, economies on the ‘periphery’ (geographically,

734
 culturally and/or in terms of links and relationships) of the already
industrialized economies tend to grow faster than more remote
countries. Presumably, those on the periphery have greater scope for
learning from the industrialized economies than those farther away, so
that we would expect the former to grow faster than the latter, ceteris
paribus. However, the invention of the Internet and reductions in travel,
transport, and communications costs have virtually eliminated the notion
of periphery in terms of geographical distances and made educational
levels and cultural factors more important in its definition.
Our discussion of the endogenous technical change also implies that
opening the domestic economy to the flow of foreign products and
foreign firms is often very advantageous for the economy. But it need
not always be to the home country’s advantage. If the expansion of
knowledge occurs mainly abroad while the destruction of productive
capacity — through the inability or unwillingness of domestic firms to
innovate at the rate attained by firms abroad — occurs at home, the
domestic economy will end up with lower output. This is especially
likely to occur in poorer countries with more limited resources for
investments in R&D and in education. Whether this effect will be
completely or more than offset by the inflows of investment and new
knowledge is a moot point. Therefore, the endogenous growth theory
does not imply that the liberalization of foreign trade will necessarily
benefit all countries equally; in fact, some countries can be net losers —
at least in the short term. For the poorer countries, even if they are not
net losers, the gains may be insufficient for their standard of living to
converge to those in the richer countries. To illustrate, the per capita
incomes of most countries in sub-Saharan Africa relative to the world
average income have declined in recent decades. Consequently, a
considerable dispersion of per capita incomes across countries, and
especially across continents, can continue to persist, even while there
may be convergence in the standards of living of other countries.

15.8 The Importance of the Inflation Rate and the

Quantity of Money for Growth
The relationship between the financial system and growth has two distinct
aspects:
(i) The effect of increases in the quantity of money and the inflation rate on
the growth rate of output.

735
(ii) The effect of the development of the financial structure on the growth
rate of output.
The short-run macroeconomic models and growth theory entail the
following two propositions for the relationship between money and
growth, and inflation and growth:
(i) The long-run growth of output is independent of money supply growth
and the quantity of money.
(ii) The long-run growth rate of output is independent of the rate of
inflation.
However, the short-run models and disequilibrium analysis (see Chapters 8
and 9) did show that higher money supply growth and inflation can in
some cases and for some time lead to higher output. As against this, high
inflation rates can be inimical to growth because they introduce distortions
and inefficiencies in the economy.16 Conversely, price stability enhances
economic certainty and promotes investment, which is conducive to
growth, so that growth rates can be negatively related to the inflation rate.
Consequently, on balance, the net empirical effects of money growth and
inflation on output growth need not deviate significantly from the above
two propositions.
The effect of changes in the quantity of money on growth can be
examined in one of two ways. It can be estimated directly by using the
data on the money supply or indirectly by examining the effect of inflation
on the growth rate of output. Among the empirical studies which report
that the growth rate is independent of the rate of monetary growth and the
inflation rate is Lucas (1996). Lucas plotted for 100 countries the
relationship between 30-year averages of the output growth rate and the
M2 growth rate. His finding was that the plots showed the former to be
independent of the latter.17 His conclusion was that, in the long run,
money is neutral for the output growth rate. This has been confirmed by
other studies. The exception to this sometimes occurs for low-inflation
countries, such as the OECD countries, which show a positive, though
weak, relationship between the growth rates of money (or the inflation
rate) and output.
To conclude, the consensus in economics is that, in general, the long-run
output growth is independent of the long-run inflation rate and the money
growth rate. However, it is not independent of innovations in and the
productivity of the financial sector.
Box 15.5: The Role of the Financial Sector in Growth: Some Conclusions
from Economic History

736
 One form of the evidence on the relationship between the financial
sector and growth comes from the economic history of the early stages
of Industrialization in countries noted for industrial innovation.
Cameron’s (1967) assessment on this was that “financial innovation,
after all, is not so very different from technical innovation. The former is
frequently necessary for realisation of the latter” (p. 12). This pattern
occurred during the Industrial Revolution in Britain in the 19th century.
The evolution and expansion of commercial banking was an essential
concomitant of the Industrial Revolution. Further, “while it is rare for
banks to finance directly a period of experimentation with a completely
novel production technique by a new, inexperienced businessman or
inventor … it is quite common for bankers to finance the expansion of
firms that have already introduced successful innovations, and also to
finance the adoption of the innovation by imitators.” (Cameron, 1967, p.
13).
The assessment by Joseph Schumpeter, an early proponent of the
importance of technological change (in current terminology, of
endogenous technical change) to development, was: “The essential
function of credit … consists in enabling the entrepreneur to withdraw
the producer’s goods which he needs from their previous employments,
by exercising a demand for them, and thereby to force the economic
system into new channels.” (Schumpeter, 1933, p. 106).
These quotes point to the critical role of the financial sector in the
innovation and restructuring processes which are elements of any
technical change, including the endogenous one, and of economic
development.18

15.9 The Role of the Financial Sector in Growth:

Recent Empirical Evidence
The financial system consists of retail banks (i.e., banks catering to
customers through deposits) and other financial institutions (such as
investment brokers and pension funds), bond and stock markets, and rules,
regulations and accounting practices to ensure the accuracy of financial
reports and prevention of fraud, etc. Firms use a combination of internal
funds — from retained profits and personal capital — and external
borrowing to finance their investment. Some industries, especially new
innovative ones, tend to be short of internal funds and depend relatively
more on external financing. This need for external financing varies among

737
firms and industries and depends on factors such as the initial project
scale, the gestation duration, internal generation through retained earnings
and the amounts needed for further investment, etc. Financial development
reduces the cost of external financing and, therefore, promotes the growth
of existing firms and industries dependent on such financing. It also
promotes the establishment of new firms and industries, which usually
account for a large part of new ideas, innovations, and breakthroughs.19
Financial development also reduces financial market imperfections, which
usually favor the internal financing and the growth of existing firms
relative to that of new firms.20 Therefore, the lower cost and easier access
to external finance provide a mechanism through which financial
development influences the establishment and growth of innovating firms,
and encourage the change and growth in the economy. Hence, endogenous
technical change tends to occur faster in countries with more developed
financial sectors.
As a major sector whose services are needed and used by all the other
sectors in the economy but which, in turn, uses the inputs and knowledge
originating in other sectors, there is a symbiotic relationship between an
efficient and competitive financial sector and the rest of the economy. This
makes it difficult to assign causality on the questions of whether the
efficiency of the financial sector emanates from the other sectors of the
economy versus the contribution that an efficient financial sector makes to
the efficiency of the latter. It also creates considerable differences of
opinion in the profession on the dynamic contribution that the expansion
of an already efficient financial sector can make to the growth —
especially in the steady state — of the economy. Empirical evidence
indicates that financial development does contribute to higher per capita
growth rates, more so through the external financing of firms and
productivity growth than through just through facilitating capital
accumulation.
In assessing the contribution of the financial system to the overall growth
of the economy, the following points should be noted.
(i) The output of the financial sector is itself part of the GDP of the
economy. Therefore, if the productivity of the financial sector grows at
a faster rate, GDP will also grow at a faster rate.
(ii) We need to differentiate between the different components of the
financial sector. Among these, commercial banks provide different
services from those provided by investment banks, investment brokers
and stock markets, so that each can have independent affects on
economic growth. More broadly, the banking system is only a subset of

738
 a developed financial sector whose other segments are also vital to
growth in the modern economy.
(iii) The contribution of the financial sector to growth rates is at least
partly through the dynamics of the economy’s industrial structure,
especially in the external financing of new firms and new products.
Adequate access to external financing is important for the financing of
innovation by small firms and plays an important role in endogenous
technical change.
(iv) Underdevelopment or breakdown of the credit structure of the
financial sector reduces the flow of external funds to firms and reduces
their production ability and the growth rate of the economy.

15.10 The Miracle of Economic Growth

Empirical studies indicate a large role for physical and human capital in
explaining the growth. However, even taking account of these and other
purely economic determinants of growth, there is still a very significant
unexplained residual in such studies. This raises the possibility that there
are also non-economic contributors to growth, such as the degree of
openness of the economy, culture, ethnic divisions, political stability or the
lack of it, natural resource endowments, chance and luck.
A word of caution is needed at this point. The precision of growth
models conveys the erroneous impression that growth is a smooth process
over time. In reality, it is highly uneven, even precarious. Part of this is
due to the uneven pattern of advances in knowledge and their
incorporation into production, which can be major sources of fluctuations
in the macroeconomy. Further, the shift of a country to a successful
development pattern — that is, going from relatively poor standards of
living to long periods of high growth rates — has often come as a surprise
and has enough fragility for the development process and its continuation
to be considered by many economists a ‘miracle ’. Growth experience
indicates that few development miracles are sustained over very long
periods, and definitely not without interruptions or crises.

15.10.1 The miracle of growth and financial crises in the

economic tigers
Explosive growth and the rapid transformation of the economy through
new technologies and new firms and industries require the rapid expansion
of the sources of external financing (i.e., through loans, bonds, and

739
equities). The unevenness, speculative nature and fragility of the process
of invention and innovation, as well as of the nature of financial
intermediation, make the economy prone to severe financial crises. The
historical experience of the miracles of growth is littered with periodic
financial crises followed by economic ones.
The major financial and economic crises of the 1990s afflicted the very
nations that had been undergoing the growth miracle — the ‘Asian tigers’
such as Thailand, Korea, Malaysia, Indonesia, etc. The crises in the
American and European stock markets in 2000–2002 affected most
strongly the very firms that were at the forefront of the new Internet and
telecommunications technologies. Most countries — though not all firms
— do manage to come out of such crises after some years and resume their
rapid growth.

15.10.2 The economic crisis of 2007–2010 in the USA

Major elements of the financial crisis in the USA during 2007–2009 were:
• Innovations of new financial products, such as derivatives, and new types
of financial firms, such as hedge funds, in the late 1990s. These allowed
layers of financial assets to be built on mortgage payments on houses, a
real asset.
• Many mortgages had been given to persons who could not meet the
mortgage payments out of their incomes but relied on rapidly rising
house prices to justify the purchase.
• House prices rose rapidly during 200–2006, with a large positive bubble
in them by 2006. They began to stabilize in 2006, and then dropped in
2007 as the bubble collapsed.
• The layering of financial assets, with asset-based corporate paper as one
of the ultimate ones, ‘hid’ the real riskiness of the financial assets, which
led to their being perceived as riskless assets. Given this perception, they
came to be held by banks, investment banks, hedge funds, pension funds,
and other investors.
Once the underlying house prices started to fall in 2007, there were
extensive mortgage defaults, which implied that the promised returns on
mortgage-based financial assets could not be met, leading to either defaults
or a drastic decrease in the value of assets based on mortgage payments,
which led to the insolvency of numerous financial institutions which had
issued them or held them. This insolvency or near-insolvency led to a
credit crisis in which banks drastically cut back on their lending to other
banks and production firms. This, in turn, led to an economy-wide

740
decrease in production and employment (a negative supply shock), as well
as to loss of consumer and business confidence, which led to reductions in
consumer and investment expenditures (a negative demand shock). The
outcome of these negative shocks was a severe recession.
Since the USA is the world’s largest economy, its financial and
economic crises can easily spread to other countries and become
worldwide, as happened with the spread of the crisis in the USA to the rest
of the world.
Extended Analysis Box 15.2: The Economic, Social, and Political
Environment for Growth: Markets,
Competition, Capitalism, and Entrepreneurship
We have already emphasized that growth economics needs to go outside
the narrow confines of growth modeling to examine the environments
that propel and encourage growth. The hallmark of such environments
has to be their fertilization of change: change in products, markets,
machines, methods of organization, workers’ knowledge and skills, etc.
The base from which the change starts is important in determining the
nature and extent of the change. Further, institutions and practices that
encourage change are important, as are the incentives to profit from
making changes. These need to promote experimentation, invention and
innovation. These arguments make the nature of education, society,
economy, and polity significant in explaining why some economies
grow fast while others do not.
Adam Smith, one of the ‘founders’ of economics in the late 18th
century, pointed out that the expansion of the size of the market,
specialization, and competition were major forces for the expansion of
production. A larger market permits specialization of labor and
machines, which increases their marginal productivity and lowers the
cost of production, resulting in more sales and more production.
Markets, therefore, are a mechanism that encourages growth.
Some economists consider the market economy — and its dynamics
— to be the core development that separates economies without
significant growth from those with growth. However, this is an
oversimplification. Markets can grow and specialization can proceed
even under a virtually static technology, as occurred in many areas in
many periods in history, without producing major technological
breakthroughs and their manifestation in the industrial revolution. Adam
Smith was writing against a background of technical change already
occurring in Britain, but not yet in most other countries, so that its better
technologies permitted lower costs, increased sales, and expansion into

741
other markets. Both technical change and the growth of markets tended
to occur more or less together over time, each promoting the other.
Technical change and the growth of markets have occurred
concurrently over the last two centuries, so that GDP growth and the
size of the market were closely correlated. This evidence raises several
interesting questions. Did one produce the other? Which caused the
other or were both caused by a third factor? A somewhat similar and
topical question on the economic developments of the last two decades
is whether the globalization of markets produced the Internet and its
expansion or whether the Internet caused, or merely permitted,
globalization to proceed as forcefully as it has done in the last decade.
The importance of risk taking and incentives, and of the
entrepreneurial spirit
Since invention and innovation by their nature require departures from
past patterns of thought and production methods, as well as products,
they tend to involve a high risk of failure and wasted effort and
resources. Incentives in the form of profits provided by the capitalist
system (especially if strengthened by patents and copyright protection)
act as a reward for such risk taking and promote innovations and
innovations.
However, some economists argue that the entrepreneurial spirit,
especially the spirit that underlies the desire to set up new enterprises, is
also essential to such activity. Societies and educational patterns that
encourage such a spirit tend to do better at achieving higher rates of
inventions and innovations over time. By comparison, traditional and
conservative societies that encourage established patterns of behavior
and pursuits and bureaucratic ones usually tend to do less well in this
respect.
The entrepreneurial spirit is usually strongest in individuals and new
small businesses than in old and very large firms. Some of these new
small firms originate with a new idea, product or process, which is
initially untried and untested in the market place, so that its owners find
it difficult to raise external capital from banks and stock and bond
markets. Their financing depends the owner’s own resources and
amounts borrowed from friends and relatives, followed, as they become
somewhat better established, by loans from financial intermediaries
providing venture capital. Failure is common among new small firms.
Education and corruption
Among the myriad issues that the environment of growth raises, two
interesting issues for discussion relate to education and corruption. On
education, should poor countries (with very limited budgets to devote to

742
 education) concentrate on providing a basic elementary level education
to a high proportion of its citizens? Or, should they limit this proportion
so as to devote some of their meagre resources to producing a cadre of
highly educated persons, who possess the knowledge that will permit
understanding and adoption of the newer technologies in the world?
Which option will promote higher growth? Economics does not provide
a definite answer to this question.
On corruption, incomes in many countries are low and corruption is
rampant. Corruption can function as a lubricant for the economy by
allowing the economic agents to get around regulations that bind and
limit the economic activity and its expansion. But they also reduce the
returns to private enterprise and channel the energy of entrepreneurs a
nd the population into wasteful activities. Economists consider the latter
very significant and believe that corruption is a major net retardant of
growth.

15.11 Empirical Evidence on the Contributors to

Growth
Our theoretical analysis of this and the last chapter on growth show that:
(i) The Solow growth model implies that the higher the standard of living
— i.e., the higher the existing capital-labor ratio and output per worker
— the lower its growth rate. Endogenous growth theories imply that this
tendency could be offset by endogenous technical change. However, the
richest countries at the frontiers of knowledge have more difficulty
innovating than other countries. Barro (2001)21 reported a positive
effect of increasing GDP per capita on the growth rates of poor
countries (with GDP per capita below US$580 in 1985 prices) but a
significantly negative one for others. The effect was strongly negative
for the richest countries. This supports the Solow model, without
denying the importance of endogenous technical change to the growth
of the standards of living.
(ii) The higher are the growth rates of physical and human capital, the
higher is the growth rate of output per worker.
(iii) The greater is the amount of government consumption (i.e., excluding
investment by the government), the lower is the growth rate. This can
occur for two reasons: crowding out of private investment, which lowers
the growth of physical capital, and the inefficiency of the government-
owned enterprises.

743
(iv) For countries not at the forefront of technology, greater openness of
the economy increases the growth rate.
(v) Higher population growth reduces the growth rate of output per capita.
It does so by reducing the growth of capital per worker for two reasons:
more children per family tend to reduce the saving and there are more
workers among whom the capital has to be shared.
(vi) Both the growth of human capital and improvements in its quality
contribute positively to growth of output per capita. Some studies have
found that the quality of education — measured by scores on
international examinations, pupil-teacher ratios, etc. — is more
important for the growth of output per capita than its quantity (i.e., how
many more students educated at the existing quality of education).
Another empirical study found that the increasing superiority of the
USA during the 20th century was due to the earlier and more rapid
expansion of its high school system in the early 20th century, while high
school education was much less common in other countries at the time
or in the USA in the 19th century. This expansion and the pragmatic
nature of schooling had produced by the middle of the 20th century a
large difference between the educational stocks of the USA relative to
other nations, including the European ones.22
(vii) The largest contributor to the growth of output is the growth of
multifactor productivity.
(viii) Economic incentives provided by the capitalist system, risk-taking
behavior and the entrepreneurial spirit are very important to continued
high levels of inventions and innovations, and of new firms set up to
bring them to the market. The extent of the entrepreneurial spirit and
risk taking tends to be much higher among societies and economies
based on the capitalist system versus those organized along centralized,
bureaucratic ones. They can also depend on the educational system.
(ix) Patents and copyright protection increase the profits to be made from
the protected item. Most economists maintain that they increase
inventions and innovations over the amount that would have occurred in
their absence, while some economists dispute this conclusion on the
basis of empirical evidence.

15.11.1 Empirical evidence on growth in recent decades

The rate of economic growth differs among countries. However, the
developed industrialized economies seemed to possess a common pattern
of declining growth rates from the mid-1970s to the mid-1990s, followed
by an increase in growth rates after the mid-1990s. This was discussed in

744
Chapters 14 in the context of Table 14.1 for the USA. Table 15.1 below,
from Baily (2003), presents another set of growth estimates23 for the USA,
and highlights the relative contributions to growth of the computer and
Internet revolutions.

Table 15.1 Growth accounting and the IT revolution.24

This table shows that the annual rate of growth of output per hour was
2.9% during 1948–1973, 1.4% during 1973–1995 and 2.79% during 1995–
2000.25 The slowdown in growth during 1973–1995 relative to that during
1948–1973 was mainly due to the rate of decrease of multifactor
productivity (MFP) growth, which decreased from 1.9% annually to 0.4%.
The resurgence of growth after 1995 was also largely due to the recovery
in the MFP growth, which was in turn mainly attributable to the
development of the IT sector (including the computer one). This sector
contributed to growth through contributions to capital’s share and to MFP
growth. It accounted for 0.4% of capital’s contribution during 1973–1995,
followed by an increase by 0.59% (to 0.99%) during 1995–2000. It also
increased the rate of growth of MFP by 0.18%, so that the increase (as
between 1973–1995 and 1995–2000) in annual output growth due to IT
developments was 0.79%. This difference compares with labor’s total
contribution of 0.02% during 1973–1995 and 0.06% during 1995–2000.
Clearly, the IT revolution has been the largest contributor to the growth of
output per hour and hence to the growth of living standards during the past
decade.

15.12 Conclusions
Conclusions from endogenous growth theory
• The political, social, geographical, and other factors are important
determinates of a country’s standard of living and its growth rate.

745
• The essential elements of endogenous technical theories are that
technical change requires economic decisions on the acquisition of new
knowledge and that new knowledge created by a firm has positive
externalities. Positive externalities mean the return to a firm creating new
knowledge is less than the social one. Therefore, the invisible hand of
competition does not produce the optimal rate of R&D expenditures or
other investments leading to invention and innovation so that they will
not yield the optimal long-run growth rate. Governmental policies that
increase the acquisition and spread of new knowledge can increase this
growth rate while inappropriate policies can lower it.
• Both human capital and physical capital are the carriers of knowledge.
They are highly correlated with each other and with the country’s growth
rate.
• Solow’s model with exogenous technical change implies that an increase
in the saving rate can only increase the steady state output but not its SS
growth rate. By comparison, the endogenous growth theory implies that
the increase in the saving rate — and thus, investment and the rate of
increase in capital intensity — can influence the pace of innovation and
the steady state rate of growth of output. Hence, governmental policies
that affect the saving rate and the rates of return to physical and human
capital can have long-run growth effects.
• Inventions and innovations are normally a result of economic decisions
and require investments. As such, both the readiness of firms to invest in
these activities and of workers to invest in their education, and the
governmental support for these activities, play important roles in
endogenous growth theories. Macroeconomic policies that promote
physical and human capital growth increase the growth rate, and some
may do so even in the steady state.
• Patents and copyright protection increase the returns to inventions and
innovations, and can enhance the latter.
• Policies that reduce the returns to the production of new knowledge slow
technical change and the growth rate.
• In the case where there is an unstable steady state with a takeoff or
threshold capital/labor ratio, policies that push the economy beyond this
point become extremely important for the economic future of the country.
Similarly, capital inflows would increase the domestic capital intensity
and push up the country’s growth rate.
• Patents and other exclusionary devices, which close off access or make
access to the knowledge of production techniques more costly tend to
lower the growth rates of the countries denied such access.
• Endogenous growth theories imply that, under certain (but not all)

746
 conditions, relatively less advanced countries can catch up to the standard
of living of more advanced countries through higher growth rates. This
catching up occurs partly through copying the knowledge of the more
advanced countries. This ability to tap into the existing stock of
knowledge in the world is a major determinant of the growth rates of
developing countries. As they catch up to the lead countries, their growth
slows, with eventual convergence in growth rates.
• However, there could also occur divergence of growth rates among
countries: over time, the rich countries tend to get richer relative to some
poorer countries. Part of this expanding gap stems from the relative
inability of the poor countries to effectively tap into the world’s stock of
ever-expanding knowledge.

Conclusions on money, the financial sector, and growth

• Money allows specialization, trade, and expansion in the size of firms to

an extent that can never occur under barter and in economies with
rudimentary monetary systems. The usage of money is an essential
element in the modern scale and complexity of production and exchange.
• In addition, an efficient financial sector allows the accumulation and
more efficient allocation of savings to investment projects, and thereby
promotes the growth of capital in the economy.
• The empirical evidence shows that there are none, or very limited,
benefits for growth from inflation or the rapid growth of the money
supply.
• The empirical evidence usually shows a significant positive relationship
between the development of the financial sector and the economy’s
growth, though the direction of causality can go either way. Therefore,
the implication of the empirical findings for monetary policy is that, from
the long-run growth perspective, the central bank should support the
greater efficiency, development, and stability of the financial sector.
• The efficiency of the financial sector is also dependent, as is of the other
sectors, on regulations, technological change, and other aspects of the
economic environment. Here again, the central bank can make important
contributions.

KEY CONCEPTS
Exogenous technical change
Innovations and inventions
Endogenous technical change

747
Endogenous growth theory
Threshold capital-labor ratio
Human capital
Externalities generated by the creation of new knowledge
International flows of knowledge
Globalization
Patents and intellectual property protection
Creative destruction
Risk-taking and entrepreneurship, and
Financial development.

SUMMARY OF CRITICAL CONCLUSIONS

• The endogenous growth theories emphasize the roles of inventions and
innovations and their production. They introduce knowledge, whether in
the form of human capital or embodied in physical capital, as an
additional input in the production of final goods. The production of new
knowledge is itself subject to diminishing marginal productivity for the
firm producing it, so that its profit maximizing production is limited.
However, such knowledge, once created, has an externality so that the
social/macroeconomic return to the production of new knowledge is
greater than the private one.
• The new knowledge created by others requires absorptive capacity,
investment, and effort, so that while some firms and countries benefit
from it, others do not do so.
• The endogenous growth theories imply that the long-run differences in
productivity among countries depend on their social, political and
economic environment, as well on the incentives to innovate and invest.
Government policies can make a difference to long-run growth,
promoting, or hindering it.
• The endogenous growth theories can explain the convergence of the
growth rates within groups of countries while allowing for the divergence
of growth rates between groups of countries.
• Endogenous growth theories explain why the return to capital in the
world economy need not decrease. They also provide a role for
governmental policies in the promotion of long-run growth.
• The usage of money is an invention with remarkable externalities for the
technology of commodity production and exchange: it increases the size
of markets, promotes specialization and trade, allows the rise of large
firms to take the advantage of economies of scale and shifts the
commodity production frontier drastically. It also increases the supply of

748
 labor to firms.
• In the long run, the growth rate of the economy is independent of the rate
of inflation and the growth rate of the quantity of money — that is,
money is neutral in the long-run growth of countries.
• The monetary sector in the modern economy is merely one component of
the financial sector. The appropriate focus of monetary growth theory
should be on the efficiency and structure of the financial sector.
• Innovations in the financial sector contribute to the growth rate of output
in the pre-SS stage and may also do so to the steady state rate.

REVIEW AND DISCUSSION QUESTIONS

1. What is meant by ‘endogenous growth’?
2. Which is more significant (a) for SS growth and (b) for pre-SS growth:
investment (without technical change) or innovation? Discuss.
3. What is meant by ‘human capital’? Is it schooling or new knowledge or
does it include both? What is the role of each one in growth? Discuss.
4. Discuss the relevance of capitalism versus totalitarianism (with a
centralized economy) to growth.
5. What types of fiscal policies can promote growth? What types can
hinder it? Give examples.
6. How do wars (a) reduce growth and (b) promote growth? Use diagrams
to support your answer and give some examples.
7. What is meant by the globalization of international trade? Discuss its
advantages and disadvantages for the less developed economies with
agriculture as the dominant economic sector.
8. What is a patent? In what ways do patents encourage and in what ways
do they discourage the growth and spread of new knowledge?
9. ‘It is vital for understanding the contribution of money to growth that
we distinguish between the quantity of money and the size and the
efficiency of the financial sector.’ Discuss.
10. LDCs have sometimes resorted to high money growth rates to finance
their development efforts in an attempt to push up their growth rates.
Why? Was it misguided in retrospect and, if so, why?
11. ‘Banking inefficiency and the excessive regulation of the banking
sector and interest rates in some of the LDCs have proved to be highly
detrimental to their growth rates.’ Discuss.
12. Define ‘hyperinflation’. How might it hinder growth?
13. In explaining the growth rate of output per capita during the past 50
years for the USA, what factors explain its decrease from the mid-1970s
to the mid-1990s, followed by its increase?

749
ADVANCED AND TECHNICAL QUESTIONS
T1. If the marginal product of physical capital is increasing for a given
firm, it can indefinitely increase its profits by increasing its physical
capital stock, so that (a) profit maximization under perfect competition
implies that the firm would use increasingly larger and larger amounts
of capital and (b) the firm would grow to meet the industry’s total
demand. This is clearly unrealistic. How does endogenous growth
theory accommodate the increasing marginal productivity of capital for
the economy while maintaining the assumption of competition among
firms?
T2. Is there a link between rapid growth through invention and innovation
and financial crises? Why? Discuss.
T3. ‘Given that saving is positively related to the rate of interest and that
an increase in the rate of inflation will increase the rate of interest, both
the saving rate and the growth rate of the economy will be positively
related to the inflation rate and the money growth rate.’ Discuss. [Hint:
modern classical economists believe that the saving rate depends on the
real interest rate.]

FOOTNOTES
1A public good is a good in which other economic agents can share with
others, without necessarily having to pay its full cost of production.
Examples of public goods include defence of the country, and a monetary
system. A pure public good is one in whose benefits each agent can share
equally, even without paying any of its cost.
2Note that sometimes the innovations are themselves major improvements
in technology and are therefore difficult to distinguish from inventions.
3Among such incentives are the protection of rights to the innovation
through patents and copyright protection.
4owever, given the limitations on these and their long-run dependence on
sporadic macroinventions, it cannot be taken for granted that they will
always cause continual SS growth in any given economy or in all
economies.
5The profit-maximizing amount is given by the equality of the marginal
cost and marginal revenue of new knowledge, treating it as an output of
the production process.
6While all firms could increase their profits if they were to collectively
increase their research and therefore the economy’s knowledge, such co-

750
operation or collusion is ruled out by the usual incentives to shirk or cheat
(i.e., not invest in R&D itself but take advantage of the results of R&D by
other firms) in the presence of externalities.
7Note that capital includes both physical and human capital.
8A competitive equilibrium is ensured by assuming that the production and
investment by a given firm are subject to diminishing private returns, even
though their social/macroeconomic returns — i.e private returns plus
externalities — may be constant or increasing.
9We have assumed the simplification that this applies to all industries
within each country. In reality, the technology (and capital intensity) of
industries varies across countries at about the same stage of development,
being more advanced (higher) for some industries in one country while
being less advanced (lower) for other industries, so that there is always
some scope for copying from other countries.
10This argument has been used in some explanations of the failure of
centralized communist economies such as those of the Soviet Union and
Eastern Europe prior to 1990 to generate economic growth rates
comparable to those of the capitalist economies of North America and
Western Europe, eventually leading to the collapse of communism in the
former.
11This danger is minimal for the already developed economies, but can be
acute for some developing economies.
12Romer, P. M. Increasing returns and long-run growth. Journal of
Political Economy, 94, October 1986, 1002–1037. See also Romer, P. M.
Endogenous technical change. Journal of Political Economy, 98, October
1990, S71–S102.
13The long-run implications for growth of the crises in their exchange
rates, stock markets, and economies generally during the latter half of the
1990s have yet to be determined. The short-term impact of these crises
was to cut down their growth rates considerably.
14Hysteresis occurs when a short-term deviation from the long-run
position of the economy has effects on the subsequent course of the
economy. In this case, the long-run position of the economy is not
independent of short-run deviations from it.
15Such importation may involve the payment of royalties to other
countries or loss of control of domestic firms to foreign
corporations.Often, initially, imitation results in poor quality copies of
foreign products, followed by quality improvements as production
experienceincreases over time.

751
16However, this effect is not considered to be large for stable rates of
inflation in the single digit range.
17Lucas further shows that the correlation for 110 countries with 30-year
averages of the data between the M2 growth and the rate of inflation is
0.95.
18As against these claims, a prominent contributor to growth theory has
claimed that economists ‘badly over-stress’ the role of the financial
system. Other economists have argued that the financial sector responds
passively to growth in the production sector.
19In some cases, as much as two-thirds of the total growth of output in the
economy comes from new firms, so that their number and development are
major elements of growth. The effect of financial development on the
growth rate of new firms is greater than that on existing firms.
20Financial development also promotes higher accounting standards,
which reduce the costs of obtaining the external finance.
21Barro, RJ. Human capital and growth. American Economic Review, 91,
May 2001, 12–17.
22Robert Barro’s study using data for about 100 countries reported
confirmation of most of these findings.
23These estimates use the growth-accounting method presented in Chapter
14.
24This table is based on Tables 1 and 2 of Martin Neil Baily,
“Distinguished Lecture on Economics in Government: The New Economy:
Post mortem or second wind? Journal of Economic Perspectives, 16,
Spring 2003, 2–22. In combining the data from these two tables, the data
used from Table 2 is that shown for the Economic Report. This table
covers the non-farm business sector.
25Since growth rates are compounded over time, a small sustained increase
in growth rates tends to represent large variations in output per capita after
some years.

752
 CHAPTER 16
Business Cycles, Crises, and The
International Transmission of
Economic Activity

Business cycles are periodic fluctuations in economic activity.

While the magnitudes of almost all economic variables fluctuate
over time, the most significant variables for defining the business
cycle are fluctuations in output and employment. The business
cycle consists of a trough, upturn, peak, and downturn — followed
by a trough and so on. In a growing economy, recessions tend to
have a much shorter duration — usually of about six quarters to
two years — than booms, which are often longer than six years.
Macroeconomics investigates the causes of business cycles and
whether monetary and fiscal policies can eliminate or moderate
business cycle fluctuations
Business cycles are periodic fluctuations in economic activity. To describe
a typical business cycle, we can start with its phase when economic
activity is increasing. This upturn/expansion in economic activity
culminates in a peak. The peak is followed by a downturn/contraction in
economic activity. The downturn reaches a bottom, which is called the
trough. The trough is a turning point for economic activity, which then
starts a new upturn, which becomes the beginning of a new cycle.
Therefore, the business cycle has four phases: upturn/expansion, peak,
downturn/contraction, and trough. Note that the description of the
business cycle could have been given starting with any one of these
phases. The peak and the trough are called the ‘turning points’ of
economic activity. The duration of a business cycle can be measured from
one peak to the next one or from a trough to the next one.
Business cycles can be defined in two different ways:
1. They can be defined in terms of absolute movements in economic
activity. To illustrate, under this definition, the downturn would be a
decrease in output and employment.

753
2. For a growing economy, it can be defined as the movement around the
trend level of economic activity. In this case, the downturn would be a
decline in the growth rates of output and employment from their trend
levels, even if there is no absolute decline in these variables.
Some economists choose the first one of these definitions, while others
choose the second one, which seems to be more appropriate for a growing
economy. Under this definition, the data would be detrended (i.e., the
actual value of the variable less its trend value is used) to arrive at the
cyclical changes in the variables.

Figure 16.1
Figure 16.1a illustrates the four stylized phases of a typical business
cycle in a growing economy and Figure 16.1b illustrates the business cycle
after the growth trend has been eliminated. Figures 16.1a and 16.1b use
(real) output as the proxy for the economic activity and plot it on the
vertical axis. Time is plotted on the horizontal axis. Figure 16.1a plots the
value of output over time as y. Since output rises over time, this figure
shows an upward trend in y. While the plot for y shows that there are

754
upward and downward deviations from this trend, the cyclical nature of
these deviations is not fully clear. This nature becomes more readily
apparent if the output data were detrended, i.e., the trend in output was
eliminated, so that the output deviations were around the ‘average’ level of
output.1 This trend level of output, designated by yT , is shown by the line
AB in Figure 16.1a. If the trend output level were subtracted from the
original data, we would get output deviations around the zero level. The
detrended data, i.e., for (y – yT), is indicated in Figure 16.1b by the line A
′B′ and shows the output levels if there had been no growth of output. A′B′
shows more clearly the four phases of the cycle in output.
Cyclical fluctuations occur in a large number of economic variables.
Economists therefore usually construct an index of economic activity,
which is a weighted composite of a select group of economic variables.
The cycle in an established Index of Economic Activity — usually
compiled and published by an economic agency — is referred to as the
reference cycle. The most important components of this index are real
GDP, employment, and the unemployment rate.
Though there are different ways of empirically defining the onset of
upturns and downturns, an oft-used empirical guidepost of a downturn is
when there occur two consecutive quarters of decline in the reference
index of economic activity. Requiring two consecutive quarters of decline
to signal, a downturn is to guard against the possibility that the decline
over one quarter was due to random, rather than cyclical, factors. While
this reduces the chance of misidentifying a random decline as a cyclical
one, it introduces a two quarter lag into the identification of when the
downturn does, in fact, start. This recognition lag has the effect of delaying
the pursuit of policies to counter the economy’s recessionary tendencies to
at least two quarters.
Business cycles are a symptom of industrial economies. While purely
agricultural economies do experience fluctuations in economic activity,
these are tied to the state of the harvest — which depends on the weather
conditions, etc. — and are not necessarily cyclical.
Variables that increase in the upturn of the reference cycle and decline in
the downturn are said to be procyclical. Those which fall during the upturn
and rise during the downturn are said to be countercyclical. Of the main
variables of interest, output and employment are procyclical, while the
unemployment rate is countercyclical.

16.1 Recessions and Booms in Economic Activity

755
Descriptions of the business cycle frequently use the terms ‘boom’ and
‘recession’. A boom describes the state of the economy when economic
activity is close to or above the trend2 level of output. The recession
describes the state of the economy when economic activity is significantly
below the trend level.
In a typical business cycle, the upturn eventually leads to the boom and
the downturn leads into the recession. Note that, under these definitions,
the terms ‘boom’ and ‘upturn’ are not synonymous; neither are ‘recession’
and ‘downturn’. In fact, given the definition of a recession as occurring
when output is below its trend level, the recession will usually include the
later part of the downturn, the trough and the early part of the upturn.
Correspondingly, the boom will include the later part of the upturn, the
peak, and the early part of the downturn. The dividing line between the
recession and the boom will be set by when the economy crosses the trend
line for output.
In order to empirically specify the periods during which the boom and
the recession occur, we need to know the trend level of real GDP. The
procedure for deriving it is explained later.

16.1.1 The popular statistical designation of a recession

As explained above, a recession can be defined as a decrease in output
below its full-employment level. If this level has a trend, one measure of
the economy being in a recession is that output is below its trend line. A
more commonly used measure requires a fall in output from its preceding
level. Since output fluctuates due to random factors, which could cause a
temporary decrease in output when there is no fundamental cause for the
fall, a popular practical definition of the start of a recession is when real
GDP falls for two quarters.
However, if we look beyond just GDP data, a recession is really a self-
reinforcing decline in overall economic activity, with falling output,
consumer expenditures, factory orders, industrial production, employment,
and other major indices of the performance of the economy, though
without a decline in them being due to purely random causes that are likely
to be reversed sooner or later. Often, the various indices begin to signal a
decline in the performance of the economy at different times: for example,
output may have declined for two quarters while employment is still
expanding, and some other indices indicate a static (rather than declining
or improving) economy.3 There are no standard methods of combining the
movements in the relevant indices to arrive at a clear indication of a
recession, so that whether the economy has really entered a recession or

756
not becomes a judgment call, so that there can be a great deal of
uncertainty, and disputes, about when and if the economy has entered a
recession.

16.1.2 Monetary and fiscal policies to combat a recession

There are two broad approaches on the issue of whether or not to actively
use monetary and fiscal policies to fight a recession or moderate a boom.
One of these is proposed by strict adherents of the modern classical school
(see Chapter 11) who view fluctuations in economic activity over the
business cycle as an essential part of a well-functioning economy, with
aggregate demand not having a significant impact on the real variables of
the economy. This view is embodied in the real business cycle theory,
presented later in this chapter. However, few central bankers,
governments, and economists view it as presenting the acceptable
approach to the pursuit of monetary and fiscal policies.
The alternative approach to business cycles rests on the view that
declines in aggregate demand can cause a recession and increases in it can
cause booms in economic activity. Further, even if the initial cause is not a
change in aggregate demand, an aggregate demand deficiency usually
emerges in a recession, which prolongs and worsens the recession, and
excess demand emerges in a boom, which prolongs and intensifies the
boom. This approach is usually accompanied by the view that serious
recessions and booms impose costs on the economy, e.g., recessions cause
output and wage losses while booms increase the inflation rate. Under this
approach to business cycles, the appropriate policy response is to smooth
out cyclical activity by pursuing expansionary monetary and fiscal policies
in recessions and restrictive ones in booms. Most central bankers,
governments, and economists adopt this view. However, both monetary
and fiscal policies are rough tools for managing the economy, with
variable degrees of impact, as well as variable lags (which for monetary
policy in developed countries often range from six months to two years)
for full impact, so that this approach does not advocate ‘fine-tuning the
economy to the extent of trying to eliminate all fluctuations in it. This is
neither feasible nor necessarily desirable. The following analysis expands
on this approach.
If a recession is due to a deficiency of aggregate demand or a demand
deficiency emerges during the recession, the demand deficiency can be
corrected by expansionary monetary and fiscal policies. An expansionary
monetary policy requires a reduction in interest rates and an increase in the
money supply, which increase aggregate demand (see Chapters 5 and 6).

757
Since most recessions have a demand deficiency, central banks usually
pursue expansionary monetary policies to fight the recession, as they did in
the economic recession of 2007-2010. Such an automatic response is built
into the Taylor rule (see Chapter 5) for setting the interest rates. Under this
rule, the central bank’s interest rate depends positively on the output gap
(i.e., current output less the full-employment one) and the deviation of
inflation from the central bank’s target inflation rate.
An expansionary fiscal policy is defined as an increase in the fiscal
deficit, which increases aggregate demand. The fiscal deficit acts as an
automatic demand-stabilizer if, without a change in tax rates and
government’s spending policies, it generates a deficit in recessions and a
surplus in business cycle booms. Setting tax rates and spending policies to
arrive at a balanced budget over the business cycle as a whole, while
producing fiscal deficits in recessions and surpluses in booms, can achieve
this result. Such a budgetary policy produces what is known as a cyclically
adjusted balanced budget. With given tax rates, as incomes fall in a
recession, tax revenues fall. Further, certain kinds of expenditures, such as
for unemployment insurance benefits and minimum income support
programs, increase in recessions. With tax revenues falling while fiscal
expenditures rise, the budget automatically goes into a deficit in a
recession, while generating a surplus in booms, so that its movement tends
to boost aggregate demand in recessions and reduce it in booms. Hence,
such an approach to budgetary policy makes the budget an automatic
stabilizer.
Note that each recession and boom has its own distinctive aspects, so
that following a set rule for all recessions and booms may not be the best
policy. Therefore, the monetary and fiscal authorities might and often do
deviate from a priori rules: for example, by pursuing a more vigorous
fiscal policy by lowering its tax rates and increasing its expenditures
during what they perceive as an exceptionally deeper and longer-lasting
recession. The recession in 2007-2008 in the U.S. economy provides an
example of such aggressive pursuit of monetary and fiscal policies. Mild
recessions are likely to evoke moderate expansionary policies, and very
mild ones may not elicit any explicit policy response.
It should also be noted that monetary and fiscal policies, whether
pursued together or independently, are not likely to totally prevent a
recession or avoid a boom for the following reasons.
i. Information on the actual course of the economy becomes available with
a lag, so that policies may not be pursued in time.
ii. There are further lags in the impact of policies on the economy.
iii. Monetary and fiscal policies tools may be inappropriate, such as for a

758
 recession caused by a negative supply shock to the economy, which
occurs when there is a rise in oil and other resource prices.
Therefore, the best hope for monetary and fiscal policies is that they
moderate the business cycle, even if they cannot eliminate it.
Fact Sheet 16.1: Output Gap in the United States, 1980-2008
In the USA, over the past 30 years, real GDP has followed a more or
less consistent pattern of growth, as indicated by its trend line, but not
without periods of boom and bust where GDP surpasses or falls short of
the trend level. The output gap, calculated as actual output minus trend
output, serves to capture this cyclicality. During the recession of the
early 1990s, the output gap was negative. This was followed by the
economic upturn of the late 1990s and the dotcom bust of the early
2000s. Finally, although an above-trend level of GDP was experienced
from 2003 to 2007 and was reflected in a decreasing output gap, the
financial and economic crisis starting in 2007 led to a fall in GDP in late
2008, which increased the output gap.

16.2 Business Cycles and the Growth Trend in Economic

Activity
Virtually all economies go through periods of long-term growth in output
and employment, so that there is a trend in economic activity. The shorter-
term cyclical fluctuations in economic activity are superimposed on this
trend. The pattern of cyclical fluctuations becomes clearer if the data are
adjusted to eliminate its growth trend. Box 16.1 explains the procedures
for detrending the data.

759
 Box 16.1: Eliminating the Trend from the Data
Most time series of variables have a trend. Adjusting the original data on
a variable to eliminate this trend clarifies the nature of the cyclical
fluctuation in the variable. The first step in doing so requires calculation
of the trend due to growth. The second step is to eliminate this trend
from the data. The third step is to plot the adjusted data with time on the
horizontal axis, and identify the four phases of the business cycle.
Deriving the trend rate of output
Economists use several different methods for calculating the trend level
of output. One simple and fairly easy-to-use method is to use the simple
average increase over a given number of quarters or over the cycle as the
proxy for the trend. Doing so over the cycle means taking the difference
between the amounts at two consecutive troughs (or peaks), dividing by
the number of quarters between them, and then adding this amount
quarter by quarter to the amount for the initial period.4 The data thus
derived can be used as the trend level of output. Subtracting the
increases due to the trend from the original data gives the detrended (i.e.,
without a trend) output levels. Plotting this derived series would show
the business cycle fluctuations around the average output level.
The unemployment rate as the proxy for the output gap
The variation of the unemployment rate around its trend can be used as a
simple proxy for the output gap, which is the variation of output around
its trend. Fact Sheet 10.4 in Chapter 10 has already shown both the trend
unemployment rate for the USA and the variation around it, and can be
reviewed at this point.

16.3 Stylized Facts of Business Cycles

The stylized facts (i.e., commonly observed patterns) on business cycles
are
i. Business cycles are national — and in some cases, international — in
scope.
ii. As mentioned above, there are four phases of the business cycle: boom,
downturn, trough, and upturn.
iii. There is co-movement between pairs of variables, with some variables
moving in the same direction while others move in an inverse one. To
illustrate, output and employment usually move in the same direction

760
 over the business cycle, while output and unemployment tend to move
in opposite directions.
iv. For a given variable, the general pattern of the movements is
persistence since an increase is usually followed by a further increase
(as happens during upturns) and a decline is usually followed by a
further decline (as happens during downturns).5 The exceptions to this
pattern of persistence occur only at the two turning points (peaks and
troughs). Persistence occurs during the downturns and the upturns.
v. Business cycles are recurrent (i.e., follow one another) but do not have
an identical pattern over time. Different cycles tend to have different
total durations, as well as different durations of upturns and downturns.
Further, the amplitudes of the variables differ over different cycles.
Therefore, each cycle is unique in some ways.
vi. Business cycles are different from seasonal fluctuations, which usually
affect some industries but not all or most. The latter occur within a year.
Business cycle movements are also different from short, erratic
movements in variables in that the former span several years and are
more widely diffused among the sectors of the economy.
vii. Downturns can be set off by a variety of causes, including negative
demand (from a fall in investment, consumer expenditure, exports,
decrease in the money supply, etc.) or supply shocks (a decrease in
productivity, increase in prices of raw materials, decrease in the supply
of credit, etc.).
viii. Compared with business cycles before 1939, cycles have become
longer in duration, with longer expansions, though with shorter
contractions. Before 1933, contractions were slightly shorter than
expansions; after 1945, expansions have become about three times
longer than contractions, which are on average now about one year.
Recessions are now shorter and shallower than they used to be before
1933.
ix. Economic activity in agriculture, in terms of output and employment, is
immune to business cycles, since it depends mainly on the weather.
Most other industries participate in business cycle activity, with durable
(consumer and producer) goods industries having much larger
amplitudes than non-durable ones. Firms’ inventories of finished goods
rise in contractions and fall in upturns. Average labor productivity (i.e.,
output per hour of workers’ time) rises in an upturn and declines in a
downturn. This is explained by variations in the effort level of workers
and labor hoarding by firms (see Okun’s rule in Chapter 8).
x. Very severe and long-lasting recessions are called depressions. The
Great Depression lasted from 1930 to 1939. It was truncated by the start

761
 of the World War II in 1939, but without the very large war
expenditures would most probably have continued for many more years.
xi. Prior to 1950, prices often used to fall in contractions; since 1950, they
rarely fall, but the rate of increase in prices (i.e., the inflation rate) tends
to fall in downturns, so that there is often some disinflation without its
becoming a fall in prices.
xii. Business cycle fluctuations in the USA, which is the largest economy
in the world, are usually transmitted, often with a lag, to other countries
through exports to the USA, which fall in contractions and rise in
expansions. Cross-country effects also occur through international
capital flows.

16.4 The Behavior of Variables Over the

Business Cycle

16.4.1 Procyclical, acyclical, and countercyclical

movements
Variables can be classified according to the direction of their movements.
A variable is said to be procyclical if it increases in the upturns and
declines in the downturns of the reference index of economic activity.
Examples of procyclical variables are output, employment, interest rates,
consumption, investment, stock market prices, etc. A given variable is said
to be countercyclical if it decreases in the upturn and increases in the
downturn of the index of economic activity. An example of a
countercyclical variable is the unemployment rate. Certain other variables
are acyclical, i.e., without a clear cyclical pattern. The real wage rate and
the real interest rate tend to be acyclical over many business cycles.

16.4.2 Leading, lagging, and coincident variables

Variables can also be classified according to the timing of their changes. A
variable is said to be coincident if the change in it occurs at the same time
(whether pro-cyclically or counter-cyclically) as the change in the index of
economic activity. Output, employment, unemployment, consumption, and
investment are normally coincident. A variable is said to be leading if the
change in it occurs somewhat earlier (whether pro-cyclically or counter-
cyclically) than the change in the index of economic activity. Net exports,
inventory investment, and stock market prices tend to be leading variables.

762
Such variables are useful as indicators for predicting the future changes in
the index of economic activity. In this role, they are known as leading
indicators. A composite index based on such variables is the Index of
Leading Indicators. This index turns down in advance of business cycle
peaks and turns up before business cycle troughs, so that it is very useful
for predicting the next turning point of economic activity in the economy.
Finally, if the change in a variable occurs after the change in the index of
economic activity, it is a lagging variable. The rate of inflation and the
interest rate often fall into this category.
Consumer sentiment and business confidence, which reflect consumers’
and firms’ expectations of their future income and profitability, have a
predominant leading pattern.

16.4.3 The volatility of variables over time

Variables can also be classified according to the degree of their volatility.
Volatility is measured by the extent of their variance over time. Among the
most volatile variables are inventory investment and the rate of growth of
the money supply. Industrial production, net exports, unemployment, and
interest rates are quite volatile but are not among the most volatile
variables. Consumption, the productivity of labor and the inflation rate are
even less volatile. To forecast the future course of the business cycle,
economists pay close attention to changes in those variables that are both
leading and highly volatile.

16.5 Business Cycles, the Full Employment

Assumption and the Classical and Keynesian
Paradigms
We have defined business cycles as fluctuations in economic activity (with
output as the proxy) and, for empirical purposes, illustrated them by
deviations around its trend level. What is the relationship between the
trend level and the full-employment (long-run equilibrium) levels of
economic activity? Macroeconomic theory provides three possibilities for
this relationship. Using output or the unemployment rate as a shorthand
measure of economic activity, these are:
1. The first approach to the relationship between the trend and full-
employments level of output assumes that the actual level of output is
always (and, therefore, in every period of time) the full-employment

763
 level. In this approach, the trend in output merely reflects (by
assumption) the time path of full-employment output. Consequently, the
deviations of actual output around its trend are not interpreted as
deviations from the full-employment level, but rather as the deviation of
full-employment output around its own trend. This case is shown in
Figure 16.2a by the curve marked (yf , y ), since this approach assumes
that y and yf are identical.
2. The second approach to the relationship between the trend and full-
employments level of output assumes that while the actual level of
output does include deviations from the full employment level (so that y
and yf are not identical), full employment is maintained on average over
each business cycle. Given the latter, the trend rate of output over each
business cycle represents the growth of full-employment output.
3. The third approach to the relationship between the trend and full-
employments level of output assumes that the full-employment is not
necessarily maintained on average over each business cycle but over a
much longer period (encompassing several business cycles) that
approximates the duration of the analytical concept of the long run. In
this case, the trend level of output over each business cycle does not
represent the growth of full-employment output. In order to find the
latter, the trend will need to be taken over the appropriate chronological
period corresponding to the long run. Assuming that the total period
shown in Figure 16.1a (presented earlier in this chapter) approximates
the long run, the curve marked AB is the trend over the whole period
and represents in this approach the path of full-employment output over
time.
Figure 16.2b (also presented in Fact Sheet 10.4 of Chapter 10) uses
unemployment (rather than output) as the proxy for economic activity and
illustrates approaches 2 and 3 by actual data for the USA. Three business
cycles are shown in this figure. Approach 2 to measuring full employment
identifies its time path with the trend line over each business cycle, and is
illustrated by the three dotted lines. Using this approach and comparing the
three dotted lines, the full-employment rate of unemployment rose over the
three successive cycles. Approach 3 to measuring full employment
identifies its time path with the trend line over the whole period (i.e., the
long-run trend line), as illustrated by the darker solid line. Under this
approach, actual unemployment was sometimes higher and sometimes
lower than the long-run levels during the first two cycles but consistently
higher in the third one.

764
 Figure 16.2
To shed some factual light on this issue, let us consider the Great
Depression through the 1930s. Unemployment throughout this decade in
virtually all industrial economies was several times higher than in the
preceding decade of the 1920s and the succeeding decades of the 1940s
and 1950s. It is hard to believe that full employment existed on average
over the 1930s or over any of its sub-periods, so that the trend line over the
1930s cannot plausibly be taken to represent the time path of full-
employment output. But the economy of the 1950s and 1960s did well
enough on average so as to be taken to have performed closer to the full-
employment scenario, so that its trend line can more plausibly be taken to
represent the time path of full-employment output. Consequently, we need
to use intuition and general knowledge of the economy to decide which of
the three approaches above is reasonable for the data period in question.
Further, the trend of output over several business cycles and a really long
period is more plausibly a better approximation to the time path of full-
employment output than the trend over one business cycle. Hence,

765
Approach 3 is the most plausible one in the light of actual experience.
The three approaches above lead to different theories of the business
cycle. Approach 1 is the most consistent one with the real business cycle
theory listed later in Section 16.7.1, which takes the view that business
cycle fluctuations are fluctuations in full-employment output. Approach 3
is the most consistent one with the general business cycle AD –AS theory
listed in Section 16.6. Under this theory, actual output can differ from full-
employment output, and the trend of output over any one business cycle is
not necessarily the time path of long-run output. On the validity of the
three approaches above and their related business cycle theories, our above
indicated preference for Approach 3 on the basis of actual experience
favors the general AD –AS theory over the real business cycle theory.
Box 16.2: Business Cycles and The Assumption of Full Employment
The preceding discussion on the relationship between the trend of actual
output and that of full-employment output raises several questions for
theory and empirical analyses to answer. Among these are:
• Is full-employment output always reached in each period?
• If full employment does not exist in every period, is the trend in output
(or economic activity) in the data over a given business cycle that of
full-employment output?
• If full employment does not exist on average even over the business
cycle, does it exist over a longer period?
• How is the full-employment level of output to be measured for actual
economies? In what period or over what time span does it occur in an
actual economy?
• Are the cyclical fluctuations ones in full-employment output or
deviations from it?
• Are recessions really periods of output below its full-employment
level? Are booms periods of output above its full-employment level?
Different answers to the above questions are given by different business
cycle theories. These answers are not merely of academic interest. As
will be shown later, they have strong implications for the pursuit of
monetary and fiscal policies to moderate the business cycle fluctuations.

16.6 The General AD –AS Theory of Business

Cycles
The general AD –AS theory claims that fluctuations in economic activity

766
can occur through the economy’s reactions to shocks initially coming from
either the demand or the supply side of the economy, or both of them.
The impact of a positive demand shock raises the level of aggregate
demand and shifts the AD curve to the right from AD0 to AD1, as shown in
the AD–AS Figure 16.3a. There are two distinct possibilities for the effects
of this shift.
(i) The economy moves away from the LRAS curve in the short run. To
illustrate this case, suppose that, in response to the increase in demand,
the economy moves from the point a to the point b along the SRAS
curve (see Chapter 8) in Figure 16.3a. In the short run, output increases,
as does the price level. However, this is a temporary effect. In the long
run, the economy reverts to the LRAS curve, so that the economy goes
from point b to the point c. Hence, output increases and then falls back
to its initial level — thus exhibiting a cyclical pattern — while the price
level continues to rise throughout.

Figure 16.3
(ii) Even in the short run, the economy continuously retains full
employment. In this case, as aggregate demand rises, the economy
moves from point a along the vertical LRAS curve to point c, so that
output does not change but only the price level rises (see Chapter 7).
The impact of a positive supply shock that raises the full-employment
level of output and shifts the LRAS curve to the right from LAS 0 to LAS 1,
is shown in the AD–AS Figure 16.3b. There are again two distinct
possibilities.
(iii) Aggregate demand does not increase, so that the economy still
maintains the AD 0 curve in Figure 16.3b. The economy moves from the
point a on the old LRAS 0 and AD 0 curves to the point d on the new
LRAS curve LRAS , and AD 0. Hence, output increases, while the price

767
 level falls.
(iv) The increase in aggregate supply, accompanied by an increase in
aggregate employment and wages and a stock market boom, causes
increases in investment and consumption (see Box 16.3). These result in
an increase in aggregate demand, so that the economy shifts to a new
AD 1 curve. In this case, the economy moves from the point a on the old
LRAS0 and AD 0 curves to the point e on the new curves LRAS 1 and AD
1. Hence, output increases, as does the price level. In the more general
case, depending on the relative strengths of the supply and demand
shifts, the price level may increase or may fall.
Therefore, in some cases, some kinds of positive demand shocks cause
output to first rise and then to fall, while raising prices (cases [i] and [ii]
above). Similarly, some kinds of negative demand shocks cause output to
first fall and then to rise, while reducing the prices. This pattern is at least
partly consistent with the positive relationship between output and
prices/inflation over the business cycle. However, positive supply shocks
cause output to rise consistently, while lowering prices/inflation (case [iii]
above). Negative supply shocks cause output to fall, while raising
prices/inflation. This pattern is not consistent with the positive
comovement between output and prices/inflation over the business cycle,
unless it is assumed that aggregate shocks cause aggregate demand shifts
in the same direction (case [iv] above). This does occur in some cases of
supply shocks (see Box 16.3 creating a stock market boom and increases
in investment and consumption), but not necessarily in all cases.
In the general case, both supply and demand shocks are likely to hit the
economy. Further, it is possible for a demand shock to create a supply
shock, and vice versa. An illustration of this possibility is given in Box
16.3.
To conclude, the use of the AD-AS analysis indicates the possibility that
economic fluctuations may be the result of supply shocks only, of demand
shocks only or of a combination of demand and supply shocks. Further, the
nature and intensity of the shocks of the different types, and of their mix,
are likely to differ over different periods, so that each business cycle is
likely to be unique in its causes, intensity and duration.
Note that the observed magnitudes of business cycles can be the
cumulative effects of a series of small shocks with overlapping effects, so
that a big shock is not needed for explaining the observed magnitudes of
most business cycle fluctuations. Further, the economy adjusts slowly to
shocks, so that the effects of shocks persist for some time, thereby

768
accounting for the persistence of positive and negative movements in
variables.
As against the preceding general AD–AS theory of business cycles, many
economists favor a narrower view: some claim that only the shocks on the
supply side of the economy are the relevant ones for explaining the
business cycles, while others claim that only the shocks on the demand
side of the economy are the relevant ones for explaining the business
cycles. Their views are discussed in the following sections on the classical
and Keynesian approaches to business cycles. Briefly, the proponents of
the modern classical school tend to emphasize the supply shocks and shifts
in the LRAS curve as the main cause of business cycles. Their views result
in the real business cycle theory in Section 16.7.1. As against this supply-
side approach, many Keynesians tend to emphasize the demand shocks
and shifts in the AD curve as the main cause of business cycles. Their
views are reflected in the multiplier-accelerator model of aggregate
demand in Section 16.8.1.
Fact Sheet 16.2 : The Structure of the AD–AS Model
The AD–AS model determines the interest rates and aggregate demand
jointly by the interaction of the commodities and financial markets. The
aggregate demand thus determined and the aggregate supply of
commodities interact to determine the real income and output, and the
price level.

16.6.1 The propagation mechanism

The propagation mechanism is the name given to the manner in which
shocks affect economic activity. To illustrate, an increase in investment by

769
a given amount causes the nominal value of output to increase through the
multiplier process, during which incomes and expenditures rise in a
sequence. This process takes time, with the increase in nominal
expenditures during a period followed by another, but smaller, increase in
the next period. This pattern creates persistence, as in the cyclical upturn,
of the change in expenditures, which, as was noted above, is a feature of
business cycles. Further, the increases in expenditures diminish with each
step of the multiplier, and eventually become insignificant. At some point
during the multiplier process set off by the original positive shock to
investment, other shocks also impact on aggregate expenditures. Some of
these shocks have positive effects and buttress the ongoing increases in
expenditures, while others have negative effects on expenditures. The
cumulative effect of the numerous shocks may lengthen the duration of the
increases in expenditures, thereby strengthening the persistence pattern, or,
at some point, cause the expenditures to decline — thereby causing a
downturn in expenditures. Because of the nature of the multiplier process,
which embodies persistence, an initial rise (fall) in expenditures would be
followed by a sequence of increases (declines) in expenditures, which are
likely to cause persistent increases (declines) in output, i.e., a boom
(downturn) in the economy.
Since the shocks tend to differ in magnitudes and sometimes in sign, the
resulting pattern of economic activity will tend to differ among the cycles
generated by them.
Box 16.3: An Illustration of Technical Change Creating Cycles through
Aggregate Demand Shifts
Consider the innovations in the data processing and communications
sectors in the 1980s and 1990s. These were clearly supply side shocks,
which gradually moved the LRAS curve to the right.6 But they also led
to a mania in the stock markets, which multiplied the stocks of the
companies in this field very many times, as well as leading to a general
exuberance that lifted the stock prices of companies in many other
fields. These stock market surges substantially increased the real wealth
(i.e., nominal market values of assets divided by the price level of
commodities) of households and led to increases in consumption. At the
same time, they placed larger amounts of funds, through new stock
offerings at grossly inflated prices, at the disposal of firms and created a
substantial increase in investment expenditures.7 These surges in
investment and consumption expenditures translated into increases in
aggregate demand. Thus, positive supply and demand increases became

770
 intertwined through the financial markets and were both essential to the
proper explanation of the long upturn of the 1990s. In retrospect (from
the perspective of 2003), for the study of the fluctuations in economic
activity, it appears that while the technical innovations had initiated the
shifts, the shift in aggregate supply was relatively much less significant
than in aggregate demand. Once the euphoria over the profitability of
the firms founded to implement the new technology started dissipating,
the mania in their stocks collapsed in 2001, as did information
technology investment, which led to a fall in aggregate demand. While
innovations continued to occur after 1999, the fall in aggregate demand
sent the economy into a downturn and then into the recession of 2001-
2002. Therefore, somewhat surprisingly, the major impact of technology
shocks on economic fluctuations seems to have occurred through
aggregate demand fluctuations, rather than because of fluctuations in the
rate of innovation and aggregate supply.

16.7 The Modern Classical Approach to Business

Cycles
Two different approaches to business cycles emerge from the modern
version of the classical paradigm, referred to as the modern classical
approach in Chapter 11. These are:
i. The first approach to business cycles emanating from the classical
paradigm uses the short-run macroeconomic model of Chapters 5 and 8,
with the Friedman and Lucas supply rules to produce the impact of
errors in price expectations on output. This approach allows the
economy to move along the SRAS curve in the short run for
unanticipated changes in aggregate demand or supply, while asserting
the existence of full employment over the long run, whose length is
taken to be the duration of the business cycle. This approach relies upon
the Friedman and Lucas supply rules (see Chapter 8) to distinguish
between the impact of anticipated and unanticipated changes in
aggregate demand. The former do not cause deviations in output from
the full-employment level. Unanticipated shifts in aggregate demand
create errors in price expectations and cause deviations in output from
the full-employment level. Similarly, unanticipated shifts in aggregate
supply can create such deviations. These unanticipated shifts in either
demand or supply set off deviations from the full-employment output
level. However, since expectational errors are self-correcting, the

771
 deviations decrease over time and the economy reverts to the full-
employment level.
ii. The second approach assumes that full employment is continuously
maintained in the economy. That is, each period’s output (level of
economic activity) is the full-employment one.
For (ii), the classical approach is represented by the long-run equilibrium
states of the AD–AS model. This representation ignores the transitory
short-run states generated in response to errors in price expectations as
being insignificant or not relevant, so that, in terms of the AD–AS
diagram, the economy always operates on the vertical LRAS curve and
never at other (non-LRAS) points on the positively sloping SRAS curve.
Along the LRAS curve, the values of the real variables — such as output,
employment, real interest rate, real wage rate, etc. — are independent of
the level of aggregate demand (see Chapter 7). Therefore, the fluctuations
or shocks to the demand-side variables (such as investment, consumption,
money supply, etc.) do not cause cycles in real economic activity. Only
shifts in the LRAS curve, i.e., in aggregate supply, can do so. A rightward
shift of this curve creates an upturn, while a leftward shift creates a
downturn.
The LRAS curve shifts in response to shifts in the supply structure of the
economy. The two main sources of these shifts are those in labor supply
and labor productivity. The latter occurs because of technical change (see
Chapters 14 and 15) — which shifts the production function — and
changes in the prices of raw materials (such as oil) and intermediate goods.
Further, in this (second) classical approach to business cycles, since the
economy continuously maintains full employment, shifts (whether
anticipated or unanticipated) in aggregate demand do not make the
economy change its level of output. Hence, according to this approach, the
cyclical fluctuations in economic activity are to be solely explained by
fluctuations of the LRAS curve, while shifts in the AD curve do not cause
business cycle fluctuations.
The business cycle theory that embodies this (second) approach is called
the real business cycle (RBC) theory, presented in the next section.
Note that under both of the classical approaches to business cycles,
anticipated shifts in aggregate demand do not cause deviations of output
from full employment,8 so that they cannot be a source of business cycle
fluctuations. The fundamental difference between these two business cycle
approaches emanating from the classical paradigm is whether the short-run
deviations are significant or are so short-lived (until expectations adjust)
that their importance can be discounted in explaining the recurrent cycles

772
in economic activity. The first approach above allows a significant impact
for a significant duration of time, while the second one denies such an
impact. The policy implications of these two classical approaches are quite
different. Under the second approach, there is no role — and, arguably, no
need — for the demand-management monetary and fiscal policies, so that
they should not be used in a futile attempt to moderate the business cycle.
Under the first approach, there would be a role for such policies if the
policy makers can anticipate the demand and supply shifts that the public
cannot. That is, the monetary and fiscal policy makers can pursue policies
to moderate the business cycle fluctuations if they possess ‘superior
information’ (i.e., more than the public does), can act fast enough, and the
policies have a fast enough impact.9 If these conditions are not met, then
the monetary and fiscal policies should not be pursued.

16.7.1 Real business cycle theory

The dominant classical approach to business cycles is the RBC model,
which originated in the 1970s. It asserts that the economy operates
continuously in long-run equilibrium — that is, along the vertical LRAS
curve — and oscillates in response to real shocks. These shocks emanate
from the production technology/function, the labor market or the
commodity market,10 but not from the money and bond markets.
The RBC theory attributes the most important cause of the real shocks to
be shifts in the production technology, primarily due to the uneven (and
lumpy) pace of technical change, or in the prices of critical ancillary inputs
such as energy. These change labor productivity (i.e., the average product
of labor). Most booms are due to positive productivity shocks, which shift
the LRAS curve to the right and lead to higher output and employment.
Recessions are usually due to unfavorable productivity shocks or increases
in energy prices, which shift the LRAS curve to the left and reduce output
and employment. Some presentations of the RBC models also include the
effect of technological progress on investment and argue that since
technological progress is uneven, it causes fluctuations in investment
spending which cause fluctuations in aggregate demand. The shifts in
technology increase labor productivity in the booms and reduce it in the
recessions. Therefore, labor productivity is procyclical, as observed in the
data.
Since real wages equal labor productivity, which rises in booms and falls
in recessions, the RBC models also include the response of labor supply to
the increase in real wages in booms and decline in recessions. Workers
decide in recessions to cut back on their labor supply when real wages are

773
lower and substitute more work for them during the following boom when
real wages will be higher. As a consequence, unemployment rates will
seem to be higher in recessions than in booms.
Some RBC models consider shifts in the labor supply curve between
booms and recessions as another source of business fluctuations. Note that
shifts in labor supply are also shifts in the real sector of the economy. An
increase in the labor supply would produce a rightward shift of the LRAS
curve and a decrease in labor supply would produce a leftward shift.
Therefore, upturns can conceivably be explained as being due to a
significant increase in workers’ desire to work (i.e., a decrease in their
demand for leisure). Conversely, a significant increase in the demand for
leisure would cause a downturn. Therefore, economic cycles in
employment and output would occur if there were significant fluctuations
in workers’ demand function for leisure. However, many economists doubt
if there are significant cyclical fluctuations in the leisure demand function
and discount the empirical relevance of the actual shifts in labor supply as
providing a satisfactory explanation of the observed magnitudes of the
cycles in economic activity.
Real shocks can also arise in the commodity market, for instance,
because of the volatility of investments and net exports, and changes in
government expenditures and the fiscal deficit.11 However, in the context
of continuous long-run equilibrium, deficits and changes in investment and
net exports cannot change the economy’s output and employment,12 which
are major components of the index of economic activity. Further, shifts in
monetary policy (i.e., in the money supply or central bank induced changes
in the interest rate) only produce changes in prices and the nominal values
of the variables, but do not change the output and other real variables.
Therefore, in the RBC theory, shifts in the IS curve and the IRT or LM
curve due to exogenous shocks to the commodity,13 money and bond
markets are not considered to be relevant to the explanation of business
cycle fluctuations. We have also discounted the empirical importance of
shifts in the demand for leisure and, therefore, of the labor supply curve, as
providing an adequate explanation of the observed business cycles. This
leaves productivity shocks as the predominant exogenous source of the
observed business cycle fluctuations in RBC theories.
The RBC theories explain especially well the following observed
elements of business cycles in the economy.
i. Employment and its productivity are positively related and procyclical.
In the RBC theories, both rise in response to positive productivity
shocks and fall in response to negative ones.

774
ii. There is a high degree of positive correlation in the timing of the
business cycle turning points, though with some leads and lags, among
countries. In the RBC theory, productivity shocks are the main source of
business cycles and tend to hit the industrialized countries at about the
same time, thereby causing this correlation.
iii. The real wage rate is procyclical. Given profit maximization by firms,
the wage is determined by the marginal productivity of labor. In the
RBC theory, labor productivity is procyclical since cycles are caused by
productivity shifts, and this produces the procyclical fluctuations in
wages.
While the RBC theories can explain the above important characteristics
of the observed business cycles, they do not satisfactorily explain several
other characteristics. Among these are
i. The observed rate of inflation is procyclical: it increases in the booms
and decreases in the recessions. According to the RBC theories, as the
LRAS curve shifts to the right, output increases while the price level
falls. Therefore, the prediction of the RBC theory is that the price level
and the inflation rate would decrease as output rises in the booms, while
the price level and inflation would increase in the recessions, in which
output falls.14 This is contrary to the observed pattern of inflation over
business cycles.
ii. In recessions, unemployment rises due to a rise in layoffs while quits
fall. This suggests that the number of jobs falls in the recession, with the
result that some workers have been laid off by firms, even though they
would have liked to stay employed. Further, the laid-off worker cannot
find other jobs, since the number of jobs in the economy as a whole has
fallen. Therefore, there exists involuntary unemployment in
recessions.15 But, according to the RBC theories, there is no involuntary
unemployment at any time. The RBC theories explain the increase in
unemployment in recessions as the result of an increase in the natural
rate of unemployment: this rate rises because workers shift toward an
increased preference for leisure or reflects a rise in voluntary
unemployment at the lower marginal productivity and wage rates during
the recession.16 Many economists dispute this and instead claim that at
least part of the observed unemployment in most recessions is
involuntary unemployment, whether or not it occurs in other phases of
the business cycle. The real business theories do not satisfactorily
explain the rise in layoffs, the fall in quits, and the resulting increase in
unemployment during recessions.
iii. Real business cycle theories fail to explain the recessions caused by a

775
 credit crisis or the bursting of a bubble in asset prices (see Chapter 2).
They cannot explain the mid-1990s economic crises and recessions in
many East Asian countries or that in the USA in 2007–2010. Clearly, no
exogenous adverse shift in technology or in workers’ preferences for
work versus leisure occurred in these cases.

16.7.2 Policy implications of the RBC theories

Chapter 7 showed that monetary and fiscal policies could not change the
real values of the variables in the long-run equilibrium of the AD–AS
model. Since RBC theories assume that the economy always stays in long-
run equilibrium (except possibly for very minor and transient departures),
these theories imply that monetary and fiscal policies will not be effective
in eliminating or moderating the cyclical fluctuations in the real variables
such as output and employment. Therefore, according to the RBC theory,
there is no justification for pursuing these policies to moderate the
business cycles. Such policies could only change the inflation rate and the
nominal, but not the real, values of the variables.
The underlying paradigm of the RBC theories is the classical one. As
explained in Chapter 11, this paradigm asserts that the economy behaves
optimally without interference by the government or the central bank.
Consistent with this, the RBC theory claims that the economy handles
disturbances in an optimal manner and does not recommend the pursuit of
monetary and fiscal policies to moderate the business cycles. This is so
even if unemployment rises substantially during a recession. However, few
governments or the central banks are willing to following this policy
recommendation during deep recessions. Nor did they do so in alleviating
the worldwide recession that occurred in 2007–2010.

16.8 Keynesian Explanations of the Business

Cycle
Keynesians tend to view the shifts in aggregate demand, rather than in
aggregate supply, as the major initiator, though not the sole one, of
business cycles. In fact, several components of aggregate demand are
highly volatile. These include investment, exports, growth rates of money
demand, supply, etc. Of these, investment depends on business confidence
and the performance of stock markets, as does money demand. Even
consumer expenditures are subject to volatility, since they depend on
consumer confidence and household wealth (see Chapter 9), which

776
includes financial assets such as bonds and stocks. The prices of these
assets, determined in the bond and stock markets, are highly susceptible to
rumours and bouts of confidence and pessimism.17 Rapid surges in stock
prices increase households’ wealth, which increases consumer
expenditures. Rapid falls in stock prices decrease consumer expenditures.
In the AD –AS diagram, since the economy does not respond to shocks in
aggregate demand by instantly reverting to its long-run equilibrium, the
volatility in the several major components of aggregate demand causes
fluctuations in output.18 The departures of output from its long-run full-
employment level were explained in Chapter 8 as short-run (equilibrium)
departures and in Chapter 9 as disequilibrium ones. Chapter 8 explained
how fluctuations in aggregate demand produce short-run fluctuations in
output and employment due to the possibility of errors in expectations (the
Friedman and Lucas Supply rules), the costs of adjusting price (sticky
price theory), and the costs of adjusting employment (implicit contract
theory), Chapter 9 explained how fluctuations in aggregate demand
produce fluctuations in output and employment due to the faster responses
of firms and workers than markets to demand shifts, especially since they
form expectations on sales and employment prospects.
When aggregate demand increases, the departures from full employment
cause output to rise above its full-employment level, along with some rise
in the price level or the inflation rate. When aggregate demand falls, output
falls below this level, along with some fall in the price level or the
inflation rate. Employment follows a similar pattern as output, while
unemployment follows a contrary pattern. Further, since the economy will
eventually adjust to its long-run equilibrium, the effects of any increase or
decrease in demand will gradually peter out, thereby increasing the
probability of a turning point.
If the shifts in the components of aggregate demand are themselves
cyclical or if they are not cyclical but set off a cyclical response by the
economy, the aggregate demand fluctuations will cause cycles in economic
activity. In these cycles, output, employment, and inflation will be
procyclical. Since inflation is procyclical in the observed business cycles,
aggregate demand shifts offer a better explanation of the procyclicality of
inflation than do aggregate supply shifts.

16.8.1 The multiplier-accelerator model of aggregate

demand
While the initial shifts in aggregate demand can come from a variety of

777
sources, some Keynesians choose to focus on the volatility of investment
as the premier cause of these shifts. For this purpose, the changes in
investment are modeled by replacing the static investment of the IS model
of Chapters 4 to 6 by a dynamic one. The most favored dynamic
investment equation is specified mathematically as

where i is the real investment,19 y is the output, t refers to the current time
period and k (the Greek letter ‘kappa’) is a parameter. The justification for
this investment function is that firms need an appropriate amount of capital
to produce their desired output, which is determined by the expected future
demand for their products. If the output needed to satisfy the future
demand (i.e., in the following period t + 1) rises, firms have to invest in the
current period in order to increase their capital.20 In the above equation, k
represents the increase in investment (in addition to that needed to make
up for depreciation of the existing capital stock) due to a unit increase in
the desired output. The value of k is positive and is less than one. This
dynamic investment function is known as the ‘accelerator’. Combined with
the multiplier explained in Chapters 4 and 5, the resulting model is known
as the multiplier-accelerator model of aggregate demand.
The general form of relationship between aggregate demand and
investment was derived in Chapters 4 and 5 as:

where α is the multiplier whose value depends on the marginal propensity

to consume, the tax rate, and the marginal propensity to import (see
Chapter 5). The value of α normally exceeds unity. This equation implies
that

where we have switched to nominal values of aggregate demand and

investment by multiplying both sides of the equation by the price level P ,
and stated the relationship in terms of changes in the variables. This
equation states that the change in nominal aggregate demand equals the
multiplier α times the change in nominal investment. Since ΔY dt equals (Y
d d
t+1 – Y t), the multiplier relationship becomes

We can now see the link between the accelerator and the multiplier
equations. If the desired output increases due to a shock or growth of the
economy, the accelerator will induce firms to incur a positive level of net

778
(i.e., in addition to that required to cover depreciation of the existing
capital stock) investment. The multiplier will use this induced increase in
investment to increase nominal aggregate demand, which increases the
desired value of output, so that the accelerator further increases desired
investment, and the process continues on.
To incorporate other sources of the volatility of expenditures, the
preceding multiplier-accelerator model can be easily extended to include
the sources of volatility of exports, consumption, and other components of
aggregate demand. To see the simple intuitive working of the multiplier-
accelerator model as specified above, we need to start with a shock to the
desired level of output. This can come from a positive shock to one of the
components (such as to investment, exports, consumption, money demand
or supply, etc.) of aggregate demand or to an increase in the output (due to
a fall in their cost of production) that firms wish to supply. To
accommodate this increase in the desired output level, firms need to build
up their capital stock, which causes an increase in their investment. For the
given increase in the desired output, the increase in investment is
determined by the accelerator function. This increase in investment will
expand aggregate demand by a multiple, as specified by the multiplier.

16.8.2 Limitations of the validity of the

multiplier-accelerator model
Note that the multiplier-accelerator model will give us the fluctuations in
aggregate demand. They cannot become corresponding fluctuation in
aggregate supply. Since the determination of actual output requires the
behavior of both demand and supply factors, we need to specify an
aggregate supply model in order to translate aggregate demand
fluctuations into output fluctuations. The early expositions by Keynesians
of the multiplier-accelerator model unrealistically assumed an infinitely
elastic supply of output such as might possibly exist when the economy is
in a deep recession. However, any such assumption is not valid for most
stages of the economy.
Therefore, we need to add at this point several considerations that are
often left out of the simple multiplier–accelerator models.
• One of these considerations is that, in most stages of the economy, any
increase in aggregate demand is likely to be split between some increase
in output and some increase in the price level. The proportion by which
each increases will depend on whether the economy is below or above its

779
 full-employment level, and how far away it is from this level.
• The second consideration relates to the functioning of the financial
markets, which affect the financing of new investment, exports and
imports, durable consumer goods, etc. Poorly developed financial
markets, as in the developing economies, or investor pessimism, which
sometimes occurs even in financially developed markets, do not allow
borrowers to raise at a reasonable or acceptable cost the capital needed
for new investment or other expenditures, and will reduce the value of the
coefficient k in the accelerator.
• A third consideration is the extent to which the investment is incurred on
domestically produced versus imported capital equipment. For example,
if all of the investment expenditure were on machines produced abroad,
as happens in many developing economies, the accelerator would
produce a quite limited impact on the domestic economy. Rather, its main
impact would be on aggregate demand in foreign economies. A similar
consideration applies to consumption expenditures. If these are mainly on
imported goods, the multiplier would be smaller than if they were on
domestically produced ones.
Extended Analysis Box 16.1: A Taxonomy of the General Reasons for
the Occurrence of Business Cycles
Business cycles can be viewed as the result of interaction between the
structure of the economy and shocks — that is, exogenous shifts —
hitting the economy. Therefore, at a very general level, we need to
understand the nature of the economy and the nature of the shocks.
In order to explain why the observed economic activity is cyclical, the
relevant nature of the economy can be classified into three extreme types
as follows.
(A) The economy generates cycles, even in the absence of shocks, by
virtue of its internal structure. An example of this would be ocean
waves, whose height changes whether a ship passes by or not.
However, shocks (such as the passage of ships) can accentuate these
fluctuations.
(B) The economy is stable on its own, but responds to shocks in a
cyclical manner, with the intensity of its response gradually
decreasing, so that if the shocks stop occurring, the economy would
sooner or later stop having cycles. Therefore, continual shocks are
needed to generate a continuing series of cycles, but the shocks
themselves need not be cyclical in nature. An intuitive example of this
is a drum whose surface — and the resulting sound — vibrates in
response to a hand hitting it. However, the vibrations gradually

780
 decrease and the drum surface becomes static — until the drummer
starts playing again.
(C) The economy responds to shocks but without fluctuations, i.e., a
single shock produces a shift in the values of the variables but without
cyclical fluctuations in them. In this case, for the observed
movements in economic activity to be cyclical, the shocks themselves
have to be cyclical in nature.
The nature of the shocks can be of two types:
(i) The shocks are random, or at least not cyclical.
(ii) The shocks are cyclical in nature.
Given these possible variations in the nature of the economy and the
shocks hitting it, we need to know the nature of the economy, the nature
and origins of shocks and how the economy responds to them. The IS–
LM/AD–AS model studied in earlier chapters essentially represents a
stable economy, which needs shocks to produce fluctuations. Therefore,
an economy with this model will only produce cycles if the shocks are
themselves cyclical in nature. Hence, this case belongs to categories (C)
and (ii) above. However, it is unlikely that the shocks have a cyclical
nature.
Keynesians economists claim that the economy is of type (B), while
the shocks may be cyclical or not. The theoretical innovation often
proposed for achieving such a structure of the economies modifies the
static investment function21 used in the IS–LM model to the dynamic
accelerator one explained above. This accelerator is part of some of the
Keynesian approaches to business cycles.
The RBC theories belong to category (B), while the individual shocks
may be cyclical or not. In this approach, the economy itself is a stable
one, so that the productivity shocks have to be cyclical in nature in order
to generate business cycles.
Note that a common element of virtually all business cycle models is
that time lags are introduced in the various functions. This is done to
produce persistence in the direction of movement.

16.9 International Linkages Among National

Business Cycles
Business cycle fluctuations in the industrialized open economies tend to
occur at about the same time, though with some leads and lags. The major

781
reasons for this positive correlation between cyclical fluctuations among
countries are:
i. Exports and imports link the commodity markets of countries. As
economic activity rises in an economy, its imports tend to rise. This
increases the exports of its trading partners, so that their aggregate
demand also rises.
ii. Shifts in technology occur because of inventions and innovations.
Knowledge of these is communicated across countries in a variety of
ways. One of these is just by observing and learning from the practices
of other countries. Another is by multinational corporations with plants
in several countries, or through technology transfer agreements.
iii. Capital by firms is raised on international capital markets, and the stock
and bond markets of countries, especially developed ones, are closely
linked. This creates a positive correlation in the ability of firms in
different countries to raise funds for investment and in the cost of these
funds.
iv. Households invest in international capital markets to a considerable
extent and the cross-country returns on their investments are positively
correlated. This globalization of investments and the impact of wealth
on consumption create a positive correlation between consumption
expenditures among countries.
v. There is a high correlation between the monetary and fiscal approaches
used by the policy makers in different countries. This is partly a result of
the evolution of economic theory and its dissemination across the
countries. In the past, periods of expansionary monetary and fiscal
policies have coincided among countries, as have periods in which the
policy makers pursued the objective of price stability.
Clearly, cyclical changes in the larger economies tend to exert greater
impact on other countries than those of smaller countries. Since the
world’s largest economy is that of the USA, it seems that ‘if the U.S.
economy sneezes, the world economy catches a cold’.

16.9.1 A world business cycle

Since a great many of the world’s larger economies tend to move in a
similar cyclical pattern, there is empirical evidence of a world business
cycle, which reflects many worldwide economic occurrences, especially
connecting the more open, developed market-oriented economies. For
these countries, these occurrences include the steady growth of the 1960s,
the recessions of the mid-1970s, early 1980s, early 1990s and early 2000s,

782
and the world recession of 2008–2010. The link among economies occurs
through exports and imports, capital flows and consumer and business
pessimism and optimism, which tend to be contagious across countries.
The impact of large economies, especially that of the USA, which is the
world’s largest, on other countries is larger than of smaller ones. This was
illustrated quite clearly in the world recession of 2008–2010, which started
with insolvency problems in U.S. housing and its financial sector, and
spread to other countries since the recession in the USA significantly
reduced other countries’ exports to the USA.
In addition to such common international experiences of cyclical
activity, individual countries will also have country-specific elements.
Such country-specific elements will be more dominant for the less open
economies. Less developed economies often fall into this category. Hence,
the economies of the more open and more developed market economies
tend to respond more to the world business cycle than those of the less
open or less developed economies.

16.10 Crises in Economic Activity

The phrase ‘crisis in economic activity’ refers to a sudden precipitous
decline in economic activity, rather than the gradual and more limited
decreases that occur at the start of the downturn of most business cycles.
Such a crisis occurred in 1930 at the start of the Great Depression and in
2008 in the USA. The economies of several East Asian countries also went
through a crisis in the mid-1990s, as illustrated from data for Malaysia and
Thailand in the following Fact Sheet. Common elements of such crises are
often a financial crisis, with several bank failures, decreases in the money
supply and especially in credit extended by the financial sectors to firms
and households, and exchange rate depreciations.
Fact Sheet 16.3 : The Asian Crisis of the Mid-1990s
Following a period of high economic growth, Thailand and Malaysia as
well as several other East-Asian countries, entered into a deep financial
crisis. When high, unsustainable asset prices fell and borrowers started
defaulting on their debt payments, a mass pull-out of lenders occurred,
which created a massive credit crunch. The massive withdrawal of
foreign investments in the countries created large deficits in capital
inflows and the balance of payments, and resulted in rapid exchange rate
depreciations. To stimulate domestic investment and consumption, and
through them, aggregate demand, countries lowered interest rates, but,

783
given the prevailing degree of pessimism, did not succeed fast enough in
restoring the investment and aggregate demand to their former levels. To
stop the precipitous depreciation of their currencies, Thailand took large
IMF loans and Malaysia pegged its exchange rate. All countries affected
underwent widespread restructuring. Most of the East-Asian countries,
including Thailand and Malaysia, eventually succeeded in stabilizing
their exchange rates and economies by 2001, though the latter ending up
with lower GDP growth rates.

16.11 Long-Term Effects of Recessions and

784
 Booms
There is definitely a link from growth to business cycles. One of the major
determinants of economic growth is technical change (see Chapters 14 and
15), whose variability over time is a major cause of the cyclical
fluctuations in economic activity, as emphasized in the RBC theory.
Conversely, the fluctuations over the business cycles can alter economic
growth. This affect can occur for a variety of reasons, of which the most
important are related to the accumulation of physical and human capital
and their fixity. The impact of the shorter-term cyclical activity on the
longer-term growth of the economy is an aspect of hysteresis — defined as
the impact of short-term economic events on the long-term output levels.
Some economists — especially among supporters of the classical
paradigm — believe that such effects are insignificant enough to be
ignored, while others — especially among the Keynesians — believe that
drawn-out recessions and booms can affect the long-term output level
and/or its growth rate. The factors that can cause hysteresis are discussed
in the following.

16.11.1 The accumulation of human capital

Recessions are periods in which jobs are relatively scarce. There are two
results of this scarcity: relatively more workers are unemployed and, of
those who get jobs, many workers have to accept jobs that do not fully
utilize their abilities and knowledge. The unemployed workers do not
acquire the skills they would have acquired through on-the-job learning.
Further, there is a deterioration of their prior skills through not getting the
opportunity to use them. Among those who find jobs, the mismatched
workers do not get to fully utilize their education and abilities, or learn the
skills they would have acquired in more appropriate jobs. As a
consequence, recessions leave behind relatively less human capital and
some mismatch of workers and jobs. The mismatch of workers may not be
fully corrected during the following boom because of the very long
number of years in which workers usually stay in their jobs. If the
recession was brief and mild, the subsequent correction is more likely to
occur than if the recession was drawn out and deep. In the latter case, the
future productivity of the affected workers would remain less because of
the recession’s reduction of human capital accumulation. However, note
that this reduction may not significantly affect the future unemployment
rate: as jobs increase after the end of the recession, the workers with the
lessened skills do get employed, though in less skilled jobs than otherwise.

785
Further, while the post-recession period will start with a smaller human
capital base, the further growth of human capital (from this smaller base)
need not be affected. This implies that while the smaller post-recession
human capital will mean a lower post-recession starting base level for
output per capita, the post-recession growth rate of output per capita need
not be affected: the latter will depend upon technical change and capital
accumulation in the post-recession period.
Long drawn-out booms in employment could have the opposite effect.
Workers with somewhat inadequate skills or capabilities get employed
and, through on-the-job learning and training, manage to acquire the skills
they would otherwise not have had the opportunity to learn. Other workers
get slotted into more skilled jobs than in the absence of the boom, and
acquire higher levels of human capital. Therefore, the boom will leave
behind higher skill levels in the economy and higher labor productivity. If
this happens, the post-boom period will start with a higher base level of
output per capita. However, this need not change the post-boom growth
rate of output per capita.
Looking at the recessions and booms occurring in sequence, as they do,
some of the affects of the recessions on the accumulation of human capital
will be reversed by the following boom, and vice versa. But there may
remain a net effect in either direction. The magnitude of this net effect is
likely to depend on the severity and duration of the recession relative to
the boom, and the adaptability of labor.

16.11.2 The accumulation of physical capital

Recessions reduce the level of investment and thereby reduce the
accumulation of physical capital. Booms enhance the level of investment
and thereby enhance the accumulation of physical capital. Therefore,
comparing recessions with periods of normal economic activity, recessions
leave behind comparatively lower physical capital per worker, so that
output per worker is smaller than otherwise at the end of the recession. If,
during the following boom, the firms do not make up for the recession’s
lower investment levels, the long-term capital/labor ratio will remain
lower. If the firms do make up for the recession’s lower investment levels,
the effects of the recession on the capital/labor ratio will be eliminated.
The latter scenario is the more likely one following most recessions,22 so
that, in general, the long-term ratio of physical capital to labor need not be
reduced by recessions. Therefore, as far as the accumulation of physical
capital is concerned, there should not, in general, be a long-term reduction
in output per capita or in its growth rate.

786
 Similar arguments apply to the impact of booms on investment, the
capital/labor ratio and output per capita.

16.11.3 Technical change

Machines embody production techniques, just as workers embody human
capital. However, there is an important difference between them. Firms
acquire machines when needed. As firms build up their capital stock in the
post-recession period, the new machines would embody the latest
technology, which could be an improvement on the technology that would
have been available in the preceding recession. Therefore, recessions do
not cause less advanced technologies to be embodied in the capital carried
over the long term.
As mentioned in Chapter 15, the early 20th century economist Joseph
Schumpeter had taken a holistic, institutionalist approach to economic
growth and business cycles. He had compared the evolution of firms over
time to that of trees in a forest. He argued that recessions are periods in
which the weakest firms — i.e., with older technology and less efficient —
are driven out of business, thereby releasing their labor force. The boom
brings in new and vigorous firms that possess a greater ability to innovate
and adopt newer technology, and are able to use labor more efficiently.
Therefore, the sequence of recessions and booms promotes technical
change and the more efficient utilization of labor, and, therefore, brings
about higher economic growth. Eliminating the recessions would eliminate
the pruning of relatively inefficient firms from the economy, and would
reduce economic growth.

16.11.4 The longer-term effects of cyclical fluctuations

through aggregate demand
Now looking at the aggregate demand side, the recession would have
reduced investment levels, which would have to correspondingly increase
in the post-recession period. Therefore, the volatility of investment will be
increased, as will be the amplitude of the fluctuations in aggregate
demand. However, this effect will occur over the business cycle itself and
need not carry over from one cycle to the next. Therefore, this is not a
longer-term effect.

16.11.5 The longer-term impact of recessions and booms

The preceding arguments imply that there can be a longer-term impact of

787
recessions and booms on labor productivity and the economy’s output.
This effect is likely to operate through the accumulation of human capital,
rather than of physical capital or the technology embodied in physical
capital. Further, this longer-term impact will be on the levels of labor
productivity and the economy’s output, rather than on their longer-term
growth rates. While the unemployment rate will be temporarily higher
(lower) than otherwise during the recession (boom), there may be no
longer-term effects of the recession (boom) on the unemployment rate.
Empirical evidence does bear out these implications. According to one
study,23 on average, recessions shift the economy’s output to a base level
lower by as much as 75% of the decline in output during the recessions.
However, the longer-term growth rate of output is not affected by the
recessions.

16.11.6 The implications of hysteresis for policy

The above empirical findings provide a prima facie case for policy makers
to attempt to moderate the recessions. However, the success of such
attempts depends on the cause of the business cycles and the lags in the
economy. Further, whether or not the policy makers choose to pursue any
stabilization policies depends on whether they believe in the Keynesian or
the classical paradigm.
If the business cycles are due to shocks to aggregate demand, the above
empirical findings and Keynesian theoretical analysis provides, in the
general case, support for the pursuit of monetary and fiscal policies to
moderate recessions by stabilizing aggregate demand. Looking through the
perspective of the classical paradigm, even if the business cycles were to
be the result of shocks to aggregate demand, their pursuit would not be
consistent with the general advocacy by the classical paradigm that the
policy makers should leave the economy alone. Further, according to this
paradigm, only unanticipated demand shocks can cause changes in real
economic activity, and policy makers can only offset such shocks if they
can anticipate these shocks but the public cannot. Furthermore, in the
dominant classical approach to business cycles — the RBC theory —
recessions are an appropriate and optimal response of the economy to
productivity shocks and monetary and fiscal policies would not be
effective in moderating the recessions. While Schumpeter’s trees-in-a-
forest approach to cycles is not a part of the classical paradigm, its
emphasis is also on technical change. It implies that recessions, at least up
to a certain duration and intensity, serve the useful purpose of getting rid
of firms that are not efficient enough, and promote more rapid innovation

788
in the economy. Eliminating the recessions would be detrimental to
economic growth and should not be attempted. Therefore, the classical
paradigm and Schumpeter’s analysis would not, in general, advocate the
pursuit of monetary and fiscal policies to moderate the recessions and
booms.
Leaving aside the confines of the paradigms, the pragmatic conclusion
on this issue has to be based on the proposition that policy makers should
attempt to stabilize the aggregate demand if doing so would moderate
business cycles, but not do so if it will not. This is more likely to occur if
the recessions are due to an aggregate demand deficiency or if this
deficiency emerges during the recession and makes it worse. Conversely,
if the (negative) shifts in aggregate supply are the cause of the recession
and/or if no demand deficiency emerges during the period, the conclusion
has to be that monetary and fiscal policies would not be advisable.24

16.12 Money, Credit and Business Cycles

16.12.1 Money supply and business cycles

The relationship between monetary policy and the output of commodities
has been extensively discussed in many of the earlier chapters, especially
Chapters 5–9. Shifts in the money supply, either because of shocks or the
pursuit of monetary policy, change aggregate demand in the same
direction. The classical and Keynesian paradigms propose two quite
different approaches to the impact of this change in demand on real output
and other components of real economic activity. In the classical paradigm,
only the unanticipated shifts in aggregate demand — and, therefore, only
unanticipated shifts in the money supply — can alter real economic
activity. Hence, unanticipated shifts in the money supply can conceivably
produce cyclical real economic activity, while the anticipated shifts in the
money supply cannot do so25 (see Chapters 7 and 8). In the Keynesian
paradigm, both anticipated and unanticipated shifts in the money supply
can alter aggregate demand and real economic activity, so that both types
of shifts can be a source of cycles in real economic activity.
Empirical evidence has not been able to reject definitely either of these
views. Some cycles, or parts of them, produce clear evidence in support of
the classical paradigm’s views, while other cycles provide clear evidence
in support of the Keynesian views. However, empirical studies do show
(though not always) that over many business cycles:

789
• There is a positive relationship between the monetary aggregate and
economic activity. Further, changes in the monetary aggregates in general
precede changes in economic activity (employment, both nominal and
real output, etc.), so that the former appear to have caused the latter.26
• The relationship between the monetary aggregates and nominal output is
one of the more stable ones among macroeconomic variables.
• The shifts in money supply — or money demand — can and sometimes
do occur independently of prior changes in real economic activity.
These empirical observations support the Keynesian contention that the
monetary sector has been a major initiator and contributor to the observed
business cycles. But the contention between the Keynesian and the
classical schools remains on the question of whether it is all of the change
in the money supply or only its unanticipated part that causes the
subsequent change in real economic activity.

16.12.2 Credit and business cycles

Firms need funds to finance their purchases of factors of production, such
as physical, capital, and labor. Some of these funds come from their own
savings and retained earnings (undistributed profits out of past profits) and
the rest from external sources. The latter include share issues, bond issues,
and loans. Financial markets and institutions are a major source of the
funds obtained through these channels. Some of the funds are long term,
e.g., when they are raised through shares and long term bonds, and some
are short term, as and when they are raised through short-term bonds and
loans. Firms especially rely on short-term sources of funds for the working
capital needed to pay their workers and finance their holdings of
inventories. While funds thus obtained are short term, there is continuous
replenishment of such funds through reissues of short-term bonds or
renewal of loans in normal conditions in the financial markets. Firms can
economise on their need for short-term working capital to some extent by
arranging trade credit, e.g., 30- to 90-day loans from manufacturers to
wholesalers and from wholesalers to retailers. The short-term sources of
funds, including trade credit, are grouped in one category as ‘credit’.
Well-functioning financial and economic systems in normal times allow
continuous roll-over of short-term bonds and loans, so that firms’
production plans do not have to be cut back because of a shortage of
credit. A credit crisis is said to occur if firms cannot obtain their usual
amounts of credit due to a shortage of funds or deterioration in their ability
to repay the funds. On the side of the providers of funds, a credit crisis

790
could be due to a failure of commercial banks, since they are the major
source of loans to firms, and of investment banks, which hold large
amounts of short-term bonds. These reduce the availability of credit in the
economy. On the side of the borrowers of funds, a deterioration in
economic conditions and a fall in the demand for firms’ products reduces
firms’ sales revenues and profits, thereby reducing their ability to re-pay
the funds in due time. This makes it more risky to lend to firms, so that
lenders reduce the amount of credit provided to firms. The resulting
decreases in the amount of working capital forces firms to cut back
production, by reducing their orders for inventories, their employment, and
even long-term investment. Similarly, cutbacks in credit to households,
e.g., for purchases of houses through mortgages, cars and household
appliances, etc., reduce the households’ expenditures on durable goods and
reduce overall consumption expenditures.
Hence, there are two effects of a crisis in credit markets. One is from the
aggregate demand side of the economy: decreases (through the relevant
multipliers) as a result of cutbacks in investment and consumption
expenditures (see Chapters 4 and 5). Therefore, a credit crisis decreases
aggregate demand in the economy which shifts the AD curve to the left in
the AD –AS diagram. The other effect is directly on aggregate supply,
which falls through the reduction in working capital and firms’
employment of labor and other inputs. This shifts the actual (not the long-
run one) aggregate supply curve AS to the left in the AD-AS diagram. The
overall impact on output and employment is thus greater than if there was
only a decline in aggregate demand (the case analyzed in Chapter 5) or in
aggregate supply only (in Chapters 7–9).
A credit crisis also has an impact on the money supply. Banks hold the
major part of the monetary base. Banks receive demand deposits from the
public and, under the fractional reserve system that operates in virtually all
countries, re-lend (though loans or purchases of short-term bonds) most of
the deposited amount, which is re-deposited by the public in the banking
system. This process creates demand deposits. The amount of demand
deposits and therefore of the money supply (of which demand deposits are
the main part of M1) created by the banks per dollar of the monetary base
depends on what proportion of the deposits is re-lent as credit to the
economy. The smaller the amount re-lent, the smaller is the money supply
for a given monetary base. A credit crisis increases the riskiness of credit
and also raises banks’ degree of risk aversion (i.e., dislike of taking on
risk), so that it reduces the amount re-lent for a given monetary base and
reduces the money supply. Hence, all other things (especially the monetary
base) being the same, a credit crisis brings about a very significant decline

791
in the money supply, which, in turn, reduces the aggregate demand (see
Chapter 6).
A credit crisis is associated with an increase in risk in lending. It also
increases the risk in holding financial assets, which include equities, so
that the demand for them tends to fall, and share prices to decline. The
more severe instances of this appear as stock market crises. This reduces
the wealth of households, which tends to produce the declines in
consumption. A stock market crisis also makes it more difficult to raise
new funds for investment in stock markets through the sale of new shares,
and reduces firms’ investment. The reductions in consumption and
investment further reduce aggregate demand and economic activity.
The credit crisis, the money supply crisis, and the stock market crisis are
elements of most financial crises, all of them collectively producing a
much greater decline in economic activity than a single one of them
would.
Therefore, the impact of a financial crisis on the economy occurs
through many directions: decrease in investment and consumption,
decrease in the money supply, and decrease in financial wealth. These
together cause a decline in aggregate demand. The credit crisis also
produces a decrease in aggregate supply. Hence, the impact of a financial
crisis is likely to be quite severe. Consequently, the more severe economic
crises, appearing as deep recessions or depressions, incorporate a deep
financial crisis.

16.12.3 Economic crises

Economic crises occur periodically as an aspect of capitalism and free-
market economies. A few of their common elements are
• Rapid increases in asset (shares, houses, etc.) in the period preceding the
crisis, so that there is clearly a price bubble (i.e., price levels far above
those justified by fundamental economic factors) in such prices. Herd
mentality builds up expectations of further increases, thereby fuelling
further increases in the demand for assets and their prices.
• Rapid increases in credit in the economy, often with the assets with
bubbles in their prices being used as collateral for loans. These fuel
investment, production, and consumption increases.
• Accommodative monetary policy with low interest rates and high growth
rates of the monetary base and the money supply.
• The eventual bursting of the price bubble in asset prices. The fall in asset
prices causes a banking and credit crisis, with reductions in credit,
investment, consumption, production, and employment. There may also

792
 be a money supply decrease, which can occurs since banks increase the
ratio of their reserves to loans and demand deposits, possibly augmented
by some bank failures.
• Transmission of crises to other countries occurs internationally through
reductions in imports and capital flows, as well as decreases in investor,
business, and consumer confidence. The extent and speed of such
transmission are greater from larger economies to their smaller trading
countries, than the other way around.

16.12.4 Some instances of financial crises

We have discussed the East-Asian Crises of the late 1990s earlier in Fact
Sheet 16.3. Hallmarks of these crises were a severe fall in asset prices, a
credit squeeze, decreases in investment and aggregate demand causing a
fall in output and rise in unemployment, and a foreign exchange crises
(i.e., drastic decrease in the foreign exchange rates) as foreign capital left
the countries. These features were shared by the financial and economic
crisis in the USA in 2007-2010, except that the USA did not experience a
foreign exchange crisis.
The financial crisis starting in the USA in 2007 started as a fall in asset
(especially housing) prices, became a crisis of the financial system, which
led to a credit crisis. It also included elements of a money supply crisis, in
which banks’ reluctance to provide credit meant that, while the central
banks of many developed economies especially that of the USA, increased
the monetary base very considerably (called quantitative easing), the
private banks mainly increased their reserves. The result was a financial
crisis, with a credit and a money supply crisis as elements. This crisis,
through its impact on aggregate demand and supply, reduced output and
employment, and sent the world economies into an economic recession,
which started in 2008.
Financial crises are an inherent aspect of the private capitalist financial
system based on trust among borrowers and lenders, and on the stability of
the economy. Such crises have occurred numerous times in numerous
countries. Besides the crisis of 2007–2010, originating in the USA and
spreading to many other countries, examples of some other crises are
Spain, starting in 1977; Norway, starting in 1987; Finland, starting in
1991; Sweden, starting in 1991; Japan, starting in 1992; United Kingdom,
starting in 1974, 1991, 1995; USA, starting in 1984; and in several East-
Asian countries in the mid-1990s.

793
16.13 The Relevance of Monetary and Fiscal
Policies to Business Cycles
Can monetary policy be used to moderate the fluctuations in real economic
activity? The answer to this question depends on whether the classical or
the Keynesian theories are relevant to the actual situation in the economy.
The different answers have already been discussed in a number of
chapters, especially Chapters 10 and 11. Briefly, if the Keynesian views
are relevant — as in a cycle due to aggregate demand shifts — monetary
policies can moderate the cycle and should be pursued to stabilize the
aggregate demand.
However, if the classical views are relevant and if the cycle is due to
unanticipated shifts in aggregate demand, monetary policies can only
successfully moderate the cycle, but under certain difficult-to-meet
conditions, such as the central bank’s superiority of information relative to
that of the private sector, and sufficiently rapid and predictable impact of
monetary policies, etc. If these conditions are not met, unanticipated
changes induced by the central bank in the money supply could accentuate
changes in economic activity in an undesired direction. Note that if the
classical views are the relevant ones and if the cycle is due to the volatility
of technology shifts or shifts in the prices of resources, then monetary
policy would not be effective in moderating the cycle.
Hence, whether monetary policies should be pursued or not depends on
the source of the particular cyclical fluctuation, the information at the
disposal of the central bank and the public, and the relative speed of the
impact of monetary policy — as knowledge of the economic theory that is
really valid for the economy and the period in question. Further, whether it
is actually pursued or not depends on the economic philosophy and beliefs
of the monetary authorities.
Note also that most central banks often use changes in their
discount/bank rate as their preferred instrument of monetary policy. These
changes induce changes in both the economy’s interest rates and in its
monetary aggregates. All these changes collectively affect economic
activity, so that the changes in the discount/bank rate become the central
bank’s instrument for its attempt to limit the business cycle fluctuations.
Chapter 5 is the relevant one for further discussion of interest rates as the
monetary policy instrument.
Fiscal policy and its effects on the economy were extensively discussed
in Chapters 5–11. Briefly, the case for the pursuit of fiscal policy is very
similar to that for the pursuit of monetary policy. However, there are some

794
differences between the case for monetary policy and that for fiscal policy.
One of these is related to the Ricardian equivalence hypothesis, which was
briefly discussed in Chapter 11. Ricardian equivalence implies that private
saving increases by the amount of the fiscal deficit, so that national saving
and aggregate demand do not change as a result of the deficit. Therefore, if
this hypothesis holds, fiscal policy cannot function as an instrument for
stabilizing the business cycle fluctuations, even if monetary policy is
effective in this role. Another difference between fiscal and monetary
policies is related to their relative speeds of implementation and impact.
The lag from the recognition of the need for policy and the implementation
of the appropriate fiscal policy depends upon the political and
administrative structures of the government of the country in question.
This lag is usually much longer for the USA, with its strong division
between the President and Congress but with joint control over the
government’s budget, than it is for most parliamentary systems, as in the
UK and Canada. As a result, fiscal policy is usually not the policy
instrument of choice for business cycle stabilization in the USA. Monetary
policy has to play that role to a greater extent than in the parliamentary
systems. The second lag that differs between fiscal and monetary policies
is the impact one. This lag is likely to be shorter for fiscal deficits than for
changes in the monetary aggregates since fiscal deficits directly increase
the aggregate expenditures while the expansionary monetary policy has to
first increase investment, which would then increase the aggregate
expenditures.
Box 16.4: Anatomy of the Financial and Economic Crisis of 2007-2010 in
the World Economy
The crisis of 2007–2010 starting in 2007 has already been discussed
earlier in Section 16.12.1. Given its importance to the world economy,
this box further elaborates on the information there. This crisis had two
distinct phases. One of these was the initial period of financial crisis,
starting in August 2007 and mainly confined in its early stage up to
October 2008 to the U.S. financial sector. This crisis followed:
• A remarkable period of financial innovation, including new types of
marketed credit instruments, leading to strong credit growth.
• Rapidly rising house prices. This increase was at least partly fuelled by
remarkably easy access to mortgages, sometimes with mortgages
exceeding the purchase price of the house and/or obtained by
borrowers who did not have the income flows to meet the mortgage
payments and relied on rapidly rising house prices to create positive

795
 equity value in the house. Such mortgages have a high risk of default
and are called ‘subprime’. Financial intermediaries securitized/bundled
mortgages and sold short-term bonds (called structured mortgage
instruments), interest on which was paid from the mortgage payments.
The funds obtained from their sale allowed the issuers to increase the
availability of mortgage funds, so that additional mortgages could be
issued. This process and its repetition resulted in a substantial increase
in credit in financial markets. Since the mortgaged-backed bonds were
issued by extremely large financial institutions and were short term,
they were viewed as riskless by the buyers. Since they offered a
slightly yield than Treasury bills, they were bought by other financial
institutions, pension funds, private individuals, and numerous other
types of investors, both in the USA and in foreign countries, as a means
of increasing the yield on their portfolios.
• The pre-2007 period was also characterized by financial euphoria or
exuberance during which investors, including large financial firms
themselves, did not look closely at the riskiness of their investments or
chose to ignore the risk. There was therefore a speculative boom not
only in houses but also in shares.
• Average house prices in the USA began to stabilize in 2006 and then
fell in 2007. This led to many mortgagees defaulting on their
mortgages payments, especially if the mortgage was sub-prime and
beyond the income availability of the mortgagee to meet the mortgage
payments. The rising defaults hurt the ability of the financial firms that
had issued short-term mortgage-backed bonds to meet the interest
payments on the bonds, putting the defaulting ones in danger of
insolvency and bankruptcy. The number of investments firms thus
affected included some of the largest and oldest ones in the USA, and
put the USA financial system as a whole in crisis.
• The banking crisis led to a sharp decline in share prices on stock
markets, both in the USA and other countries, leading to a major
decrease in the net worth of households and firms.
• The policy response of the U.S. government to the potential or actual
collapse of major financial firms was for the U.S. government and the
Fed to rescue (bail out) some, though not all, of the investment houses
in risk of bankruptcy by lending close to a trillion dollars to such
institutions.
• By late 2008, the crisis in the financial system had led to the affected,
as well as other, financial institutions to suspend or drastically reduce
their credit to other financial firms, as well as to firms engaged in
production and households. It also became difficult for firms to raise

796
 funds by the issue of bonds in the bond markets. Therefore, the
financial crisis had created a credit crisis or crunch by the last quarter
of 2008.
• Firms use short-term credit in the form of loans, issue of short-term
bonds and trade credit to finance their short-term working capital
needs. The credit crisis drastically decreased the supply of credit and
led firms to cut down their employment, and production and
investment. It also led consumers to cut down their spending. These led
to decreases in both aggregate demand and aggregate supply. The
result was the evolution of the financial crisis into a broad-based
economic one by late 2008. The U.S. economy and that of many other
industrial economies entered a recession, with falling output and rising
unemployment rates in the last quarter of 2008. Growth rates fell in late
2008 and continued to be negative through 2009 and early 2010 in
most Western countries. While those of China and India remained
positive, they declined significantly from the growth rates preceding
2007. While some countries began to experience increases in their
GDP in 2009, many aspects of the worldwide recession continued into
2010.

The policy responses

• The fiscal policy response of the U.S. government was to increase its
spending and deficits to an unprecedented extent, as well as to call on
the governments of other countries to do so.
• The monetary policy response of the Fed was to reduce the Federal
Funds discount rate in a series of steps until this rate was between 0%
and 0.25%. The Fed also injected large increases in the monetary base
through open market operations and also by encouraging commercial
banks to borrow from it.
• Another policy action of the Fed and the government acting in concert
was to rescue individual financial firms and provide loans or equity
capital to other firms, such as General Motors.
• With the high degree of globalization of capital markets in the
preceding decades, the financial crisis in the USA precipitated a
financial crisis in many other countries, especially European ones.
Further, with the increased globalization of international trade and
production in the preceding decades, the U.S. economic crisis spread to
other countries through cutbacks in imports by the USA and became a
worldwide economic recession. There was a push by the USA and
other countries for concerted international action by the major trading

797
 nations on rescuing the financial sectors and stimulating world
aggregate demand.
• It was recognized that regulatory and market failures had led to the
financial crisis. The overhaul of the regulatory framework of financial
institutions is being planned and will inevitably lead to increased
surveillance and regulation by the Fed, as by central banks in many
other countries. It is quite likely that new regulations will require
higher ratios of capital to liabilities, as well as limit risk-taking by
financial intermediaries.

16.14 Inflation and Unemployment

Some booms produce relatively low unemployment and high inflation
rates. This especially occurs during booms based on the expansion of
aggregate demand due to money supply increases beyond the full capacity
of the economy. If the inflation rate reaches undesirable levels, the public
begins to demand its reduction. But, doing this through drastic reductions
in the money supply has often forced the economy into a downturn and
raised unemployment rates. Therefore, it is often asked whether the
reduction in the inflation rate though contractionary policies, especially
monetary ones, will necessarily cause a recession, with an increase in the
unemployment rate.
The underlying analyses of the relationships between money supply
reductions, inflation, and unemployment have been presented in several of
the earlier chapters, with Chapter 10 being the most relevant one. Chapter
10 had presented the relevant analyses in general and in the context of the
Phillips’ curve (PC) and the Expectations Augmented Phillips’ curve
(EAPC).
The EAPC is based on the deviation of the actual unemployment rate
from the natural rate due to errors in price expectations.27 Its equation was

where u* is the short-run (SR) unemployment rate, un is the natural one

(i.e., the long-run equilibrium one), π is the actual inflation rate and πe is
the expected one for the period. Starting from a situation in which the
expected and actual inflation rates were identical, as were the actual and
natural unemployment rates, the EAPC implies the following.
(a) If the policy-induced decrease in inflation is anticipated, the economy
will continue to maintain the natural unemployment rate. Therefore,

798
 unemployment will not increase.
(b) If the policy-induced decrease in inflation is not anticipated, so that the
actual inflation rate falls below the expected one, unemployment will
rise above its natural rate in the short term (i.e., until expectations
adjust). But only the unanticipated part of the decline in inflation will
cause a rise in the inflation rate. Therefore, the best anti-inflationary
policy will be one in which the central bank manages to convince the
public of its lowered inflation targets. This implication has led some
economists to claim that the resulting increase in the unemployment rate
is really due to the failure of the central bank to adequately inform and
convince the public of its lowered inflation targets.
(c) Stabilizing the inflation rate at the existing level, when inflation was
expected to accelerate, would raise the unemployment rate.
(d) According to the EAPC, there is no cost in terms of output and
unemployment of maintaining a high inflation rate as long as the actual
and expected inflation rates are identical. However, Chapter 10 had
pointed out that there would nevertheless be other costs, so that central
banks often choose not to leave the economy with continuously high
inflation rates.
Most economists believe that expectations adjust only gradually to a
decline in the actual inflation rates, in which case the EAPC’s implication
is that unemployment will rise during the deflationary adjustment period.
Keynesians do not fully subscribe to the EAPC, which is based on the
assumption of short-run equilibrium and errors in price expectations. In the
Keynesian paradigm, reductions of the money supply and therefore, of
aggregate demand, would send the economy into disequilibrium (see
Chapter 9) and raise unemployment, irrespective of the relationship
between the actual and expected inflation rates. This increase in
unemployment is again a short-term one and occurs while the economy
takes its time to reach the new long-run equilibrium.
Empirical experience has generally been that policy-induced reductions
in the money supply, sufficient enough to significantly reduce the
aggregate demand, do lower the economy’s real output and raise its
unemployment rate in the short term, but not necessarily in the long term.
This has happened in countries with high credibility of the central bank’s
targets, as well as in countries where this credibility was low. It has
especially happened in countries where there was a fast reduction in the
money supply, though it has also tended to occur in countries that adopted
gradual reductions. Therefore, reducing the inflation rate should normally
be expected to raise the unemployment rate in the short term. Stated
differently, the reduction in the inflation rate will come at the cost of

799
higher unemployment and lower levels of real economic activity in the
short term. Hence, the policy makers have to make their decision after an
evaluation of the relative costs and benefits of lower inflation and higher
unemployment.

16.15 Conclusions
• Business cycles are periodic fluctuations in economic activity and are
empirically measured by the movement of a composite index of
economic activity. Business cycles are endemic to industrial economies.
• A cycle consists of a peak, downturn, trough, and an upturn. A recession
is the part of the cycle when economic activity is below its trend level. A
boom is the part when economic activity is above its trend level.
• Business cycles can result from the response of the economy to shocks to
aggregate demand and/or to aggregate supply.
• The classical paradigm explains business cycles as representing the
response of the economy in the short run to unanticipated demand shocks
or to supply shocks. It emphasizes shocks to the real sectors of the
economy as the main source of business cycles.
• The RBC theory focuses on fluctuations in labor productivity due to the
variation over time in the occurrence of technical change as the major
source of business cycle fluctuations.
• The Keynesian paradigm explains business cycles by aggregate demand
shocks, including anticipated ones, and supply shocks, but emphasizes
aggregate demand volatility as the dominant cause of business cycle
fluctuations.
• The accelerator-multiplier model generates cyclical fluctuations in
aggregate demand.
• Cyclical fluctuations in economic activity in open, industrialized
economies are interrelated.
• Long and deep recessions have the potential of reducing the base level of
output per capita for the future, while long booms have the potential of
raising the base level of output per capita for the future.
• Economic crises are an inherent feature of capitalist free-market
economies. They usually include or may originate with financial and
credit crises.
• Deep and long-lasting recessions are called depressions.
• In a crisis, monetary and fiscal policies can be used to bolster aggregate
demand and the availability of credit. The increasing integration (through
exports and capital flows) of the economies of industrialized countries in
the globalization process has speeded up and increasingly coordinated the

800
 transmission of crises and business cycles among the countries.

KEY CONCEPTS

Trend level of output versus

full-employment output
level
Peak, downturn, trough, upturn
Recession, boom
Procyclical and countercyclical
variables
Index of economic activity
Index of leading indicators
Shocks
Credit crisis
Propagation mechanism
Real business cycle theory, and
Multiplier-accelerator model.

SUMMARY OF CRITICAL CONCLUSIONS

• In principle, the cyclical fluctuations in economic activity can be due to

shocks to the demand side or the supply side of the economy. The shocks
to aggregate demand can come from shocks to the commodity, money, or
bond markets.
• Classical economists emphasize the shocks to the supply side from shifts
in productivity or labor supply, or the short-run response of the supply
structure, through the SRAS curve, to unanticipated shifts in aggregate
demand.
• The RBC is part of the classical paradigm and emphasizes shifts in labor
productivity coming from changes in technology as the predominant
cause of the observed business cycles.
• Keynesian economists allow both shifts in aggregate supply and demand
to initiate the cyclical fluctuations, but consider the latter to be dominant
source of the observed business cycles. Further, both anticipated and
unanticipated shocks to aggregate demand can initiate business cycles.
• Technology, capital markets, and trade links create positive

801
 comovements in cyclical activity among countries, though with some
leads and lags.
• Recessions (booms) have the potential for leaving long-term reductions
(increases) in the post-recession (boom) level of output per capita, but are
less likely to affect its growth rate.
• A credit crisis can cause a reduction in both aggregate demand and
aggregate supply, thereby resulting in a severe economic crisis.
• Globalization has speeded up and coordinated the transmission of crises
and business cycle across countries.

REVIEW AND DISCUSSION QUESTIONS

1. Specify the four phases of business cycles and describe them.

2. What can cause business cycles in the full-employment (i.e., the long-
run equilibrium) output level? Can cycles occur in the growth rate of
output?
3. Show, using the AD–AS diagram, how changes in (a) the money supply
(b) investment, might cause cycles in real output.
4. Show, using the AD–AS diagram, how changes in (a) the technology of
production, (b) the demand for leisure, might cause changes in real
output.
5. Explain the real business cycle theory and its policy implications for
preventing or moderating the business cycles.
6. Explain the Keynesian approach to business cycles and its policy
implications for preventing or moderating the business cycles.
7. Why do many economists believe that the real business cycle theory
fails to adequately explain the recessions? What seems to be the main
reason for this failure?
8. Explain the multiplier-accelerator model and show how shocks to
investment can cause business cycles.
9. What causes the positive relationship in cyclical economic activity
among countries?
10. Can recessions leave long-term economic effects? How, and of what
kind? What is the name given to such effects? Define this term.
11. Discuss the role, if any, of monetary and fiscal policies in moderating
the business cycles. If this is possible, should they do so?
12. What is a credit crisis and how can it cause an economic crisis?

802
 ADVANCED AND TECHNICAL QUESTIONS

T1. ‘Governments and central banks should not do anything to alleviate

the rise in unemployment during recessions.’ Discuss the alternative
answers that different economic theories provide to this statement.
T2. ‘Even if firms fail on a large scale due to a fall in aggregate demand
during recessions, such failure is the mechanism by which the
competitive functioning of the economy weeds out inefficient firms. The
failing firms do not deserve to be bailed out. Productivity will be higher
if such firms close down’. Discuss.
T3. Should large firms (e.g., General Motors and Chrysler in the
automobile industry; AIG, Citigroup, Lehman Brothers, etc., in the
financial industry of the USA) faced with heavy losses due to a drastic
shift away in the demand for their particular models/products, poor
management, management’s failure to anticipate demand shifts, and/or
excessive risk-taking, etc., be bailed out by massive government
grants/aid/subsidies? Discuss.
T4. Discuss the role of innovations and inventions in increasing aggregate
demand and causing stock market booms and crises. Are the latter an
essential result of the former? If so, why?

1 This average could be taken separately over each cycle or over a longer
period. The method of detrending the data and whether the period for
averaging should be the cycle or longer are discussed later in this chapter.
2 Note that the term ‘trend’ in the definitions of booms and recessions is
sometimes replaced by the term ‘full employment’. The relationship
between trend and full employment output is discussed later.
3 Real GDP data also exclude price movements, so that nominal GDP is
divided by the price level to compute real GDP. This can give an
erroneous picture about the performance of an economy that is heavily
dependent on exports. For an example, suppose that exports prices rise
while the amount exported falls but by a smaller percentage. While real
GDP will show a decline in economic activity (production), the nominal
income from exports would rise, which would raise consumer
expenditures, so that GDP and consumer expenditures would give different
signals on whether economic activity has declined.
4 Alternatively, the compound growth rate can be calculated over the given
number of quarters and the data adjusted to remove this growth rate.

803
5 Note that random fluctuations do not generally exhibit persistence in the
above sense.
6 They led to an upsurge in the rate of growth of total factor productivity,
especially after 1995.
7 Besides the heavy investment by the information technology
corporations, which were the major beneficiaries of the stock market
mania, there was also a widespread increase in information technology
investment throughout the various sectors of the economy.
8 See Chapter 8.
9 See Chapter 11 for this analysis.
10 Including foreign trade.
11 Classical economists who believe in Ricardian equivalence (see Chapter
11) discount the importance of fiscal deficits in shifting cycles in the shifts
of the IS curve and, therefore, creating fluctuations in real interest rates
and other variables of the economy.
12 As shown in Chapter 7, they only shift the interest rate in the long-run
equilibrium but do not change output.
13 The uneven changes in investment due to the uneven technological
progress are endogenous, so that some RBC models encompass them.
14 Note that under the ceteris paribus clause, the money supply is held
constant.
15 A feature of such unemployment in recessions is that the layoffs of
workers by firms rise, as do the number of workers on recall — and
obviously willing to resume work as soon as the employer calls them back.
16 If the increase in unemployment in recessions were due to an increase in
voluntary unemployment only, then quits (by workers of jobs) should
increase in the recessions. They do not do so. Rather, firms are reducing
the number of jobs offered and lay off some workers.
17 John Maynard Keynes referred to these psychological factors as ‘animal
spirits’. Alan Greenspan, Chairman of the U.S. Federal Reserve System at
the time of the stock (especially of dot.com companies) mania of the late
1990s, referred to the rapid multiple increases in the prices of Internet,
computer, and other communication stocks, as ‘irrational exuberance’.
18 Remember that for the Keynesians, both anticipated and unanticipated
increases in aggregate demand would increase output, and vice versa (see
Chapter 11).
19 This investment is in addition to that required for the maintenance of the
existing capital stock.

804
20 The interest rate can be introduced into this investment function, so that
changes in the interest rate will modify the time path of investment.
21 This function was that investment depends only on the interest rate.
22 The reason for this is that investment is the intended change in the
desired capital stock. The desired capital stock depends on the expected
demand for commodities. As this demand recovers in the post-recession
period, the firms will increase the investment to build up their capital stock
to its long-term desired level. However, this takes time, so that there would
be some lingering effects until the capital stock reaches the desired level.
23 Charles Nelson and Charles Plosser. “Trends and random walks in
macroeconomic time series: some evidence and implications”. Journal of
Monetary Economics , September 1982.
24 If they are pursued, there could be benefits but also costs, but the net
benefit is more likely to be negative.
25 In the classical paradigm, anticipated money supply changes are
‘neutral’ in their impact on the long-run equilibrium values of the real
variables. This proposition was discussed in Chapters 8 and 11 under the
concept of the neutrality of money.
26 Conversely, this implies that the changes in output and employment are
not the cause of all of the changes in the monetary aggregates.
27 Its two forms were the Friedman and the Lucas supply rules.

805
 Index

The index that appeared in the print version of this title does not match the pages in
your eBook. Please use the search function on your eReading device to search for terms
of interest. For your reference, the terms that appear in the print index are listed below.

45 degree diagram

accelerator-multiplier model
acyclical variables
AD curve
open economy
AD-AS Analysis
Adam Smith
adjustment costs
and business cycle
and fiscal policy
and monetary policy
and persistence
of changing employment
of changing output
of changing prices
of output
of prices
types of
adjustments
following a demand shock
following a supply shock
aggregate demand
deficiency
impact on output in long run
multiplier-accelerator model
under fixed exchange rates
under interest rate targeting

806
aggregate demand deficiency
aggregate demand equation
derivation
aggregate expenditures
aggregate, demand
Asian tigers
assets, foreign
asymmetrical effects of policies
Austrian School
automatic stabilizer
autonomous consumption
autonomous investment
autonomous investment multiplier
autonomous tax revenue multipliyer
balance of payments
accounting
and flows ofintereste payments
and unilateral flows
deficits
deficits in
economic
equilibrium
equilibrium in
surplus in
balance of payments on capital account
balance of payments on current account
balance of trade
balanced budget
cyclically adjusted
balanced budget multiplier
bank rate
reserves
bank rate
bank reserves
barter economy
behavioral relationships
bills of exchange
black money
bond
definition
demand for

807
 long-term
market
market for
mortgaged-backed
nominal
open market operations in
present discounted value (PDV) of
prices of
real
short-term
speculative (portfolio)
speculative demand
Treasury bills
bond financing of deficit
surplus
boom
BP curve
and IRP
and LM curve
slope
under IRP
Bretton Woods system
bubbles
and business cycles
in asset prices
in house prices
business confidence
business cycle
AD-AS theory
and credit
and demand shocks
and fiscal policy
and full employment
and monetary policy
and money
and persistence
and technical change
boom
contraction
counter-cyclical pattern
cyclical pattern

808
 downturn
international linkages
Keynesian business cycle theory
Keynesian theory
modern classical approach
monetary policy
peak
propagation mechanism
RBC model
real business cycle theory
recession
stylized facts
supply shocks
trough
upturn
world
capital
contribution to growth
financial flows
human
outflows of
physical
capital drain
and growth
capital flows
and growth
nominal
capital markets
capital stock
investment
physical
capital, in endogenous growth theory
capital-labor ratio and growth
takeoff
threshold
capitalisim
and economic crises
capitalist economies
central bank
and government deposits
borrowing from central bank

809
 control over the money supply
discount rate
instruments
central bank independence
centralized economies
classical approach misconceptions
traditional
classical model modern
new
classical paradigm
and business cycles
and conservatism
and small government
assumptions of
impact of Great Depression on
classical/Solow growth model
closed economy
commerical banks
borrowing from central bank
demand for reserves
reserve requirements
commodities
commodity market
45 degree diagram
and capital flows
equilibrium
equilibrium condition
equilibrium in open economy
IS-LM analysis
open economy
commodity sector
open economy model
communism
communist economies
consol
consumer confidence
and precautionary saving
consumer price index (CPI)
consumption
and income
and interest rate

810
 and wealth
autonomous
definition
expenditures
function
functional symbol of
marginal propensity to consume
shocks to
convergence
of living standards
coordination failure
copyright
Core inflation
cost of living clauses
and wage contracts
counter-cyclical
countercyclical variables
coupon payment
credit
and business cycles
and effective supply
shocks to
credit crises
and money supply
credit crisis
and money supply
credit supply
crises
Asian
credit
economic
financial
examples of
foreign exchange
and development
money supply
stock market
transmission of
crisis
credit
economic

811
 in USA of 2007-2010
economic 2007-2010
economic activity
financial
in USA in 2007-2010
crowding out
and interest rate
and output
vs crowding-in
crowding-in
currency
currency boards
vs central bank
current account balance
cyclical pattern
David Hume
David Richardo
debt forgiveness
deficient demand
definition bonds
closed economy
commodities
goods
money
open economy
demand deficiency
and involuntary unemployment
demand deposits
creation of
demand shocks
from supply shocks
demand, aggregate
actual
deficiency
deficient
dynamic analysis
effective
excess
impact on output in long run
multiplier-accelerator model
notional

812
 shocks to
under interest rate targeting
demand-deficient economy
effectiveness of fiscal policy
effectiveness of monetary policy
policies in
depressions, economic
disagreements in economics
discount factor
rate
discount rate
and differentials in interest rates
changes in
discouraged workers
disequilibrium analysis
and effectiveness of fiscal policies in
and effectiveness of monetary policies in
in the economy
output in
disinflation
cold turkey policy
gradulist policy
disposable income
divergence
of growth rates
of living standards
dollarization
and productivity differences
with dual currencies
Dow Jones industrial index
dual currency system
dynamics
economic crisis of 2007-2010
economics as a science
economics, an art rather than a science
economy large
impact of
efficiency wage
employment
and expectational errors
and labor supply increase

813
 and productivity increase
equilibrium
in long-run equilibrium
endogenous variables
entrepreneurship
and growth
equilibrium
below full employment
conditions
deviations from
disequilibrium
in commodity market
in labor market
long run
short run
stable
unstable
equilibrium condition
equilibrium prices
transparency of
ex post investment
excess demand
exchange rate
appreciation
depreciation
devaluation
effective
fixed
fixed and flexible
fixed/pegged
flexibility/floating
flexible
forward
managed
nominal
overshooting
persistence of movements in
real
spot
exogenous variables
expansionary fiscal policy

814
expectation
and dynamic adjustments
erros in
formation of
rational
revisions in
expectations agumented Phillips curve
and business cycle
expectations augmented employment analysis
expectations augmented employment function
expectations augmented output analysis
expectations augmented output equation
expectations augmented Phillips curve
and expectations
vs Phillips curve
expected rate of inflation
export function
exports
multiplier
net
shocks to
external balance
ex ante investment
Federal Funds rate
financial crisis
financial innovation
financial sector and growth
empirical evidence
historical experience
financial sector, informal
fiscal
deficit
policy
surplus
fiscal deficit
bond-financing of
Money-financing of
fiscal multiplier open economy
under interest targeting
fiscal policy
and aggregate demand

815
 and business cycle
and long run output
and price level
determinants of
fiscal policy
impact on long-run output
in a recession
in demand deficient economy
ineffectiveness in long run
lags
proactive role
under fixed exchange rates
under flexible exchange rates
vs monetary policy in long run
fiscal policy and stabilization
fiscal responsibility
Fisher equation
Fisher equation on interest rates
fixed exchange rates
and aggregate demand
and policies
costs and benefits
general equilibrium
flexible prices and wages
flexible/floating exchange rate
foreign aid
and growth
foreign exchange
and foreign currencies
and gold
and SDRs
and supply of
demand elasticities
demand for
from capital flows
from commodity flows
from interest payments
from net transfers
market
supply elasticities
supply of

816
foreign exchange market
demand curve
equilibrium in
foreign exchange market equation
foreign exchange reserves and balance of payments
change in
foreign exchange, flows
foreign investment and growth
forward contract
forward market
free enterprise system
free markets ideology
Friedman supply rule
and clearance in labor market
and inflation expectations
policy implications
validity
Friedman, Milton
and the modern classical school
and the Monetarists
vs the Keynesians
full employment
deviations from
fundamental value
gap
external
fiscal
saving
GDP growth rates, world
GDP growth, USA
general equilibrium microeconomic model
globalization
and national autonomy
and technical changes
benefits from
causes of
effects of
full
lossess from
measuring
mercantilism

817
 truncated
gold
Gold Exchange Standard
Gold Standard
goods
in the open economy
non-traded
traded
government
expenditures
role of
spending
transfers
government expenditures
multiplier
versus government spending
Great Depression
Gross Domestic Product (GDP) as a measure of welfare
deflator
doulble counting
GDP at factor cost
GDP at market prices
GDP at value added
per capita
Gross National Product (GNP)
growth
a miracle
accounting equation
and benefits from international linkages
and entrepreneurship
and financial sector
and human capital
and inflation
and money
and price stability
and quantity of money
and size of markets
continuous SS
contribution of inputs to
contributions to
contributors to

818
 early industrial stage
empirical evidence
historical experience
Malthusian stage
modern stage
political and social environment
post-Malthusian stage
post-modern
sources of
with labor saving technical change
growth and financial sector
growth rate
calculation
convergence
historical
pre-SS
steady state
growth theory
and capitalist system
assumptions
boundaries of
endogenous
growth theory, endogenous
and policies
growth trap
Herd mentality
high real wage
hot money
household confidence
human capital
in growth theory
hyperinflation
costs of
hysteresis
effects of high unemployment
identity
between national saving and investment
national income
versus theory
implicit contract
and labor hoarding

819
 and MPL
and variations in labor effort
theory
Implicit employment contracts
import function
imports
multiplier
income
disposable
uses of
income share of
indebtedness
external
index of economic activity
induced investment
industrial product price index (IPPI)
inflation actual
and unemployment
benefits of
core
cost
costs of anticipated
costs ofunanticipated
deflation
demand push
derivation of
determinants of
disinflation
indexation for
targeting
inflation rate
actual
and economic blocks
and growth
anticipated
convergence of
expected
unanticipated
world
innovations
innovative process

820
 and creative destruction
intellectual property (IP) protection
interest payments flows of
outflow of
interest rate
actual
actual real rate of interest
as a target
differentials
differentials in
expected
expected real
in growth theory
in IS-LM analysis
long run
long-term
market
medium-term
nominal
real
Interest Rate Parity (IRP)
and exchange rates
and stocks
as a theory of the exchange Rate
benchmark theory
covered
incorporating risk
invalidity
short-run IRP
uncovered
interest rate target
interest rate targeting
internal balance
internal rate of return
International Monetary Fund (IMF)
as a world central bank
capital flows liberalization
conditionality
policies
stand-by arrangement
surveillance

821
 trade liberalization
inventions
macro
inventories
unintended changes
investment
accounting definition
accounting measure of
autonomous
definition in macroeconomics
ex ante/intended
ex post/actual
function
in growth model
induced
multiplier
shocks to
unintended
investment multiplier open economy
under interest targeting
involuntary unemployment
due to a demand deficiency
due to a high real wage
due to supply shocks
measuring
IRP
IRT Curve
Irving Fisher
IS curve
shifts in
and exchange rate changes
open economy
shifts in
slope of
open economy shifts in
slope of
IS curve/equation
IS equation
IS equation, open economy
IS-LM analysis
and aggregate demand

822
 under fixed exchange rates
IS-LM curves
and aggregate demand
and output
IS-LM equation, derivation
J curve
experience of
policy implications
Keynes
John Maynard
Keynes, John Maynard
The General Theory
Keynesian model
monetary and fiscal policies in
neo
new
Keynesian paradigm
and business cycle
Keynesian paradigm themes
knowledge
cost of production
dissemination
dissemination across countries
externalities
generation
knowledge, new
labor
and technical change
contribution to growth
demand
effective demand
force
income share of
negative growth rate of
notional demand
participation rate of women
participation rates
supply
variability in growth rates
labor demand derivation

823
 function
matched
labor force
in growth theory
maximum
non-employment rate
participation rate
shifts in
participation rate of women
shifts in growth rate
labor hoarding
labor market, sluggish adjustments
labor markets
dynamic adjustments in
labor productivity shocks to
labor supply
and interest rate
and wage contracts
backward bending
function
intertemporal
matched
lag in effect of monetary policy
long and variable
lagging variable
laissez faire
Laspeyres index
leading indicators
liberal policies
liberal socialism
liquidity money supply trap
liquidity trap
liquidity, world
LM curve
movements along
shifts in
LM equation
derivation
instability of
loanable funds theory
long run

824
 definition
long term
long-term interest rate
LRAS curve
and anticipated demand increases
Lucas Robert
Lucas supply rule
and clearance in commodity market
and hyper-inflation
and inflation expectations
and price expectations
and relative price
policy implications
validity
macroeconomic models, markets in
Malthus
growth theory
Malthusian growth model
Malthusian stage
post
Malthusian trap
marginal product of labor and real wage rate
derivation
marginal product of labor curve
general shape
marginal product of labor, derivation
marginal productivity of capital (MPK)
increasing
private
social
marginal propensity to consume
market (nominal) rates of interest
market clearance
continuous
market coordinator
market failure
market interest rate
Marxism
medium-term interest rate
menu costs
and horizontal supply curve

825
mercantilism
MI
model
modern classical model
modernization
and globalization
monetarism
monetarists
monetarists and friedman
monetary base
multiplier
policy
monetary aggregates Canada
UK
USA
monetary base
and commercial banks
multiplier
monetary base multiplier
monetary policy
an art rather than a science
and aggregate demand
and business cycle
and business cycles
and interest rate targeting
and long run output
and long-run output
and price level
and stabilization
and sterilization
and sticky prices
discretionary
feedback rule
impact on aggregate demand
in a recession
in deficient demand economy
in demand deficient economy
in disequilibrium
ineffectiveness in long run
interest rate target
interest rate target vs money supply target

826
 money supply target
proactive role
Taylor rule
under fixed exchange rates
under flexible exchange rates
vs fiscal policy
monetary policy interest rate
monetary targeting
money
definitions of
demand
financing of the fiscal surplus
functions of
growth and inflation
market
market and bond market
nominal value of
portfolio demand
portfolio demand and wealth
real value of
speculative demand
stock
supply
transactions demand
versus bonds
money and growth
money demand
and financial wealth
function
instability
motives
real
shocks to
speculative
transactions
volatility
money illusion
money market
disequilibrium
equilibrium
money market equilibrium

827
 under interest rate targeting
money supper under interest rate targeting
money supply
and credit crisis
and output
as a target
central bank control
determination
determination of
formula
instability of
trap for money supply
world
money, neutrality of
money-financing of the surplus
moral hazard
multifactor productivity
multiplier
and distribution of income
fiscal
for target interest rate
interest rate target
investment
lag in
policy
multiplier, investment
multiplier-accelerator model
Mundell-Fleming Model
NAIRU
national expenditures sources of
national income
identity
use of
national income identity
national saving
natural rate long term
short term
neoclassical growth model
neoclassical model
neoKeynesian economics
net experts

828
 capital
commodities
net exports
capital
commodities
Net National Product (NNP)
net tax revenues
new classical model
new Keynesian model
nominal bonds
nominal output
nominal value of money
non-employment rate
non-government organizations (NGOs)
notional demand functions
notional supply functions
official settlements balance
oil prices
impact of
Okun’s rule
and discouraged workers
and work effort
open economy
small
open market operations
output
actual
and aggregate demand
and expectational errors
and inflation
and oil prices
and productivity increase
components of actual
derivation of
effective
equilibrium
full-employment
level
long run
short run
short run deviations

829
 targeting
output deviations
transient and self-correcting
output gap
output real
output supply curve horizontal
long run
short run
under menu costs
Output-price adjustment function
overnight lending rates
overshooting
of depreciations
Paasche index
paradigms
paradigms in macroeconomics classical
Keynesian
paradox of thrift
parameters
participation rate
ofwomen
patents
and globalization
peak
perpetual bond
Phillips curve
and monetary policy
as a policy trade-off
expectations augmented
historical experience
instability of
shifts in
vs expectations augmented Phillips curve
physical capital
policies
and superior information
anticipated
cold turkey policies
discretionary
gradualist
pro-active

830
 proactive
random
rules
stabilization
systematic
unanticipated
policy dilemma
policy options
in deficient demand economy
policy variables
population
growth rates
PPP and IRP combined
PPP exchange rate
present discounted value
price adjustment equation under sticky prices
price bubble
price expectations
error in
price index
price level
derivation for long run
derivation of
price of the bond
prices
adjustments in
flexible and dynamic adjustments in
prices, sticky
private saving
proactive policies
procyclical variables
production function
and labor efficiency
Cobb-Douglas
general
in endogenous growth theory
in growth theory
linear
oil in the
quadratic
short term

831
 under implicit contracts
with labor saving technical change
with neutral technical change
with technology
productivity
differential growth
and fixed exchange rates
total factor
public debt
Purchasing Power Parity (PPP)
absolute
and exchange rates
long run relative
relative
long run
short run
short run relative
under fixed exchange rates
purchasing power parity (PPP) and IRP
quantity equation
and high inflation rates
as an indentity
quantity theory
an equilibrium condition
transactions approach
rate of return nominal
real
rational expectations
and systematic policies
in long run
random errors in
Reaganomics
real bonds
Real Business Cycle (RBC) theory
policy implications
real business cycle (RBC) theory
real value of money
recession
of 1973-75
of1980
of 2007-10

832
recession of 1973-1975
recession of 2001-2002
recontracting
reference cycle
reserve base
reserve requirements
residual, in growth theory
Ricardian equivalence
Ricardian equivalence theorem
R&D
R&D expenditures
sacrifice ratio
saving
and interest rates
function
in growth theory
national
private
uses of
versus savings deposits
saving rate high
in growth theory
shifts in
in growth theory
participation rate
saving vs investment
saving, national, uses of
saving, private
savings deposits
Say’s Law
Schumpeter, Joseph
securitization
seigniorage
share
prices
shocks
short run
aggregate supply
short term
short-run equilibrium
short-run macroeconomic

833
small government
small versus large government
socialism
Solow growth model
policy implications
with technical change
Solow model
with labor saving technical change
Special Drawing Rights (SDRs)
SRAS curve
and Lucas supply rule
and unanticipated demand increases
Friedman supply rule
SRAS rule
and sticky prices
SS growth
and saving rate
and increasing returns
and labor force
instability of
under technical change
vs level effects
SS growth rate stable
under increasing MPK
unstable
St. Louis School
stability
stability of equilibrium
stabilization
under fixed exchange rates
stabilization policies
and aggregate demand
stable equilibrium
stagflation
and oil price increases
staggered wage contracts
standard of living
and endogneous technical change
and participation rate of women
convergence
divergence

834
 historical experience
sterilization
sticky price
and SRAS curve
theory
stylized facts
impact of demand on output
of 2007-10 crisis
subprime mortgages
supply equation
and sticky prices
supply shocks
permanent
unanticipated
supply side policies
supply, aggregate
actual
effective
long run
notional
shocks to
short-run
T-bill rate
takeoff capital-labor ratio
tax rate
marginal
tax revenues
autonomous taxes
net tax revenues
Taylor rule
technical change
and globalization
calculation
endogenous
exogenous
growth rate of
in Solow model
labor saving
negative externalities
negative impact
other estimates

835
 positive externalities
Solow’s estimates
technology clusters
Thatcherism
theories/models in economics
theory
model
trade
dynamic benefits
static benefits
traded goods
transfer payments
flows of
Treasury bills
trend rate of output
trough
truncated globalization
uncertainty
unemployment
actual
actual rate
and errors in price expectations
and inflation
causes of high
changes in
components
costs of
cyclical
disequilibrium
due to a demand deficiency
due to a high real wage
due to supply shocks
effective
equilibrium level of
frictional
full-employment level
impact of monetary and fiscal policies
impact of policies on frictional
impact of policies on structural
in long run
in long-run equilibrium

836
 in short run
insider-outsider theory
involuntary
due to credit crisis
due to deficient demand
long run
measuring involuntary
natural
natural rate
and expectational equilibrium
estimation
variations in
natural rate of
rate
search
seasonal
shifts in natural rate of
short run
structural
trend of
trend rate
voluntary
unemployment rate
unintended investment
unstable equilibrium
upturn
variables
endogenous
exogenous
policy
velocity
velocity of circulation of money world
wage
and price expectations
contracts
rate
reservation wage
wage contracts
adjustment of
mathematical derivations
wage rate

837
and involuntary unemployment
and productivity increase
equilibrium
high
nominal
real
wages, reduction in, and instability
wage rate, adjustments in
wealth
lifetime
Wicksell, Knut
work effort

838
目录
Title Page 2
Copyright 3
Dedication 5
Contents 6
Preface 30
About the Author 41
Glossary of Symbols 42
Useful Mathematical Symbols and Formulae for Economics 44
Part I Introduction to Macroeconomics 49
1 Output, Unemployment, and the Basic Concepts 50
1.1 Introduction to Macroeconomics 50
1.1.1 The nature of macroeconomic analysis 51
The classification of goods in macroeconomics 51
Closed economy versus open economy analysis 51
Short-run macroeconomic theories, growth theories and
52
business cycle theories
1.2 The Classification of Economic Agents and Markets in Short-
52
Run Macroeconomic Models
Box 1.1: The Analytical Devices of the Short-Run Versus
54
Long-Run Analysis
Real Time Concepts: The Short Term and the Long Term 54
1.3 Introduction to AD-AS Analysis 55
1.3.1 The supply of commodities 55
The long-run (equilibrium) aggregate supply of commodities 55
The short-run aggregate supply of commodities 56
1.3.2 The components of aggregate demand 57
1.3.3 The diagrammatic AD-AS analysis 57
The possibility of disequilibrium in the economy 58
The actual aggregate supply of commodities 58
The impact of an increase in aggregate demand 58
The impact of an increase in the long-run productive capacity 59
of the economy

839
 The stability of equilibrium 60
1.4 The Relationship between Output, Employment, and
60
Unemployment in the Economy
The labor force 61
Unemployment 61
1.5 Measures of National Output and Expenditures 62
1.5.1 Gross domestic product (GDP) 63
Fact Sheet 1.1: Global Trends in GDP Per Capita, 1950–2003 63
1.5.2 Gross national product (GNP) 64
1.5.3 Net domestic product (NDP) and net national product
64
(NNP)
Fact Sheet 1.2: Comparing GDP and GNP, 1965–2004 64
1.5.4 Measuring GDP 65
Box 1.2: GDP per capita as a Measure of the Standard of
69
Living
GDP as a Measure of Welfare 69
1.6 Measuring the Price Level and the Rate of Inflation 70
1.6.1 Measures of the price level 70
1.6.2 The inflation rate 72
Fact Sheet 1.3: Inflation in Canada, 1915–2007 72
1.6.3 Core inflation 73
Mathematical Box 1.1: Calculation of the Price Index and
73
Growth Rates
The construction of a price index: an illustration 73
Differences between Laspeyres and Paasche indices 74
Separating the rate of growth of nominal income into the real
75
growth rate and the inflation rate
1.6.4 Deriving the rate of inflation from a price index 76
Box 1.3: Mathematical Formulae to Learn 77
Fact Sheet 1.4: Measures of Inflation for the USA, 1960–
78
2007
1.6.5 Disinflation versus deflation 79
1.7 Nominal Versus Real Output 79
1.8 The Economic Relationship between Real Output and Inflation 80
1.9 The Nature of Economic Relationships 81
Box 1.4: Definition of Equilibrium 81

840
 Equilibrium conditions versus identities 82
Equilibrium versus disequilibrium 82
Stable versus unstable equilibrium 82
1.10 Exogenous and Endogenous Variables and the Concept of
83
Shocks in Macroeconomics
1.10.1 An illustration 83
1.10.2 Shocks 84
1.10.3 Multipliers 84
1.11 Growth Theory 85
1.11.1 Growth of the standard of living 85
1.12 Business Cycle Theories 86
Fact Sheet 1.5: Booms and Recessions in the USA Since
87
1960
Box 1.5: The Fundamental Role of Economics as a Science 88
Theories/models in economics 89
Economics as the ‘premier social science’ 89
1.13 Paradigms in Macroeconomics 90
1.14 Economic and Political Systems: Organization of the
90
Macroeconomy
1.14.1 Capitalism 91
1.14.2 Marxism and communism 91
1.14.3 Socialism 92
1.15 Conclusions 93
Key Concepts 94
Critical Conclusions 94
Review and Discussion Questions 95
Advanced and Technical Questions 96
2 Money, Prices, Interest Rates, and Fiscal Deficits 100
2.1 What Is Money and What Does It Do? 100
2.1.1 The functions of money 101
2.1.2 The practical definitions of money 101
2.2 Money Supply and Money Stock 102
2.3 The Nominal Versus the Real Value of the Money Supply 103
2.4 Bonds and Stocks in Macroeconomics 104
2.5 The Definition of the Money Market in Macroeconomics 105

841
2.6 A Brief History of the Definition of Money 105
2.7 The Current Definitions of Money and Related Concepts 107
Extended Analysis Box 2.1: Current Meanings of the
107
Symbols for the Monetary Aggregates in Selected Countries
The monetary aggregates for USA 107
The monetary aggregates for the Canada 108
The monetary aggregates for the UK 108
Fact Sheet 2.1: Monetary Aggregates of the USA 109
2.8 The Monetary Base and Bank Reserves 109
2.8.1 The relationship between the monetary base and the
110
money supply
2.9 The Quantity Equation 111
2.9.1 The quantity equation in growth rates 112
2.9.2 The implications of the quantity equation for a
112
persistently high inflation rate
Fact Sheet 2.2: Money Growth and Inflation in the USA,
113
1960–2008
2.10 The Quantity Theory 113
Extended Analysis Box 2.2: The Difference between the
114
Quantity Theory and the Quantity Equation
2.10.1 The transactions approach to the quantity theory 114
The adjustment period relevant to the quantity theory 115
Is velocity constant over time? 116
Fact Sheet 2.3: Velocity of Money in the USA, 1960–2008 117
2.11 The Definitions of Monetary and Fiscal Policies 117
2.12 The Central Bank and Monetary Policy 118
2.13 The Economic Aspects of the Government and Fiscal Policy 118
2.13.1 The financing of fiscal deficits/surpluses and changes
120
in the money supply
The implications of the independence of the central bank
121
from the government for financing deficits
2.13.2 The public debt 121
2.13.3 The selective nature of government expenditures,
122
taxes, and subsidies
2.14 Interest Rates in the Economy 122
2.14.1 The Fisher equation on interest rates 123

842
 Fact Sheet 2.4: Nominal and Real Interest Rates in the USA, 125
1982–2008
2.14.2 The concept of present discounted value (PDV) of a bond 126
2.14.3 Bubbles in asset prices 128
The importance of asset bubbles for output and business
128
cycles
Box 2.1: The Determination of Stock Prices 129
Bubbles in house and land prices 130
A bubble in tulip bulb prices! 130
Conclusions 130
Key Concepts 132
Summary of Critical Conclusions 132
Review and Discussion Questions 133
Advanced and Technical Questions 134
3 Introduction to the Open Economy: Exchange Rates
138
and the Balance of Payments
3.1 Exchange Rates 139
3.1.1 Three concepts of exchange rates 139
The (nominal) exchange rate 139
The real exchange rate 140
The effective exchange rate 141
3.2 Fixed, Flexible, and Managed Exchanged Rates 142
3.3 Purchasing Power Parity (PPP) as a Theory of the
143
Exchange Rate
3.3.1 PPP at the level of a single commodity 143
Absolute PPP among countries 143
Extended Analysis Box 3.1: Does PPP Apply in the Real
144
World? An Illustration
3.4 Relative PPP and Shifts in the Relative Efficiency of
145
Economies
3.4.1 Long-run changes in relative PPP 147
3.4.2 Short-run changes in relative PPP 147
3.4.3 Implications of short-run PPP for exchange rates and
148
inflation rates
Box 3.1: International Comparisons of Standards of Living in
148
Terms of PPP

843
3.5 Interest Rate Parity (IRP) and the Determination of the
150
Exchange Rate
3.5.1 The benchmark IRP theory 150
3.5.2 The benchmark IRP as a determinant of the domestic
152
interest rate
3.5.3 The benchmark IRP as a theory of the exchange rate
152
under flexible exchange rates
3.5.4 The role of speculative returns to stocks in capital flows
153
and IRP
3.5.5 Extending IRP to incorporate risk factors and risk
153
aversion
Fact Sheet 3.1: Interest Rate Differentials Between Countries 154
3.5.6 The relative importance of PPP and IRP in determining
155
exchange rates
3.6 The Balance of Payments 155
3.6.1 The components of the balance of payments 158
Fact Sheet 3.2: United States Balance of Payments, 1976–
158
2008
3.6.2 Equilibrium in the balance of payments 159
3.6.3 The change in foreign exchange reserves 160
Fact Sheet 3.3: US Foreign Exchange Reserves and Balance
160
of Payments, 2005–2008
Equilibrium in the balance of payments 161
Foreign exchange reserves and short-term bonds 161
3.7 The Balance of Payments in an Accounting Sense 161
3.8 The Market for Foreign Exchange and the Changes in Foreign
163
Exchange Reserves
3.8.1 The demand and supply of foreign exchange 163
3.8.2 Equilibrium in the foreign exchange market 164
Extended Analysis Box 3.2: The Demand and Supply of
164
Foreign Exchange Stated in Foreign Currencies
3.9 The Market Determination of the Nominal Exchange Rate 165
Fact Sheet 3.4: Exchange Rates against the US Dollar, 1980–
166
2008
3.9.1 Hot money 167
Box 3.2: National Policies on the Balance of Payments and
167
Accumulation of Foreign Exchange Reserves

844
 3.10 The Persistence of Balance of Payments Deficits and
168
Surpluses
Conclusions 168
Key Concepts 169
Summary of Critical Conclusions 170
Extended Analysis Box 3.1.A: Comparison of the Actual and
170
PPP Costs of the Big Mac
Review and Discussion Questions 171
Advanced and Technical Questions 172
Part II Short-run Macroeconomics 176
4 Determinants of Aggregate Demand: The Commodity
177
Market of the Closed Economy
4.1 Symbols Used 177
4.2 The Commodity Sector of the Closed Economy 178
4.2.1 Uses of national income 178
4.2.2 Sources of national expenditures 179
4.2.3 Equilibrium in the commodity market 180
4.2.4 National saving 180
4.2.5 The relationship between saving and investment 181
4.2.6 The (physical) capital stock 182
Box 4.1: An Unjustified Oversimplification for a Modern
182
Economy
4.3 The Two Uses of Private Saving and the Drag of Deficits on
183
Investment
4.4 Disequilibrium and the Role of Changes in Inventories in the
184
Adjustment to Equilibrium
4.4.1 The role of unintended changes in inventories 185
Extended Analysis Box 4.1: A Simplified Diagrammatic
186
Analysis: The 45° Diagram
4.4.2 Limitations of the 45° diagram for macroeconomic
188
analysis
4.5 For the Macroeconomic Analysis of the Closed Economy, Is
There a National Income Identity and One Between National 188
Saving and Investment?
4.5.1 An accounting national income identity 189
Extended Analysis Box 4.2: The Distinction Between the 190

845
 Meanings of Investment in Macroeconomics 190

The actual change in the capital stock and the definitions of

191
investment
4.6 Demand Behavior in the Commodity Market 191
4.6.1 Consumption expenditures 192
Fact Sheet 4.1: Consumption and Disposable Income in the
193
USA, 1980–2008
Fact Sheet 4.2: Interest Rates and Saving in the USA, 1985–
195
2008
4.6.2 The saving function 194
Extended Analysis Box 4.3: The Dependence of
196
Consumption on Wealth, Interest Rates
The concept of lifetime wealth as the present discounted
196
value of future incomes
4.6.3 Investment expenditures 199
4.6.4 Government expenditures and tax revenues 202
Fact Sheet 4.3: Interest Rates and Investment in the USA,
200
1960–2008
Extended Analysis Box 4.4: A more Realistic Investment
201
Function
The impact of business confidence on investment 201
Fact Sheet 4.4: USA Fiscal Deficit, 1962–2008 204
4.6.5 The commodity market and the price level 204
4.7 The Commodity Market Model: The IS Equation/Curve 205
The impact of investment fluctuations on income: a partial
206
investment multiplier
Box 4.2: The Mechanism of the Investment Multiplier: An
207
Illustration
The impact of fiscal policy on income: partial fiscal policy
208
multipliers
Box 4.3: The Impact of a Balanced Budget: The Balanced
208
Budget Multiplier
A word of caution on IS multipliers 209
The IS curve 209
Shifts in the IS curve versus movements along it 210
Extended Analysis Box 4.5: The Slope of the IS Curve 211

846
 Key Concepts 212
Summary of Critical Conclusions 213
Review and Discussion Questions 213
Advanced and Technical Questions 214
5 Aggregate Demand in the Open Economy under an
219
Interest Rate Target: Is-IRT Analysis
5.1 The Number of Goods and Markets in the Open Economy 220
5.2 The Foreign Exchange Sector of the Open Economy and the
220
Balance of Payments, Review
5.2.1 Net interest payments and net transfer payments 223
5.2.2 Relationship between nominal and real interest rates 223
5.3 The Commodity Market of the Open Economy 224
Fact Sheet 5.1: Components of Aggregate Demand for USA,
226
Canada, and Thailand, 2007
5.3.1 The uses of private saving in the open economy 226
5.3.2 Three gaps: saving, fiscal, and external 227
Fact Sheet 5.2: The uses of private saving in the USA, 1980–
228
2008
5.3.3 Commodity market equilibrium and capital flows 229
Extended Analysis Box 5.1: Financing high levels of
domestic investment during the development stages and 230
foreign exchange crises
Foreign exchange crises and debt forgiveness 231
5.3.4 The open economy IS relationship 232
The complete model of the commodity sector for the open
233
economy
Box 5.1: Derivation of the Open Economy is Equation 235
5.3.5 The open economy IS curve 237
5.3.6 Shifts in the open economy IS curve 237
Extended Analysis Box 5.2: The Mathematical Derivation of
238
the Slope of the IS Curve
Changes in the slope of the IS curve as the economy becomes
238
more open
Comparing the magnitudes of the fiscal multipliers for open
239
economies
The stability of aggregate demand in more open economies 239

847
 The stability of aggregate demand in more open economies 239
5.3.7 The impact of exchange rate changes on the IS curve 239
5.4 The Formulation of Monetary Policy 241
Box 5.2: Money Supply, the Stock of Money and Their
241
Relationship with the Interest Rate
5.4.1 Monetary policy through interest rate targeting 243
5.4.2 Rules versus discretion in setting the target interest rate 244
5.4.3 Diagrammatic depiction of the interest rate target 245
5.5 The Determination of Aggregate Demand under Interest Rate
246
Targeting
5.5.1 Diagrammatic derivation of the AD curve 247
5.5.2 The downward slope of the AD curve in the open
248
economy under an exogenous interest rate target: a reiteration
5.6 The Policy Multipliers for the Open Economy Aggregate
248
Demand
5.6.1 The impact of investment fluctuations on aggregate
249
demand: the investment multiplier
5.6.2 The impact of fiscal policy on aggregate demand: the
249
fiscal multipliers
5.6.3 The impact of monetary policy on aggregate demand:
249
the target rate multiplier
5.6.4 The distribution of incomes, consumption patterns, and
250
the size of the multiplier
5.6.5 The impact of exports on domestic aggregate demand 250
5.6.6 The lag in the impact of changes in the interest rate
251
target
5.7 Diagrammatic Analysis of Fiscal and Monetary Policies 251
5.8 The IS, IRT Curves and the Determination of Output: A
252
Caveat
5.9 But What about the Monetary Sector and Its Money Demand
252
and Supply Functions?
5.9.1 The demand for money 253
5.9.2 Ensuring equilibrium in the money market under an
254
interest rate target and the determination of the money supply
5.10 Managing Aggregate Demand in the Open Economy:
255
Monetary and Fiscal Policies under Flexible Exchange Rates
5.10.1 The effectiveness of fiscal policy for the open

848
 5.10.2 The effectiveness of monetary policy for the open
256
economy under interest rate targeting
5.10.3 Limitations on the effectiveness of a policy of interest
257
rate targeting
5.11 Internal Versus External Balance 258
Extended Analysis Box 5.3: The Mundell-Fleming Model of
258
the Open Economy
5.12 Conclusions 259
Key Concepts 260
Summary of Critical Conclusions 261
Review and Discussion Questions 261
Advanced and Technical Questions 263
6 Aggregate Demand Under a Money Supply Operating
268
Target: IS-LM Analysis
6.1 Monetary Policy 269
6.1.1 Reasons for choosing interest rate targeting over money
269
supply targeting, or vice versa
Lags in the impact of money supply changes versus those of
269
interest rate changes
Instability of money demand 270
The informal financial sector and black money in developing
270
economies
6.1.2 Choosing the monetary aggregate as the target variable 271
6.2 The Demand for Money 272
6.3 The Motives for Holding Money 274
6.3.1 The volatility of the speculative demand for money 275
6.3.2 Other reasons for the volatility of the demand for
275
money
6.4 The Standard Money Demand Function 276
Extended Analysis Box 6.1: Special Cases of Money Demand 277
6.4.1 Equilibrium in the money market 279
Disequilibrium in the money market 280
6.5 The LM Equation 280
6.5.1 The LM curve 281
6.5.2 Shifts in the LM curve 281

849
 6.5.2 Shifts in the LM curve 281
6.5.3 Shifts in the LM curve versus movements along it 282
6.5.4 Final comments on the LM curve 283
Extended Analysis Box 6.2: The potential general shape of
283
the LM curve
6.6 Deriving the Aggregate Demand for Commodities by
284
Combining the IS and LM Curves
6.6.1 The IS equation for the commodity market equilibrium 284
6.6.2 The relationship between the nominal and real interest
285
rates
6.6.3 Diagrammatic determination of aggregate demand 286
The IS, LM curves and the determination of output: a caveat 287
6.7. The Impact of Expansionary Monetary and Fiscal Policies on
287
Aggregate Demand
6.8 Bringing Aggregate Supply into the Open Economy Analysis,
288
a Preview of Chapters 7 to 9
6.8.1 The impact of an increase in aggregate demand on the
288
quantity supplied and the price level
6.8.2 The impact of monetary and fiscal policies on
289
equilibrium output and price level in the open economy
The short-run impact of aggregate demand changes on output
289
and the price level
The long-run impact of aggregate demand changes on output
290
and the price level
6.8.3 Disequilibrium in the domestic economy and
290
stabilization through monetary and fiscal policies
6.8.4 Summation on the roles of monetary and fiscal policies
291
under flexible exchange rates
Extended Analysis Box 6.3: The Mundell-Fleming Model of
292
the Open Economy
6.9 The Central Bank's Control Over the Money Supply 293
6.10 The Central Bank's Instruments for Changing the Monetary
293
Base
6.10.1 Open market operations 294
6.10.2 Shifting government deposits between the central bank
294
and the commercial banks in Canada
6.10.3 A mechanism for commercial banks to change the
295

850
 6.11 The Central Banks' Control Over the Monetary Base 296
Multiplier through Reserve Requirements
6.12 The Impact of Discount Rate Changes on the Economy 296
6.12.1 The central bank's discount rate and interest rate
296
differentials in the economy
6.13 The Determination of the Money Supply 297
Extended Analysis Box 6.4: The Creation of Demand
298
Deposits
6.14 A Common Money Supply Formula for M1 300
6.14.1 The monetary base multiplier 301
6.14.2 Numerical examples 301
6.15 Conclusions 302
Key Concepts 302
Critical Conclusions 303
Appendix 303
Determination of the LM curve 303
Derivation of aggregate demand for the closed economy from
304
its IS and LM equations
Intuition 305
Simplifying the AD equation for further analysis 305
Determination of the long-run price level under an exogenous
306
money supply
Derivation of aggregate demand for the open economy from
306
its IS and LM equations
Review and Discussion Questions 307
Advanced and Technical Questions 308
7 Full-Employment Output and the Natural Rate of
312
Unemployment
7.1 The Production Function 313
Fact Sheet 7.1: Diminishing Marginal Product of Labor 314
Mathematical Box 7.1: Examples of Production Functions
316
and the Derivation of the Marginal Product of Labor
The general procedure for deriving the MPL from the
316
production function
7.2 The Labor Market 318
7.2.1 Demand for labor 319

851
 7.2.1 Demand for labor 319
Diagrammatic analysis 320
7.2.2 Supply of labor 320
Extended Analysis Box 7.1: The Intertemporal Analysis of
322
Labor Supply
The Backward Bending Labor Supply Curve Over Long
324
Periods
7.3 The Long-Run Equilibrium Levels of Employment and Output 325
7.3.1 The impact on long-run output of an increase in the
325
price level or inflation rate
7.3.2 The diagrammatic analysis of employment and output 326
The effect on output of an improvement in labor productivity 326
7.3.3 The impact of aggregate demand on long-run output 326
The pursuit of fiscal policies 327
The pursuit of expansionary monetary policies 327
7.3.4 The ineffectiveness of monetary and fiscal policies in
327
LR equilibrium
Mathematical Box 7.2: The Derivation of the Demand for
328
Labor, Employment, and Output
The role of aggregate demand in determining output and the
329
price level, an illustration
The impact of expansionary monetary and fiscal policies 330
7.4 The Effects of an Increase in the MPL on LR Output and
330
Employment
7.4.1 Diagrammatic analysis of the impact of an improvement
331
in technology on output and employment
7.5 The effects of an Increase in Labor Supply on LR Output,
331
Employment and Price Level
Mathematical Box 7.3: The Effect of an Improvement in
333
Technology
7.6 Conclusions on Changes in the Equilibrium Levels of
334
Employment and Output
7.7 Full Employment and the Full-Employment Level of Output:
335
Definitions and Conditions
7.8 The Role of Supply-Side Policies in Changing Full-
336
Employment Output
7.9 Stagflation 337

852
 Box 7.1: The Impact of Oil Price Shocks on Full Employment 338
and Full-Employment Output in Oil Importing Countries
The impact of oil price increases on oil producing countries 339
Can expansionary monetary and fiscal policies eliminate this
340
stagflation?
Extended Analysis Box 7.2: The Role of Demand-Side
340
Policies in Changing Full-Employment Output
An implicit assumption in the above conclusions 341
7.10 Crowding Out of Investment by Fiscal Deficits, Given the LR
341
Supply of Output in the Closed Economy
7.11 The Actual Level of Output in the Economy 343
7.11.1 Changes in the actual rate of output over time 344
7.12 The Rate of Unemployment 345
7.12.1 The LR equilibrium (natural) rate of unemployment 345
Shifts in the natural rate of unemployment 346
The lack of impact of monetary and fiscal policies on the
346
natural rate of unemployment
7.13 The Long-Run Equilibrium (Natural) Rate of Interest 347
The ineffectiveness of monetary policy and inflation in
348
changing the long-run real interest rate
7.14 Conclusions 348
Key Concepts 349
Summary of Critical Conclusions 349
Review and Discussion Questions 350
Advanced and Technical Questions 351
8 Output in the Short Run: The Role of Expectations and
356
Adjustment Costs
8.1 The Role of Uncertainty and Errors in Expectations 358
8.1.1 The theory of rational expectations 358
8.1.2 Random errors and their predictability 359
8.1.3 Application of rational expectations to monetary and
359
fiscal policies
8.2 Price Expectations and the Labor Market: Output and
360
Employment in the Context of Wage Contracts
8.2.1 Labor supply in wage negotiations during the first stage 360
8.2.2 Employment during the contract period (the production

853
 8.2.2 Employment during the contract period (the production 361
stage)
8.2.3 Diagrammatic analysis 362
The effect of a proportional increase in the expected and
364
actual price levels
Conclusions from the contract wage analysis of production 364
The expectations augmented employment and output supply
365
curves
Implications of the expectations augmented employment and
output supply equations for variations in employment and 365
output over the business cycle
Box 8.1: Errors in Price Expectations, the Duration of the
366
Wage Contract and Cost-of-Living Clauses
8.3 Friedman's Expectations Augmented Employment and Output
367
Rules
Extended Mathematical Analysis Box 8.1: Labor Demand
368
when There are Expectational Errors in the
The expectational equilibrium level of employment 369
The impact of expectational errors on employment 369
The Implications for Employment and Output of Wage
370
Contracts with Expectational Errors in Prices
8.4 Lucas Supply Rule with Errors in Price Expectations in
370
Commodity Markets
8.4.1 Firms' responses to errors in expectations 373
Extended Analysis Box 8.2: Diagrammatic Analysis:
Comparing the FSR and LSR Curves (FSR and LSR) and 373
the LRAS Curve
8.5 The Implications of the FSR and LSR for the Impact of
375
Anticipated Demand Increases
8.6 The Implications of the FSR and the LSR for the Impact of
375
Unanticipated Demand Increases
8.6.1 The impact of revisions in anticipations and real time 376
8.6.2 Practical aspects of the FSR and LSR analysis 376
Extended Analysis Box 8.3: The Implications of the FSR and
378
LSR for the Impact of Monetary and Fiscal Policies
The Implications of Rational Expectations for Systematic
379
Monetary and Fiscal Policies

854
 when Private Demand is Volatile 380

The Scope for Monetary Policies and the Political Economy

381
of the Government's Budget
8.7 FSR and LSR: The Impact of Anticipated Permanent Supply
382
Changes on the Economy
8.8 The Impact of Unanticipated but Permanent Supply Changes
382
on the Economy
Box 8.2: Cost of Living Clauses 384
8.9 The Short- and Long-Term Relationships between Output and
384
Inflation
8.10 Empirical Validity of the FSR and the LSR 385
8.11 Types of Adjustment Costs and Their Impact on Output 386
8.11.1 Firms' responses to increases in the demand for their
387
products
8.12 Menu Costs and Price Stickiness as the Explanation of the
388
SRAS Curve
8.12.1 Aggregate supply in the sticky-price hypothesis 390
8.12.2 Monetary policy and the sticky-price SRAS curve 391
8.13 Costs of Adjusting Employment: Implicit Contracts as an
391
Explanation of the SRAS Curve
8.13.1 Variations in work effort over the short term 392
Box 8.3: Work Effort in Restaurants During the Day 393
Work Effort by Students over the Term and the Calendar
393
Year
8.13.2 The Flexibility of the Employment-Output Nexus in
394
the Short Term
Long-term labor contracts and labor hoarding 394
8.13.3 Okun's Rule: The Relationship between
395
Unemployment and Output Changes
8.14 The Implications of Adjustment Costs for Persistence of
396
Output and Employment Fluctuations Over the Business Cycle
8.15 Implications of Adjustment Costs for the Impact of Monetary
398
and Fiscal Policies
8.16 Stagflation and the Recessions of 1973–1975 and 1980 398
Box 8.4: Empirical Evidence on Price Changes, Wage
399
Contracts and Output Response

855
 8.17 Stylized Facts and Intuition on the Theories Explaining
399
Variations in Output
8.18 Conclusions 401
Key Concepts 403
Summary of Critical Conclusions 403
Review and Discussion Questions 404
Advanced and Technical Questions 406
9 Actual Output, Disequilibrium, and the Interaction
413
among Markets
Fact Sheet 9.1: Money Growth and Output Growth in USA, 1960–
416
2008
9.1 The Relevance of Expectations on Variables Other Than Prices 416
9.2 Shocks to Aggregate Demand and Supply 417
Examples of shocks to consumption 418
Examples of shocks to investment and net exports 419
Shocks to money demand and supply 420
Shocks to the cost and availability of credit 420
9.2.1 Shocks to the aggregate supply of commodities 421
9.3 An Analysis of the Disequilibrium Following a Demand Shock 421
9.3.1 The assumptions for disequilibrium analysis 421
Effective aggregate demand and supply 422
Extended Analysis Box 9.1: The Dynamics Required for the
423
Maintenance of a Full-Employment Scenario
9.3.2 A plausible dynamic scenario 424
Initial effects of the emergence of a demand deficiency 424
Secondary effects of the demand deficiency 424
The transition from the old equilibrium to the new one 425
Output-price adjustment function [Optional] 425
Box 9.1: The Nature and Critical Role of Expectations in
426
Dynamic Adjustments
9.4 Diagrammatic Analysis Following a Fall in Aggregate
428
Demand
Effective demand for labor 429
9.5 Disequilibrium with Flexible Prices and Wages 429
Extended Analysis Box 9.2: The Relative Rapidity of

856
 Firms and Consumers in Adjusting Production, Employment, 430
and Prices

The failure or sluggishness of the labor market in adjusting

nominal wages versus the faster responses of firms and 431
workers in adjusting employment and consumption
9.6 If the Real-World, Real-Time Economy is not on its LRAS or
432
SRAS Curve, Where will it be?
9.7 Can the Economy Get Stuck Below Full Employment? 433
Impact of the disequilibrium on the long run: hysteresis 433
9.8 Optimal Monetary and Fiscal Policies for the Demand-
433
Deficient Economy
The use of expansionary monetary policies 434
Limits to the effectiveness of monetary policies in a demand-
435
deficient recession
The use of fiscal policy: increases in government
435
expenditures and/or cuts in taxes
The policy dilemma 436
Fact Sheet 9.2: Economies in Disequilibrium: USA During
437
the Great Depression
Recapitulation of the roles of monetary and fiscal policies in
438
a demand-deficient economy
Extended Analysis Box 9.3: Disagreements Among
Economists on the Appropriate Policies in Real Time for a 439
Real-World Economy
9.9 Excess Demand18 in the Economy and Appropriate Policies 440
9.10 Asymmetry in the Economy's Responses to Deficient and
441
Excess Demand
9.11 An Analysis of Disequilibrium Following a Supply Shock 441
9.11.1 Supply shocks from labor productivity 441
The scope for monetary and fiscal policies 443
Box 9.2: Demand Shocks Emanating from Shifts in Long-
443
Run Supply
9.11.2 The impact of a decline in credit supply on the
444
effective supply of commodities
9.12 'Crowding-Out' or 'Crowding-In' of Investment by Fiscal
445
Expenditures in a Demand-Deficient Economy?
9.13 Disequilibrium (Involuntary Unemployment) in the Labor

857
 9.13 Disequilibrium (Involuntary Unemployment) in the Labor 446
Market due to a High Real Wage
Expansionary monetary policy as a means of lowering the
447
real wage and restoring equilibrium in the labor market
Extended Analysis Box 9.4: Is the preceding demand-
deficient analysis a reflection of market failure or of markets' 448
sluggishness in adjustment of prices and nominal wages?
9.14 The Dual Implications of a High Saving Rate 449
The paradox of a high saving rate — also known as the
449
paradox of thrift
9.15 Conclusions 449
Key Concepts 451
Summary of Critical Conclusions 452
Review and Discussion Questions 453
Advanced and Technical Questions 454
10 Employment, Unemployment, and Inflation 461
10.1 Definitions of the Labor Force and Labor Supply 461
10.1.1 Frictional unemployment and actual employment in
463
labor market equilibrium
The notions of matched labor supply and matched labor
465
demand in the presence offrictional unemployment
10.1.2 Discouraged workers 465
10.1.3 Other determinants of the equilibrium level of
466
unemployment
Box 10.1: The Labor Force Participation Rate 467
The non-employment rate 468
Shifts in labor force participation rates in recent decades 468
Fact Sheet 10.1: Participation Rates in Canada and the United
468
States, Male and Female, 1980–2008
10.2 The Components of Unemployment 469
The diagrammatic depiction of the components of
470
unemployment
10.3 Involuntary Unemployment 471
Involuntary unemployment due to a high real wage 471
Involuntary unemployment due to the emergence of a
471
demand deficiency

858
 The variation in the number of discouraged workers with the
472
state of the economy
10.4 The Actual Rate of Unemployment 473
Fact Sheet 10.2: Unemployment Rates in Selected Countries,
474
1985–2008
10.4.1 The division of unemployment into natural and
475
cyclical unemployment
10.4.2 Changes in the actual rate of unemployment over time 476
Short-run and long-run variations in the natural rate of
477
unemployment
Fact Sheet 10.4: 477
Estimating the natural rate of unemployment 478
Fact Sheet 10.5: Unemployment and Trend Unemployment in
478
the United States, 1980–2008
10.5 Policies to Reduce Structural Unemployment 479
Extended Analysis Box 10.1: Changes in Structural
479
Unemployment
The impact of fiscal and monetary policies on structural
480
unemployment
10.6 Policies to Reduce Frictional Unemployment 481
10.7 Measuring Involuntary Unemployment 481
Two ways of measuring involuntary unemployment 481
Monetary and fiscal policies to reduce involuntary
482
unemployment
Box 10.2: Causes High Unemployment in Canada and Europe
482
Relative to the USA
10.8 The Costs of Unemployment 484
10.9 Relationship Between Unemployment and the Inflation Rate
485
from the AD-AS Analysis
10.10 Phillips Curve (PC) 486
10.10.1 Use of the Phillips curve as a policy trade-off 487
10.10.2 Instability of the Phillips' curve 488
10.11 Deviations of the SR Equilibrium Unemployment Rate from
488
the Natural One
Fact Sheet 10.6: Historical Phillips Curves in the United
489
States, 1960–1995

859
 10.11.1 Friedman and the expectations augmented Phillips
curve (EAPC) 489

10.11.2 Expectational equilibrium and the natural rate 491

Extended Analysis Box 10.2: The Friedman AND Lucas
492
Supply Rules
Expectational errors and the commodity markets: The Lucas
493
supply rule
10.12 The Implications of the EAPC for Shifts in the Phillips
493
Curve and for Policy
The relationship between the Phillips' curve and the EAPC 494
10.13 The EAPC and the Non-Accelerating Inflation Rate of
495
Unemployment
10.14 The Implications of the EAPC for the Impact of Anticipated
495
and Unanticipated Demand Increases
10.15 The Determinants of Inflation 496
Fact Sheet 10.7: 496
Another way of stating the determinants of inflation 498
10.16 The Costs of Inflation 498
10.16.1 Inflation under perfect competition with fully
anticipated inflation for all future periods: the costs of 498
inflation in the analytical long-run case
An intermediate scenario: anticipated inflation in the
500
commodity and labor markets for some quarters ahead
10.16.2 The costs of unanticipated inflation 501
A caveat 502
Box 10.3: The Impact of Hyperinflation on Long-Term
Output Growth and the Standard of Living: Practical 502
Experience in Contrast to Theory
10.17 The Sacrifice Ratio 503
Fact Sheet 10.8: Sacrifice Ratio in the United States, 1950–
503
2008
10.17.1 Gradualist versus Cold Turkey Policies of
504
Disinflation
10.17.2 Indexation to the rate of inflation: should nominal
505
wages and interest rates be indexed?
10.18 Conclusions 505
Key Concepts 508

860
 Summary of Critical Conclusions 508
Review and Discussion Questions 509
Advanced and Technical Questions 510
11 Paradigms in Macroeconomics 515
Stylized Facts on Money, Prices, and Output 517
11.1 An Analogy for the Two Main Paradigms in
518
Macroeconomics
11.1.1 The approach of the classical paradigm to the
519
pathology of the economy
11.1.2 The approach of the Keynesian paradigm to the
520
pathology of the economy
Extended Analysis Box 11.1: The Fundamental Assumptions
520
of the Classical Paradigm
11.2 Defining and Demarcating the Models of the Classical
522
Paradigm
11.2.1 The traditional classical approach 523
Extended Analysis Box 11.2: The Founders of the Classical
524
Tradition in Macroeconomics
David Hume (1711–1776) 524
Adam Smith (1723–1790) 525
David Ricardo (1772–1823) 526
11.2.2 The neoclassical model 526
Box 11.1: Some Major Misconceptions about Traditional
Classical and Neoclassical Approaches in the Classical 527
Paradigm
11.2.3 The 1970s monetarism 528
11.2.4 The modern classical model 530
11.2.5 The new classical model 531
Extended Analysis Box 11.3: The Founders of the Classical
532
Approach in the Modern Period
Milton Friedman (1912–2006) 532
Robert Lucas (1937-) 532
Extended Analysis Box 11.4: The Economic Contributions of
533
Milton Friedman
11.2.6 Milton Friedman and the Keynesians 534
11.2.7 The relationship between the Monetarists and
535

861
 Friedman 535

11.2.8 Friedman and the modern classical school 536

11.3 The Keynesian Paradigm 537
Extended Analysis Box 11.5: The Founders of the Keynesian
538
Paradigm
Knut Wicksell (1851–1926) 538
John Maynard Keynes (1883–1946) 539
11.3.1 Development of the Keynesian schools after Keynes 540
11.3.2 Frequent themes in the Keynesianparadigm 540
New Keynesian economics 541
11.3.3 The variety of Keynesian models 541
Monetary and fiscal policy effects in Keynesian models 542
11.3.4 The critical role of dynamic analysis when aggregate
543
demand falls
Box 11.2: The Keynesian-Neoclassical Synthesis on
544
Aggregate Demand
11.4 The Reformulation of Keynesian Approaches 544
11.4.1 NeoKeynesian building blocks for the reformulation of
544
Keynesian macroeconomics
11.4.2 Taylor rule and its incorporation into Keynesian
545
macroeconomics
11.4.3 The new Keynesian model 546
11.5 The Credit and Economic Crisis of 2007–2010: Which
546
Theory Can Explain It?
11.6 Which Macroeconomic Paradigm Should One Believe In and
548
Use?
Box 11.3: The Anatomy of Two Quite Different Recessions:
549
1973–1975 and 2001–2002
11.7 Paradigms and Policies 549
11.7.1 Stabilization versus pro-active policies 549
11.7.2 Rules versus discretion in the pursuit of policies 550
11.8 The Role of the Government in the Macroeconomy 550
11.8.1 Evolution of ideas about the role of the government 551
11.8.2 Classical and Keynesian approaches and the debates
552
on the size of the government

862
 Keynesian versus classical economics in relation to the real
555
world and real time
Key Concepts 557
Summary of Critical Conclusions 557
Review and Discussion Questions 558
Advanced and Technical Questions 559
Part III Topics in Open Economy Macroeconomics 563
12 The Foreign Exchange Market, IMF, and
564
Globalization
12.1 Review of PPP 565
12.2 Review of Interest Rate Parity (IRP) Theory 566
12.2.1 IRP as a theory of the interest rate 566
The importance of stock market returns for capital flows: a
568
problem for the IRP theory
Fact Sheet 12.1: Interest Rate Differentials between Countries 568
Simplified form of the benchmark IRP for short-run analysis 569
Covered versus uncovered IRP 569
12.2.2 IRP as a theory of the exchange rate 571
The inconsistency between the IRP's implications and
572
common observations
12.2.3 The impact of expectations on exchange rates 574
12.3 PPP and IRP Combined 575
Long run analysis 575
Short run analysis 575
12.4 The Market for Foreign Exchange, a Review 576
Fact Sheet 12.2: Interest Rates and Net Capital Flows in the
578
USA, 1985–2008
12.4.1 The relationship between the balance of payments and
579
(D$ — S$)
12.4.2 Diagrammatic analysis 579
12.5 Demand and Supply Elasticities 581
12.6 The J Curve for the Value of Net Exports in Real Time 582
12.6.1 Limitations of the commodity-based exchange market
584
analysis
12.6.2 Policy implications of the J curve 584

863
 12.6.2 Policy implications of the J curve 584
Extended Analysis Box 12.1: The General Experience of
585
Industrialized Countries and LDCs on the J Curve
Implications of the above analysis for movements of
586
exchange rates under flexible exchange rates
12.7 Historical Experience with Flexible Versus Fixed Exchange
587
Rates
12.8 Measures to Reduce Balance of Payments Deficits 588
12.9 The International Monetary Fund (IMF) 590
12.9.1 The IMF and disequilibrium in the balance of
591
payments of the member countries
Extended Analysis Box 12.2: IMF Conditionality 591
For whose benefit are the IMF policies? 592
12.9.2 The IMF and capital flows 592
12.9.3 Special drawing rights (SDRs) 593
Extended Analysis Box 12.3: The World's Demand for
594
Liquidity
Exchange and financial crises 594
Is the IMF a central banker for the world? 595
Does the world economy need a central bank? 596
The rising demand for the IMF to create a world currency 596
12.10 The International Transmission of Crises and Business
597
Cycles
12.11 Economic Globalization 597
12.11.1 Causes of the push for global markets 599
12.11.2 Measures of the extent of globalization 600
12.11.3 Other aspects of globalization 602
12.11.4 Globalization and modernization 602
12.11.5 The persistence of nationalistic mercantilism —
603
globalization limited to truncated globalization
12.11.6 The possibility of net losses from globalization 604
12.11.7 Effects of globalization on commodity and capital
605
flows
12.12 Conclusions 606
Key Concepts 608
Summary of Critical Conclusions 609

864
 Advanced and Technical Questions 611
13 The Open Economy Under a Fixed Exchange Rate
617
Regime
13.1 The Balance of Payments (BP) Equilibrium Curve Under
618
Fixed Exchange Rates
13.2 The IS-LM Model for the Open Economy Under Fixed
619
Exchange Rates
13.2.1 The slope of the BP curve 620
13.2.2 The relative slopes of the LM and BP curves 620
Extended Analysis Box 13.1: Differences between the IS-LM
621
Diagrams for the Flexible and Fixed Exchange Rates
13.2.3 General equilibrium under fixed exchange rates, a
621
diagrammatic treatment
13.3 Managing Aggregate Demand in the Open Economy:
622
Monetary and Fiscal Policies Under Fixed Exchange Rates
13.3.1 The impact of an expansionary fiscal policy on
623
aggregate demand
13.3.2 The impact of an expansionary monetary policy on
623
aggregate demand
Extended Analysis Box 13.2: Monetary and Fiscal Policies in
the Case of Interest-Insensitive Capital Flows and Fixed 624
Exchange Rates
13.4 Bringing Aggregate Supply into the IS-LM Analysis with a
625
Fixed Exchange Rate
13.5 Macroeconomic Equilibrium in the Small Open Economy
626
with Full Employment and IRP
13.5.1 Fiscal policies under IRP and full employment 627
13.5.2 Monetary policy under IRP and full employment 628
13.6 PPP, the World Rate of Inflation and the Convergence in
628
National Inflation Rates
13.6.1 The world rate of inflation in the fixed exchange rate
629
case
Box 13.1: The impact of a large economy on smaller
630
economies
13.7 Economic Blocs, Fixed Exchange Rates and the Convergence
630
in Inflation Rates
13.8 Disequilibrium in the Domestic Economy and Stabilization

865
 13.8 Disequilibrium in the Domestic Economy and Stabilization
through Monetary and Fiscal Policies: The Fixed Exchange Rate 631
Case with Interest Sensitive Capital Flows
A caveat 632
Extended Analysis Box 13.3: Disequilibrium in the Domestic
Economy and Stabilization through Monetary and Fiscal
632
Policies: The Fixed Exchange Rate Case with Interest
Insensitive Capital Flows
13.9 The Need for Monetary and Fiscal Policies under Fixed
632
Exchange Rates
13.9.1 Internal versus external balance 633
13.9.2 A note of caution on the use of monetary and fiscal
634
policies
13.10 The Effect of Relatively Different Improvements in
634
Productivity at Home and Abroad under Fixed Exchange Rates
13.11 The Political Economy of Exchange Rates: The Choice
635
between Fixed and Flexible Exchange Rate Regimes
13.12 Other Tools for Handling Balance of Payments Deficits 637
Extended Analysis Box 13.4: The Gold Standard and the Gold
638
Exchange Standard
13.13 Summary of the Costs and Benefits of a Fixed Exchange
640
Rate
13.14 Dollarization 641
13.14.1 Dollarization and different productivity growth rates 643
Extended Analysis Box 13.5: A Dual Currency System:
644
Dollarization Along with a Separate National Currency
13.15 Currency Boards 645
13.16 Conclusions 646
Key Concepts 647
Summary of Critical Conclusions 647
Review and Discussion Questions 648
Advanced and Technical Questions 649
Part IV Growth Economics 653
14 Classical Growth Theory 654
Fact Sheet 14.1: Long-Term Real GDP Growth in Selected
655
Countries, 1960–2003

866
Extended Analysis Box 14.1: Setting the Boundaries of
658
Macroeconomic Growth Theory
14.1 The Classical (Solow's) Growth Model's Assumptions 659
14.1.1 The technology of production 659
14.1.2 Saving, investment and the change in the capital stock 660
14.1.3 Labor force growth 661
14.2 The Analysis of the Solow Model 661
14.3 Diagrammatic Analysis of the Solow Model 663
14.3.1 The SS growth rate of output 664
Extended Analysis Box 14.2: Variability of the Saving Rate 664
14.3.2 The impact of shifts in the saving rate 665
Mathematical Box 14.1: Numerical example A 667
14.3.3 The impact of shifts in the labor force growth rate 668
Mathematical Box 14.2: Numerical example B 669
14.4 The Growth Implications of a More General Production
671
Function
14.5 Convergence Versus Divergence in Output per capita among
673
Countries
14.6 Assessing the Importance of Technical Change in the Solow
674
Model16
14.6.1 Solow's estimates of the contribution of technical
677
change to the improvement in living standards
Mathematical Box 14.3: Numerical Examples on Technical
677
Change
14.6.2 The residual as the unexplained component of growth 678
Mathematical Box 14.4: Detailed Estimation of the
679
Contributions to Economic Growth
Estimating the returns to different educational levels 679
14.7 Some Empirical Findings on the Contributions to Growth 682
14.8 The Solow Growth Model with Exogenous Changes in the
683
Quality of Labor
14.8.1 Diagrammatic analysis 685
14.8.2 The role of human capital in changing the quality of
686
labor
Extended Analysis Box 14.3: The Variability of the Labor
Force Growth Rate 687

867
 Force Growth Rate
14.8.3 Technical change with falling or negative labor force
688
growth rates
14.8.4 Shifts in participation rates 688
Fact Sheet 14.3: Participation Rates in Canada and the USA,
689
Male and Female, 1980–2008
Box 14.2: The Implications of Increases in the Labor Force
689
Participation of Women
14.9 SS Growth versus Level Effects of Exogenous Shifts 691
14.10 The Implications of the Solow Analyses for Policy 692
Fact Sheet 14.4: Real GDP Per Capita, 2003 Population
692
Growth Rates
Box 14.3: Historical Growth Patterns, Labor force Growth,
693
and Technology Shifts
14.10.1 Malthus's theory of economic growth and the
695
Malthusian stage
14.10.2 The post-Malthusian stage 695
14.10.3 The modern growth stage 696
14.11 Conclusions 698
Key Concepts 699
Summary of Critical Conclusions 699
Review and Discussion Questions 700
Advanced and Technical Questions 701
15 Advanced Topics in Growth Theory 711
15.1 Technology, Knowledge and Externalities 713
15.1.1 The distinction between endogenous and exogenous
713
technical change
15.1.2 The definition of capital in endogenous growth
714
theories
15.1.3 The externalities of new knowledge 715
Box 15.1: Inventions, Innovations, and Patents 715
Innovations versus inventions as forms of technical change 715
15.1.4 Patents and the protection of copyright, intellectual
717
property and trade secrets Recent changes in patent protection
15.2 The Production of New Knowledge 719
15.2.1 Diagrammatic analysis 720

868
15.4 Diagrammatic Analysis 723
15.4.1 Growth rates and capital flows among countries 725
15.4.2 Convergence and divergence in living standards
726
between countries
15.4.3 Continuous growth in the advanced economies 726
Box 15.2: Foreign Aid and Investment — and Capital Drain 726
15.5 The Historical Experience of Growth 727
Box 15.3: Economic Globalization and Endogenous
728
Technical Change
Technology Clusters 728
Globalization and Patents 729
15.6 Human Capital 729
Box 15.4: The Innovative Process and Its Creative
730
Destruction
15.7 The Implications of Endogenous Growth Theories for
732
Macroeconomic Policies
Extended Analysis Box 15.1: International Linkages and
734
Growth
15.8 The Importance of the Inflation Rate and the Quantity of
735
Money for Growth
Box 15.5: The Role of the Financial Sector in Growth: Some
736
Conclusions from Economic History
15.9 The Role of the Financial Sector in Growth: Recent Empirical
737
Evidence
15.10 The Miracle of Economic Growth 739
15.10.1 The miracle of growth and financial crises in the
739
economic tigers
15.10.2 The economic crisis of 2007–2010 in the USA 740
Extended Analysis Box 15.2: The Economic, Social, and
Political Environment for Growth: Markets, Competition, 741
Capitalism, and Entrepreneurship
15.11 Empirical Evidence on the Contributors to Growth 743
15.11.1 Empirical evidence on growth in recent decades 744
15.12 Conclusions 745
Conclusions from endogenous growth theory 745
Conclusions on money, the financial sector, and growth 747

869
 Key Concepts 747

Summary of Critical Conclusions 748

Review and Discussion Questions 749
Advanced and Technical Questions 750
16 Business Cycles, Crises, and The International
753
Transmission of Economic Activity
16.1 Recessions and Booms in Economic Activity 755
16.1.1 The popular statistical designation of a recession 756
16.1.2 Monetary and fiscal policies to combat a recession 757
Fact Sheet 16.1: Output Gap in the United States (USA),
759
1980–2008
16.2 Business Cycles and the Growth Trend in Economic Activity 759
Box 16.1: Eliminating the Trend from the Data 760
Deriving the trend rate of output 760
The unemployment rate as the proxy for the output gap 760
16.3 Stylized Facts of Business Cycles 760
16.4 The Behavior of Variables Over the Business Cycle 762
16.4.1 Procyclical, acyclical, and countercyclical movements 762
16.4.2 Leading, lagging, and coincident variables 762
16.4.3 The volatility of variables over time 763
16.5 Business Cycles, the Full Employment Assumption and the
763
Classical and Keynesian Paradigms
Box 16.2: Business Cycles and The Assumption of Full
766
Employment
16.6 The General AD-AS Theory of Business Cycles 766
Fact Sheet 16.2: The Structure of the AD-AS Model 769
16.6.1 The propagation mechanism 769
Box 16.3: An Illustration of Technical Change Creating
770
Cycles through Aggregate Demand Shifts
16.7 The Modern Classical Approach to Business Cycles 771
16.7.1 Real business cycle theory 773
16.7.2 Policy implications of the RBC theories 776
16.8 Keynesian Explanations of the Business Cycle 776
16.8.1 The multiplier-accelerator model of aggregate demand 777
16.8.2 Limitations of the validity of the multiplier-accelerator

870
 model
Extended Analysis Box 16.1: A Taxonomy of the General
780
Reasons for the Occurrence of Business Cycles
16.9 International Linkages Among National Business Cycles 781
16.9.1 A world business cycle 782
16.10 Crises in Economic Activity 783
Fact Sheet 16.3: The Asian Crisis of the Mid-1990s 783
16.11 Long-Term Effects of Recessions and Booms 784
16.11.1 The accumulation of human capital 785
16.11.2 The accumulation of physical capital 786
16.11.3 Technical change 787
16.11.4 The longer-term effects of cyclical fluctuations
787
through aggregate demand
16.11.5 The longer-term impact of recessions and booms 787
16.11.6 The implications of hysteresis for policy 788
16.12 Money, Credit and Business Cycles 789
16.12.1 Money supply and business cycles 789
16.12.2 Credit and business cycles 790
16.12.3 Economic crises 792
16.12.4 Some instances of financial crises 793
16.13 The Relevance of Monetary and Fiscal Policies to Business
794
Cycles
Box 16.4: Anatomy of the Financial and Economic Crisis of
795
2007–2010 in the World Economy
The policy responses 797
16.14 Inflation and Unemployment 798
16.15 Conclusions 800
Key Concepts 801
Summary of Critical Conclusions 801
Review and Discussion Questions 802
Advanced and Technical Questions 803
Index 806

871