BALANCE-OF-PAYMENTSTHEORY

AND THE UNITED KINGDOM EXPERIENCE

A.P. Thirlwall,
Balance-of-Payments Theory and
the United Kingdom Experience

Other books by the author

Growth and Development: with Special Reference to Developing

Economies, Second Edition
Inflation, Saving and Growth in Developing Economies
Financing Economic Development
Regional Growth and Unemployment in the United Kingdom
(with R. Dixon)
Keynes and International Monetary Relations (editor)
Keynes and Laissez-Faire (editor)
Keynes and the Bloomsbury Group (editor with D. Crabtree)
Keynes as a Policy Adviser (editor)
Balance-o f-Payments
Theory and the
United Kingdom
Experience
SECOND EDITION

A. P. THIRLWALL
Professor ofApplied Economics
University of Kent at Canterbury

M
© A. P. Thirlwall 1980, 1982

All rights reserved. No part of this publication

may be reproduced or transmitted, in any form
or by any means, without permission.

First edition 1980

Second edition 1982

Published by
THE MACMILLAN PRESS LTD
London and Basingstoke
Companies and representatives
throughout the world

ISB N 978-0-333-34307-4 ISBN 978-1-349-06534-9 (eBook)

DOI 10.1007/978-1-349-06534-9

Typeset in Great Britain by

REPRODUCTION DRAWINGS LTD
Sutton, Surrey

The paperback edition of this book is sold subject to the condition that it shall
not, by way of trade or otherwise, be lent, re-sold, hired out. or otherwise
circulated without the publisher's prior consent in any form of binding or cover
other than that in which it is published and without a similar condition including
this condition being imposed on the subsequent purchaser.
To Merlyn Rees, M.P.,
who first taught me to think and write
Contents
Preface to the First Edition xi
Preface to the Second Edition xiv
Acknowledgements xv

1 Balance-of-Payments Accounting and the Foreign-

Exchange Market 1
Concepts of the balance of payments 1
The accounting balance of payments 2
Balance-of-payments equilibrium 5
More on the concept of equilibrium 19
The foreign-exchange market 22
The demand for dollars 23
The supply of dollars 26
Capital movements and speculation 31
The forward rate of exchange 33
Types of exchange-rate regime 37
Fixed versus flexible exchange rates 43

2 The Balance of Payments and the National Economy 48

The integration of the balance of payments into the
national accounts 48
The conflict between balance-of-payments equilibrium
and other objectivesof economic policy 50
Income equilibrium in an open economy and the
foreign-trade multiplier 54
The foreign -trade multiplier relating imports to
expenditure 57
The 'new' foreign-trade multiplier with foreign
repercussions 61
viii Contents
Balance-of-payments adjustment 65
The classical price-specie flow mechanism 66
The elasticity approach 67
The absorption approach 69
The monetary approach 69

3 The Elasticity Approach to the Balance of Payments 72

The Marshall-Lerner condition derived 72
Devaluation and the response of-firms 78
Lags and the J·curve effect 83
Devaluation and inflation 85
Devaluation and the terms of trade 92
The difficulties of measuring price elasticities 93

4 The Absorption Approach to the Balance of

Payments 98
The direct effect of devaluation on income 100
The direct effect of devaluation on absorption 101
The interaction between changes in income and
changes in absorption 103
Strengths and dangers of the absorption approach 105
The monetary aspects of balance-of-payments deficits 107
Domestic credit expansion 108

5 The Monetary Approach to the Balance of

Payments 110

6 Simultaneous Internal and External Balance 118

Internal and external balance under flexible exchange
rates with no capital movements 120
Internal and external balance under fixed exchange
rates with capital movements 124
Contents ix
Internal and external balance in an fS-LM curve
framework 129
The use of monetary and fiscal policy under alternative
exchange-rate regimes 131
The assignment of policies in the United Kingdom 132

7 A History of the U.K. Balance of Payments 134

1854 to 1939 134
1940 to 1950 140
1951 to 1959 143
1960 to 1969 148
Devaluation, 1967 154
1970 to 1980 157
Appendix 7.1: an economic analysis of the United
Kingdom's payments difficulties in the 1960s - the
prelude to devaluation, 1967 166
Appendix 7.2: a polemic on floating the pound , 1972 181

8 Import Functions 194

The determinants of imports 194
Imports and capacity 196
Imports and labour-market bottlenecks 200
Imports and income (estimates of income elasticities) 204
Imports and relative prices (estimates of price elasticities) 210
Import penetration 212

9 Export Functions 221

The determinants of exports 221
Exports and capacity 222
Exports and world income (estimates of income
elasticities) 229
Exports and relative prices (estimates of price elasticities) 230
The United Kingdom's declining share of world exports 238
Appendix : commodity and geographic composition of U.K.
exports and imports 244
x Contents

10 The Balance-of-Payments Equilibrium Growth Rate 250

The determination of the balance-of-payments equilibrium
growth rate 251
A fundamental law of growth 254

11 Export-led Growth 259

The importance of export-led growth 259
Models of export-led growth 263
An export-led growth model with a balance-of-payments
constraint 26~

12 The Balance of Payments as a Structural Problem 273

Non-price factors determining exports and imports 276
North Sea oil and the balance of payments 280
The balance of payments and de-industrialisation 283
Import controls 286

Notes 288
References 297
Name Index 309
Subject Index 312
Preface to the First Edition
This book has two main purposes: first, to provide an elementary
exposition of the balance of payments and of balance-of-payments
adjustment theory ; and second, to challenge the relevance of con-
ventional adjustment mechanisms as solutions to the United Kingdom 's
weak balance of payments.
Even before I started to do my own research and thinking about
the balance of payments I was sceptical of the view that countries'
balance-of-payments problems are primarily a function of adverse
relative price movements which can be rectified by adjustments to the
exchange rate . Where the scepticism originates I am not sure but it
has spilt over into dissatisfaction with most of the textbooks on the
balance of payments, and with their treatment of balance-of-payments
adjustment theory, which more or less take for granted, with various
degrees of qualification, that relative price adjustments through a
change in the exchange rate will secure for a country simultaneous
internal and external equilibrium . This was the driving intellectual
force in fashion when the world moved to floating exchange rates
in 1973 .
My scepticism has been most acute in the case of the United
Kingdom , which has experienced persistent difficulty in reconciling
balance-of-payments equilibrium with full employment and a higher
growth rate , and where, as it turns out, devaluation and currency
depreciation have been singularly unsuccessful in reconciling the
conflict.
My dissatisfaction is not confined to traditional neoclassical price-
adjustment theory . It extends to the more recent monetary approach
to the balance of payments which gives the impression, whether it
intends to or not , that all balance-of-payments difficulties have their
origin in monetary disequilibrium and that monetary correction is a
sufficient condition for internal and external balance.
xii Preface to the First Edition

It is argued strongly in the book, particularly in Chapters 10 and

11, that the balance of payments must be viewed in a growth context ,
and that at the heart of the United Kingdom's economic difficulties
is the slow rate of growth of exports, which cannot be raised perm-
anently, if at all, by exchange-rate depreciation, and is not amenable
to monetary manipulation. The problem requires real economic
policies of a structural nature related to the wider characteristics of
goods , such as their quality, design and reliability, marketing and
delivery . Exchange-rate depreciation simply ossifies the industrial
structure and makes manufacturers temporarily more competitive
in the export of goods with given characteristics which were the source
of the payments weakness in the first place. As I argue in Chapter 10,
J am convinced in my own mind that the United Kingdom's inferior
economic record in relation to other industrialised countries, particu-
larly since 1950, is largely a function of its balance-of-payments
strait-jacket caused by the slow rate of growth of exports relative to
other countries, and until this underlying weakness is rectified the
United Kingdom will be condemned to a low growth rate and the
spectre of rising unemployment whatever other domestic policies
are pursued .
Apart from being an unusual (but I hope appealing and convincing)
mixture of textbook and polemic , the book has two other features
which distinguish it from traditional texts. First, it is argued strongly
that the focus of balance-of-payments policy ought to be on the current
account. To focus elsewhere, as the new monetary approach to the
balance of payments urges, can store up trouble for countries in the
future by delaying the real economic adjustments that a country must
make ultimately . Second, for the first time in a textbook, the foreign-
trade multiplier is presented relating imports to expenditure - this
is obviously more realistic than relating imports to income if the import
coefficients of various components of domestic expenditure differ.
Finally, an attempt has been made to organise the book in such a
way that it can be used both as a preliminary text on balance-of-
payments theory and also as a source of reference on the performance
of the U.K. balance of payments which can be updated in subsequent
editions if the book proves useful to students.
In writing the book I am in debt to a number of organisations and
individuals. The Social Science Research Council financed my research
on labour-market bottlenecks and the balance of payments. During
the project several other ideas were developed . The Treasury and the
Preface to the First Edition xiii

Department of Trade and Industry provided me with statistics.

Mr John Goy and Mrs Julia Stevens of the University of Kent Library
obtained and checked references for me . The editors of Banca
Nazionale del Lavoro Quarterly Review, Applied Economics, the
Oxford Bulletin of Economics and Statistics and the National
Westminster Bank Quarterly Review gave me permission to reproduce
certain material which first appeared in their journals. Mrs June Harrison
typed the manuscript with superb efficiency. Finally, my colleagues
Richard Disney , John Craven , Charles Kennedy , Bob Dixon and Thea
Sinclair, helped to improve the book with their constructive comments
on various chapters. To all these I am deeply grateful, but they bear
no responsibility for the deficiencies which remain .

1980 A.P. T.
Preface to the Second
Edition
This revised edition, two years after the book's initial publication , is
designed to bring up to date all the tables relating to the balance of pay-
ments, imports and exports, the commodity and geographic compo sition
of trade , the exchange rate , and so on, and also the text relating to the
history of the United Kingdom balance of payments. The tables and text
now extend to the year 1980 . The new edition also gives me the oppor-
tunity to correct some minor printing errors that occurred on initial
publication.
For teaching purposes, the format of the book remains exactly the
same. As far as the central thesis of the hook is concerned, nothing has
changed to alter my conviction that the United Kingdom's industrial
weakness requires a positive and aggressive foreign trade strategy. If
anything, the relentless growth of import penetration and the loss of
jobs in manufacturing strengthen that conviction .
I am very grateful to my secretary, Hilary Daly, for help in preparing
the manuscript for press.

1982 A.P.T.
Acknowledgements
The author and publishers wish to thank the follow ing, who have kindly
given permission for the use of copyright material :

Basil Blackwell Publisher Ltd, for a table from Oxford Bulletin of

Economics and Statistics (November 1977); Cambridge University Press,
for tables from Abstracts of British Historical Statistics , by Mitchell and
Deane ; The Controller of Her Majesty's Stationery Office, for tables
from U.K. Balance of Payments, Key Statistics, and Economic Progress
Reports; Uoyds Bank Ltd , for a table from Lloyds Bank Rev iew
(January 1975), by M. Panic ; National and Commercial Banking Group
Ltd , for a table from 'Floating Rates - Expectations and Experience' ,
by Fred Hirsch and David Higham, reproduced in Three Banks Review
(June 1974); National Westminster Bank Ltd , 'for a table based on
figures from National Westminster Bank Quarterly Review (May 1975);
Organization for Economic Co-operation and Development, for a table
derived from OEeD Manpower and Labour Force Statistics ; Martin
Robertson & Company, for an extract from Modem Capitalism: Its
Growth and Transformation, by J. Cornwall (I 977).
1
Balance-of-Payments
Accounting and the
Foreign-Exchange Market
CONCEPTS OF THE BALANCE OF PAYMENTS

There are two distinct concepts of the balance of payments which must
not be confused : the market balance of payments, and the accounting
balance of payments. The market balance of payments refers to the
balance of supply and demand for a country 's currency in the foreign-
exchange market at a given rate of exchange . If the exchange rate is
fixed , the market balance of payments would be in balance only by
chance . If it is not in balance and the exchange rate must be main-
tained , the mone tary authorities would have to inte rvene to achieve
balance by buying their own currency with foreign exchange if the
home currency were in excess supply or by selling thei r own cur rency
for foreign exchange if the home currency were in excess demand . If
the exchange rate is allowed to float freely, however, the market
balance of payments must always balance because the exchange rate is
the price which equates the supply and demand for a currency in the
foreign-exchange market .
The accounting balance of payments is a record of all the financial
transactions in goods, services and capital assets which have taken place
between a country 's residents and the residents of other countries
within an arbitrary accounting period, normally one year. Like most
accounting records , this concept of the balance of payment is an
ex post concept, the record of accounts being compiled by double-
entry book-keeping. For there to be a deficit or surplus on the balance
of payments, using this concept of the balance of payments, certain
2 Balance-of-Payments Theory and U.K. Experience

sections of the account must be taken . Which sections should be looked

at is considered later when we examine the question of what is meant
by balance-of-payments equilibrium , and how equilibrium should be
defined for the sensible conduct of economic policy .
Given these two distinct concepts of the balance of payments, the
meaning of a balance-of-payments deficit or surplus will be different
in each case , and movements in the different balances might be quite
unrelated . In particular it must be stressed, and this will be a constant
theme throughout the book , that a market balance of payments does
not guarantee a balance-of-payments equilibrium in any objective sense .
It depends how equilibrium is defined within the framework of the
accounting balance of payments. As we shall see later, only if there is
no concern about the balance between the exchange of real goods and
services and complete indifference to the value of the exchange rate can
the market balance of payments be said to give balance-of-payments
equilibrium within the accounting framework . This feature of the
book distinguishes it from many others in which the belief is expressed
that if, exchange rates are left to find their own level, giving a market
balance of payments, there will be a balance-of-payments equilibrium
obviating the need for corrective action . This view is not only naive in
principle, because there are many different definitions of equilibrium,
but is contradicted by the facts - no more so than by the continual
corrective discretionary actions of governments since floating exchange
rates became prevalent in the world economy.

THE ACCOUNTING BALANCE OF PAYMENTS

The accounting balance of payments is a record of a country's financial

transactions with the rest of the world in an accounting period,
normally one year . It is constructed on the principle of double-entry
book-keeping, and as such always balances . For every credit there is a
debit, and for every debit there is a credit. The financial transactions
recorded include payments and receipts for physical goods (imports and
exports) , which are called visible items; payments and receipts for
services, which are called invisible imports and exports, and capital
transactions. Any receipt from the rest of the world which increases
the net claim on foreigners is recorded as a credit in a country's balance
of payments, but the net claim on foreigners itself is recorded as a
debit. Similarly , any payment to the rest of the world which increases
Balance-of-Payments Accounting 3
a country's net liabilities to foreigners is recorded as a debit in a
country's balance of payments, but the increase in liabilities to
foreigners itself is recorded as a credit. To see how the principle of
double-entry book-keeping works, making the accounting balance of
payments always balance, let us give some examples. First, the account-
ing balance-of-payments statement must be divided to distinguish
between transactions on current account 'above the line', and trans-
actions on capital account 'below the line'. Items above the line in the
current account affect a country's current level of national income .
Items in the capital account below the line affect a country's debtor-
creditor status, or its wealth position . Exports of goods and services
(both visible and invisible) will appear as a credit item in the current
account above the line because the demand for a country's goods
abroad affects current national income. By the principle of double-
entry book-keeping, however, the increase in the net claim on
foreigners that the exports represent will be recorded as a debit in the
capital account. For example , if payment for £100-worth of exports
from the United Kingdom to the United States is made by the U.S.
importer making a dollar deposit in New York in favour of a British
bank, the £100 of exports would be recorded as a credit above . the
line but as a capital outflow below the line . Similarly, imports of goods
and services will appear as a debit item in the current account of the
balance of payments above the line because the demand for another
country's goods also affects a country's current national income . The
increase in the net liabilities to foreigners that imports represent will
be recorded as a credit in the capital account. For example , if payment
for £ 1DO·worth of imports from the United States is made using sterling
cheques drawn on London banks, the £ 1DO-worth of imports would be
recorded as a debit above the line but as a capital inflow below the line.
Unilateral transfers between one country and another, such as
government transfers and private remittances , will also be recorded
twice even though no real goods are exchanged. For example, if a
gift is sent by an individual to a resident in another country , the gift
would be recorded as a debit in the country of origin and as a credit
in the recipient country, and then would be 'accounted for' on the
credit side of the account in the country of origin and on the debit
side of the account in the country of destination .
Capital movements involving financial transactions are also recorded
on both sides of the account below the line. The export of capital
(investment abroad) is treated as a debit, while the payment itself is
4 Balance-of-Payments Theory and U'K. Experience

recorded as a credit representing an increase in the net liabilities to

foreigners . Conversely, imports of capital are recorded as a credit and
the receipt itself is recorded as a debit involving an increase in the
net claim on foreigners.
The sale of gold and foreign currencies by the monetary authorities
is recorded as a credit, with the corresponding receipt recorded as a
debit.
With each transaction recorded twice, the sum total of debits and
credits should be equal. In other words the current and capital accounts
of the balance of payments should sum to zero. In equation form :

current account + capital account =0

or

(X - M) + (LTC + STC + G) = 0 (1.1 )

where X is exports of goods and services (visible and invisible), M is

imports of goods and services, LTC is net long-term capital flows,
STC is net short -term capital flows , and G is the net increase in
gold and foreign currency reserves (a net loss is recorded as credit and a
net gain is recorded as a debit) . It can now be clearly seen that if the
current account is in surplus, this represents a net claim on foreigners,
and this shows up in the capital account as a debit equivalent to the
export of capital or the accumulation of reserves . Similarly, if the
current account is in deficit, with a country importing more than it
exports, the country must either be borrowing abroad or losing
reserves, which show up as a credit on the capital account. A real
change on current account must always be matched by a corresponding
financial transfer on the capital account.
In practice, because of errors, omissions and so-called 'leads' and
'lags', the current and capital accounts never sum exactly to zero, and
it is this which gives rise to the balancing item in the balance of pay-
ments. The balancing item is the difference between the total value of
transactions recorded and the amounts of the supply of and demand for
foreign exchange recorded (in the United Kingdom , by the Bank of
England) . Errors may arise if the real and monetary sections of the
balance-of-payments accounts are compiled independently, as happens
to a large extent in the United Kingdom . Trade and long-term invest-
ment flows are calculated mainly from customs records, while
 Balance-of-Payments Accounting 5
monetary movements are largely derived from reports by the banking
system . Omissions may arise through the sheer difficulty of recording
all transactions, particularly short-term capital movements. Leads and
lags refer to the difference in timing between the recording of trans-
actions in goods and the recording of transactions in currencies . Traders
in goods and services, for example , may delay or hasten payment
depending on expectations about the future value of the exchange rate .
Within an accounting period this may cause a discrepancy between the
value of goods recorded and the value of currency transactions.

BALANCE·OF·PAYMENTS EQUILIBRIUM

While in a book-keeping sense the balance of payments must balance, in

an economic sense the accounts may be in severe imbalance because
individual components of the accounts may be in surplus or deficit.
This leads to the notion of balance- of-payments equilibrium/
disequilibrium, which is defined in terms of a selected group of items
within the accounting balance . The question is, which group of items?
To answer this it is helpful to ask the prior question, why does a
country bother to construct a balance-of-payments account in the
first place if it always balances? One answer is that a country wishes to
know its changing economic strength relative to other countries for the
sensible conduct of economic policies, and one measure of this is what
is happening to its balance of payments. Looked at in this way, what is
happening to exports and imports on the current account of the
balance of payments is of crucial importance because changes in
exports and imports affect a country's national income. Exports rising
relative to imports will be expansionary , while imports rising relative to
exports will be contractionary . What is happening to the balance of
overseas investment is also important because this affects employment
and the level of future income . If deficits arise on the current account
of the balance of payments, they have to be financed, and the balance-
of-payments account will show this . The accounts are useful for this
purpose, too . In fact , while there are a number of ways of defining
balance- of-payments equilibrium within the accounting balance, there
would seem to be two criteria, alluded to above, which are important
in deciding on which items to focus within the accounts . First, what
is the best measure of equilibrium for the conduct of economic policy
which is designed to maximise the long-run welfare of society? Second ,
6 Balance-of-Payments Theory and U.K. Experience

which measure shows up those items which are contributing to a

balance-of-payments surplus or deficit on the one hand and financing
the surplus or deficit on the other? Let us look at possible definitions
of balance-of-payments equilibrium in terms of these two criteria.

The Balance of Trade

The balance of trade measures the difference between earnings from

visible exports and visible imports . While it is interesting to see what
this balance is, and to note that for the United Kingdom it has been
in deficit for most of the nineteenth and twentieth centuries, focus on
the balance of trade as a measure of equilibrium would be rather
narrow, especially in a country like the United Kingdom which over the
years has developed a comparative advantage in the provision of
financial services and other 'invisible' exports.

The Balance on Current Account

The current-account balance is the difference between total exports

and total imports ('visibles' and 'invisibles'). During the mercantilist
period in Europe, equilibrium was defined in terms of the current-
account baiance. Taking the components of equation (lJ), balance-
of-payments equilibrium would be represented as

X-M=O (1.2)

Using this concept of equilibrium no account is taken of the balance

of capital movements or changes in gold and foreign currency reserves.
In mercantilist times the objective of economic policy was to maximise
the current-account surplus in order to accumulate gold . This, it was
argued, was the means to prosperity. While we now know better, focus
on the current account as a measure of equilibrium has much to
recommend it, particularly from the point of view of the conduct of
economic policy. It is what happens to the current account which
affects most directly the real income of a country and the living stan-
dards of its people, and which, if in deficit, may give rise to problems
of financing which then have implications for the future conduct of
economic policy . If the current account of the balance of payments is
integrated into the national accounts, it can be seen that the
emergence of a deficit implies that a country's expenditure exceeds its
Balance-of-Payments Accounting 7

income, and this cannot go on for ever unless financed by a continual

inflow of long-term capital investment. As debts grew, it would be
increasingly difficult to finance a current-account deficit by the issue
of monetary assets .
National income is defined as

y = C + / + (X - M) ( 1.3)

where Y is income, C is total consumption, / is investment and (X -- M)

is the current-account balance of payments. In a closed economy total
expenditure (C + f) could not exceed income (Y) . In an open economy
expenditure can exceed income if there is an import surplus (M > X) .
An import surplus (a current-account deficit) implies that a country
is 'living beyond its means' . Rearranging equation (I .3), another way
of saying the same thing would be

S-/=X-M (I.4)

where S (= Y - C) is saving. In a closed economy saving must equal

investment (in an accounting sense) . In an open economy, however,
saving and investment may diverge depending on the foreign balance.
An import surplus (M > X) implies investment in excess of saving
(financed by foreign borrowing), but this also cannot go on indefinitely
unless there is a continual inflow of direct long-term capital investment.
The danger of a country not worrying about the state of the current-
account balance, and being willing to finance a deficit on current
account by any form of capital inflow, is that if the capital inflows are
highly liquid, the country is put at the mercy of international specu-
lators, and this may then reduce the freedom of government action in
other important fields of economic and social policy. Likewise , if there
is recourse to international financial institutions such as the
International Monetary Fund (LM.F .) to support a deficit on current
account, freedom of economic action by government may also be
reduced , as it has been in the United Kingdom on several occasions in
the past.

The Basic Balance

The fact that a current-account deficit may be financed on a con-

tinuing basis by the inflow of direct long-term capital investment from
abroad gives rise to the concept of the basic balance , which is another
8 Balance-ofPayments Theory and UiK, Experience

possible measure of balance-of-payments equilibrium . This concept of

equilibrium was used by the United States until 1955 and by the U.K.
authorities up to September 1970 . Like the current balance, the basic
balance measures underlying trends in the balance of payments in
relation to the performance of the real economy , abstracting from
volatile short-term capital movements.. Using the concept of the basic
balance , equilibrium would be defined as

X -M + LTC= 0 (1.5)

As long as LTC is not volatile , and domestic economic policy is not

constrained by it, there is little to choose between defining equilibrium
either in terms of the current balance or in terms of the basic balance
from the point of view of the conduct of economic policy .
From the basic balance to the final balance of zero there has been a
long-standing discussion and controversy as to which liquid assets and
liabilities should be considered as contributing to the surplus or deficit
on the balance of payments and which should be regarded as financing
it. In the United Kingdom prior to 1970 all monetary movements,
including private short-term capital movements, were regarded as
accommodating or financing the deficit or surplus (including the
balancing item) because balance-of-payments equilibrium was defined
in terms of the basic balance . However, it is now the practice in the
United Kingdom and elsewhere to consider private short-term capital
movements (and the balancing item , in the United Kingdom) as con-
tributing to the deficit or surplus, with only official monetary trans-
actions accommodating the resulting deficit or surplus. To reflect this
change of view, the measure of equilibrium in use in the United Kingdom
up to 1980 was the balance for official financing (or total currency
flow) and is now the concept of total official financing , both of which
are akin to the concept formerly in use in the United States called the
balance of official reserve transactions.'

Total Currency Flow 2

Extending equation (1.5), the total currency flow measure of equili-

brium is defined as

X -M + LTC + STCp = 0 (1.6)

Balance-of-Payments Accounting 9

where short-term capital flows are now divided into private flows (P)
and official or compensating flows. Thus private short-term capital
movements become part of the balance-of-payments deficit or surplus
'above the line' to be financed 'below the line' by official accommo-
dating transfers. This concept of equilibrium gives a measure of the
financial pressure on the monetary authorities from the point of view
of maintaining the external value of the currency. Under floating
exchange rates the measure takes on less relevance because some of the
pressure on the exchange rate may be released by allowing the exchange
rate to fall, and the amount of official financing will be correspond-
ingly smaller. Indeed , if the exchange rate were completely free to
float , the balance of payments would always be in equilibrium in the
total currency flow sense and there would be no need for official
financing and the holding of gold and foreign currency reserves. There
would be a coincidence between the market balance of payments and
the accounting balance of payments. However, to claim an equilibrium
in the total currency flow sense when the exchange rate is free to find
its own level, the authorities would have to be completely indifferent
to the value of the exchange rate and the balance on current account.
The authorities rarely are indifferent, and currencies in practice are not
allowed to float freely. A disequilibrium balance of payments in the
total currency flow sense can then be observed . But the disequilibrium
may be very much smaller than disequilibrium in the basic balance or
current balance if a basic or current-account deficit is being matched by
a surplus of private short-term capital inflows .
The International Monetary Fund expressed the view as early as
1948 that it was of great importance to know the financial pressures on
the monetary authorities resulting from international transactions , and
how they are met. In fact , the distinction between autonomous and
compensating (or accommodating) transactions was largely developed
by the I.M.F . The distinction was supposed to provide a measure of
payments problems more relevant to the activities of the I.M.F. than
the state of the current balance of payments alone . From the individual
country's point of view, however, the danger of focusing on all
autonomous (private) transactions, as distinct from those on the
current account alone, is that the real economic problems of the
country might be disguised by large private short-term capital inflows.
Alternatively, the real economic problems may be made to appear
worse than they actually are by speculative capital outflows which
10 Balance-of-Payments Theory and U.K. Experience

bear no relation to any adverse movements in the real economy . Thus

a measure of equilibrium which indicates the pressure on the monetary
authorities in the foreign-exchange market is not necessarily the best
measure of equilibrium for focusing on problems in the real sector of
the economy. Coppock and Metcalfe (1974) have argued with respect
to the new presentation of the U.K. balance-of-payments accounts that
'problems might arise if a basic deficit were obscured for a number of
years by forces leading to a relatively persistent short-term capital
inflow which was then reversed, leaving the official reserves and
borrowing facilities vulnerable to a speculative crisis' . This is the danger
of the total currency flow measure of balance-of-payments equilibrium.
The official reasons given by the Central Statistical Office in favour
of the changed presentation from the concept of basic balance to the
total currency flow were fourfold; first, the volatility of certain capital
flows previously regarded as being long-term ones , making the distinction
between long-term and short-term capital movements arbitrary and mis-
leading ; second, the growing importance of directly related short- and
long-term flows, for example United Kingdom direct investment over-
seas financed by short-term borrowing in foreign currencies which might
be regarded as joint autonomous transactions yet were shown under the
old presentation in separate accounts; third, the growing tendency for
export and import trade credit to reflect long-term elements even though
these transactions were by convention regarded as short-term accommo-
dating movements; and fourth, it was thought that the balancing item
was difficult to fit into the old classification and could prove awkward
if the size of the balancing item was large. Table 1.1 summarises a
comparison of the old presentation of the accounts with the new, and
Table 1.2 gives an historical summary of the U.K. balance of payments
up to the present time based on the new presentation of accounts .
Tables 1.3 and 1.4 show the composition of the current and capital ac-
count.
Table 1.2 is useful not only for seeing the magnitude of the different
balances but also the divergences between the different measures of
balance-of-payments equilibrium. Particularly striking between the
years 1963 and 1968 is the much larger deficit on the balance for
official financing than on the current account of the balance of pay-
ments, reflecting a net outflow of short- and long-term capital . It is
a question that we shall have to consider later whether the real econ -
omy ought to be allowed to be upset by exchange-rate changes and
 TABLE 1.1

Summary of old and new presentations of the balance-of-payments accounts

Pre-1970 1970-80 Post-1980

Visible balance Visible balance Visible balance

+ Invisible balance + Invisible balance + Invisible balance
Current balance Current balance Current Balance
+ Balance of long-term capital + Balance of private and other + Total investment and other
autonomous capital flows capital transactions
Basic balance + Balancing item + Balancing item
+ Balancing item = Total currency flow + Allocation of Special Drawing Rights
+ Balance of monetary movements + Allocation of Special Drawing Rights Gold subscription to I.M.F .
o - Gold subscription to I.M.F . = Total Official Financing
+ Total official financing + Methods of Financing
=0 o

....
....
 TABLE 12

Summary balance of payments [Em.]

1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969

CURRENT ACCOUNT
Visible balance - 115 -401 - 140 - 100 -11 9 -543 - 260 - 108 -599 - 7 12 -209
Invisibles
Services balance +86 +8 +21 +18 -31 -50 -36 +3 1 +168 +357 +422
Interest , profits and dividends balance +262 +233 +254 +334 +398 +394 +435 +388 +378 +334 +498
Transfers balance -61 - 68 -88 -97 - 123 -159 - 169 - 18 1 - 216 - 223 -206
Invisibles balance +287 +173 +187 +255 +244 + 185 +230 +238 +330 +468 +714

CURRENT BALANCE +172 - 228 +47 +155 +125 - 358 -30 +13u -269 - 244 +505

Investment and other capital transac tio ns - 108 +286 - 3 16 -3 -100 - 3 11 -317 - 580 - 504 - 760 - 176
E.E .A. loss on forward commitments - - - - - - - - - 105 -251
Gold subscription to I.M.F . (-) - 58 - 32 - - - - - - 44

Official financing
Net transaction s with overseas mo netary
aut horities - 79 - 116 +370 -375 +5 +573 +599 +625 +556 +1296 - 699
Foreign currency borrowing (net) - - - - - - - - - - +56
Official reserves (dra wing on +/additions
to - ) +119 - 177 - 31 +183 +53 +122 -246 - 34 +1 15 +114 -44

Balancing item -46 +267 -70 +40 -83 -26 -6 - 97 +207 - 155 +358
 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

CURRENT ACCOUNT
Visible balance - 34 +190 - 748 - 2586 -5 351 -3333 - 3929 - 2284 -1542 - 34 58 +1178
Invisibles
Services balance +481 +625 +701 +786 +1083 +15 19 +2443 +3254 +3711 +4039 +4188
Interest , profits and dividends balance +554 +502 +538 +1257 +1415 +773 +1365 +104 +592 +846 - 38
Transfers balance -178 -193 -244 -438 - 420 - 480 - 760 - 1115 - 1822 - 2290 - 2122
Invisibles balance +857 +934 +995 +1605 +2078 +1812 +3048 +2243 +2481 +2595 +2028

CURRENT BALANCE +823 +1124 +247 - 98 1 - 3273 - 1521 - 881 - 41 +939 - 863 +3206

Capital transfers - - - - 59 -7 5
Investment and other capital transactions +545 +1790 - 684 +166 +1594 +134 - 3073 +4212 -4260 +2177 -147 5
Allocation of S.D.R.s (+) +171 +125 +124 - - - - - - +195 +180
Gold subscription to I.M.F . (-) - 38

Official financing
Net transactions with overseas monetary
authorities -1295 - 1817 +499 - - +984 +1113 - 1016 - 596 - 140
Foreign currency borrowing ( net) - +82 - +999 +175 1 +810 +1791 +1114 - 187 -2 50 - 941
Official reserves (drawings on +/additi ons
to -) - 125 -1536 +692 -228 -105 +655 +853 - 9 588 +2329 - 1059 - 291

Balancing item - 81 +232 - 828 +103 +108 -78 +326 +3190 +2195 +396 -539

Source : Central Statistical Office , United Kingdom Balance of Payments: 1981 Edition (London: H.M.S.O., 1981).
.....
\..u
.......
~

TABLE 1.3

Composition of the current account (£m. )

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

CREDITS
Exports ( Lo.b.) 8150 9043 9437 11937 16394 19330 25191 31728 35063 40678 47389

Services
General government 51 59 72 104 110 139 215 241 318 342 397
Private sector and publiccorporations
Sea transport 1357 161 5 1607 2055 2665 2651 3233 3433 3149 3804 3816
Civil aviation 316 354 410 480 625 780 1049 1203 1455 175 5 2210
Travel 432 500 576 726 898 1218 1768 2352 2507 2797 2965
Financial services 439 471 529 60 1 785 1025 1302 1391 1542 1601 1595
Other services 849 966 1094 1330 1653 2054 2672 3224 3671 4206 4826

Interest. profits and dividends

Generalgovernment 54 91 156 176 237 266 253 384 691 8 16 943
Private sector and publiccorporations 1398 1395 1592 2647 3049 2575 3718 3648 4457 7063 7261
Transfers
Generalgovernment - - - 67 132 366 253 298 439 550 958
Private sector 230 246 264 320 357 393 537 614 729 760 793

Total invisibles 5126 5697 6300 8506 10511 11467 15000 16788 18958 23694 25764

Total credits 13276 14740 15737 20443 26905 30797 40191 48516 54021 64372 73 153
 DEBITS
Imports ( f.o .b.) 8[84 8853 10[85 14523 21745 22663 29120 34012 36605 44136 46211

Services
General government 360 374 423 513 629 709 882 965 1016 1141 1[88
Private sector and pub lic corpo rations
Sea transp ort 1437 1673 1688 2160 2776 2568 3161 3350 3167 3683 3681
Civil aviation 270 303 340 415 540 675 840 984 1176 1467 1815
Tra vel 382 442 535 695 703 917 1068 1186 1549 2109 2757
Othe r services 514 548 601 727 1005 1479 1845 2105 2023 2066 2180

Interest , profits and dividends

General government 323 295 298 375 589 780 901 1117 1306 1392 1598
Private sector and publi c corporations 575 689 912 119[ 1282 1288 1705 2811 3250 5641 6644
Transfers
General government 177 205 210 427 455 736 1041 1413 2142 2608 2790
Private sector 231 234 298 398 454 503 509 614 848 992 1083

Total invisibles 4269 4763 5305 6901 8433 9655 11952 14545 16477 21099 23736

To tal debi ts 12453 136[6 15490 21424 30[78 32318 41072 48557 53082 65235 69947

....
v,
....0-
T ABLE 1.3 (con td)

BALANCES
Visible balance -34 +190 - 748 - 2586 -535 1 -3333 - 3929 -2284 - 1542 -3458 +1 178

Services
General government -309 - 315 -35 1 - 409 -5 19 -5 70 - 667 -724 - 698 - 799 - 79 1
Private sector and public corporations
Sea transport - 80 - 58 -8 1 - 105 - III +83 +72 +83 - 18 +121 + 135
Civil aviation +46 +51 +70 +65 +85 +105 +209 +219 +279 +288 +395
Travel +50 +58 +4 1 +31 +195 +30 1 +700 +1166 +958 +688 +208
Financial services +439 +471 +529 +601 +785 + 1025 +1302 +139 1 +1542 +1601 +1595
Other services +335 +418 +493 +603 +648 +575 +827 +1119 +1648 +2140 +2646

Inte rest, profits and dividends

General government - 269 -204 - 142 - 199 -352 -5 14 - 648 -733 -61 5 - 576 - 655
Private sector and public corporations +823 +706 +680 +1456 +1767 +1287 +20 13 +837 +1207 +1422 +617
Transfers
General government - 177 - 205 - 210 - 360 - 323 -370 - 788 - 1115 -1703 - 2058 - 1832
Private sector -I +12 - 34 - 78 - 97 - 110 +28 - -11 9 -232 -290

Invisible balance +857 +934 +995 +1605 +2078 +1812 +3048 +2243 +248 1 +2595 +2028
of which: private sector and public
corporations: services and I.P.D. +1613 +1646 +1732 +266 1 +3369 +3376 +5123 +48 15 +5616 +6260 +5596

Current balance +823 +1124 +247 - 981 - 3273 - 152 1 -88 1 - 41 +939 - 863 +3206

Source: Cent ral Stat istical Office, United Kingdom Balance of Paym ents: 1981 Edition ( London: H.M.S.O., 1981).
 TABLE 1.4
A naly sis of capital transactions and offic ial financ ing [Em.] "

1970 1971 1972 1973 1974 197 5 1976 1977 1978 1979 1980

Item
number

CURRENT BALANCE 1 +823 +1124 +247 -981 -3273 - 1521 -88 1 - 41 +939 -863 +3206

CAPITAL TRANSFERS 2 - - - - 59 -75

INVESTMENT AND orasa CAPITAL TRANSACTIONS

Overseas investment in United Kingdom
Direct 3 +363 +450 +403 +734 +854 +6 15 +799 +1326 +1292 +1818 +2094
lnvestmen t by oil companies 4 +259 +282 +78 +382 +924 +883 +819 +1131 +666 +1215 +1714
Portfolio 5 +83 +286 +290 +386 +323 -4 +435 +1853 - 85 +1253 +841
of which , British governmen t stocks 5b -1 0 +82 +108 +42 +105 -1 3 +116 +97 9 -3 +929 +568
MiseeUaneousinvestment 6 -2 -3 -4 -5 +103 +20 +35 +89 +35 +21

Total overseas investment in U.K. 7 +703 +10 15 +772 +1497 +2204 +1514 +208 8 +4 399 +1908 +4307 +4649

U.K. private investment overseas

Direct 8 -546 -676 -737 -1621 -1575 -11 71 -21 45 - 1885 -2740 -2788 -2 569
Investment by oil companiesand
miscellaneous investment 9 -151 - 139 -61 -415 -298 - 137 - 214 - 46 1 -82 1 -2858 - 1364
Portfolio 10 -1 32 - 45 -604 +276 +725 -59 +90 +12 - 1073 - 909 -2958

Total U.K. private investment overse.. 11 - 829 -860 - 1402 -1760 - 1148 - 1367 -2269 -2334 - 46 34 - 6555 -689 1

Official long-term capital 12 -206 -274 -254 - 255 -287 - 291 - 160 -303 - 336 - 40 1 -125

Import credit 13 +14 +54 +198 +349 +88 +59 +39 +297 +349 +72 -238
Export credi t 14 -395 -287 -409 - 552 - 810 - 577 - 1103 - 635 -920 - 849 -90 7

Foreign curren cy borrow ing or lending abroad

.... by U.K. banks 15 +472 +471 +471 +535 -295 +253 - 106 +364 - 433 +1623 +2024
'I
.....
00
TABLE 1.4 (contd)

Exchange reserves in sterling :

British government stocks 16 +63 +55 +65 +74 - 124 +7 +14 +5 -li S +248 +930
Banking and money market liabilities , etc . 17 +130 +658 +222 +87 +1534 -624 -1421 -24 -4 +531 +316
Other external banking and money market
liabilities in ster ling 18 +266 +709 - 91 -7 +148 +550 +255 +1481 +293 +2581 +2569

External sterling lending by U.K. banks 19 +124 +39 -241 +35 +53 +94 -348 +58 -507 +198 - 2462

Other external borrowing or lending

U.K. public secto r 20 -2 +4 +7 +19 +72 +56 +27 +750 +22 -7 -165
U.K. private sector 21 +143 +140 +139 +320 +279 +260 +109 +98 +66 +497 -938

Othe r transactio ns 22 +62 +66 -161 -176 - 120 +200 -198 +56 +51 -68 - 237

Total investment and ot her capital

transactions 23 +545 +1790 -684 +166 +1594 +134 -3073 +4212 -4260 +2177 -1475

Allocation of S.D.R.s 24 +171 +125 +124 - - - - +195 +180

Gold subscription 10 I.M.F . 25 -38

OFFICIAL FINANCING
Nel transactions with overseas
monetary authorities 26 - 1295 -1817 +449 - - +984 +1113 - 1016 -596 -140
Foreign curren cy bo rrowing (net) 27 - +82 - +999 +1751 +810 +1791 +1114 - 187 - 250 - 94 1
Official reserves (drawing on +/add itions to -) 28 -125 - 1536 +692 - 228 -lOS +655 +853 - 9588 +2329 - 1059 -291
Total official financing 29 -1420 -3271 +1141 +77 1 +1645 +1465 +3628 -736 1 +1 126 -1905 -13 72

Balancing item 30 -81 +232 - 828 +103 +108 - 78 +326 +3 190 +219 5 +396 - 539

·Assets: increase- / decrease +, Liabilities: increase +/decrease-

Sourc e: Centra l Statistical Office, United Kingdom Balance of Payments : 1981 Edition (Lo ndon : H.M.S.O., 1981).
Balance-of-Payments Accounting 19

other policies to accommodate the fickleness of those who deal in

capital . More recently , in 1974 and 1975 , the opposite was the case,
with a huge current-account deficit being partly met by net inflows
of private capital. Some would argue that the real economy should
have been adjusted immediately in response to the changes that brought
about the deficit in 1974 , and this would then have made the economy
much less subject to future speculative attacks. Table 1.2 also shows
how balance-of-payments deficits arising from all autonomous trans-
actions may be financed . There are basically three methods : support
from overseas monetary authorities ; government borrowing from
institutions like the I.M.F. and public-sector borrowing; and drawing
on official reserves. The extent to which recourse has been made to
these different methods in the United Kingdom can be seen from
Tables 1.2 and 1.4. The balance-of-payments experience of the United
Kingdom historically, and in more recent years , is discussed in Chapter 7.

MORE ON THE CONCEPT OF EQUILIBRIUM

The arbitrariness over the choice of components within the accounting

balance of payments for the measurement of balance-of-payments
equilibrium has led to such statements by economists as 'the balance
of payments deficit cannot be measured, it can only be analysed'.
While there may be agreement with this sentiment, and agreement that
more than one number should be given, none the less some definition
of equilibrium is required for the conduct of economic policy. Other-
wise, how is balance-of-payments adjustment policy to be gauged?
But having chosen the most appropriate measure of equilibrium for the
purpose at hand, is equilibrium or disequilibrium then diagnosed
regardless of the economic policies in effect at the time and independ-
ently of the achievement of other domestic objectives? Kindleberger
(1969) comes out boldly with his definition of equilibrium as 'that
state of the balance of payments of a country .. . for a given set of
parameters, which can be sustained without intervention'. This gets
close to the balance for official financing definition of equilibrium,
holding domestic policy constant : but should there be no constraints
on the parameters and other domestic goals, e.g. on the level of interest
rates, tariffs, employment, and so on? Suppose equilibrium on this
definition is achieved because very high interest rates are leading to a
20 Balance-of-Payments Theory and UiK, Experience

large inflow of private short-term capital, and at the same time high
interest rates are choking off domestic investment. Is that equilibrium?
Or suppose that equilibrium is achieved solely as a result of a very
high tariff on imports, or at the expense of high unemployment
depressing the demand for imports. Is that equilibrium?
The question being raised here is, should equilibrium on the balance
of payments be defined regardless of the level of restrictions and the
achievement of other domestic goals? Economists seem to be divided
on this matter. Some say no, but some say yes because specifying levels
of restrictions and other goals infuses value judgements into the con-
cept of equilibrium. Machlup (1950) calls such a practice 'disguised
politics' . He argues that 'by infusing a value judgement, a political
philosophy or programme, or a rejection of a programme or policy,
into the concept of equilibrium designed for economic analysis, the
analyst commits the fallacy of implicit evaluation or disguised politics'.
While it may be conceded that attaching to equilibrium some level of
restriction or some other defined goal is to make a value judgement ,
and that equilibrium , conceived analytically , is compatible with any
level of restriction, it cannot make sense for the purpose of the conduct
of economic policy to measure the balance of payments without regard
to the level of restrictions and the achievement of other goals. It is the
easiest thing in the world to achieve balance-of-payments equilibrium
on any definit ion by becoming a siege economy or running the
economy at less than full employment, but at what price? One can
sympathise with both sides of the argument, but ultimately a recon-
ciliation is called for. One such reconciliation is to use the actual
balance of payments in comparison with an estimate of the full-
employment, restriction-free balance of payments, and this would high.
light more clearly the policies that need to be pursued for the recon-
ciliation of objectives. This is the approach preferred by Meade (1951),
who draws the distinction between the 'true' and 'potential' balance of
payments, the latter being the full-employment, restriction-free balance
and the 'true' balance being the balance of autonomous trade and
transfers which must be matched by accommodating transfers . Meade
says 'it is, of course, this potential deficit (or surplus) which is the
proper measure of balance of payments disequilibrium'. Adopting this
reconciliation, there will be a different equilibrium for each level of
restriction and for each level of employment.
The same questions that have been raised about the meaning of
balance-of-payments equilibrium in the presence of restrictions, and
Balance-of-Payments Accounting 21
when economies have other goals, can also be asked about the mean-
ing of exchange-rate equilibrium : that is, about the market balance of
payments. Under fixed exchange rates, if the monetary authorities are
intervening in the market to maintain the fixed parity, the exchange
rate cannot be an equilibrium rate in the sense that there is no tendency
for it to change . But similarly, even if the exchange rate is free to find
its own level, it is hardly proper to call the prevailing rate in the free
market an equilibrium rate if at the same time trade is highly restricted
and there is depression and unemployment at home . Like the balls in
Marshall's bowl , everything in the economic system is determined
simultaneously and nothing is determined unless everything else is given.
Thus there will be a different equilibrium exchange rate for each given
set of the other variables and parameters in the economic system . If
other factors are taken into consideration, there can be no such thing as
the equilibrium exchange rate, just as there can be no unique measure
of balance- of-payments equilibrium .
Finally, we return to the question of whether equilibrium in the
foreign-exchange market gives balance-of-payments equilibrium . The
first point to make is that there is not a complete coincidence of
the items covered by the market balance of payments and the items
covered by the accounting balance of payments because the latter
includes some transactions which never pass through the foreign.
exchange market , for example unilateral transfers in kind, barter
deals and direct investments of plant and machinery. Ignoring this
factor, however, it is sometimes argued that the market balance of
payments gives an accurate indication of balance-of-payments equili-
brium or disequilibrium in the accounting sense, with the implication
that if only the exchange rate were left to find its own level, giving a
market equilibrium, this would automatically achieve balance-of-
payments equilibrium within the accounting balance. The argument
is that when the supply and demand for foreign exchange balances at
the current exchange rate, this must mean that autonomous trans-
actions requiring foreign currency payments are equal to autonomous
transactions involving foreign receipts. In the new presentation of the
U.K. balance-of-payments accounts this is tantamount to saying that
when the foreign-exchange market is in equilibrium the balance for
official financing will be zero, and thus there will be balance-of-
payments equilibrium. As was argued earlier, however, this balance-of-
payments equilibrium may be the result of net inflows of private short-
term capital , perhaps of a speculative nature, which have little, if
22 Balance-ofPayments Theory and U.K. Experience

anything, to do with the state of the real economy. An equilibrium

exchange rate in a free market may coincide with balance-of-payments
equilibrium in one sense but not in another . Changes in the balance
of payments which affect people's real living standards are changes in
the current balance and this may move into disequilibrium as the
exchange market moves towards equilibrium. When balance-of-
payments adjustment policy is considered in detail later, particularly
in Chapter 3, we will discuss many ways in which movements in the
exchange rate towards a market-clearing equilibrium may be disequili-
brating for the current balance of payments and, indeed , for the
economy as a whole .
If movements in the exchange rate do cause disequilibrium between
other items in the accounting balance of payments, this in turn will
have repercussions on the exchange rate, and there is really no way of
knowing what the long-run equilibrium exchange rate is. A cumulative
appreciation or depreciation of the exchange rate may be set in motion,
with the rate moving from one temporary equilibrium to another in an
upward or downward spiral. Inflation may produce a spiral of de-
preciation ; so may speculation against a currency. Most balance-of-
payments adjustment theory is about the movement from one tem-
porary equilibrium to another , with very little discussion about the
path of the exchange rate or the behaviour of the domestic economy
between the temporary equilibria . The assumption seems to be that the
path of the exchange rate to long-run equilibrium is smooth, but from
the experience of the world economy in the 1970s this assumption
must surely be challenged. The very idea that there is a long-run equili-
brium to which exchange rates will move may itself be a chimera .

THE FOREIGN·EXCHANGE MARKET

Behaviour in the foreign-exchange market depends on the supply of and

demand for currencies. Under a system of fixed exchange rates or
managed floating in which the monetary authorities wish to preserve
the exchange rate at a particular level, a positive total currency flow
will increase the level of foreign-currency reserves because the supply of
foreign currency to a country will exceed the demand . Conversely
a negative total currency flow will lead to a decrease in foreign-currency
reserves because the demand for foreign currency exceeds the supply.
Under a system of freely floating exchange rates a positive total
currency flow will lead to an appreciation of a country's exchange rate
Balance-of-Payments Accounting 23
(i.e. a fall in the home price per unit of foreign currency), and a neg-
ative total currency flow will lead to a depreciation of a country's
exchange rate (i.e. a rise in the home price per unit of foreign currency 3 ).
A good deal of balance-of-payments theory, and particularly the theory
of balance-of-payments adjustment policy, under both fixed and
flexible exchange rates is concerned with the influence of different
policies on the level of foreign-currency reserves and whether a policy
will rectify a loss of reserves in the case of a balance-of-payments deficit
or whether the policy will be destabilising . Here, therefore , we focus
on the foreign-exchange market , outlining the determinants of the
supply of and demand for foreign exchange, and the conditions for
the market to be stable in response to a change in the home price of
foreign exchange.
The shape and position of the supply and demand curves for foreign
currency are determined by the nature and magnitude of the underlying
transactions. Taking the current account of the balance of payments we
can show how the curves can be derived from the supply and demand
curves for imports and exports. For illustrative purposes we take the
dollar ($ ) as the foreign currency and consider the supply and demand
for dollars in relation to the sterling (£) price of dollars. In working out
what happens when a change in the exchange rate takes place it matters
whether the analysis is carried out measuring the effects in the home
currency (sterling) or in the foreign currency . Since the normal balance-
of-payments problem is one of rectifying a foreign-exchange shortage ,
the analysis here will be undertaken measuring changes in terms of
dollars. If interest were in the domestic repercussions of exchange-rate
changes, however, it would be more appropriate to measure the effect
of exchange-rate changes in terms of the home currency , as
Joan Robinson (1937) did in her early classic paper on the foreign
exchanges. At that time, however, the economic problem was recession
rather than balance-of-payments difficulties. We concentrate first on
the demand for and supply of foreign currency for trading purposes.

THE DEMAND FOR DOLLARS

The quantity demanded of dollars in relation to the sterling price of

dollars can be derived from the U.K. demand for imports from the
United States and the U.S. supply curve of imports. If it is assumed for
simplicity that the supply curve of U.S. goods is infinitely elastic (so
24 Balance-ofPayments Theory and U.K. Experience

that the dollar price of imports does not rise as the demand for imports
from the United States increases), the elasticity of demand for dollars
with respect to a change in the sterling price of dollars will equal the
U.K. elasticity of demand for U.S. goods.
Imports may be either competitive or non-competitive with domes-
tic goods . Competitive imports fill the gap between the domestic
demand for a commodity and its domestic supply. For most of these
goods the demand for imports will be more elastic than the demand for
the goods that the imports compete with (importables) because as the
domestic price of the goods falls not only does domestic demand
increase but also supply contracts. Exceptions to this would be if
domestic supply is completely inelastic or if domestic supply becomes
non-existent because the price is too low . Figure 1.1 illustrates these
propositions.

p
5

(a )

FIGURE 1.1

Take the importable commodity, steel. Figure 1.1 (a) shows the
domestic demand for steel and the domestic supply in relation to price .
Figure 1.1 (b) shows the demand for imported steel. At price PI the
total domestic demand for steel is met by domestic supply . At prices
lower than PI the demand for steel exceeds the domestic supply, and
the demand for steel imports to make up the difference is greater than
the increase in demand for steel itself because domestic supply con -
tracts. At price Pz no domestic supplier is willing to supply steel , and
the demand for imported steel then has the same elasticity as the
Balance-of-Payments Accounting 25
demand for the importable, as shown by the same slope of the two
demand curves below P2 • Above P2 , however, the demand curve for
imported steel is more elastic than the total demand for steel. It is
easily shown that the elasticity of demand for competitive imports
is a weighted sum of the domestic demand and supply elasticities. Let
M= D -S and

dD dS)
(
M ar; dP m

where Em is the elasticity of demand for imports and Pm is the dom-

estic price of imports. Therefore

E
m
= (!!-M"')M
x +(~- x s )
d '\
m

where d m is the domestic demand elasticity, and sm is the domestic

supply elasticity . The smaller the amount imported relative to total
domestic demand, the higher Em will be relative to d m . Only if there
is no domestic supply will the elasticity of demand for imports equal
the domestic demand elasticity for the commodity. For non-
competitive imports the demand curve will depend on the good's own
characteristics and the structural relationships that exist between
domestic output and necessary inputs that must be imported .
Aggregating the demands for all imports and converting into foreign
currency ($) gives the demand for foreign exchange at a given rate of
exchange . As the home price of foreign exchange falls, the demand for
foreign exchange may be expected to rise, the elasticity of demand
depending on the elasticity of demand for imports. The demand for
dollars in relation to the sterling price of dollars is shown in Figure 1.2.
In interpreting the curve it is helpful to remember that a fall in the
sterling price of the dollar implies an appreciation of sterling and a
depreciation of the dollar, which means that the sterling price of dollar
goods will fall in the home market and more imports and dollars
will be demanded. Conversely, a rise in the sterling price of dollars
means a depreciation of sterling and an appreciation of the dollar,
which will raise the sterling price of dollar goods and reduce the
demand for imports and dollars.
26 Balance-of-Payments Theory and U.K. Experience

Sterling price
of $
D~

~D Demand for $

FIGURE 1.2

THE SUPPL Y OF DOLLARS

On to Figure 1.2 can be superimposed the supply curve of dollars to

give an equilibrium sterling price of dollars at which the supply of and
demand for dollars is equal in the foreign-exchange market. The supply
of dollars depends on the U.S. demand for U.K. exports or, in other
words, on the U.S. demand to import goods from the United Kingdom.
If the demand for U.K. goods in the U.S. market is of unitary elasticity
(if demand changes in the same proportion as the dollar price of U.K.
goods does), the supply of dollars will remain unchanged with respect to
a change in the sterling price of dollars. The supply curve of dollars
will be vertical (SISI in Figure 1.3). If the elasticity of demand for
U.K. goods is greater than unity (i.e . if the demand changes more than
in proportion to price), the supply curve of dollars will be positively
sloped in relation to the sterling price of dollars (S2S2) ' If the elas-
ticity of demand for U.K. goods is less than unity (i .e. if the demand
changes less than in proportion to price), the supply curve of dollars
will be negatively sloped in relation to the sterlirig price of dollars
(S3S3)'
In interpreting the supply curves it is again helpful to remember
what a change in the sterling price of the dollar means. If the sterling
price of the dollar rises, this implies a depreciation of sterling and an
appreciation of the dollar. The dollar price of U.K. goods in the U.S.
market will fall if the sterling price of goods remains unchanged. If
the demand for U.K. goods then rises in proportion to the fall in the
dollar price, the supply of dollars will remain unchanged (supply curve
Balance-of-Payments Accounting 27

Sterling price
of$ 5,
52

52

5,
Supply of $

FIGURE 1.3

SISI ; if the demand for U.K. goods rises more than in proportion,
the supply of dollars increases (S2S2) ; and if the demand for U.K.
goods rises less than in proportion, the supply of dollars decreases
(S3S3)' The argument is reversed for a fall in the sterling price of
dollars, or an appreciation of sterling .
Combining Figures 1.2 and 1.3 the equilibrium sterling price of
dollars is obtained at PI in Figure 104 . In Figure 104 two negatively
sloped supply curves of dollars are drawn , one steeper (S3aS3a) than
the demand curve for dollars and the other flatter (S3bS3b). The
importance of this difference will become apparent when we consider
the conditions for the foreign-exchange market to be stable .
From Figure 104 it is clear that if the sterling price of the dollar
is free to vary, as under a system of freely floating exchange rates,
the foreign-exchange market will be cleared . The market balance of
payments will be in equilibrium and the total currency flow will be
zero . The theory of how exchange-rate changes are supposed to pre-
serve balance -of-payments equilibrium in an accounting sense is then
as follows . Starting from the equilibrium rate of exchange PI an in-
crease in import demand in the United Kingdom will increase the
demand for dollars and cause the U.K. balance of payments to go into
deficit. In 'normal' circumstances this causes the sterling price of the
dollar to rise, which will make the sterling price of imports from the
United States dearer and the dollar price of U.K. exports to the United
States cheaper, which should tend to rectify the imbalance . Conversely ,
28 Balance-of-Payments Theory and U.K. Experience

Sterling price
Of $
D 5~
\
5,
\ 5,
\
\

P, I------~~

, D
53.
5,
Supply of and demand for $

FIGURE 104

an increase in the demand for U.K. exports will increase the supply of
dollars and cause the balance of payments to go into surplus, reducing
the sterling price of dollars. This will make imports cheaper and exports
dearer , which again should tend to rectify the imbalance.
It is clear from Figure lA , however, that a change in the exchange
rate will not rectify imbalance in all circumstances ; it depends on the
relative slopes of the demand and supply curves of foreign exchange.
Starting from a position of imbalance where the demand for dollars
exceeds the supply, a rise in the sterling price of dollars will restore
balance as long as the demand for dollars is reduced by more than
the supply . This will be so as long as the supply curve of dollars is
either positively sloped or not more negatively sloped than the demand
curve. Likewise, starting from a position of imbalance where the supply
of dollars exceeds the demand, a fall in the sterling price of dollars will
restore balance as long as the demand for dollars is Increased by more
than the supply. This condition will also be met as long as the supply
curve of dollars is either positively sloped or not flatter than the
demand curve if it is negatively sloped. Stability of the exchange rate
when there is disequilibrium in the market thus depends on the relative
slopes of the supply and demand curves for foreign exchange. In Figure
lA, if the supply curve of dollars is SJ bSJ b a rise in the sterling price
of dollars , when there is an excess demand for dollars, would create
even greater excess demand, and a fall in the sterling price of dollars
when there is an excess supply of dollars would create an even greater
Balance-of-Payments Accounting 29
excess supply. In these circumstances variations in the exchange rate
will not produce equilibrium and the market will be unstable. Note that
for stability it does not matter that the demand for U.K. exports is
inelastic, provided that the inelasticity does not give a supply curve of
dollars with a flatter negative slope than the demand curve for dollars.
We shall derive more precisely in Chapter 3 the precise elasticity con-
ditions with respect to the demand for exports and imports for a de-
preciation of the exchange rate to improve the balance of payments
starting from equilibrium (the so -called Marshall-Lerner condition).
It is possible, of course, for the elasticity of demand for U.K .
exports to the United State to be elastic over one range of prices and
inelastic over another range of prices , so that the foreign-exchange
market is stable at one rate of exchange but unstable at another rate
of exchange. Suppose, for example , that the demand for U.K. exports
is elastic at a low sterling price of dollars (i .e. at a high dollar price
for U.K. goods) but inelastic at a high sterling price of dollars (i.e .
at a low dollar price for U.K. goods). The supply curve of dollars
would then look like that drawn in Figure 1.5, bending back on itself
at the point of unitary elasticity of demand for exports (point L). PI
is a stable equilibrium exchange rate , but any depreciation of the
currency beyond point L will either reduce a surplus (from L to P2 ) ,
or increase a deficit (beyond P2 ) . A freely floating rate will tend to
move away in either direction from A (to PI in the downward
direction).
The discussion so far has been about 'static' stability and instability .
The foreign-exchange market may also be sub ject to 'dynamic ' in-
stability caused by the supply and demand curves themselves shifting

Sterling price D
of $

5 D

Supply and demand for $

FIG URE 1.5

30 Balance-of-Payments Theory and U.K. Experience
in a destabilising way as the exchange rate changes . Indeed , this type
of instability may be more common in practice. Speculation , and
domestic inflation sparked off by the rapid depreciation of a currency ,
may be sources of dynamic instability. Consider Figure 1.6, which
shows the supply and demand for dollars in relation to the sterling
price of dollars.

Sterling price D,
-,
of $
D , ,, s
,

P, ,
,
P ,
<,

s D,
D

Supply and demand fo r $

FIGURE 1.6

Suppose that at exchange rate P there is an excess demand for

dollars equal to xy. To rectify this disequilibrium sterling is either
devalued or allowed to depreciate to PI ' If the demand and supply
curves (DD and SS) do not shift, equilibrium will be achieved . How-
ever, suppose that depreciation of the currency generates so much
domestic inflation that the supply curve of dollars from the sale of
exports shifts from SS to SISI ' At the exchange rate PI there is
now a greater excess demand for dollars, ab, than there was at P.
Alternatively, suppose currency depreciation generates so much
pessimism among currency traders that instead of demanding more
pounds and less dollars, they demand more dollars, shifting the demand
curve for dollars from DD to DID I • At the exchange rate PI, with the
old supply curve SS, there is now a greater excess demand for dollars,
be, than there was at P. If both the supply and demand curves shift,
the excess demand for dollars at the lower sterling exchange rate would
be ac, greater than before the depreciation. It should also be
mentioned, of course , that income expansion in the 'depreciating'
country will also increase the demand for dollars , offsetting some of the
gain from depreciation.
 Balance-of-Payments Accounting 31
CAPITAL MOVEMENTS AND SPECULATION

So far we have been concerned with the supply of and demand for
foreign exchange arising from the international exchange of goods and
services. Transactions in capital and currency speculation also give rise
to the supply of and demand for foreign exchange and affect the
current (spot) rate of exchange. The wish of residents of a country to
export capital and acquire foreign assets such as bonds , equities, prop-
erty, and so on, leads to a demand for foreign exchange, while the
desire of foreigners to acquire capital assets in the home country adds
to the supply of foreign exchange. The effect of a net inflow of capital
on the exchange rate will be to cause the exchange rate to appreciate
in the same way as a trade surplus ; and a net outflow of capital will
cause the exchange rate to depreciate in the same way as a trade deficit.
If the exchange rate is adjustable, or free to vary of its own accord ,
the buying and selling of currencies for profit will cause variations in
the demand for and supply of foreign currency. The essence of specu-
lative activity , as it is called , is that traders will buy a currency when its
price is 'low ' if they expect the price to rise, and sell a currency when
its price is 'high' if they expect the price to fall. If speculators forecast
correctly , they make a profit and , so it is argued, stabilise the exchange
rate at the same time by adding to demand when price is 'low ' and
adding to supply when price is 'high'. The rate of return from specu-
lative activity can be very high. An example will illustrate . Suppose the
current price of a dollar is 50 pence and the speculator expects the
price to rise to 52 pence, so he buys. If he held the dollars for a week,
before selling, and assuming no costs , he would make a return of 4
per cent on the week, or approximately 200 per cent per annum.
The notion that speculation will on balance be stabilising rests on
the assumption that speculators are profit-maximisers who must buy
cheaper than they sell, and if they do the current exchange rate will be
brought closer to its long-run average. For speculation to be destabi-
lising, it is argued, speculators would have to lose money and would
eventually go out of business. In his classic paper Friedman (1953)
says: 'people who argue that speculation is generally destabilising
seldom realise that this is largely equivalent to saying that speculators
lose money, since speculation can be destabilising in general only
if speculators on the average sell when the currency is low in price and
buy when it is high'. Friedman also admits , however , that 'it does not ,
of course , follow , that speculation is not destabilising ; professional
speculators might, on the average, make money while a changing body
32 Balance-of-Payments Theory and U.K. Experience

of amateurs regularly lose large sums'. He also adds that his remarks
that speculation must be stabilising if it is to be profitable are

a simplified generalisation of a complex problem . . . A full analysis

encounters difficulties in separating 'speculative' from other trans-
actions, defining precisely and satisfactorily 'destabilising specu-
lation' and taking account of the effects of the mere existence of a
system of flexible rates as contrasted with the effects of actual
speculative transactions under such a system .

It can in fact be shown that profitability is neither a necessary nor a

sufficient condition for speculation to be stabilising. It is not necessary
because perfectly competitive speculation under perfect foresight
would lead to perfect stability of the exchange rate but zero profits.
It is not sufficient, either, if the exchange market exhibits multiple
equilibria in the absence of speculation with unstable regions in
between (like Figure 1.5 above if the supply curve were to bend back
on itself again and cut the demand curve from below). In this case
profitable speculation could cause fluctuations that would not have
occurred in its absence . Thus profitable speculation need not always be
stabilising, and destabilising speculation can be profitable.
Ultimately the question of whether speculation is stabilising or not
is an empirical one : the matter cannot be decided by theoretical reason-
ing alone. The empirical evidence, however, may be very difficult to
interpret. The major difficulty is knowing what a freely fluctuating
exchange rate would have been in the absence of speculation . How do
we know, for example, whether speculation has been stabilising or
destabilising since the general abandonment of fixed exchange rates in
the world economy in 1973? The fact that exchange rates have shown
instability since 1973 may be the result of destabilising speculation,
or may reflect the fact that the 'non-speculative' rate is subject to
variation which was not allowed to express itself under the former fixed
exchange-rate system . The instability could also be part of a process of
adjustment towards a more stable long-run equilibrium after a period of
exchange-rate rigidity .
Whether or not speculative activity is stabilising in the long run,
there is no doubt that speculative activity can be destabilising and very
harmful to economies in the short run, especially when the speculative
activity starts a spiral of depreciation necessitating domestic economic
policy inappropriate to the functioning of the real economy. If specu-
Balance-of-Payments Accounting 33
lative capital movements could be controlled, the harm done to the
domestic economy would constitute a powerful argument for doing so .
lt is, of course, absolutely vital to be able to control capital movements
in a fixed exchange-rate system unless the foreign-currency reserves can
take the strain of a speculative outflow. One of the reasons the par-
value system of exchange rates established at Bretton Woods broke
down from 1971 onwards was the inability of countries to defend the
par values of their currencies in the face of movements of short-term
capital, the volume of which grew so rapidly in the 1960s (Lamfalussy,
1976) . But those who argued that under a system of flexible exchange
rates movements around the trend would be limited (because specu-
lation would be stabilising?) seem to have been confounded by
experience , at least that of the 1970s.
Government transactions also affect the supply of and demand for
foreign currency. Governments provide foreign currency to other
countries, thus depreciating their own currency, and themselves receive
foreign loans, which appreciate their currency. The central banks of
countries also intervene on behalf of governments to stabilise exchange
rates - buying foreign currencies to stop their own currency from
appreciating and selling foreign currencies to prevent their own
currency from depreciating.

THE FORWARD RATE OF EXCHANGE

The rate of exchange that we have been considering so far is the so-
called 'spot' rate of exchange, which is determined in the free market
by the immediate supply of and demand for currencies. Many inter-
national transactions, however , do not require immediate settlement
but instead involve contracts to pay in the future . To facilitate the
implementation of contractual arrangements involving future obli-
gations there exists in addition a forward exchange market at which
traders can buy and sell currencies 'forward' at an agreed rate . Thus
whereas in the spot market the buyers and sellers of foreign currency
accept the ruling market price, in the forward market the buyers and
sellers agree to a price in advance, sometimes more than a year into the
future but usually less. The two markets, and the two rates of
exchange , are linked by three main factors : interest-rate differences
between countries ; speculation; and 'hedging' and 'covering'.
If there were no cost or risk in the transfer of funds from one
34 Balance-of-Payments Theory and U.K. Experience

country to another , and interest rates were the same at home and
abroad, there would be no difference between the spot and forward
rates of exchange within and between countries. For example , if the
forward rate was higher than the spot rate , it would pay to purchase
curren cy spot and sell it simultaneously at the higher forward rate ,
investing the funds in a foreign bank (paying the same interest rate as
domestic banks) until the forward contract expired. By a process of
arbitrage, i.e. the switching of funds between the two markets , the two
rates would be equalised.
If the rate of interest on monetary deposits differs between two
countries, however, the forward rate will tend to diverge from the spot
rate, being higher (at a premium) in the country with the lower interest
rate, and lower (at a discount) in the country with the higher interest
rate . As a first-order approximation, ignoring transfer charges, risks
and other factors affecting the forward rate, the currency will sell
forward at a premium or discount equal to the interest-rate differ-
ential.
Suppose that the spot rate between the dollar and the pound is
$ 2 to £1, and that the interest rate is higher in the United Kingdom
than in the United States. It would then pay investors to switch funds
from the United States to the United Kingdom provided that the
interest gain is not offset by a lower rate of conversion of sterling to
dollars when the investment matures than when the original investment
in sterling assets was made. To ensure against this contingency investors
will sell pounds (and buy dollars) forward at an agreed rate . The for-
ward rate of exchange will fall below the spot rate of $ 2 to £1 , and
funds will tend to move from the United States to the United Kingdom
until the differential yield on investment in the United Kingdom is
just offset by the forward exchange conversion loss in the United
Kingdom . The equilibrium forward rate of exchange giving 'interest-
rate parity' between the two currencies is given by the formula

(1 .7)

where R F is the forward rate of exchange in the United Kingdom,

R s is the spot rate, ius is the interest rate in the United States, and
i UK is the interest rate in the United Kingdom. For example , supposing
iUS = 0.05;i UK = 0.1 andR S = $2 :£1 , then
Balance-of-Payments Accounting 35

R = (I + 0.05) x $2 $2.1 "" $1.9 :£1

F
(I + 0.1) £1 £l.l

As a rate of discount against spot sterling this is (2 - 1.9)/2 "" 0 .05 ,

or 5 per cent. If the discount were less, it would continue to pay U.S.
investors to switch their funds to the United Kingdom. In practice
funds move when the difference between the forward discount or
premium and the interest differential is minutely small, perhaps as
narrow as 1/100 of 1 per cent.
If the interest-rate differential between the two countries were the
only factor affecting the relationship between the forward and spot
rates of exchange, the discount (or premium) on sterling should equal
the interest-rate difference. In practice other factors also influence the
forward rate of exchange and its relation to the spot rate .
Risk and liquidity considerations will affect the forward rate of
exchange and its relationship to the spot rate . Importers who have to
pay in foreign currency in the future, and exporters who expect to be
paid in foreign currency in the future, may wish to 'cover' against the
risk of exchange-rate changes. For example , a U.K. exporter may
expect to be paid $100 in three months' time but be worried that in
the meantime the dollar will depreciate and sterling appreciate so that
he would get less pounds for his $100. He may therefore decide to sell
forward his $100 at an agreed rate knowing exactly how much sterling
he will then receive in three months' time when he delivers the $100.
Selling dollars forward in anticipation of a depreciation of the dollar
will influence the forward rate of exchange . Likewise, a U.K. importer
may have to pay a supplier in dollars in three months' time and be
worried that the dollar will appreciate and sterling depreciate in the
meantime , involving him in a higher sterling payment for a given
amount of dollars . He may therefore decide to buy dollars forward
knowing exactly how much sterling he must pay in three months'
time when he collects the dollars. The importer may , of course, decide
to pay immediately or expect sterling to appreciate, in which case it
would be the supplier who would presumably attempt to cover.
Expectations that traders hold about future exchange-rate movements
are a major contributor to leads and lags and to the balancing item in
the balance-of-payments accounts that we discussed earlier.
Forward cover is also sought by business firms holding foreign
assets who are obliged for accounting reasons to convert the value of
their foreign assets and liabilities at the current spot rate of exchange .
36 Balance-of-Payments Theory and U'K. Experience

To avoid exchange risks firms may continually cover. This practice is

called 'hedging', and tends to be insensitive to expected exchange -
rate changes .
Speculation is another factor that affects the relation between the
spot and forward rates of exchange . If a speculator believes that the
future spot rate will be higher than the current forward rate , the 'bull'
speculator will buy forward in order to sell spot at the higher rate
than he bought forward. For example, if the current forward rate is
$1 .9 to £1 and the expected spot rate is $2 to £1 , a profit will be made
by speculators who agree to buy sterling forward at $1 .9 per t, and
who then sell the sterling for $2 per £. The forward rate will tend to
move up towards the expected spot rate, and if this encourages buying
in the spot market, expectations about the future spot rate will become
self-fulfilling . Conversely, a speculator who expects the spot rate in the
future to be below the current forward rate will sell sterling forward
in order to buy spot at a lower rate than he sold forward . Suppose the
current forward rate is $1.9 to £1 and the expected spot rate is $1.8
to £1 . A profit will be made by speculators who agree to sell sterling
forward at $1.9 per £ , and who then buy sterling back again at $1.8 per
£. In this case the forward rate will tend to move down towards the
expected spot rate , and if this encourages selling in the spot market,
expe ctations about the future spot rate will again be self-fulfilling.
If the spot rate of exchange is fixed within limits but the forward
rate is free to vary without limit (as was the case under the Articles of
Agreement of the I.M.F .), a situation may arise where the covering
operations of commercial traders in forward markets depreciates the
forward rate of exchange in excess of the interest-rate differential.
It then becomes profitable to buy foreign currency spot and sell it
forward, exerting considerable pressure on the rate of exchange of the
home country, which under a system of fixed exchange rates must be
preserved by the monetary authorities using its reserves and selling
foreign currency.
Alternatively , the monetary authorities may intervene to support
the forward rate of exchange by forward sales of foreign exchange to
reduce the forward discount for domestic currency . The authorities
would, of course , have to deliver the foreign exchange at a later date
but in the process some foreign-exchange loss that would otherwise
have occurred might be prevented. If, however, the intervention is
unsuccessful , and the currency has to be devalued (or depreciates
further under a system of flexible exchange rates), the monetary
Balance-of-Payments Accounting 37
authorities stand to lose heavily. The U.K. Exchange Equalisation
Account lost nearly £400 million in its attempt to prop up the forward
value of sterling before the devaluation of sterling in November 1967.
Some would argue that many of the U.K. foreign-exchange crises have
not been predominantly of a purely speculative nature but the result
of interest arbitrage arising from the spread between the spot and for-
ward rates of exchange. In theory stabilisation of the spot rate should
give stability to the forward rate if arbitrage funds are perfectly elastic
with respect to the (covered) interest differential. In practice, however ,
they are not ; and there may be controls. In these circumstances it may
be preferable for the authorities to stabilise the forward rate in order
to keep the spot rate stable.

TYPES OF EXCHANGE- RATE REGIME

Four main types of exchange-rate regime can be distinguished, all of

which have operated in one form or another. Proposals for alternative
exchange-rate regimes fall within one or other of the four main cate-
gories of regime distinguished, though some schemes comprise features
of more than one of the main types . The four main types of regime that
we shall discuss briefly are as follows : (I) rigidly fixed exchange rates
epitomised by the old gold bullion standard ; (ii) the adjustable peg or
par-value system, which was the system which governed international
monetary relations between 1944 and the early 1970s, and is some-
times referred to as the gold exchange standard ; (iii) managed exchange-
rate flexibility, which is the exchange-rate regime currently practised by
most countries; and (iv) freely floating exchange rates, which, with the
exception of some European countries in the nineteenth century,
have not been in operation anywhere in the world for any length of
time .

(i) Rigidly Fixed Exchange Rates: the Gold Bullion Standard

Under a fixed exchange-rate system, the use of exchange-rate changes

for balance-of-payments adjustment is ruled out by definition. Balance-
of-payments adjustment must take place through price and income
adjustments in the domestic economy. The gold bullion standard,
which was in operation in its purest form between 1880 and 1914 , was
a fixed exchange-rate system under which balance-of-payments adjust-
38 Balance-of-Payments Theory and U.K. Experience
ment was supposed to be achieved automatically through internal
adjustment provided certain rules were adhered to . Under the gold
bullion standard each country fixed its currency in terms of gold,
each country's currency was convertible into gold," and international
settlements were made in gold. The exchange rate between two
countries could not vary outside the so-called 'gold-export points'
determined by the cost of shipping gold from one country to another.
Suppose, for example, that the same amount of gold is bought and sold
by the monetary authorities in country X for $2 and in country Y for
£1. The exchange rate between country X and country Y would then
be $2 to £1 . Now suppose that country Y is in balance-of-payments
deficit and there is an excess supply of pounds. There will be a
tendency for the exchange rate to fall below $2 to £1 . But the move-
ment cannot go far since people in country Y with debts to pay to
creditors in country X could always take £1 to their own monetary
authority, obtain gold with it, export it to country X and get $ for it.
There are costs involved in doing this, however, so that it would only be
profitable to export gold from Y to X if the exchange rate falls outside
the 'cost margin' around the fixed rate . For example, suppose the cost
of obtaining gold in Y, shipping it to X and converting it into dollars is
2 cents per $2 of gold. It would then pay to export gold to obtain
dollars only if the rate of exchange fell below $1.98 to £1. Otherwise
more dolIars could be obtained per pound by converting pounds into
dollars directly. $1.98 is referred to as country Y's gold-export point or
X's gold-import point. Similarly , if it cost 2 cents per $2 to export gold
from X to Y, it would pay to export gold to obtain pounds only if the
rate of exchange rose above $2 .02 to £1. This would be X's gold-export
point or Y's gold-import point. Thus under the gold standard the
exchange rate may vary within the gold points . Outside these points ,
however, there will be no variation in the exchange rate; instead, gold
will flow to accommodate any surplus or deficit on the balance of
payments.
Then , according to the rules of the gold standard, if country Y is
in balance-of-payments deficit, gold should flow from country Y to
country X, causing first a contraction of the money supply in Yand
an expansion of the money supply in X . As a consequence interest
rates should rise in Y, and wages and prices fall, while interest rates
should fall in X , and wages and prices rise, until external equilibrium
is achieved. External equilibrium does not , of course, guarantee internal
Balance-of-Payments Accounting 39
balance in the sense of full employment unless there is an appropriate
downward adjustment of the money-wage rate in relat ion to the price
level in the deficit country . For both external and internal balance in
the deficit country, a reduction in real wages is required . The gold
standard mechanism uses, in effect, financial policy for external balance
and wage flexibility for internal balance .
The workings of the old gold standard were never perfect and were
modified through time by several factors . Research shows that actual
gold movements in relation to the size of balance-of-payments deficits
were very small. First, there was never a one-to-one correspondence
between changes in the supply of gold and changes in a country's
money supply. There never could be except under a strict gold specie
system where the money supply itself consists of gold , or where the
money supply is backed 100 per cent by gold . In practice small changes
in gold brought about large changes in the money supply , explaining
why large balance-of-payments deficits were accompanied by gold
movements which were very small in comparison . Second, income
adjustments frequently took place, at times of bad harvests for
example, reducing or eliminating deficits that would otherwise have
arisen and necessitated the export of gold. Third, larger and larger
capital flows occurred in response to interest-rate differences, helping
to finance balance-of-payments deficits on current account without the
export of gold .
The fixed exchange-rate system imposed by the old gold standard
ultimately broke down in 1931 , largely because the rules of the gold
standard were not being adhered to.' Countries such as the United
States and France which were accumulating gold were not pursuing
sufficiently expansionary policies, and Brita in, which was losing gold ,
was not willing to pursue further deflationary act ion to stop the gold
outflow. With her gold reserves virtually exhausted , Britain suspended
the gold standard, and for a time some of the major currencies floated
against each other. The period was one of great international economic
turmoil characterised by world depression, a lack of international
liquidity and acceptable reserve currencies for countries to finance
balance-of-payments disequilibrium, and trade restrictions for balance-
of-payments protection which worsened the world depression . It was
to overcome these features of the international economy of the 1930s
which led to the Bretton Woods agreement and the establishment of
the International Monetary Fund in 1944.
40 Balance-of-Payments Theory and U.K. Experience

(ii) The Adjustable Peg or Par-Value System

The agreement at Bretton Woods in 1944 re-established fixed (but

adjustable) exchange rates between currencies in the form of so-called
'par values', defined in terms of gold. Countries were obliged to inter-
vene in the market to maintain the exchange rate within I per cent
either side of its par value. Provision was made, however , that par values
could be adjusted by up to 10 per cent without permission of the
I.M.F. and by more than 10 per cent if it was convinced that the bal-
ance of payments of the country concerned was in fundamental dis-
equilibrium. Fundamental disequilibrium was never defined precisely.
The system evolved was a gold exchange standard in which domestic
currencies were not convertible into gold as under a gold bullion
standard, but in which the monetary authorities undertook to buy and
sell the currency of another country which was itself operating a gold
bullion standard. Under the Bretton Woods system national currencies
were convertible into the dollar , and the dollar itself was convertible
into gold at $35 per ounce . The par value of sterling in relation to the
dollar was $4 .03 to £1. In effect the dollar replaced gold as the reserve
asset of the international monetary system .
In theory balance-of-payments adjustment under a gold exchange
system is supposed to work in the same way as under the gold standard ,
with deficit countries allowing prices to fall and interest rates to rise,
while surplus countries allow prices to rise and interest rates to fall.
In the Articles of Agreement of the Bretton Woods system , however,
it was clearly recognised that internal price flexibility could not be
relied on and that if provisions were not made it would be income and
employment that would bear the brunt of adjustment, as in the inter-
war years. Two provisions were made to safeguard full employment.
First, a country in temporary disequilibrium could borrow from the
I.M.F. to tide it over its difficulties without having to pursue internal
adjustment policies which threatened employment. Second , provision
was made for exchange-rate adjustment - in practice this was rarely
used. Here we cannot go into the many reasons, some no doubt of a
political and psychological nature , why countries were reluctant to alter
the par value of their currencies - which was quite permissible under
the I.M.F. Articles of Agreement. Perhaps, however, their economic
instinct and judgement against adjustment was correct, for variable
exchange rates can only rectify payments disequilibrium without
sacrificing employment if there is sufficient flexibility of the real-wage
Balance-ofPayments Accounting 41

rate, i.e, wage-earners must accept a cut in real wages which a reduction
in the real terms of trade through devaluation implies . If money wages
chase prices upwards, there can be no change in the real terms of trade
and the balance-of-payments disequilibrium remains . Despite the
criticisms levelled against countries in the 1950s and 1960s for main-
taining fixed exchange rates at the apparent sacrifice of domestic
employment, the movement from fixed to flexible exchange rates in
the 1970s has not been conspicuously successful in reconciling pay -
ments balance with full employment and has undoubtedly exacerbated
inflation.
The workability of a gold exchange standard depends both on the
maintenance of confidence in the currency that is being used as the
reserve asset, and also on convertibility into gold. Under the Bretton
Woods system the dollar became the key currency , and because of a
general shortage of world liquidity it was convenient for the rest of
the world to allow the United States to run payments deficits to
augment the supply of international reserves. As dollar liabilities
began to mount, however , increasingly the ability of the United States
to maintain the convertibility of the dollar into gold became to be
doubted. The loss of confidence in the dollar eventually became such
that on IS August 1971 the convertibility of the dollar into gold was
suspended and the gold exchange standard that had governed inter-
national monetary transactions for over twenty-five years was formall y
ended. For a short time currencies floated against each other until
par values were restored again by the Smithsonian Agreement in
December 1971 which raised the price of gold to $38 per ounce and
devalued the dollar correspondingly by approximately 8 per cent , and
widened the permissible fluctuations around the par values from I per
cent to 2.25 per cent. The wider bands, however , were not sufficient
to stop speculation against the weak currencies, particularly sterling ,
and an era of floating exchange rates was ushered in in 1972 with the
United Kingdom's decision to float the pound . Other countries followed
rapidly, and by mid-1973 most of the major currencies were floating
against each other.
In a world of capital movements one of the intrinsic problems of an
adjustable peg system with little flexibility is containing speculative
pressure against a currency which traders feel will ultimately have to be
devalued. Speculators are provided , in effect , with a system in which
they can hardly lose if the pressure becomes so great : a so-called 'one-
way option' for speculators. Whatever stability a fixed exchange-rate
42 Balance-of-Payments Theory and U.K. Experience
system might give to international monetary relations, it has to contend
with this fundamental drawback and the problem it poses for con-
fidence in the reserve currencies of the system.
In an attempt to preserve the virtues of pegged exchange rates, while
avoiding the speculative pressure that can build up with the threat of
devaluation, several schemes have been developed by economists which
might make the adjustable peg system workable by introducing more
flexibility into it. One such scheme, attributed to Professor Machlup of
Princeton University and developed by Williamson (1965), is the
crawling peg. Under the crawling peg a country would maintain its
pegged exchange rate within agreed margins at a level equal to the
moving average of the market exchange rate over an agreed previous
time period . This would allow a country's currency to drift gradually
lower if circumstances warranted, and at the same time would avoid
both the upheaval of devaluation under the adjustable peg system and
the possibility of excessive depreciation under free floating. To avoid
speculation against the currency the interest rate could be raised by a
margin equal to the permitted rate of depreciation.
Other objective indicators, other than a moving average of the past
rates of exchange, could be used to introduce more flexibility into
exchange rates while preserving some semblance of order. Changes in
the level of reserves, the domestic rate of inflation, the level of employ-
ment , etc ., could all be used to trigger off exchange-rate changes
according to agreed rules (Kenen, 1975 ; Machlup, 1973). All the
proposals here would involve automatic adjustment of a pegged rate
according to some agreed formula.

(iii) Managed Floating

Under managed floating there are no pegs and no parities that the
authorities are obliged to preserve. Instead the currency is free to
float but the authorities intervene to avoid what they regard to be
undesirable consequences of excessive appreciation or depreciation .
A weak currency may lead to excessive depreciation which the
authorities may wish to avoid because of its repercussions on the domes-
tic price of imports and the internal cost structure. Alternatively,
countries with a strong currency may wish to avoid appreciation if
they want to accumulate reserves and are not unduly worried about the
effect of reserve changes on the money supply. A country may even
wish to engineer the depreciation of its cur rency that would otherwise
Balance-of-Payments Accounting 43
appreciate if the foreign-exchange market were left to operate freely .
To operate a managed float requires that the monetary authorities add
to the supply of or demand for foreign exchange as circumstances
warrant in order to achieve the exchange rate desired . The limits to
which a country can manage a floating rate depends on the volume of
reserves it has (to defend a depreciating currency), and its ability to
control the money supply, if need be, as it accumulates reserves (to
prevent an appreciating currency). Countries may also experience
international pressure to let the market operate freely, particularly
surplus countries that accumulate reserves instead of allowing the
exchange rate to appreciate.

(;1)) Freely Floating Exchange Rates

Under freely floating exchange rates the exchange rate is left to find
its own level in the market without any official intervention . Balance-
of-payments equilibrium is supposed to be achieved automatically, but
only in a net currency flow sense. In an economic system in which the
authorities are indifferent to the exchange rate, there is no need for
international reserves. In practice , however, countries are not indiffer-
ent to the value of their currency in relation to others, and in the recent
past free floating has never been adopted for any length of time as an
exchange-rate regime. It stands at the other end of the spectrum to the
rigidly fixed parities of the old gold standard, and probably neither of
these extremes will ever be adopted in the future.

FIXED VERSUS FLEXIBLE EXCHANGE RATES

The exchange rate system established at Bretton Woods survived for

over twenty-five years, finally disintegrating between 1971 and 1973.
While fixed exchange rates provided an environment of comparative
stability in which international monetary relations were conducted in
the post-war years , the system came under considerable strain at
particular periods, no more so than in the United States and the United
Kingdom in the later years of the 1960s. Even before the 'writing was
on the wall', at least as far as the United States was concerned, many
economists had sung the virtues of flexible exchange rates as an alterna-
tive to fixed exchange rates, particularly as a means of reconciling
internal and external balance , avoiding excessive speculation against a
44 Balance-of-Payments Theory and U.K. Experience

currency, and minimising the international transmission of business

cycles. We shall consider some of the arguments in favour of flexible
exchange rates and against fixed rates, and then turn to some of the
disadvantages of flexible rates, especially in the light of the inter-
national experience since 1972. The experience of the United Kingdom
under floating exchange rates since 1972 will be left to Chapter 7.
Under fixed exchange rates a change in the relative price of imports
and exports is ruled out as a means of rectifying balance-of-payments
disequilibrium. If a deficit is the problem, either interest-rate policy
must be used to improve the capital account, or domestic demand-
management policy must be pursued to reduce domestic prices, or out-
put if prices are inflexible downwards, to achieve balance-of-payments
equilibrium. A major argument for exchange -rate flexibility has always
been that it may avoid the necessity of restricting demand and dampen-
ing economic activity in order to correct a balance-of-payments deficit.
Of course , exchange depreciation at full employment also requires
domestic demand contraction in order to release resources to improve
the balance of payments," but provided resources can be switched
easily and quickly into the traded goods sector to meet the demand for
exports and import substitutes there is no reason why unemployment
should rise. Flexible exchange rates possess the potential virtue , there-
fore , that they may be able to reconcile more easily the desire of
most economies for simultaneous internal and external balance.
A second potential advantage of flexible exchange rates is supposed
to be that it gives the central bank of a country much more room for
manoeuvre, in the sense of enabling it to pursue an independent mone-
tary policy, free from the constraint of excessive variability of the
reserves. Under fixed exchange rates the money supply is endogenous
because balance -of-payments surpluses and deficits reflect themselves
in changes in reserves, which form part of the base of the domestic
money supply . Unless the effects of reserve changes can be sterilised
by open-market operations (i .e, by the buying and selling of
securities), the money supply may be difficult to control. Moreover, if
capital flows are very interest elastic, a government will find difficulty,
for example , in raising interest rates above those in the rest of the world
without attracting more capital flows, so that an independent monetary
policy becomes difficult. By contrast, under freely floating exchange
rates in which balance-of-payments deficits and surpluses reflect them -
selves in exchange -rate changes, the money supply is not dependent on
the balance of payments, and the effect of capital movements will be
Balance-of-Payments Accounting 45
on the exchange rate, making an independent monetary policy possible .
Flexible exchange rates are also supposed to reduce the need for
international reserves and to obviate the necessity of borrowing abroad
to maintain the exchange value of the currency if it comes under
speculative attack. Under fixed exchange rates a fundamentally weak
currency is liable to strong speculation against it, because if it is feared
that the fixed rate will have to be altered, there is only one way it can
go, and that is down . Fixed exchange rates may thus lead to a great
deal of instability on the capital account of the balance of payments
which flexible exchange rates might avoid. Indeed, as we indicated
before, the theoretical expectation is that in a system in which the
exchange rate is free to vary, speculation will, on balance, be stabilising .
Any initial depreciation, it is claimed, will induce private speculators
to move funds into the depreciating currency. This serves to limit
speculation against the currency , to arrest the depreciation, to finance
the deficit on the balance of payments until the quantities of exports
and imports have had time to adjust to relative price changes, and to
save foreign-currency reserves.
A third major advantage claimed for flexible exchange rates is that
they insulate countries from instability in other countries. Under fixed
exchange rates, as one country booms causing its balance of payments
to go into deficit, other countries' balances of payments move into
surplus, which exerts an expansionary influence. Thus the international
business cycle is propagated . Under flexible exchange rates the effect
of a change in activity in one country on activity in another country
works through changes in the terms of trade (assuming flexible ex-
change rates immediately rectify balance-of-payments disequilibria
arising from changes in activity). If one country expands, its terms of
trade deteriorates, while in other countries the terms of trade improve.
If expenditure does not rise by as much as real income in countries
experiencing a terms-of-trade improvement, activity will contract,
thus lending stability to the world economy. The argument hinges
crucially, however, on flexible exchange rates rectifying payments
imbalances, and on expenditure not increasing (falling) by as much as
real income increases (falls) as a result of an improvement (deterior-
ation) in the terms of trade . It should also be mentioned that stability
is defined narrowly in terms of movements in real economic activity.
What happens to the price level, however, is another aspect of economic
stability which should not be forgotten . It used to be argued that
flexible exchange rates would also tend to be less inflationary than
46 Balance-of-Payments Theory and U.K. Experience

fixed rates for the world economy as a whole because they would not
only insulate countries against outside disturbances but would also
avoid the expansionary monetary consequences of countries forced into
balance-of-payments surplus by deficits elsewhere in the system. In the
1960s the United States was seen as a powerful source of inflationary
pressure in the world economy because of its enormous balance-of-
payments deficit at the time , caused largely by overseas investment
and the finance of the war in Vietnam . However, proponents of flexible
exchange rates who argued that such a system would act as a weapon
against inflationary tendencies in the world have been remarkably (and
understandably) silent since 1972.
There are several reasons why flexible exchange rates may on
balance be more inflationary than a system of fixed rates, and this is
one of the arguments against floating exchange rates. First, in countries
experiencing exchange-rate depreciation the domestic price of imports
will rise, which increases costs directly and indirectly , and hence
domestic prices, to the extent that prices are based on costs . Admit-
tedly countries experiencing exchange-rate appreciation should find
their import prices falling, but domestic prices are notoriously sticky
downwards. Thus there tends to be asymmetry in the system with
costs and prices rising more in 'depreciating' countries than they fall
(if at all) in 'appreciating' count ries. A second factor to bear in mind
is that exchange-rate flexibility tends to weaken the domestic con-
straints against inflation. Whereas under a system of fixed exchange
rates the necessity to control the economy is apparent from the neces-
sity to support the exchange rate, under a flexible rate the resolve
may be weakened if there exists the belief that the balance of payments
will look after itself whatever the state of the domestic economy . This
facile view seems to have prevailed in the United Kingdom between
1972 and 1974 .
It is also sometimes suggested that flexible exchange rates will
increase the world price level by introducing greater uncertainty into
the economic system and thereby increasing costs of production
because of the higher cost of foreign-exchange insurance. Dr Witteveen.
a former Managing Director of the International Monetary Fund.
expressed the view in 1974 that floating exchange rates had probably
been inflationary for the world economy between 1972 and 1974
because of the risks attached to international transactions and to the
production of internationally traded goods, not all of which can be
easily or cheaply insured against. If trade is discouraged because of
Balance-of-Payments Accounting 47

exchange-rate uncertainty, this might also raise world costs and prices
by reducing international specialisation . There is no firm evidence,
however, that this is so.
In a comprehensive article surveying inflation under fixed and
flexible exchange rates, Crockett and Goldstein (1976) reached the
over-all conclusion that flexible rates probably impart a net inflationary
bias but argue that this should not necessarily imply that there would
be a clear welfare gain from choosing the alternative system . First,
they say countries ought to be allowed to choose their inflation-
unemployment trade-off. Second, they make the point that it is mainly
unanticipated inflation that causes welfare loss, so that unanticipated
inflation would also have to fall with the real inflation rate for the
less inflationary system to confer a welfare benefit.
The case for flexible exchange rates may also be questioned on other
grounds. It is not clear that flexible exchange rates do rectify payments
imbalances , either promptly or permanently , unless there is continual
appreciation or depreciation of countries' exchange rates. Neither is it
clear that they reduce the need to hold international reserves. Further-
more, the experience of exchange-rate flexibility in the 1930s and since
1972 does not inspire confidence that speculation is always stabilising.
Initial movements in the exchange rate based on changes in real econ-
omic variables have been frequently aggravated by speculative capital
flows with no real economic justification, necessitating as much official
intervention in the foreign-exchange market and the domestic economy
as under fixed exchange rates.
The arguments for and against fixed and flexible exchange rates are
finely balanced. An individual country must choose the system that
serves it best consistent with its position and responsibilities in an
interdependent world economy . After the experience of floating
exchange rates since 1972, however, there can no longer be an over-
whelming presumption that floating rates are superior to some form of
adjustable peg system - contrary to the dominant academic view during
the Bretton Woods era of the 1950s and 1960s. Many of the presumed
advantages of floating exchange rates have failed to materialise .
2
The Balance of Payments
and the National
Economy!
THE INTEGRATION OF THE BALANCE OF PAYMENTS INTO THE
NATIONAL ACCOUNTS

In a closed economy with government activity there are three types of

expenditure which generate income - private consumption , private
investment and government expenditure - and three ways in which
income may be disposed of - by consumption, saving and tax pay-
ments. In a closed economy total expenditure can fall short of income
but cannot exceed it.
In an open economy income is generated by consumption , invest-
ment, government expenditure and exports , each of which may have
an import content which will not generate income domestically. Hence

Y=C+/+G+X -M (2.1)

where Y is income, C is private consumption, / is private investment,

G is government expenditure, X is exports and M is imports. The dis-
posal of income may be written as

Y=C+S+T (2 .2)

where S is saving and T is tax payments. Combining equations (2.1)

and (2.2) gives the condition for income equilibrium :

(2.3)
 The Balance ofPayments and the National Economy 49
In an open economy expenditure may exceed domestic income if
imports exceed exports. By the use of these three equations the
relationship of the balance of payments to the functioning of the
economy may be seen. Let us give some simple examples . Suppose
that the government's budget is balanced so that G = T; then the
plans of the private sector to invest in excess of saving would lead to
an excess of imports over exports and the plans could be realised
by a balance-of-payments deficit financed by foreign borrow ing.
Alternatively, suppose the private sector is in balance (I = S); then
the plans of government to spend in excess of taxation would also lead
to a balance-of-payments deficit and the plans could be realised if the
deficit can be financed . The so-called New Cambridge theory ascribed
the United Kingdom 's balance-of-payments deficits in the 1960s and
early 1970s to government budget deficits, basing the theory on the
empirical observation that the private sector of the economy stays
roughly in balance ," No sooner had the theory been expounded,
however, than it was confounded by contrary empirical evidence,
and the theory died a quick death . It remains true, however, that the
performance of the current account of the balance of payments is a
function of expenditure decisions in both the private and public sectors
of the economy; and , of course, in an accounting sense the difference
between imports and exports must be equal to the difference between
saving plus taxation and investment plus government spending .
In this chapter we shall be concerned with three main questions .
First , we shall consider the conflict between balance-of-payments
equilibrium and the achievement of other goals of economic policy .
Second, we shall be concerned with how trade alters the condition for
income equilibrium in the macro economy , and how, through the
foreign-trade multiplier, fluctuations in income may be transmitted
from one country to another. Here a departure is made from the
traditional approach to the determination of income equilibrium in an
open economy, which relates imports to income via the marginal pro-
pensity to import. Following the approach recently developed by
Kennedy and Thirlwall (l979a, 1979b), imports will be related to
expenditure. Relating imports to income assumes implicitly either
that all expenditures have the same ratio of imports to expenditure or
that all imports are consumption goods. Only on these assumptions
will the marginal propensity to import be independent of the type of
expenditure which generates the multiplier process. In practice, how -
ever, it is extremely unlikely that the ratio of imports to expenditure
will be the same for all items of expenditure, or that all imports will be
50 Balance-of-Payments Theory and U.K. Experience

consumption goods , so that relating imports to income may be a

serious mis-specification of the income multiplier. As far as I am aware
this is the first textbook to present a simple expos ition of the income
multiplier which relates imports to expenditure , and it is hoped that the
approach will be widely adopted in the teaching of the foreign-trade
multiplier. The model is also extended to the two-country case to
examine the condition for equilibrium when foreign repercussions are
considered . Finally, a brief introduction will be given in this chapter to
the theory of balance-of-payments adjustment before considering more
fully the major approaches to balance-of-payments adjustment policy
in later , separate chapters.

THE CONFLICT BE1WEEN BALANCE.QF-PAYMENTS EQUILIBRIUM AND

OTHER OBJECTIVES OF ECONOMIC POLICY

The discussion here abstracts from the possibility of exchange-rate

changes and the adoption of policies to improve the capital account of
the balance of payments. These considerations are taken up in subse-
quent chapters.
Balance-of-payments equilibrium , however defined , is not desired for
its own sake. There is nothing particularly virtuous or conducive to
welfare in balance-of-payments equilibrium itself. Balance-of-payments
equilibrium is an intermediate objective , along with others, towards
the final objective of most economic societies of maximising welfare
per head . The other main macroeconomic objectives along with
balance-of-payments equilibrium are full employment , faster growth
and stable prices - the 'magic quadrilateral', as Joan Robinson once
described them. A balance-of-payments surplus means that residents
of a country are forgoing immediate welfare by exporting more goods
and services than they import, thus in effect transferring welfare to
other countries. A balance-of-payments deficit, while it may lead to a
temporary increase in welfare , will require rectification in the long
run which may mean the sacrifice of other domestic goals. There is no
conflict between balance-of-payments equilibrium and stable prices,
but there may be a conflict between balance-of-payments equilibrium
and both full employment and faster growth.
If full employment and balance-of-payments equilibrium conflict,
full employment may have to be sacrificed to achieve balance-of-
payments equilibrium if other policy measures are ruled out or fail.
 The Balance ofPayments and the National Economy 51

Economic-policy discussion to improve the U.K. balance of payments

has been dominated, under fixed and floating exchange rates , by the
consideration of policies to reconcile the conflict between full employ-
ment and balance-of-payments equilibrium . Assuming fixed exchange
rates, a conflict may exist for the following reasons. The level of
employment is determined by the relationship between national ex-
penditure and the output capacity of an economy. The higher the level
of expenditure in relation to capacity output, the higher the level of
employment. As expenditure increases, however, and the level of out-
put approaches capacity, imports are likely to rise. Some imports ,
mainly those without domestic substitutes, will tend to be required
in some fixed proportion to domestic expenditure, and these will rise
as expenditure and employment rise. Other imports, which have dom-
estic substitutes, will rise slowly at first but then will become
increasingly sensitive to the pressure of demand because at full employ-
ment domestic output reaches its capacity level and demand can only
be met by imports. At the same time exports may fall, or at least not
match the increase in imports, because production for the home market
becomes more profitable . Thus , as the economy moves towards full
employment, the balance of payments may be expected to deteriorate,
with the possibility that at the level of full employment (however a
country defines this elusive concept) there may" be a balance-of-pay-
ments deficit. The conflict may be represented by a trade-off curve, the
slope of which shows how much the balance of payments deteriorates
as the level of employment increases and the unemployment rate falls.
An empirical estimate of the trade-off curve is made by the author for
the period 1958 to 1966 in Figure 7.1 of Chapter 7. A hypothetical
curve is shown in Figure 2.1. In this particular example the trade-off
curve (see Figure 2.1) has been drawn cutting the horizontal axis at
5 per cent unemployment. This means that if full employment were to
be defined as 2 or 3 per cent unemployment , a conflict would exist
between full employment and balance-of-payments equilibrium. At this
level of unemployment the balance of payments would move into
deficit. Cutting expenditure to achieve balance-of-payments equilibrium
would mean a sacrifice of full employment.
It may , of course , be possible to shift the trade-off curve inwards to
reconcile the conflict. Devaluation is supposed to act in this way .
Ruling out exchange-rate changes, however , such policies as the better
marketing of exports, improved productivity, import tariffs and export
subsidies could all potentially improve the balance of payments at a
52 Balance-of-Payments Theory and U.K. Experience

Balance-of-
payments
deficit

Ot---+---+----+--+-""""toc--+----+----
2 3

Balance-of-
payments
surplus

FIGURE 2.1
given level of unemployment. It can also be shown that the position
of the trade-off curve will depend on the degree' of demand
disequilibrium between markets in the economy . If some markets have
excess demand, and others have excess supply, and if imports are
increasingly sensitive to the pressure of demand, the aggregate level of
imports will be higher for any given level of aggregate demand (or
unemployment) than if all markets were in equilibrium at the given
level of aggregate demand (see Thirlwall and White, 1974 ; and Hughes
and Thirlwall, 1979) . In Chapter 8 evidence is given to suggest that
bottlenecks in particular sectors of the U.K. economy may have raised
the import bill in the 1960s and 1970s by as much as £500 million to
£1000 million per annum at the pressure of demand prevailing.
The conflict between faster growth and balance-of-payments equili-
brium, again assuming fixed exchange rates, arises for the same type of
reasons as the conflict between full employment and balance-of-
payments equilibrium. The attempt to grow faster from a position of
below full employment increases the import bill, and the attempt to
raise the capacity growth rate by expanding demand in excess of the
capacity growth rate, in the hope that this will encourage investment,
will increase the import bill even more . There are only two ways in
which expenditure in excess of output in the domestic economy can
be eliminated, either by prices rising to equilibrate demand and supply,
or by imports rising (or exports falling). One of the reasons why a
country may want to grow faster, and certainly why the United
Kingdom has suffered a growth neurosis since the Second World War,
is that its growth rate is inferior to that of other countries. Whether a
country can grow at the rate of other countries without running into
 The Balance ofPayments and the National Economy 53
balance-of-payments difficulties depends primarily on the income
elasticity of demand for its imports and exports. The income elasticity
of demand for imports measures the proportional change in demand for
imports with respect to a proportional change in the country's income .
The income elasticity of demand for exports measures the proportional
change in the demand for a country's exports with respect to a pro-
portional change in other countries' incomes. If a country's income
elasticity of demand for imports is no greater than the income elas-
ticity of demand for its exports, it is possible (other things remaining
the same) for a country to grow at the same rate as other countries
without imports growing faster than exports and the balance of pay-
ments deteriorating. If, however, a country's import elasticity is higher
than its export elasticity, it cannot grow faster than other countries
without the balance of payments deteriorating, even if it possessed the
capacity to do so. For example, suppose that other countries are grow-
ing at 5 per cent per annum, that the income elasticity of demand for
a country's exports is unity, and the country's income elasticity of
demand for imports is 2. The country's exports will be growing at 5 per
cent (5 x I), but if the country's growth of output is 5 per cent , its
imports would grow at 10 per cent (5 x 2). To keep the growth of
imports in line with exports the growth of output would need to be
constrained to equal the growth of exports divided by the income
elasticity of demand for imports, i.e. to 5 -;. 2 = 2.5 per cent per annum.
This illustrative example is not far removed from the experience of the
United Kingdom in the years since the Second World War. The rest of
the world has been growing roughly twice as fast as the U.K. economy,
and the U.K. growth rate has had to be constrained to an average of
about 2.5 per cent per annum in order to maintain balance-of-payments
equilibrium . The capacity growth rate of the U.K. economy has been
slightly above 2.5 per cent, but this higher rate has not been sustainable
over a long period because of the balance-of-payments consequences.
The result has been a slow but steady rise in the underlying rate of
unemployment. In one sense it is fortunate that the United Kingdom
has not possessed the capacity to grow at the rate of other countries,
because if it had the potential growth of output would not have been
achievable and there would have been more unemployment. Instead of
the gap between the growth of actual and potential output being, on
average, 0.5 per cent it would have been closer to 2.5 per cent, with the
consequence of much greater surplus capacity. In Chapter 10 it is shown
how closely the growth rate of several industrial countries approximates
54 Balance-of-Payments Theory and U'K, Experience

to their rate of growth of exports divided by their income elasticity of

demands for imports, and throughout the book the thesis is argued that
at the heart of the United Kingdom's inability to grow faster , and its
weak balance of payments, is its slow rate of export growth caused by
the low income elasticity of demand for U.K. exports in world markets.

INCOME EQUILIBRIUM IN AN OPEN ECONOMY AND THE FOREIGN-

TRADE MULTIPLIER

Earlier we presented the equations which integrate the balance of

payments with the functioning of the national economy . In using these
equations for the determination of the equilibrium level of income, all
the magnitudes must be interpreted in an ex ante sense. The expendi-
tures and disposals of income are planned (p) . For simplicity, govern -
ment spending and taxation are now ignored :

(2.4)
and

In a closed economy (with no exports and imports) the condition for

income equilibrium is the familiar Keynesian condition Ip = Sp . In an
open economy the condition for income equilibrium becomes

(2.5)

or

(2.6)

In words, plans to save and import must equal planned investment plus
export demand . If plans to save and to buy foreign goods (the purchase
of which does not generate income at home) exceed plans to invest and
for foreigners to buy home goods (the purchase of which does generate
income at home), income will fall, thus reducing saving and imports,
until plans to save and import equal once again plans to invest and
export. Conversely, if plans to save and import fall short of plans to
 The Balance ofPayments and the National Economy 55

invest and export, income will rise, raising saving and imports, until
plans to save and import are equal once again to plans to invest and
export. The mechanism by which plans to save and import are brought
into line with plans to invest and export, once a disturbance takes
place, is the so-called multiplier mechanism, which is the relationship
between a change in planned (autonomous) expenditure and the
corresponding change in income . To derive the multiplier we first
have to derive an expression for the equilibrium level of income in the
open economy . The usual method of approach is to assume that
planned investment and exports are autonomous (i.e. not dependent on
income), and that planned saving and planned imports are both
functions of lncome." Thus from equation (2 .6) we have :

lp = lp (2.7)
Xp = X p (2.8)

Sp = Sap + sY (2 .9)
and
M p = Map + mY (2.10)

where a bar over a variable denotes 'autonomous', s is the marginal

prope~sity to save, m is the marginal propensity to import , and Sap
and Map represent autonomous saving and imports, respectively .
Substituting equations (2 .7) to (2.10) into equation (2.6) gives:

Sap + sY + Map + mY = t; + X p (2 .11)

Rearranging , and solving for the equilibrium level of income :

t; + X P - Sap - Map
Y= (2.12)
s+m
The multiplier , which is defined as the relation between a change in
autonomous expenditure and the resulting change in income, can be
der ived with respect to any of the elements of autonomous expenditure
in the numerator of equation (2 .12) . For the same increase in invest-
ment or export demand . for example . the change in income is

AY A:
(2 .13)
s+m
56 Balance-ofPayments Theory and U.K. Experience

For an increase in autonomous saving or imports (which would exert

a negative effect on income) :

- I
(2.14)
s+m

I/(s + m) is called the foreign-trade multiplier. It contrasts with the

multiplier for a closed economy of I /s (which can be derived from
equation (2.11) , eliminating imports and exports from the equation).
Since 0 < m < I, the multiplier with foreign trade will be less than the
multiplier without foreign trade. The explanation is that some
additional expenditure leaks into imports (as well as into saving), and
this does not generate any further income at home. To show how the
multiplier process brings income back into equilibrium following an
autonomous disturbance , let us work a simple numerical example .
Suppose s = 0.1 and m = 0.1 , giving a foreign-trade multiplier of
1/0.2 (= S). Start with income in equilibrium at £IOOm. and then
assume an autonomous increase in investment of £Im . raising planned
investment plus export demand in excess of planned saving plus imports.
With a multiplier of S, an increase in investment of £1m. will increase
income by iSm . Out of the iSm . increase in income , however , there
will be increased saving of iSm. x 0 .1 = £O.5m. and increased import
demand of iSm. x 0.1 = £O.Sm. so that the total level of planned
saving plus imports increases by £Im . - exactly equal to the planned
increase in autonomous expenditure. Income will remain at a new
equilibrium level of £IOSm. where planned saving and imports equal
planned investment and exports, until equilibrium is again disturbed.
In practice the propensity to save in the United Kingdom is about
O.1S (in the long run) and the propensity to import is about 0 .2.
These values would give a closed-economy multiplier of 6.6 and a
foreign-trade multiplier of 2.9, a substantial difference . The multiplier
in reality is much lower than 2.9 because additional income also leaks
into taxation (the sensitivity of tax revenue to national income is
about 0 .2).
In practice, also, the multiplier will vary according to the magnitude
of the import content of the initial autonomous expenditure change ,
but this is ignored by the traditional approach to the foreign-trade
multiplier which relates imports to income . For example, suppose that
there is an increase in autonomous investment of £lOOm. but £SOm.
of this expenditure is on investment goods imported from abroad . The
The Balance ofPayments and the National Economy 57
initial stimulus to the domestic economy is only £SDm . Compare this
with an inc rease in the demand for exports of £1 DDm . which has an
import content of only £IDm . In this case the initial stimulus to the
domestic economy is £9Dm. and the total effect on income via the
multiplier will be greater than for an equal increase in investment.
To cope with this weakness of the traditional approach to the
foreign-trade multiplier , however, it is not sufficient simply to subtract
the proportion of imports in autonomous expenditure from the numer-
ator of the multiplier formula and work on in the same way . The
denominator of the multiplier must also be changed . The induced
leakage of income into imports which dampens the process of income
generation is, strictly speaking, the import content of induced con-
sumption, and only by chance will m, the marginal propensity to
import , be a measure of this when imports are related to the different
components of expenditure which generate income . A complete over-
haul of the presentation of the foreign-trade multiplier is called for in
which imports are related to the different components of expenditure .
It can be shown that the two approaches of relating imports to income
and imports to expenditure are only equivalent in the very special case
of all items of autonomous expenditure having the same ratio of
imports to expenditure (i.e. the same import coefficient). A special
case of this , in turn , would be if autonomous expenditure had no
import content , in which case the import coefficients would be zero
and all imports would be induced consumption-good imports. In
developing the new foreign-trade multiplier the model will also be
extended to the two-country case where changes in income in one
country affect activity in the other, which in turn affects activity in
the first country. The multiplier and the condition for income equili -
brium will be affected.

THE FOREIGN-TRADE MULTIPLIER RELATING IMPORTS TO

EXPENDITURE

We start with the income .equation :

Y=C+/+X-M (2.15)

Let consumption (like saving earlier) be partly autonomous (C) and

partly dependent on income (cY) , where c is the marginal propensity
58 Balance-of-Payments Theory and U'K. Experience

to consume. Let f and X be autonomous, as before . Now make imports

a function of expenditure (not income) and assume , as is the case in
practice, that each type of expenditure has an import content. The
import function can then be written as

(2.16)

where M is autonomous imports, Aae is the import coefficient of

autonomous consumption, Ae is the import coefficient of induced
consumption, Ai is the import coefficient of investment , Ax is the
import coefficient of exports, and the import coefficient (A) is defined
generally as the ratio of imports to expenditure .
Substituting equation (2 .16) into (2.15) gives the expression for
income of

y = c + cY + T + X - M - Aaee - Ae cY - Ai 1- Ax X
(2 .17)
or

-M + (I -- Aae)C+ (I - Ai)T+ (I - Ax)X

Y=
1 - c (1 - Ae )
(2.18)

The income multiplier with respect to any change in autonomous

expenditure in the numerator of equation (2.18) can be derived in the
normal way . With respect to a change in autonomous investment we
have :

D. Y I - Ai
(2.19)
M 1-c(I-Xc)
Equation (2.19) contrasts with the conventional foreign-trade multi-
plier in which imports are related to income through the marginal pro-
pensity to import, of(see equation (2.13» :

1- c + m (2.20)

where 1 -- c = s.
The Balance ofPayments and the National Econ om y 59
The result in equation (2.20) ignores the import conten t o f the
autonomous expenditure , or , as we shall prove below, assumes the
import coefficient of investment and consumption to be the same.
Note also from the more realistic formulation of the foreign-trade
multiplier that it is clearly wrong to treat the effect of an autonomous
increase in imports as the exact obverse of an autonomous increase in
exports which is what the conventional approach assumes . From
equation (2.1 8) the multiplier with respect to an autonomous change
in imports is

-1
- -- - - (2 .21 )
I - c (I - Ac )

while the multiplier with respect to an autunomous change in exports is

1 --. Ax
(2.22)
I .- c (I - Ac )

The effe ct of an equal change in aut on om ou s exp orts and imports on

income is not zero as one would be led to suppose from the conven -
tional multiplier fo rm ulation . For an equal incre ase in autonomous
exports and imports, income will fall by - Ax /[1 - c(1 - Ac}] times
the equal change in exports and imp orts: and for an equal decrease in
autonomous exports and imports. income will rise by Axl [1 -- c
(1 - Ac}1 times the equal change in exports and imports .
The conventional foreign-trade multiplier result with respect to a
change in autonomous expenditure will be equivalent to our more
realistic result if and only if all components of autonomous expenditure
have the same import coefficient and there is no change in autonomous
imports . A special case o f th is would be if all the import coefficients
attached to autonomous expenditure are zero , in which case all imports
would be induced consumption-good imports . The pro of is as follows.
Totally differentiating equation (2 .16) and assuming dM = 0, we have :

(2.23)

and dividing through by dY:

dM Aac de + Ai dT + AxdX
m -- - - - - - - --- + Acc
dY dY (2 .24)
60 Balance-ofPayments Theory and U'K. Experience

Now from equation (2 .18) we have by total differentiation (assuming

dM= 0) :

Substituting (2 .25) into (2 .24) gives:

AacdC + Aidl + Ax dX
m -=------,,-
(I - Aac) de + (I - Ai) d/ + (I - Ax) dX
(2.26)

m will be independent of dC, dland dX if and only if

(2 .27)

On that assumption we have:

A
m = -- -- [I - c (I - A )] + A c
I _A c c (2 .28)

If equation (2.28) is substituted for m in the conventional formulation

of the multiplier - equation (2 .20) - we obtain, after simplification :

I .. A
- - ----
I -c + m 1 - c (I - Ac ) (2 .29)

which is the formulation of the multiplier which results from relating

imports to expenditure. In the special case of autonomous expenditure
having no import content, so that '11.= 0, it is clear from equation (2.28)
that the conventional multiplier result would also be obtained since
m = Acc, so that the multiplier (1 - '11.)/ [1 -- C (I - Ac)] would reduce
to 1/(1 - c + m). Many treatments of the traditional foreign-trade
multiplier which relate imports to income say explicitly that the
assumption is being made that all imports are consumption goods . It
has not been explicitly recognised, however, that the same result would
obtain if the import coefficients attached to all items of autonomous
expenditure were the same. In practice neither circumstance is likely
The Balance of Payments and the National Economy 61
to prevail, and there must therefore be strong grounds for discarding
the traditional approach in favour of relating imports to expenditure ,
particularly as it is so easy to do . The difference made to the value
of the multiplier could be very great in countries, such as developing
ones, where the import coefficients differ substantially . Typically
in a developing country the import coefficient of investment expen-
diture , for example, is very high, while the import coefficient of other
expenditures might be much lower. In the case of the United Kingdom
the difference made is not that great because different components of
expenditure have roughly similar import coefficients. Turning to the
'Input-Output Tables for the United Kingdom (I 971)' we find Aac =
Ac = 0.169 , Ai = 0.210 , A.x = 0.229 , and that the import coefficient of
total final expenditure (includ ing public expenditure and expenditure
on stocks) is 0.174 . Using the traditional foreign-trade multiplier with
c = 0.8 and m = 0 .21 gives an income multiplier of 2.44 . Relating
imports to expenditure the following multipliers are obtained :

6Y 6Y 6Y
- = 2.44 ; -= = 232 ; -= = 2.26
6C t:J AX

While the difference is not large between the traditional and the new
results , or between the various expenditure multipliers, even an error
of 0.2 in the assumed value of the multiplier could mean a vast absolute
difference in the prediction of the change in income resulting from a
forecast change in some item of expenditure. It can also be shown
(Kennedy and Thirlwall , 1979a) that the new formulation of the
multiplier is more stable than the conventional formulation with
respect to variations in common parameter values. For purposes of
policy-making and prediction it is always better to work with a relation
which is as stable as possible so as to minimise the effects of errors in
the parameters.

THE 'NEW' FOREIGN-TRADEMULTIPLIER WITH FOREIGN

REPERCUSSIONS

Up to now we have considered the foreign-trade multiplier without

regard to the fact that countries are linked together by trade and that
income changes in one country will affect income in other countries ,
which in turn will feed back on the country experiencing the initial
62 Balance-of-Payments Theory and U.K. Experience

income change . Consider a two-country model in which the exports

of country a are the imports of country b, and the exports of country
b are the imports of country a. The expressions for the equilibrium
level of income , and the income multiplier, must be modified . We now
have a household sector for both countries and a firm sector for both
countries disaggregated into the production of consumption goods,
investment goods and exports. The export function of both countries
has two components: an autonomous component equivalent to the
autonomous import term of the other country ; and an 'induced'
component related to expenditure in the other country . We will show
that in the case of foreign repercussions , the multiplier with respect
to an autonomous increase in exports is not the same as the multi-
plier with respect to an autonomous increase in investment. The model
is as follows ." For country a:

Ya = (I - Aca ) c, + (I . Aia) fa + (I - Ax a ) Xa .- Ma
(2.30)

(2.31 )

(2.32)

(2.33)

For country b:

Yh = (I - Acb) Cb + (I . Aib)fb + (1 .- Axb)X b -Mb

(2.34)

(235)

(2.36)

(2.37)

Rearranging equations (2.30) to (2.37), and solving simultaneously

(by matrix methods), expressions for Ya and Y b can be obtained which
 The Balance ofPayments and the National Economy 63
incorporate the feedback effects from one country to another arising
from the fact that the exports of one country are the imports of the
other. Taking country a, the multiplier can then be derived with respect
to any change in autonomous expenditure in the system . With respect
to a change in investment in country a we obtain:

(2.38)

This multiplier is extremely complex to interpret, but note that if the

import coefficient of the different components of expenditure is the
same, the result reduces to

(2.39)

This result differs from the earlier investment multiplier result without
foreign repercussions (equation (2.19)) by an additional term in the
denominator . Since all the coefficients that comprise the term lie
between zero and unity, and the term has a negative sign attached, the
effect of the additional term is to increase the value of the multiplier -
as one would expect. The new multiplier with foreign repercussions
may be compared with the conventional result using the marginal
propensity to import, as (for example) in Brooman (I970) , which is:

I - ca + m a - m a mb
I - cb + mb (2.40)

where m a and mb are the marginal propensities to import in the two

countries. It can be readily shown that the conventional foreign-trade
multiplier result with foreign repercussions will only be equal to the
64 Balance-of-Payments Theory and U.K. Experience
new formulation if all components of autonomous expenditure have
the same import coefficient, A.s
The case of an autonomous increase in exports in country a will give
a different multiplier result from that in equation (2.38), even if the
import coefficient attached to autonomous exports is the same as that
attached to autonomous investment. The reason is that an autonomous
increase in a's exports means an autonomous increase in b's imports,
i.e. LlXa = tiMb oThis reduces b's income directly, and thus the feed-
back effects on a's exports and income will be less than in the case of
an autonomous increase in investment in country a. Using the same
system of equations as before, the change in income in country a with
respect to a change in autonomous exports in a is

ar, ar,
--+
aXa aMb
(2.41)

Note again that if the import coefficient of different components of

expenditure is the same, the result reduces to

LlYa ar, ar, (2.42)

--=--+ =
LlXa aXa aMb
Axa AXb
1 - ca ( 1 - Aca) - --::..:..::...----"-"---
1 - cb (1 - Acb)
This may be compared with the conventional result, as in Brooman
(1970), of
I -

rnb rna
1- ca + rna -
1 -cb + rnb (2.43)
The Balance ofPayments and the National Economy 65
Both the new and the conventional results give a lower value for the
multiplierwith respect to an autonomous change in exports than with
respect to an autonomous change in investment , as predicted . There is
now an additional negative term in the numerator. It is no longer true,
however, that the conventional result will be equal to the new result
even if all items of autonomous expenditure have the same import
coefficient. Even under these restrictive assumptions equations (2.42)
and (2.43) will differ by the term cb in the second expression of the
numerator of equation (2.42). Since 0 < cb < I, the multiplier using
the new formulation will be higher than that given by the conventional
treatment of imports , as in Brooman (J 970) for example . The reason
is that as autonomous impo rts rise in country b, consumption in b
falls less than income in b, and therefore the second- and subsequent -
round effects on a's exports are less than t hey would be if b's imports
and a's exports were linked to income in b rather than to expenditure
in b. This is a new insight derived from the new multiplier formulation .

BALANCE-OF-PAYMENTS ADJUSTMENT

We conclude this chapter with an introductory discussion of different

approaches to the theory of balance-of-payments adjustment , as a
prelude to a fuller discussion of the different approaches in the chapters
to follow . In the earlier discussion it was argued that under fixed
exchange rates balance-of-payments difficulties may require internal
price and income adjustments involving the sacrifice of domestic goals.
Internal price and income adjustment is just one of several methods of
adjustment to balance-of-payments disequilibrium . Exchange-rate
variation is another possible adjustment mechanism . For a long time
after the collapse of the international gold standard in 1931 , economic
analysis of the balance of payments was dominated by the so-called
elasticity approach to balance -of-payments adjustment, with attention
centred on the effect of exchange-rate changes and the elasticity of
demand for exports and imports in international trade . When Keynesian
theory brought income analysis to the fore, models of balance-of-
payments adjustment were developed incorporating both the price and
income effects of changes in exchange rates . There was, however,
a general dissatisfaction with the partial-equilibrium framework of the
elasticity approach, and in response the absorption approach to
balance-of-payments adjustment was developed , and this argued that
66 Balance-of-Payments Theory and U'K. Experience
movements in the balance of payments can only be understood in
relation to how policies affect the total functioning of the economy -
specifically how expenditure is affected relative to output. In recent
years, although balance-of-payments policy in the United Kingdom and
other countries has been dominated by exchange -rate changes, the
professional literature has come to be dominated by the so-called
'monetary' approach to the balance of payments, which is essentially
an extension of the absorption approach, stressing balance-of-payments
deficits as a monetary phenomenon to be corrected by monetary
means . After a brief survey of the different approaches to balance-of-
payments adjustment, we shall look more closely at the elasticity
approach, the absorption approach and the monetary approach in
separate chapters. We shall be critical of the assumptions of the elas-
ticity approach , and cast doubt on whether the monetary approach is
useful for the study of balance-of-payments problems that originate
in the real economy. We conclude that the absorption approach is
probably the most useful and versatile. However, these approaches to
balance-of-payments adjustment do not come to grips with perman-
ently improving the balance of payments in a growing economy or
raising permanently a country's balance-of-payments equilibrium
growth rate, which is the fundamental problem facing the U.K.
economy. The reason why these approaches do not cope with the
fundamental problem is that they do not focus on the non-quantifiable
adjustments in the real economy which are needed to raise the income
elasticity of demand for exports in order to raise the rate of growth
of exports. The approaches either stress price adjustment through
exchange-rate changes or monetary adjustment, and these do not
offer a solution to raising permanently the rate of growth of exports,
unless they induce structural change and enable producers to break into
fast-growing markets.

THE CLASSICAL PRICE-SPECIE FLOW MECHANISM

The classical price-specie flow mechanism describes the automatic

adjustment that is supposed to have taken place as a result of payments
imbalance under the gold standard. A key part of the mechanism is the
assumed rise and fall of the price level according to whether the balance
of payments is in surplus or deficit. A surplus would lead to the ac-
The Balance ofPayments and the National Economy 67
cumulation of gold and an expansion of the domestic money supply,
causing prices to rise and the balance -of-payments surplus to be re-
duced. A deficit would lead to a loss of gold and a contraction of the
domestic money supply, causing prices to fall and the balance-of-
payments deficit to be improved . David Hume used this mechanism
to refute the mercantilist belief that a country could achieve a per-
sistent balance-of-trade surplus by the mercantilist policies of trade
protection and export promotion .
The classical monetary theory assumed a system of fixed exchange
rates , the rates being fixed by the gold points (as discussed in Chapter
I) . The theory is the classical precursor of the new monetary theories
of the balance of payments which analyse the balance of payments in
terms of the relation between the supply of and demand for money .
The classical approach to balance-of-payments adjustment contained
no analysis, however, of the effect of exchange-rate changes as a sub-
stitute in a fixed-price world for flexible domestic prices , which was
more relevant to the period after 1931 when the gold standard coll-
apsed as a system of international payments.

THE ELASTICITY APPROACH

The elasticity approach to balance-of-payments adjustment was de-

veloped in the 1930s largely by Joan Robinson (I937) in response
to the need for a theory of balance-of-payments adjustment under
flexible exchange rates . The elasticity approach and the Keynesian
multiplier extension of it are concerned with three questions : What
are the conditions for currency depreciation (devaluation) to improve
a country's balance of payments on current account? What will be the
effect of currency depreciation on the level of domestic activity. and
how will this affect the balance of payments and the conditions for de-
preciation to be successful? Finally, what will be the effect of de-
valuation on the terms of trade of the devaluing country?
The question of whether devaluation will rectify a balance-of-pay-
ments deficit is conducted within the framework of partial-equilibrium
analysis focusing on the price elasticities of demand for exports and
imports. At the simplest level the elasticity of supply of exports and
imports is assumed infinite and income is held constant. Within this
framework the so-called Marshall-Lerner condition for a successful
68 Balance-of-Payments Theory and U'K, Experience
devaluation can be derived , and it states that the sum of the price
elasticities of demand for exports and imports (measured in a common
currency) should exceed unity . Despite its shortcomings as a method of
approach, and the unreal assumptions on which the analysis is based,
those who defend the use of currency depreciation to rectify balance-
of-payments deficits still cling strongly to favourable evidence relating
to the price elasticities of demand for exports and imports in support
of their case. We shall consider in detail in Chapter 3 the derivat ion of
the Marshall-Lerner condition ; the shortcomings of the elasticity
approach, and the unfavourable effects that devaluation can have on
the domestic economy , especially on the rate of inflation .
Devaluation not only alters the relative price of traded and non -
traded goods , by making exports cheaper in foreign currency and
imports dearer in domestic currency , but also raises income by a
multiple of the net expansion of demand if devaluation switches
demand from foreign to home-produced goods . Traditional elasticity
analysis assumed no change in domestic income, or assumed that
income was stabilised by the monetary authorities. Neither assumption
was satisfactory. The immediate reaction in the light of Keynesian
theory was to add a multiplier effect to the initial impact effect of
the exchange-rate change. Th is does not alter the Marshall-Lerner
condition for a successful devaluation , holding autonomous expenditure
constant in money terms, but weakens the magnitude of the impact of
the exchange-rate change . In Chapter 3 there is a more detailed discus-
sion of this point.
Tagging the income effects of devaluation on to the elasticity
approach , however, does not come to grips with the fundamental
weakness of the elasticity approach, which is that it is partial-equilibrium
analysis and confines attention to the effect of exchange-rate changes
within the markets for exports and imports alone . But price changes in
these two markets will have ripple effects throughout the economic
system which will feed back to the export and import markets. The
operation of the income multiplier will itself have relative price effects.
Monetary effects are also ignored . It was increasingly recognised that
the balance of payments is an aggregate phenomenon and that partial-
equilibrium analysis is an inadequate framework for understanding
aggregate phenomena in general . The response to this major worry
was the development of the absorption approach to the balance of
payments pioneered by Alexander (1952), but suggested earlier by
Harrod (1947).
 The Balance ofPayments and the National Economy 69
THE ABSORPTION APPROACH

Alexander's argument , and the essence of the absorption approach

to the balance of payments, is that the most fruitful way to diagnose
the balance of payments and to evaluate policies to rectify surpluses and
deficits is to consider the relation between imports and exports and the
functioning of the economy as a whole . In fact if the elasticities used
by the elasticity approach are defined as total elasticities and not
partial elasticities (because other things do not remain constant outside
the foreign-trade sector when exchange rates change) , this is what the
elasticity approach also amounts to . The absorption approach views the
balance of payments as the outcome of the difference between a
country's expenditure and its income, and sta tes that balance-of-
payments policy will only improve the balance of payments if expen-
diture is reduced relative to income (or income is raised relative to
expenditure). At the heart of the absorption approach is the account-
ing identity which we have already come across in Chapter I , i.e.
B = X - M = Y - E, where E is total domestic expenditure. Since the
approach deals with an accounting identity , care must be exercised in
deducing the cause of balance-of-payments deficits . Because a deficit
must mean that expenditure exceeds income, this does not imply that
the deficit is caused by decisions to spend in excess of income - as a
result, for example , of excessive monetary expansion . The cause of the
deficit may be the result of a gradual deterioration in the quality or
relative price of exports , which would also show up in the accounts
as expenditure in excess of income , but the cause would be a fall in
output. A belief in the usefulness of the absorption approach to an
understanding of balance-of-payments difficulties and for evaluating
the efficacy of corrective policies does not imply a belief that balance-
of-payments deficits are necessarily a monetary phenomenon caused by
excessive monetary expansion , which is the argument propounded by
Johnson (1958) in his elaboration of the absorption approach . The
absorption approach, or rather his elaboration of it , is indeed the
precursor of the new monetary theories of the balance of payments
and of the monetary approach to balance-of-payments adjustment.

THE MONETARY APPROACH

A clear statement of the monetary approach to the balance of pay-

ments is not easy because disciples of the approach do not all propound
70 Balance-of-Payments Theory and U.K. Experience

the same gospel. The essence of the approach would seem to be, how -
ever, that the balance of payments must be looked at as a whole (the
current plus the capital account) and that international monetary
movements must be considered as the outcome of stock disequilibrium
between the supply of and demand for money within a country ; an
excess demand for money leading to an inflow of international reserves
and a balance-of-payments surplus, and an excess supply of money
leading to a loss of reserves and a balance-of-payments deficit. Thus ,
according to Johnson (1977b) :

the essential difference between the monetary approach and the

other post-Keynesian approaches .. . is that the monetary approach
formulates the problem of the balance of payments as-a monetary
phenomenon to be analysed with the tools of monetary theory,
whereas the other approaches formulate it as a residual difference
between real flows determined by other flows and real relative
prices .

This leads Johnson (1977a) to assert :

all balance of payments disequilibria are monetary in essence .

So-called 'structural' deficits or surpluses simply cannot exist. . ..
Similarly any assertion that real changes cause balance of payments
deficits is correct only if the real change in question is accompanied
by policies involving either the running down of reserves or borrow-
ing on commercial terms.

These are challenging statements and will be scrutinised in Chapter 5 .

The policy implications that Johnson lists are then as follows . Since
balance-of-payments deficits represent a monetary stock disequilibrium
between the supply of and demand for money, balance-of-payments
disequilibrium must inevitably be transitory because ultimately there
must be self-correcting monetary consequences. If the natural adjust-
ment processes cannot be allowed to work themselves out , there must
be deliberate monetary contraction. Devaluation, tariff policy and
other expenditure-switching policies are simply substitutes for mone -
tary contraction and will improve the balance of payments only if
they increase the demand for money, for example through a real-
balance effect," by raising the domestic price level. All balance of-
payments policies , it is argued, must be analysed in terms of their
The Balance ofPayments and the National Economy 71

effect on reducing monetary disequilibrium . In this monetary frame-

work the effect of devaluation does not depend directly on the elas-
ticities of demand for exports and imports but on whether the price
effects of devaluation produce a reduction in real expenditure relative
to income by increasing the nominal demand for money. It also follows
from the monetary approach that the effects of balance-of-payments
policies, as well as balance-of-payments disequilibrium itself, must be
transient in nature, unless stock disequilibrium is continually recreated
by domestic credit changes.
3
The Elasticity Approach to
the Balance of Payments
THE MARSHALL-LERNER CONDITION DERIVED

As indicate d in the last chapter the elasticity approach to the analysis

of balanc e-of-payments adjustment based on the Marshall-Lerner
condition rests on several restrictive assumptions. First , the analysis
is founded upon partial equilibrium in the sense that it considers only
the effect o f exchange -rate variations in the market for exports and
imports, and everything else is held constant , so that the position of
the demand curves for exports and imports themselves are held con-
stant. In practice everything else will not remain constant. Exchange-
rate changes will have price effects elsewhere in the system which will
shift th e demand curves for exports and imports. Income will also
change, affecting the demand curves for exports and impo rts . A second
restrictive assumption is that all relevant elasticities of supply of output
are assumed to be infinite so that the price of exports in the home
currency does not rise as demand increases , the price of foreign goods
that compete with exports does not fall as demand for them falls, the
price of imports in foreign currency does not fall as the demand for
imports falls, and the price of domestic goods competing with imports
does not rise as the demand for import substitutes increases . There are
four elasticities of supply to consider : the elasticity of supply of
exports; the elasticity of supply of foreign goods that compete with
exports ; the foreign elasticity of supply of imports; and the elasticity
o f supply of home goods that compete with imports. The basic
Marshall-Lerner condition for a successful currency depreciation
assumes all four supply elasticities to be infinite. The simple formula
The Elasticity Approach to the Balance ofPayments 73
can be modified to incorporate the elasti city of supply of imports and
exports, but the elasticity of supply of goods that compete with im-
ports and exports is still assumed to be infinitely elastic . Third , the
elasticity approach ignores the monetary effects of exchange-rate
changes. FinalIy , it is assumed that trade is initially balanced and that
the change in the exchange rate is a small one. The Marshall-Lerner
condition is easily mod ified to cover the case where trade is initially
unbalanced , but the smalI-change assumption is necessary so that
second-order interaction terms arising from changes in multiplicative
variables can be ignored.
On the above assumptions the MarshalI-Lemer condition states that
devaluation will improve the balance of payments on current account if

(3.1 )

where Em is the price elasticity of demand for imports , and Ex is the

price elasticity of demand for exports.
The Marshall-Lerner condition can be derived in a number of ways ,
and with respect to measurement in foreign or domestic currency . Since
the essence of a balance-of-payments problem is a shortage of foreign
currency, it is more appropriate to conduct the analysis measuring
exports and imports in units of foreign currency. If the focus of
attention were on the impact of exchange-rate changes on domestic
income and employment , it might be more appropriate to work with
units of domestic currency. A simple formal proof of the MarshalI-
Lerner condition is as follows . Let

B= pX -M (3 .2)

where B is the balance of payments measured in foreign currency,

X is exports measured in domestic currency , p is the exchange rate,
i.e. the foreign price of a unit of domestic currency , and M is imports
measured in foreign currency. Devaluation will improve the balance of
payments if dB/dp < O. Differentiating equation (3 .2) with respect
to a change in p (i.e. taking a smalI change in the exchange rate) :

dB ax
-- = X +P -
dp ap
74 Balance-ofPayments Theory and U'K . Experience

Now -(!!.. x ax) measures the price elasticity of demand for exports
X ap
(Ex) assuming that the domestic price of exports remains unchanged .

(!!.
Ex > 0 if exports respond positively to a fall in the exchange rate.
Similarly, x ai!'!)
measures the price elasticity of demand for
M ap
imports (Em) assuming that the foreign price of imports remains the
same . Em > 0 if imports fall with a fall in the exchange rate . As long
as trade is initially balanced with M = pX, we have, from (3.3) .

dB
- = X(l-t'x - Em ) (3.4)
dp

so that dB/dp < 0 if Ex + Em > 1. In words, the balance of payments

will improve with a fall in the exchange rate if the sum of the price
elasticities of demand for imports and exports exceeds unity .
The result can be grasped intuitively by appreciating that if export
volume were to rise proportionately with a reduction in the foreign
price of exports (t'x = 1), the balance of payments would remain
unchanged, and any reduction in imports as a result of a higher dom-
estic price of imports (Em > 0) would be sufficient to improve the
balance of payments. Under the conditions of the theorem, any com-
bination of export and import demand elasticities is sufficient provided
their sum exceeds unity .
The Marshall-Lerner condition can also be illustrated diagrammatic-
ally . Take the two-country model of the United Kingdom and the
United States used in Chapter 1. Figure 3.1 represents the supply of
and demand for U.K. exports in relation to the dollar price of exports.
The supply is assumed infmitely elastic, and the demand negatively
related to the dollar price . Assume that trade is initially balanced at
the prevailing exchange rate . Devaluation reduces the dollar price of
exports from P to PI , causing a shift in the supply curve from SS to
SIS I and a movement along the demand curve from A to B. The
level of dollar receipts from exports before devaluation is OPAQ,
after devaluation it is OPtBQt . Whether the new level of foreign-
exchange receipts is greater than the old depends on the price
elasticity of demand for exports. If Ex < I, receipts go down ;
if Ex = I, receipts stay the same; and if Ex > 1, receipts increase.
 The Elasticity Approach to the Balance ofPayments 75
$ price of o
exports

P 5 5

P, 5, 5,

o 0 0, Supply and demand

for exports

FIGURE 3.1

$ price of 0
imports 0"
\
\

P 5 0, C 5
\
I
I ,
,
-,

"OJ 0
o 0,0 Supply and demand
for imports

FIGURE 3 .2

Figure 3 .2 represents the supply of and demand for U.K. imports.

The supply is assumed infinitely elastic and the demand negatively
related to the dollar price. Devaluation itself does not alter the dollar
price of imports but will raise the sterling price . The effect of devalu-
ation is thus to reduce the demand for imports at any given dollar price
provided demand with respect to the sterling price is not totally in-
elastic . The demand curve shifts leftwards from DD to D,D, and the
quantity of imports demanded falls from Q to Q, . Dollar expenditure
on imports will fall from OPCQ to OPDQI . The total change in foreign-
exchange receipts depends on the behaviour of both exports and
76 Balance-of-Payments Theory and U'K, Experience

imports. If the demand for exports is totally inelastic, the demand for
imports would have to be of at least unitary elasticity for the loss of
foreign-exchange earnings from exports to be offset. At the other
extreme, if the demand for imports is totally inelastic . the demand
for exports would have to be of at least unitary elasticity to prevent
foreign-exchange earnings from falling. Any combination of export and
import elasticities in between is possible provided their sum exceeds
unity.
The percentage improvement in foreign-currency receipts resulting
from devaluation will be approximately equal to the sum of the per-
centage increase in the volume of exports plus the percentage decrease
in the volume of imports minus the percentage decline in the foreign
price of the domestic currency. Th is may be expressed as

~r: (dX/X + dM/M _ I) 100 (3.5)

p dplp dp[p

For a small (l per cent) devaluation of the exchange rate , the Marshall-
Lerner condition for an improvement in foreign-exchange receipts is
easily seen as Ex + Em - I > O,orEx + Em > I .
If the balance of payments is not initially balanced , it is clear from
equation (3.3) that the Marshall-Lerner condition must be modified .
Measured in terms of foreign currency, the condition for a successful
devaluation starting from imbalance is

(3.6)

Thus the larger the initial deficit (M > pX), the less stringent is the
elasticity condition for a successful devaluation .'
As indicated above, the Marshall-Lerner condition also hinges
crucially on the assumption that four elasticities of supply are infinite .
Robinson (1937) has incorporated the elasticity of supply of exports
and imports into the Marshall-Lerner condition. The algebra is tedious
and we will simply give the result. The condition for the balance of
payments to improve, following a devaluation , is

(3 .7)
The Elasticity Approach to the Balance ofPayments 77
Equation (3.7) reduces to the Marshall-Lerner condition of Ex + Em
> 1 if Sx = Sm = 00 , and pX = M.2 If the elasticity of supply of foreign
goods that compete with domestic exports and the elasticity of supply
of domestic goods that compete with imports are not infinite, the
demand curves for imports and exports themselves will be affected , and
this remains a limitation of the Marshall-Lerner condition along with
other ceteris paribus assumptions . The importance of holding the
supply and demand curves for foreign currency constant in deriving
the elasticity conditions for a successful devaluation was shown in
Figure 1.6 in Chapter 1.
In conditions when resources are unemployed , as (for example) in
the 1930s when the elasticity approach was first formulated , it may
not be unreasonable to assume that supply elasticities are infinite .
In conditions close to full employment the assumption is much more
questionable . Also , in conditions of unemployment devaluation will
raise income in the devaluing country, which will then increase the
demand for imports, and decrease the demand for exports if income
falls in the non -devaluing countries. To incorporate income effects
into the Marshall-Lerner condition it is necessary to make imports
and exports not only a funct ion of variations in the exchange rate
but of income too. The incorporation of the income effects of de-
preciation does not alter the Marshall-Lerner condition for a successful
devaluation, but it will lower the magnitude of the effect of depreci-
ation on the balance of payments. To show this let :

Y =A + (1 - s) Y + B[p (3 .8)

where Y is money income ; A is autonomous expenditure in money

terms ; B is the balance of payments measured in foreign currency and
is converted into domestic currency by the exchange rate,p ;ands is the
marginal propensity to save. From equation (3.4) let :

dB = X(l-Ex - Em ) dp -m(dYp) (3 .9)

where m is the marginal propensity to import in the devaluing country .

Now assuming no change in the money value of autonomous ex-
penditure (dA = 0), we have from equation (3 .8) : dY = dB/sp . Sub-
stituting this result into (3 .9) gives:

dB = _s- X(l -Ex -Em)dp

s+m (3 .10)3
78 Balance-of-Payments Theory and UiK . Experience

The sign condition for the price elasticities to improve the balance of
payments is clearly not affected (the price elasticities must still sum to
greater than unity) , but since sj(s + m) is a fract ion , the magnitude of
the change in dB will be smaller than if the expansionary effects are
ignored . The Marshall-Lerner condition could not be more stringent
(holding A constant) since if the ~ncome effects outweighed the
'elasticity' effects, so as to worsen the balance of payments, the de-
preciation of the currency could not be expansionary.
A seemingly contrary result has been derived by Harberger (1950)
and Stern (1973), among others , which suggests that the income effects
of devaluation alter the Marshall-Lerner condition , making it more
stringent. The explanation is that these models hold real autonomous
expenditure constant allowing autonomous expenditure in money
terms to vary in response to devaluation , so that. in a two-country
model, money expenditure rises in the devaluing country and falls in
the non-devaluing country . This raises imports in the devaluing country
and reduces the exports of the devaluing country to the non-devaluing
country . Allowing for these effects , it can be shown (e .g. see Stern
(1973» that the condition for depreciation to improve the balance
of payments (starting from equilibrium) becomes :

(3 .11 )

where m 1 is the marginal propensity to import in the devaluing country

and m2 is the marginal propensity to import in the other country.
Thus while the traditional result derived by Robinson (1937) and
Brown (1942) holds autonomous expenditure constant in money
terms , the result above holds real expenditure constant and allows
money expenditure to vary. It is not obvious which specification should
be preferred a priori, or which is the most realistic. Presumably if the
government wishes the devaluation to be as successful as possible it will
try to prevent money expenditure from rising, in which case the
traditional result would hold . On the other hand, if a successful de-
valuation is interpreted to mean a devaluation which improves the
balance of payments without real income falling, the Harberger-Stern
condition would have to be satisfied .

DEVALUATION AND THE RESPONSE OF FIRMS

What actually happens to the volume of exports and imports in practice

as exchange rates change depends very much on the feasibility of a
The Elasticity Approach to the Balance ofPayments 79
supply response to meet greater export demand and the demand for
import substitutes, and how manufacturers respond to a change in the
value of the currency with respect to the pricing of their products.
For devaluation to inc rease exports , either the foreign price must be
lowered or the home price raised to encourage firms to produce more
for export given the increase in the profitability of exports. For de-
valuation to lower imports the sterling price of imports must be raised ,
or foreign supply discouraged. A number of factors , however. may
prevent the adjustment of prices and may nullify the effectiveness of
exchange-rate changes. The structure of the market in which firms
operate may make it unwise to alter prices, for example under
oligopoly if firms are uncertain about rivals' reactions . Commodities
may be subject to international price agreements and cartel arrange -
ments . There are costs associated with altering prices; accounting costs
and loss of goodwill . There may be fear by exporters that other
countries will retaliate. Exports and imports way be traded on the
basis of long-term contracts signed in terms of domestic currency. This
is particularly likely to be true in the case of heavy capital goods, with
the unfortunate side-effect that if there is the expectation by manu-
facturers of higher home costs as a result of .devaluation, this will
result in a much higher contract price for exports than would other-
wise have been the case. According to many industrialists, devaluation
reduces competitiveness on a continuous basis for a substantial pro-
portion of U.K. exports." Another factor to bear in mind is that ex-
porters may invoice orders in the home currency leaving overseas agents
to price the product in foreign currency. In this case, which is said
to be the typical practice in U.K. industry, the gain to the fum in terms
of profits or greater sales volume becomes very uncertain .
A survey in the Banker (September 1976) reported that 81 per cent
of U.K. goods are denominated in sterling . If this is so , and the prices
of goods are held constant in foreign currency, the retailers, whole-
salers, importers and agents will share the profits from devaluation.
This will have beneficial consequences if it encourages the stocking
of U.K. goods but the supply response may be weaker than if the
extra profit accrued to the firms themselves .
It is clear that there may be a clash between the national interest
and the interests of the firm in the firm's reaction to a devaluation .
If conditions are such that it is most profitable to keep the foreign
price of the product the same and to raise the home price, this will
not increase exports directly - only indirectly to the extent that the
higher profitability of exports induces a greater export effort. The
80 Balance-ofPayments Theory and U'K. Experience

effort may not be forthcoming, however , if there are supply difficulties ,

with the domestic economy so buoyant that the profitability of home
sales is still greater than the profitability of exports. In a study of U.K.
firms by Gribbin (1971) , prior to the 1967 devaluation of sterling,
it was found that the profit rate on export sales was lower than on
home sales for a significant proportion of firms in the sample , owing
largely to tougher price competition i'n foreign markets. No relation
was found, however, between differences in the rate of profit on home
and export sales and the rate of growth of exports of firms .
Let us consider in more detail the theoretical expectation of the
response of firms to devaluation in different market conditions, and
then examine some of the recent empirical evidence . We shall draw
largely on the work of Hague, Oakeshott and Strain (1974), based on
the 1967 devaluation , and of Holmes (1978). Assume that firms
attempt to maximise profits and that they export all the ir output to
one market, and consider how a firm will set its export prices after a
devaluation under the four different market structures of perfect com-
petition, monopoly, monopolistic competition and oligopoly.
Under perfect competition there is no pricing decision for the firm
to make . The demand curve facing the firm is perfectly elastic at the
given world market price . The firm should raise the home price of
the product by the full amount of the devaluation and increase the
volume of exports to profit from the increase in the home price .
According to the Marshall-Lerner condition incorporating supply
elasticities (equation (3.7)) , the greater the demand elasticity, the
greater the supply elasticity should be for a successful devaluation .
Under a monopoly in the export market the pricing decision is
complicated . In Figure 3.3 the curves labelled ' I ' represent the average
and marginal revenue curves from exports before devaluation , and the
curves labelled '2' represent the average and marginal revenue curves
after devaluation . To simplify the analysis we assume constant average
costs . Before devaluation the firm maximises export profits at X I at
price PI (in home currency), where marginal revenue (MR I) equals
marginal cost (MC). Devaluation raises the home price of foreign
currency, and therefore increases the home-price equivalent of each
foreign price . With no increased foreign competition this shifts the
average and marginal revenue curves upwards by the full amount of the
devaluation to AR 2 and MR 2 respectively . The firm now has a choice :
it could maximise profits with an export level of X 2 , raising the home
 The Elasticity Approach to the Balance ofPayments 81
Price (in home
currency)
I,
P3 - - - -\-i'
II "
P2 - - - -1--1,
II I '
P 1 I-----'r-~~ - - -('-
\ I I'
\1
f----+---4-~:------"--------ACMC
,, =
,
AR,AR 2

X, Quantity of exports

FIGURE 3.3

price from PI to P 2 which is not equal to the full amount of the de-
valuation ; or it could keep the level of exports the same at X I and
raise the home price by the full amount of the devaluation to P 3 (which
would not maximise profits). Alternatively, it could keep price at
PI and increase the quantity of exports to X 3 (which would not
maximise profits either). If the monopolist is uncertain about demand
conditions and believes that the demand curve facing him is very
inelastic, he may well prefer to do nothing and stay at Xl with a home
price of P3 . The sacrifice of profits would be minimal. According to
equation (3 .7) , the lower the demand elasticity, the lower the supply
elasticity should be for a successful devaluation . Any export volume
response should take the form of firms attempting to sell more at the
same foreign currency price by improved design, marketing , delivery
and so on.
Under monopolistic competition in the export market, with no
increase in foreign competition, the choices facing firms will be the
same as those in the case of the monopolist discussed above . How
one firm reacts will largely depend on how other firms react. If one
or two firms decide to increase exports , others might try to follow
suit, and there may be a general price-cutting war. On the other hand,
if the profitability of exports before devaluation was unsatisfactory ,
firms may decide to raise the home price of exports, though not by
the full extent of the devaluation.
82 Balance-of-Payments Theory and U.K. Experience
Under oligopoly, because of the strong interdependence between
producers and the uncertainty among producers over how others
will react if one firm alters price , foreign currency prices are likely
to be sticky and oligopolists will take the benefit of devaluation in
higher home prices, leaving output unchanged.
In none of the cases examined can we necessarily expect a strong
quantity response to devaluation , and in some cases it would not be
desirable. In most cases it would seem to pay firms to raise the home
price of their exports and reap the benefits of devaluation in the form
of a higher profit per unit of exports. This will only benefit the balance
of payments to the extent that the desire of firms to export is in-
creased and more can be sold without lowering the foreign price .
Apparently , however, many firms regard the difference between the
profitability of exports and home sales as a meaningless distinction .
For most firms, production for one market supports production for the
other, in the sense of helping to keep average costs of production down,
and therefore the profitability of sectors is interdependent. The com-
parative profitability of exports is irrelevant to the export decision.
This conclusion emerges strongly from a study by Hovell (1968)
of fifty U.K. companies producing agricultural machinery , mechanical
handling equipment and te xtile machinery .
As far as pricing is concerned, Hague , Oakeshott and Strain (1974)
found in their stud y of nineteen firms following the 1967 U.K. de-
valuation that the majority of the firms contemplated only the two
extreme policies of either no change in price or the full change equal
to the percentage devaluation, when a compromise would probably
have been optimal depending on the demand elasticity for the output
and the capacity to supply. Most firms , however , appeared to be
ignorant of the elasticity of the demand curve facing them and had
little idea about competitors' reactions . In over one-half of the firms
the same decision following devaluation was taken with respect to all
export markets , regardless of differences between markets. Only two
firms had worked out a plan of action in anticipation of devaluation
and none seemed quite sure what the national interest required,
whether to reduce foreign prices or to raise sterling prices . On the
import side it appears that none of the firms practised import sub-
stitution as a result of the increase in the home price of imports mainly
because of a lack of domestic substitutes and because import sub-
stitutes were also raised in price by home suppliers.
In another study of twenty-nine firms in the electrical and mech-
The Elasticity Approach to the Balance ofPayments 83
anical engineering industries conducted by Rosendale (1973) at the
National Institute of Economic and Social Research, seventeen firms
claimed that they benefited from the 1967 devaluation in terms of
higher sales revenue for periods of one to three years, after which
rising costs eliminated the advantage : 30 per cent of the forty-one
products distinguished were raised in price by the full amount of the
devaluation (Le. there was no direct volume response), and in 50
per cent of cases there was no rise in sterling prices .
A study by Holmes (1978) considers the behaviour of fifty-four
large companies and their pricing behaviour following the 1967 de-
valuation and exchange depreciation since 1972 . Many of the large
exporters operate in oligopolistic markets, and his findings accord
with our earlier theoretical prediction that where this is so foreign
prices will not be cut for fear of retaliation . Supply constraints also
militate against cutting foreign prices.

LAGS AND THE J-eURVE EFFECT

Even if a volume response is forthcoming , and the Marshall-Lerner

condition is ultimately satisfied , there may be a long lapse of time
before the quantities adjust sufficiently to offset the change in the
price of foreign exchange, making the balance of payments worse
before it gets better. As the balance of payments worsens, domestic
policy and the achievement of other goals may be upset considerably .
In the first place devaluation may not immediately affect the relative
prices of traded and non-traded goods if foreign trade is subject to
forward contracts . If export prices are fixed in sterling and imports in
foreign currency, under contract for some months ahead, foreign-
exchange receipts will suffer as a result of devaluation and the balance
of payments will worsen . Once relative prices have changed (if they
change) there will still be recognition lags, decision lags, delivery
lags and production lags, all of which may produce in the short term
a less than proportionate response in exports and imports to the fall
in the value of the currency. Artus (1973) has calculated the average
export delivery delays from three countries in the fourth quarter of
1971 to be : for machinery, 400 days for West Germany , 297 days for
the United Kingdom, and 107 days for the United States; and for
machine tools, 308 days for the United Kingdom and 217 days for the
United States. In the work of Junz and Rhomberg (1973) exchange-
rate changes only seem to affect export-market shares significantly
84 Balance-ofPayments Theory and U'K, Experience
after a lag of some three years . The experience of the United Kingdom
will be discussed more fully later in the empirical chapters. There is also
the phenomenon of physical leads and lags (in addition to the monetary
leads and lags discussed in Chapter I). If devaluation is seen by traders
as a prelude to further devaluation (or currency depreciation is seen as
a prelude to further depreciation under a system of floating exchange
rates) , domestic importers will accelerate orders through fear of having
to pay more for goods in the home currency later , while foreign im-
porters delay their orders for the exports of the devaluing country in
the hope of buying them more cheaply later in terms of their own
currency. This will cause the volume effect from devaluation (at least
in the short term) to be perverse, worsening the balance of payments.
The short -term worsening of the balance of payments resulting
from devaluation for the reasons mentioned above is sometimes
described as the i -curve effect since the plotting of the balance of
payments against time traces out such a shape (see Figure 3.4) .5

Balance -of -
payments
surplus

c
Ol-----~'---------__,_-
time
A

D
Balance-of-
payments
deficit

FIG U R E 3.4

The time span that elapses before the perverse effects of devaluation
are overcome could be quite long . There is not only the time taken for
the balance of payments to register improvement to be taken into
account (A to B) but also the time taken for the losses of devaluation
between A and B to be recouped (B to C). Estimates for the United
Kingdom by the Treasury suggest that the initial deterioration in the
current account lasts about two quarters (A to D) and that the cumu-
lative loss is eliminated within a year. Masera's study (1974) of the
1967 U.K. devaluation, discussed in more detail in Chapter 7, suggests
about six to seven quarters for losses to be made good . But moving
up the curve cannot be guaranteed . In the context of the elasticity
The Elasticity Approach to the Balance ofPayments 85
approach to the balance of payments, the deficit will not improve
if the Marshall-Lerner condition is not satisfied . It will not improve
either if the rise in the home price of imports so affects the internal
price level that the competitive advantage conferred by devaluation
is completely offset or turned into a competitive disadvantage. There
are a number of plausible models which predict that in the absence of
strong counter-inflationary domestic policy the internal price level
may rise by the full amount of the devaluation. The models which
predict this outcome come from quite diverse schools of economic
thought, from Keynesians on the one hand to international monetarists
on the other. The experience of the U.K. economy since 1972 could be
invoked in support of such models. Two representative models of
devaluation and inflat ion are considered below .

DEVALUATION AND INFLATION

The model to be developed first is an elaboration of a model presented

by Wilson (1976) which starts by making the average price of final out-
put equal to the sum of wage costs, profits and import costs per unit of
output :

W+R +M
P=
Q (3.12)

where P is price , W is total wage costs , R is profit, M is total import

costs and Q is output. Totally differentiating equation (3.12) and
dividing by P gives

dP d(W/Q) d(R/Q) d(M/Q)

---'---+ +
P (W+R+M)/Q (W+R+M)/Q (W+R+M)/Q
(3.13)

Now

W+R+M W W+R+M R W+R+M M W+R+M

---=- x--- =- x ---=- x
Q Q W Q R Q M
86 Balance-of-Payments Theory and UK, Experience

so that

dP d(W/Q) W d(R/Q) R d(M/Q)

= --- x ---+ - - - x - - -+ - - -
P W/Q W+R+M R/Q W+R+M M/Q
M
x - - -- -
W+R+M (3.14)
or

dP dw W dr R dm M
- =-x-+ - x - + - x-
P w 0 rOm 0 (3.15)
where dwlw is the rate of change of wage costs per unit of output,
dr]r is the rate of change of profits per unit of output, dmfm is the rate
of change of import costs per unit of output, and 0 is the money value
of total output. Thus the domestic rate of inflation is equal to the
weighted sum of the rates at which the different 'costs' of output rise,
where the weights are the proportional shares of those costs in the
money value of total output.
The initial effect of devaluation (which raises the home price of
imports) is to raise internal prices by

dm M
- x -
m 0 (3.16)

This is the first-round effect. The relationship between the magnitude

of devaluation and the rise in import prices is discussed later. For the
moment it is assumed to be equiproportional.
Now assume that wages chase prices upwards and that the profit
margin is a fixed proportion of import costs and wages so that profits
per unit of output remain unchanged . The second -round effect on
prices is then

~~(W +R)=dm x~
w 0 m 0 (3 .17)

If wages then chase second-round price effects, the third -round effect
on prices is

e:: x-
M
o (W;R)J (W;R)
(3 .18)
The Elasticity Approach to the Balance ofPayments 87

and so on. The ultimate rate of increase in prices arising from the
increase in import prices is

~':. = dm x!!!.- ~ + W+ R + (~ + R_)2 ... + (lY~~) nJ

Pm OL 0 0 0
(3.19)

(3 .20)

M
[~oJ
dm
- x-
m 0 (3 .21)

Therefore

dP dm
--
P m (3.22)
that is, the internal price level increases by the same amount as the
rise in import prices.
Whether the internal price level rises in the same proportion as the
devaluation itself depends on two factors : first, the extent to which
import prices rise with respect to the devaluation ; and second, on
whether wage increases match price increases equiproportionately.
The home price of imports will rise in the same proportion as the rise
in the home price of foreign currency if (i) other countries do not alter
their exchange rate , (ii) suppliers do not alter the foreign price of
imports, and (iii) either the own price elasticity of demand for imports
is zero or the own price elasticity of supply of imports is infinite. Thus

(3.23)
where k is the 'exchange rate expressed in home currency per unit of
foreign currency , Pf is the foreign price, and K = [I - (em/sm)]-I ,
where em is the own price elasticity of demand for imports and sm
is the own price elasticity of supply of imports , and 0 < K < 1. (see
Goldstein , 1974; and Branson , 1972). The relation between dmlm and
dkfk is ultimately an empirical issue . To allow for the fact that the
relation may not be equ iproportional, let dmlm = A (dk/k) for sub-
88 Balance-of-Payments Theory and UiK. Experience

stitution in equation (3.16).

Also, how much domestic prices change depends on the precise
response of wages to prices. Ultimately this is also an empirical issue.
To allow for the fact that the relation may not be equiproportionallet
dwjw = r [(dmlm) (MIO)] for substitution in equation (3.17) . Using
the substitutions for dmlm and dwfw in equations (3.16) and (3.17)
and computing the price effects round by round gives the ultimate
rate of increase in prices arising from devaluation of

dP dk 0M{ I + 'Y (w0 R)+ [(W

Ak x
R\J2
'Y -0 )
+ +
P

(3 .24)

Thus

dk M
d.P
P
A -x -
k 0
----- - - -
[ I - 'Y (11- MIO) J (3 .25)
If A = 'Y = I , equation (3 .25) reduces to equation (3.22) . From the
model more or less pessimistic conclusions can be drawn about the
inflationary consequences of devaluation depending on the actual
values attaching to A and 'Y. As far as Ais concerned, there is no reason
why import prices should rise more than the devaluation, so that
o < A .;;; 1. It is possible, however, that 'Y may be greater than unity,
causing domestic prices to rise by more than in proportion to the
devaluation. For example, suppose that wage-earners gear their wage
demands not to the over-all increase in prices resulting from devaluation
but to the price of food, which may rise by more because the pro-
portion of total food imported is higher than the proportion of total
output imported . In that case 'Y will be greater than unity and, de-
pending on A, values of 'Y not far above unity could make inflation
'explosive'. Inflationary expectations may also cause 'Y to exceed unity .
Progressive taxation may also push 'Y above unity, since to protect
net income against a rise in prices the demand for gross income has to
be greater than the rise in prices. On the other hand, employers may
not give in to wage demands , which will act as a moderating influence.
Wages policy may also temporarily dampen wage claims, or the mone-
The Elasticity Approach to the Balance ofPayments 89
tary authorities may refuse to finance excessive wage claims so that
claims are moderated through the fear of unemployment.
The evidence for the United Kingdom following the 1967 devalu-
ation, which raised the sterling price of the dollar by 16.7 per cent,
is that import prices rose over all by between 13 and 14 per cent .
Barker (l968) estimates a 13.9 per cent increase over all using an
input-output framework, and the National Institute of Economic
and Social Research (I972) estimated an over-all increase of 13.5
per cent. This would imply an estimate for " of 0 .8. Estimates of
r, in variously specified wage-price equations, generally centre around
a value of between 0.7 and unity, rarely exceeding unity . If it is
assumed that " = 0 .8 and r = 1, prices would ultimately rise by 80 per
cent of the devaluation. If r is not equal to unity , it can be seen from
equation (3.25) that the rate at which prices rise as a proportion of the
devaluation will depend on import costs as a proportion of the value
of output (M/O) . Goldstein (1974) has attempted to estimate em-
pirically the effect of exchange-rate changes on aggregate wage and
price behaviour in the United Kingdom using a simple two-equation
model which incorporates the relationships outlined in the model
above, namely : the effect of a change in import prices on domestic
prices (assuming that import prices rise by the full amount of the
devaluation , " = 1); the effect of prices on wages; and the effect
of wages on prices. The two-equation model is

dw dP
- = ao + al U + a2 -
w P (3.26)

and

dP dw dm
- = b o + b, - + b2 Q + b 3 -
P w m (3.27)

where U is the percentage level of unemployment, Q is productivity

growth, and the other variables are as defined above. The relationship
between the change in import prices and the rate of inflation is given by

~ (dP/P)
a(dm/m) (3.28)
90 Balance-of-Payments Theory and U'K. Experience
In our model elaborated above b 3 is constrained to the value of the
ratio M/O (see equation.(3 .16)), and b 1 is constrained to the value of
(W + R)/O (see equation (3.17)), so that b 1 + b 3 = 1. The relationship
between dP/P and dmlm then depends on a2 . If a2 = I, dP/P = dmjm ,
and the increase in prices is equal to the devaluation. If a2 < I, prices
rise less than the devaluation ; and if a2 > 1, inflation is possibly ex-
plosive. In Goldstein's empirically estimated model for the period
1954-71 using quarterly data the estimates for the post-devaluation
period are b 3 = 0.19 , b , = 0 .76, and a2 = 0.56 . Substituting these
values into equation (3.28) gives the result that prices rise by only 33
per cent of the rise in import prices , and by a lesser percentage of the
devaluation if import prices do not rise by the full amount of the
devaluation . These estimates appear to be very much on the low side
owing to the low estimate of a2 . In recent years, at least, a2 has been
close to unity.
Harrod (l967b) has described devaluation as the most potent known
instrument of domestic price inflation which has such sorry effects
on human misery; and it has long been recognised that the inflationary
effects of exchange-rate depreciation may counteract any relative
price advantage conferred . Meade (1951) in his early pioneering work
recognised that 'it would be useless to turn to the mechanism of
variable exchange rates [to rectify imbalance I unless there were
sufficient flexibility of real wage rates' ; in other words devaluation will
be ineffective if wages chase prices upwards . Henderson (1949) also
expressed fears that all devaluation might do would be to promote
internal inflation. Even Friedman (1953), a leading advocate of float-
ing exchange rates , has conceded that the inflationary repercussions of
exchange depreciation may be an objection to floating exchange
rates in a particular country at a particular time .
The international monetarist school of economists, based in the
United Kingdom at the London Business School, is also sceptical of
exchange-rate adjustment. It argues that exchange depreciation is
irrelevant to balance-of-payments adjustment not only because balance-
of-payments deficits and policies must necessarily be transient, as in the
monetary approach to the balance of payments (see Chapter 5), but also
because domestic prices will ultimately rise by the extent of the de-
valuation. Ball, Burns and Laury (1977) of the London Business School
develop three models, each of which predicts that, with free collective
bargaining, wages and prices are likely to rise eventually by the full
extent of the devaluation. They are wholly sceptical of the elasticity
approach , at least as far as imports are concerned. The authors say :
The Elasticity Approach to the Balance ofPayments 91

we have been unable for a number of years to establish any signifi-

cant relationship between relative prices and the demand for imported
goods in the U.K. Any effect on the import of goods resulting from
changes in the exchange rate affect the level of imports only through
their expenditure reducing effects - there are no substitution effects.

Their first model is the same as the model empirically tested by

Goldstein using equations (3 .26) and (3 .27) . Their second model is
structurally the same as the first model , except that the G.D.P. deflator
rather than consumer prices enters the wage equation with a coefficient
of unity, and the G.D.P. deflator is a weighted average of changes in
unit labour costs and world prices . Their third model is often referred
to as the 'Scandinavian ' model of inflation where the price of traded
goods adjusts to the level of world prices, and wages in both the traded -
and non -traded-goods sectors adjust to the traded-goods price level.
so that any change in exchange rates will ultimately increase the dom-
estic price level by an equivalent amount. The model is as follows. First,
let

dP
P (3.29)

where PT is the price of traded goods , PN T is the price of non-traded

goods, and Al is the share of traded output in total output. Second

(3.30)

where Pw is the world price level, and k is the exchange rate (i .e. the
home currency price of foreign currencies). Third

(3.31 )

where wNT is wages in the non-traded-goods sector, and QNT is

productivity in the non-traded-goods sector. Assume
92 Balance-of-Payments Theory and U.K. Experience

dWNT
WNT (3.32)

Substituting (3.32) into (3.31) and the result into (3 .29) gives:

~ = x, dP T + (I-AI)
P PT
r
l PT
Q
QT
Q
dP T + {d T _ d N
QN;f
rt ]
(3.33)

Remembering that dPT/P T = (dP w/P w)k, the model says that the
rate of price increase will be equal to the rate of increase in the world
price index measured in local currency plus a proportion (I - Ad of
the gap between productivity growth in the traded- and non-traded-
goods sectors. The influence of world prices is dominant and any fall in
the exchange rate will ultimately increase the domestic price level by
an equal amount (unless productivity growth in the non-traded-goods
sector is markedly higher than in the traded-goods sector).

DEVALUATION AND THE TERMS OF TRADE

The elasticity approach to balance-of-payments adjustment has also

traditionally been concerned with the effects of devaluation on the
barter terms of trade ." The effect that devaluation has on the terms of
trade is also important for other theoretical approaches to balance-of-
payments adjustment , particularly the absorption approach . The
matter can be considered quite briefly because no a priori prediction
can be given as to whether the terms of trade will deteriorate or im-
prove . It was originally assumed that the terms of trade would change
in the same direction and to the same extent as the devaluation . But
this argument ignores the income, and hence expenditure , effects of
devaluation, which will be partly a function of the price elasticities of
demand. It can be shown (e.g . Stern, 1973) that the barter terms of
trade will worsen or improve with a devaluation depending on whether
the product of the elasticities of supply of exports and imports (Sx Sm)
is greater or less than the product of the elasticity of demand for
exports and imports (Ex Em) ' If Sx Sm > Ex Em the terms of trade
will worsen and if Sx Sm < Ex Em the terms of trade will improve .
It was originally argued that because a country is more specialised
in exports than imports, and given conditions of unemployment in
 The Elasticity Approach to the Balance ofPayments 93
which the elasticity of supply of exports is likely to be high , the terms
of trade are likely to deteriorate. In practice the relevant elasticit ies will
vary from country to country, and from circumstance to circumstance,
and no generalisation can be made .
It is also relevant to mention that if the interest is in how devalu-
ation affects real income through a terms-of-trade effect, it is not the
barter terms of trade which are important but the single factoral terms
of trade . This is equal to the net barter terms of trade adjusted for
changes in the productivity of resources employed in the export sector
in order to measure the total change in import-buying power. This
question is considered more fully in the next chapter on the absorption
approach to the balance of payments.

THE DIFFICULTIES OF MEASURING PRICE ELASTICITIES

No discussion of the elasticity approach to devaluation would be

complete without mentioning the practical difficulties of measuring
the price elasticity of demand for exports and imports on which the
Marshall-Lerner condition depends so heavily. The difficulties are
numerous. Orcutt (1950) has enumerated systematically the sources
of bias that can creep into the estimat ion of price elasticities of
demand . Orcutt orientated his argument towards those factors that
may bias empirical estimates of demand elasticities downwards, mainly
in an attempt to counter the elasticity pessimism prevalent at the time .
Some of the points he makes are valid econometric ones. In the case
of some of the factors the bias could go either way . In other cases
subsequent analysis and research suggest that the point may not be an
important one. Some other factors not mentioned by Orcutt may bias
the measurement of price elasticities upwards .
Orcutt lists five factors biasing downwards empirically estimated
price elasticities of demand for internationally traded goods : simul-
taneous-equation bias; aggregation bias; bias due to random observation
errors in the price indices; short-run elasticities less than long-run
elasticities; and elasticities lower for small price changes than for large
ones. Let us consider these factors in turn.
The possibility of simultaneous-equation bias arises, as in all supply
and demand analysis, because demand depends on price and price
depends on quantity demanded if supply curves are less than perfectly
elastic. Consider Figure 3.5. If the demand and supply curves shift
94 Balance-of-Payments Theory and U.K. Experience
together, the observed PQ points will lie on the dashed-line EE , which ,
if interpreted as a demand curve, would in this example give a much
lower price elasticity of demand than that given by the shape of the
true demand curve DD . The problem referred to here would not arise
if the price of im ports (or ex ports) could be regarded as predetermined
rather than endogenous. Th is could be assumed for a small country ,
which would then be a price -taker in world markets , or if the supply is
perfectly elastic, in which case a shift in demand could not influence
price and there would be no simultaneous-equation bias. Another
approach which enables price to be treated as predetermined is to
assume that the system is recursive so that cause and effect are assumed
not to occur simultaneously. If price cannot be considered predeter-
mined, the problem of simultaneous-equation bias can be minimised by
making sure in the estimation procedure that it is the demand curve
that is being estimated, and that no supply response to price is being
picked up through shifts in demand . The way to do this is to consider
explicitly all possible factors affecting demand so that the demand
curve can be considered fixed. Then if there have been shifts in supply
caused by factors other than those that shift demand , it can be assumed
that the estimated PQ relationship is a demand relationship . In Figure
3.5 , if DD is held constant, the shift in supply from SS to SIS I , which
causes price to rise from Y to X , will trace out the true demand curve .
Simultaneous-equation estimation techniques can also be used to
minim ise bias. When these are used the estimated elasticities are not

o s

'----------------0

FIGURE 3.5
The Elasticity Approach to the Balance ofPayments 95
significantly higher than those obtained by cruder estimation pro-
cedures. It is now generally conceded that downward simultaneous-
equation bias is much less than first suspected, probably because no
country is so big relative to world demand for world price to be
affected by its demand.
Aggregation bias refers to the fact that there tends to be a cor-
relation in practice between the price elasticity of goods and the
observed range of price variation . This means that those goods which
are elastic have a smaller impact on the aggregate price index than the
importance of the imports warrants, and the elasticity estimates using
aggregate data will be unduly influenced by goods with a low price
elasticity. The aggregate estimate will tend to be biased downwards.
A 'true' aggregate price elasticity incorporating all the disaggregated
information, and which when multiplied by the aggregate price change
gives an estimate of the total quantity change free of aggregation bias,
is (see Magee, 1974)

ep = I epj (mj/M) -
j
dPj/Pj
-
dP/P (3.34)

where ep is the aggregate price elasticity, epj is the individual product

price elasticity, m;lM is the share of product demand in total demand,
P is the aggregate price level, and Pj is the individual product price.
Equation (3.34) can be derived in the following way. Take the case
of imports. From a typical aggregate import demand function we can
write:

dM dY dP
M = "v Y + ep P
(3.35)

where M is aggregate imports, Y is aggregate income, and ey is the

income elasticity of demand for imports. From the disaggregated

.z.,
import demand functions we have:

dM =~dmj (dy;\(mj)+~epj (dPj~j)

M I mj I Yj) M I r, J\M
(3.36)

For equation (3.35) to be compatible with equation (3.36) we must

have:
96 Balance-of-Payments Theory and U.K. Experience

(3 .37)

and

(3.38)

From equation (3.38) the aggregate price elasticity consistent with the
disaggregated data is

'\ dP;/P;
ep =Le p ; (m;/M) - -
i dP/P (3.39)

which is equation (3.34). ep is the weighted sum of the individual price

elasticities only if (dP;/P;) /(dP/P) = I for all i, or if ep ; and (dP;/P;)/
(dP/P) are not correlated . If they are negatively correlated , as Orcutt
suggests, the aggregate elasticity estimate will be biased downwards.
Opinions differ as to the likely strength of this correlation. The
problem can , of course , be avoided by analysing the elasticity of
individual commodities separately and taking a weighted average.
The difference between this result and the aggregate result would be
one test of bias. Prais (1962) disaggregated U.S. imports into five major
commodity groups and estimated the appropriate parameters. The
weighted average of the separately estimated elasticities was 1.6,
exactly equal to the estimate using aggregate price and quantity data,
suggesting no aggregation bias. Ball and Marwah (1962) , in their study
of U.S. imports, also concluded the same. On the other hand, Barker
(1970b) fmds for the United Kingdom that relative price effects
measured in an aggregate import demand function are both much
smaller and less significant than those measured using disaggregated
data. For the period 1955 to 1966 the directly estimated aggregate
price elasticity of demand for imports is approximately -0.1 . The
aggregate estimate derived from a weighted average of disaggregated
(five groupings) data is close to -0 .7. Barker also makes the point that
in analysing the effects of devaluation, pre-devaluation estimates of
elasticities from aggregate data cannot be relied upon to hold in the
post-devaluation periods because devaluation will alter the values of
(dP;/P;)/(dP/P) . Import shares (m;/M) may also alter.
The Elasticity Approach to the Balance ofPayments 97
On the question of observation errors as a source of bias, it is true
that observation errors in the price series will bias the price elasticity
towards zero, but only if the errors in the quantity variable are un -
correlated with the price (and income) variables . If the errors in the
quantity variable are negatively correlated with the price variable
(which they might be if the quantity variable has been obtained by
dividing a value series by the price series subject to random errors), the
estimated demand elasticity will exceed the actual elasticity in absolute
value if the estimate is less than unity .
Short -run elasticity estimates may be lower than long-run elasticities,
but this is hardly a 'source of bias' . Long-run estimates can be made .
Elasticity estimates for small price changes may be lower than for
large price changes because there are economic and psychological costs
in commod ity substitution, and because the perception of price changes
is more apparent in the case of large changes . However, evidence
gathered by Goldstein and Kahn (I976) for twelve countries over the
period 1955-73 does not support the hypothesis .
In response to Orcutt's five points, therefore, it can be concluded
that while biases may exist in the aggregate price-elasticity estimates,
they may go either way depending on the data used and on the
equation specification . There is also an additional important factor
which may give an upward bias, and that is the operation of non-price
factors . If markets are cleared, as many are, by non-price factors
which are correlated with prices, but these other variables are not
included in the estimating equations, the price variable will pick up
these effects and make the elasticity of demand with respect to price
look greater than it is.
Finally , we make the point (stressed in Chapter I) that the elas-
ticity of demand for imports that have domestic substitutes is a
weighted average of the domestic demand and supply elasticities of the
commodity in question . If the domestic supply elasticity of the corn-
modity varies, so will the elasticity of demand for imports. This is really
the starting-point of the absorption approach to the balance of pay-
ments which argues that the response of the balance of payments to
price changes depends on supply as well as on demand , and more
broadly on the functioning of the total economy.
4
The Absorption Approach
to the Balance of
Payments
The weaknesses of the elasticity approach to balance-of-payments
adjustment can be summed up by saying that it is partial-equilibrium
analysis ; it ignores supply conditions and cost changes as a result of
devaluation. and it tends to neglect the income and expenditure effects
of exchange-rate changes. At the very least the elasticities used by the
approach ought to be total elasticities, not partial elasticities . But
taking the total elasticities of exports and imports is tantamount to
examining the relation between the balance of payments and the
functioning of the economy as a whole. This insight is the starting-
point of the absorption approach to the balance of payments which was
originally developed by Alexander (1952) and subsequently elaborated
on by Johnson (1958), though, arguably, with misleading conclusions .'
The absorption approach consists of regarding the balance of pay-
ments not simply as the excess of residents' receipts from foreigners
over residents' payments to foreigners but rather as the excess of
residents' total receipts over total payments. Formally

(4.1)

where R F is receipts by residents from foreigners , and PF is payments

by residents to foreigners. Since, however, all payments by residents
to residents (RR) are simultaneously receipts by residents from resi-
dents (PR),B can be written as
B = R F + R R - PF - PR (4.2)
The Absorption Approach to the Balance ofPayments 99
Hence

B=R-P (4.3)

where R is total receipts by residents, and P is total payments by

residents.
The absorption approach can either be applied to the balance of
payments as a whole or to the balance of payments on current account.
In the latter case the balance of payments is the difference between
national income and national expenditure. Taking the national income
equation Y = C + J + X - M, and labelling total expenditure A (for
absorption), we have

B=X-M= Y-A (4.4)

The balance of payments on current account is the difference between

national output (income) and national expenditure. Within this frame-
work any policy for balance-of-payments correction can be evaluated in
terms of whether it raises Y relative to A, because this is the condition
for balance-of-payments improvement. Since from the income equation ,
Y - C equals saving (S), the balance of payments can also be expressed
as

B=X-M=S-l (4.5)

and any balance-of-payments correction policy can also be evaluated

in terms of whether it raises saving relative to investment.
Policies to raise Yare termed expenditure-switching policies and
must not be accompanied by an equal rise in A if the balance of pay-
ments is to improve . Devaluation, tariffs, quotas on imports, subsidies
to exports, and price and quantity adjustments of all kinds to increase
exports and reduce imports are all examples of expenditure-switching
policies. At full employment, when Y cannot increase, expenditure-
switching must be accompanied by reductions in A if the balance of
payments is to improve? Otherwise there would be no resources to
devote to meeting the increased demand for exports and import substi-
tutes. Reducing A by itself, of course, would cause unemployment.
Policies to reduce A are called expenditure-reducing and must not be
accompanied by an equivalent fall in Y if the balance of payments is to
improve. Expenditure-reducing policies accompanying expenditure.
100 Balance-of-Payments Theory and UiK. Experience

switching policies at full employment must reduce expenditure on

traded goods, otherwise expenditure-switching will not be successful.
All the factors on which the success of a devaluation depends can be
analysed within the framework of the absorption approach without the
need for elasticity estimates and ceterisparibus assumptions.
Let us now consider in greater detail the effect of a devaluation
within the framework of the absorption approach . Since B = Y - A .
DB = L\Y - L\A . First , devaluation will have a direct effect on
income (L\YD ) . It will also have a direct effect on absorption (L\A D )
plus an indirect effect through the change in income (i.e. 0:L\ YD '
where 0: is the propensity to absorb) . 3 Thus

(4.6)

= L\YD (1 -- 0:) - L\AD (4 .7)

The condition for devaluation to improve the balance of payments

is

(4 .8)

There are thus two relations to consider : first, the direct effect of
devaluation on income and the value of 0:; and second, the direct
effect of devaluation on absorption .

THE DIRECT EfFECT OF DEVALUATION ON INCOME

There are at least three important direct effects of devaluation on

income : an idle resource effe ct ; a terms-of-trade effect ; and a resource-
reallocation effect. If there are idle resources, the effect of devaluation
will be to increase real income as demand is switched to home-produced
goods . How much income increases will depend on the degree of import
substitution , the propensity of other countries to import , and the value
of the foreign-trade multiplier. The idle resource effect on income will
improve the balance of payments , however, only if 0: < 1. If the pro -
pens ity to absorb is greater than unity, the balance of payments will
worsen . The propensity to absorb comprises mainly the sum of the
The Absorption Approach to the Balance ofPayments 101
propensities to consume and to invest as income changes. If the pro-
pensity to consume is high, a positive propensity to invest could cause
a to exceed unity . Alexander seemed to be of this view.
The effect of devaluation on the terms of trade can go either way .
As we saw in Chapter 3, the terms of trade will improve or deteriorate
in the devaluing country according to whether the product of the
elasticity of supply of exports and imports is less than or greater than
the product of the two elasticities of demand . If the terms of trade
improve, the effect on income and the balance of payments will depend
on the same factors as a positive idle resource effect. If the terms of
trade deteriorate, real income will fall and the balance of payments will
worsen if a < 1, but improve if a > 1. It is clearly fallacious to argue ,
as under the elasticity approach, that a decline in the terms of trade
will necessarily improve the balance of payments if the Marshall-
Lerner condition is satisfied , since income may fall by more than
absorption. The real-income change is not confined to the change in
the buying power of exports over imports. What happens to produc-
tivity in the export sector is also important. This is the notion of the
single factoral terms of trade . The net barter terms of trade may de-
teriorate, but real income may rise because of an increase in produc-
tivity in the export sector.
Machlup (1956) also points out that there are substitution effects
as well as income effects as the terms of trade alter. These are price -
induced changes in absorption which are not included in a . They must
be considered alongside other factors which affect absorption directly
as devaluation takes place.
Finally, there may be a resource-reallocation effect favourable to
income if the exchange rate has been previously overvalued and if
trade restrictions are removed at the same time . Overvaluation of a
currency in effect subsidises the production of non-traded goods
relative to traded goods . If productivity is lower in the non -traded-
goods sector, devaluation of the exchange rate will shift resources from
the lower- to the higher-productivity sector. This means that a re-
duction in absorption at full employment may not be necessary for the
trade balance to improve.

THE DIRECT EFFECT OF DEVALUATION ON ABSORPTION

If there is full employment and income cannot increase, or if a ;;;. 1

so that induced absorption increases by as much as or more than the
102 Balance-of-Payments Theory and U'K. Experience

increase in income, then any favourable impact on the balance of

payments from devaluation must come from a direct reduction in
absorption. The primary direct effect of devaluation on absorption
may be expected to be the dampening effect of rising prices on con -
sumption which may come about through a variety of mechanisms such
as a real-balance effect, income redistr ibution, and money illusion .
In addition interest rates must be expected to rise. The real-balance
effect refers to the desire of people to hold a constant proportion of
their real income in the form of real-money balances. If the value of
real-money balances is eroded by rising prices, people will attempt to
accumulate more nominal balance, which they do by reducing ex-
penditure out of real income . Hence the operation of a real-balance
effect will tend to reduce absorption directly. Rising prices will also
tend to redistribute income towards groups with higher marginal
propensities to save and this will also cause a fall in consumption .
Income will tend to be redistributed (i) from weak groups who are
generally poorer to strong (richer) groups who can defend themselves
against rising prices; (ii) from wages to profits, particularly if devalu-
ation makes the export sector more profitable and wages do not adjust
fully to price increases; and (iii) from taxpayers to governments, which
will increase saving if the marginal propensity to save of governments is
higher than that of taxpayers. Money illusion , which may exist in the
short run, will cause real expenditure to fall as people continue to
spend the same amount in money terms even though prices have risen,
or as they fail to increase their money expenditure in proportion to the
rise in prices, which is perhaps more likely ." Interest rates will tend to
rise with inflation because of a reduction in the real value of money
supply and increased uncertainty affecting the demand for money .
Rising interest rates might be expected to reduce consumption and
investment expenditure.
Real absorption in the economy will also be affected directly by the
change in the home price of imports. If the elasticity of demand for
imports is less than unity , there will be more spending in domestic
currency on imports, and real expenditure on domestic goods will
fall. A recent contemporary example of this type of effect was the
oil price increase of 1973-4 which threw the developed countries of
the world into prolonged recession. The cutback in imports, however ,
was not in general sufficient to compensate for the fourfold increase
in the price of oil or to compensate for lost exports resulting from the
slow-down of world trade . While absorption fell, income fell by more .
The Absorption Approach to the Balance ofPayments 103
It is possible for devaluation to be deflationary if the increased ex-
penditure on imports in domestic currency (if imports are inelastic in
demand) exceeds the increased domestic currency value of exports.
There are other forces which may increase absorption and worsen
the balance of payments. Expectations about future price rises may
increase absorption . Wages may rise faster than prices, and the velocity
of circulation of money may rise, thus adding to aggregate monetary
demand.

THE INTERACfION BETWEEN CHANGES IN INCOME AND CHANGES IN

ABSORPTIONs

We have considered the direct effect of devaluation on income and

absorption and also the indirect effect of changes in income on ab-
sorption via the marginal propensity to absorb (ex). Changes in abo
sorption, however, will also affect income . Equation (4.7) omits this
consideration, and it is a weakness of Alexander's original analysis. A
model is required which incorporates both the effect of changes in
income on absorption, and the effect of changes in absorption on
income . When the latter effect is incorporated into equation (4 .7)
the condition for the balance of payments to improve as a result of
devaluation can be interpreted in the same way as before, but the
magnitude of the change in the balance of payments will be different.
To complete the model by incorporating the income effects of changes
in absorption, let the total change in income be the sum of the direct
effect of devaluation on income (AYD) and the indirect effect (AY/)
brought about by the change in absorption :

(4 .9)

and let

(4.10)

where (j is the proportion of the change in absorption that falls on

domestic goods (0 < (j.;;; 1). Similarly, let the total change in absorp-
tion be the sum of the direct effect of devaluation on absorption
(AA D ) and the indirect effect (AA 1) brought about by the change
in income :
104 Balance-of-Payments Theory and U.K. Experience

(4 .11)

where, as before ,

(4.12)

Substituting (4.10) into (4.9) and (4.12) into (4.11) gives

(4.13)

and

(4.14)

and solving (4.14) and (4.13) simultaneously we have

AYD + I3AA D
AY= - - -- -
1 -130: (4.15)

and

(4.16)

The change in the balance of payments resulting from devaluation is

thus ."

AYD(1-a)-AA D(1-I3)
AS = AY - AA =- - - - - - --
1 - fJa (4.17)

and the balance of payments will improve if

>
(4.18)

Comparing equation (4.17) with equation (4.7) , if devaluation increases

income directly, the balance of payments will again only improve if 0: < 1.
The magnitude of the improvement now depends, however, on fJa.
The Absorption Approach to the Balance ofPayment s 105

If 0 < {jcx < I, the magnitude of the improvement will be greater

than in the simple model. If {jcx > I, however, the system will be un-
stable. On the absorption side, if devaluation decreases absorption
directly , this will improve the balance of payments provided {j < I .
This condition will be met as long as the fall in absorption is not wholly
on domestic output. The magnitude of the improvement depends on
the value of (I - {j)/(l - (h) . If this is less than unity, the impact of a
reduction in absorption on the balance of payments will be less than in
the simple model. Again if {jcx > I , the model is unstable .

STRENGTHS AND DANGERS OF TIlE ABSORPTION APPROACH

While the absorption approach to the analysis of balance-of-payments

adjustment may be analytically superior to the elasticity approach , it is
clearly not the case that elasticities and relative price changes are un -
important , or that the information needed to evaluate whether devalu-
ation will be successful in rectifying balance -of-payments disequilibrium
is any easier to come by . The conventional supply and demand elas-
ticities, and what happens to the relative price of traded and non-
traded goods, affect both income and absorption , but by how much
and in what direction depends on many unknowns . What is the extent
of the idle resource effect , and the terms-of-trade effect, on income?
What is the value of the propensity to absorb? By how much does a
rise in the internal price level dampen absorption , if at all? What is
the value of {j? In using the absorption approach to the analysis of
devaluation, answers to all these questions, and many more , must be
known .
Perhaps the major contribution of the absorption approach to
balance-of-payments correction has been to highlight more clearly the
policy prerequisites for balance -of-payments adjustment. In particular ,
it stresses the fact, in a way that the elasticity approach does not do
(at least explicitly), that at full employment devaluation must be
accompanied by expenditure-reducing policies if devaluation is to be
successful. Otherwise there are no available resources to supply more
exports and import substitutes unless productivity is raised . It is not a
coincidence that whereas the elasticity approach was developed, and
found favour, in conditions of unemployment prior to the Second
World War, the absorption approach has been developed in the full-
employment conditions of the post-war era .
106 Balance-of-Payments Theory and U.K. Experience

Many attempts have been made to synthesise the elasticity and

absorption approaches to the balance of payments, notably by
Alexander (I959), Michaely (I960) and Tsiang (I 961), but the syn-
thesis is not necessary if it is recognised that elasticities and relative
price changes affect income and absorption . The absorption approach
embraces the elasticity approach. It must do so since it works with the
identity B = Y - A . It is certainly misleading to attempt a synthesis
by modifying the elasticity formula to accommodate income effects,
because the two effects cannot strictly be dichotomised . Relative
price changes, combined with elasticities, affect income, and income
changes affect relative prices and elasticities . In the last resort both
the absorption and elasticity approaches depend on relative price
movements, so that the two approaches, as Michaely (1960) has argued,
are not in conflict :

an increase in the ratio of international to domestic prices, which

is essential for a decrease in the import surplus according to the
relative prices approach, can take place if and only if there is a
decrease in absorption, and a decrease in absorption can occur
only if there is an increase in the general price level. Hence the two
approaches to the analysis of devaluation must lead to the same
conclusions.

The major danger in employing the absorption approach to the balance

of payments is that the cause of balance-of-payments disequilibrium
may be misinterpreted , leading to the incorrect policy prescriptions if
the goals of economic policy are to be achieved simultaneously. Since
the equation B = Y - A is an identity derived from the national income
accounts , it is incorrect to infer, as many have done," that balance-of-
payments deficits are necessarily caused by plans to spend in excess of
plans to produce . Ex ante Y and A may be in balance but the balance
of payments may move into deficit because of a deterioration in
competitiveness or because of sectoral difficulties , with some sectors
experiencing excess supply and unable to export. Ex post both these
problems would show up in the national accounts as Y < A . The reason
would not be, however, that plans to spend had exceeded income , but
that income had fallen. In addition a government wishing to maintain
full employment may have expanded demand to accommodate the
balance -of-payments deficit. The implied excess of payments by resi-
dents over receipts by residents has unfortunately led followers of the
The Absorption Approach to the Balance ofPayments 107

absorption approach to view balance-of-payments deficits necessarily

as a monetary phenomenon in a causal sense . But it is clear from what
has just been stated that this is not necessarily so . The test of whether
excess monetary expansion is the cause of a balance-of-payments
deficit is to observe what happens to unemployment when monetary
expenditure is reduced . If the deficit is eliminated without causing
unemployment, this is prima facie evidence that the deficit is caused
by excess monetary demand . If unemployment rises, the origin of the
deficit must lie elsewhere . It is, of course, a golden rule in economics
that causation can never be inferred from identities without adequate
theorising and without due regard to the facts of any particular case.

THE MONETARY ASPECTS OF BALANCE-OF-PAYMENTS DEFICITS

All this is not to deny that balance-of-payments disequilibrium may

have monetary origins, or will have monetary repercussions . The
other major contribution of the absorption approach to the balance of
payments has been to highlight the monetary aspects of balance-of-
payments deficits, neglected by the elasticity approach. Indeed, the
absorption approach is the forerunner of the new monetary approach
to the balance of payments. As we saw earlier, a deficit must mean
an excess of payments by residents to foreigners over receipts by
residents from foreigners . This means that there must be a net purchase
of foreign exchange from the foreign-exchange authority or a run-down
of foreign assets . This must imply in turn that the stock of privately
held money must be decreasing, that the cash balances of residents are
being run down, and domestic money is being transferred to the
foreign-exchange authority . If the monetary authorities do not increase
the supply of money. there exists what Johnson (1958) has called a
stock deficit, which should ultimately be self-correcting as interest rates
rise, credit tightens, and aggregate expenditure falls. There may. how-
ever, be a severe loss of reserves before the deficit is corrected, and
more unemployment if the deficit is not caused by an excess supply
of money in the first place . Stock deficits are supposed to be tem-
porary , yet in a growing economy it is possible to conceive of a con-
tinuous stock deficit caused by the desire of residents to purchase
foreign assets on a continuous basis as their resources increase through
time .
If the monetary authorities respond to the depletion of cash bal-
108 Balance-of-Payments Theory and u.K. Experience

ances by creating credit , the stock deficit will not be self-correcting

and the balance-of-payments deficit will persist. This is called a flow
deficit. Flow deficits need correcting by expenditure-reducing policies
if the cause of the deficit is excess monetary demand, or by expen-
diture-switching and structural policies if the credit creation is designed
to preserve full employment in the face of an 'autonomous' deterior-
ation in the balance of payments.
In the new monetary theory of the balance of payments, which is
examined and evaluated more fully in the next chapter, the bond
market and the capital account of the balance of payments are intro-
duced . Now if there is excess demand in the goods market, there must
be exce ss supply in the money and bond markets combined ." If the
bond market is in equilibrium, then we have the situation discussed
above in which there is an excess supply of money which leads to a loss
of foreign-exchange reserves. If the money market is in equilibrium,
however, the excess demand for goods must mean an excess supply of
bonds, leading to a surplus on the capital account of the balance of
payments. A current-account deficit on the balance of payments thus
only leads to balance-of-payments disequilibrium in the total currency
llow (or balance for official financing) sense if there is an excess supply
of money, in which case reserves are lost. If there is an excess supply
of bonds, the current account is financed by a capital inllow . The neat-
ness of the approach, and the apparent policy implications , should not
be allowed to obscure the fact, however, that the practical difficulty
of diagnosing the cause of the balance-of-payments deficit still remains.
An excess supply of money is not necessarily the cause of the deficit,
and, if it is not, the deficit can only be controlled or rectified by
monetary policy at the expense of domestic employment. In many
instances an excess supply of money may merely be a symptom of a
much more fundamental source of balance-of-payments difficulty, such
as domestic supply constraints or an autonomous loss of export
markets.

DOMESTIC CREDIT EXPANSION

Changes in a country's domestic money supply are made up of changes

in the level of foreign-exchange reserves and the level of domestic
credit expansion (D.C .E.) . If the economy is running a balance-of-
payments deficit , the loss of reserves reduces any increase in the dam-
The Absorption Approach to the Balance ofPayments 109

estic money stock that is taking place and understates the way in which
domestic monetary policy is affecting the economy and the balance
of payments. In the late 1960s D.C.E. targets were imposed in the
United Kingdom in an attempt to make monetary policy more rigorous .
Setting a D.C.E. target means that the money supply must vary
positively with changes in the reserves; expand when the balance of
payments is in surplus , and contract when the balance of payments
is in deficit - very much as under the old gold standard. When a country's
currency is weak the policy has some merit as a quasi-automatic adjust -
ment device, though it could lead to excessive deflation . Similarly ,
when a currency is strong D.C.E. targets may lead to an excessive
increase in the money supply, and consequent inflationary pressure .
In 1977, in the United Kingdom, when the sterling exchange rate
strengthened, D.C.E. targets were relegated to the background in favour
of targets for the broad-based definition of money , M3. A strong
assumption of the monetary approach to the balance of payments,
examined in the next chapter, is that the money supply varies positively
with the level of foreign-exchange reserves, and that the effect of
reserve changes on the money supply cannot be sterilised by the mone-
tary authorities using open-market operations.
5
The Monetary Approach
to the Balance of
Payments
As we said in the introductory remarks in Chapter 2, the focus of
the monetary approach to the balance of payments is on the balance
of payments as a whole (the current and the capital account) so that a
balance-of-payments disequilibrium is equivalent to a change in the
level of international reserves. The essence of the argument is that
balance-of-payments disequilibrium must be considered as the outcome
of stock disequilibrium between the supply of and demand for mon ey .
Balance-of-payments difficulties are a monetary phenomenon which can
be corre cted by monetary adjustment. Traditional balance -of-payments
adjustment policies can only be successful to the extent that they
eliminate the stock disequilibr ium between the supply of and demand
for money . Let us develop a formal model of the monetary approach ,
and outline its assumpt ions , as a prelude to evaluating its usefulness
in contributing to an understanding of balance-of-payments problems
and their solution. The model outlined here draws on the presentation
by Hahn (1977) in his review of the Frenkel and Johnson (1976)
volume on The Monetary Approach to the Balance of Payments. The
monetary approach assumes that exchange rates are pegged, that the
economy is in long-run full-employment equilibrium, that the demand
for money is a stable function of income, that changes in the money
supply do not affect real variables, that in the long run a country's
price level and interest rate converge on the world level because of the
high elasticity of substitution between goods in international trade and
highly mobile capital, and that the changes in the money supply
The Mon etary Approach to the Balance ofPayments 111
brought about by changes in the level of foreign-exchange reserves are
not sterilised by the monetary authorities . The assumptions are strong
by any standards and their significance will be considered later.
The ex ante excess demand for goods can be expressed as

Xg = A - Y + B (5.1 )

where A is the total expenditure of all agents in the economy , Y is

income, and B is the balance of payments. Now consider in the first
instance the case of an economy where the only asset is money , and
denote the excess demand for money by X m . The budget constraint
of all agents taken together, by Walras's Law, leads to :

(5 .2)

If Xg = 0, then B = X m : that is, a balance-of-payments surplus implies

an excess demand for money, and a balance-of-payments deficit implies
an excess supply of money . 1 If A - Y is zero ex ante, it may be temp-
ting to treat the excess supply of money as the cause of the deficit, but
this could be misleading, for reasons to be outlined later. If A - Y is
treated as zero ex post, then B = X m is a consistency condition in
long-run equilibrium .
The excess demand for (or supply of) money is the difference
between the demand for money, assumed to be a stable function of
income , and the supply of money of domestic origin , both defined in a
stock sense . Thus :

Xm=k(Y)-M (5.3)

where M is the domestically determined component of the money

supply controlled by the domestic monetary authorities. An excess
demand for money, associated with a balance-of-payments surplus ,
leads to an increase in the supply of money through the accumulation
of international reserves, assuming that the effect of an increase
in reserves on the domestic money supply is not sterilised by the
monetary authorities using open-market operations and selling bonds.
As the supply of money increases, the excess demand for money is
eliminated and the balance-of-payments surplus tends to zero . Con-
versely, an excess supply of money associated with a balance-of-pay -
ments deficit leads to a de crease in the money supply through the loss
112 Balance-of-Payments Theory and U'K. Experience

of international reserves , again assuming no sterilisation of the reserve

loss by open-market operations, and the balance-of-payments deficit
tends to zero . Thus, if changes in the reserves are not sterilised , it is
a fundamental proposition of the monetary approach that a balance-
of-payments disequilibrium is a temporary phenomenon representing
a stock disequilibrium in the money market which will ultimately be
self-correcting. That this is the main feature of the monetary approach
is clear from Johnson (1972) , who contends that the basic assumption
of traditional balance-of-payments theory is that the monetary con-
sequences of deficits or surpluses are sterilised so that a deficit or
surplus is treated as a flow equilibrium . By contrast , he says , the
new monetary approach assumes (asserts!) that the monetary inflows
and outflows associated with surpluses or deficits are not sterilised,
or cannot be within the period relevant to policy analysis - but in-
stead influence the domestic money supply . Since the demand for
money is a stock demand, not a flow demand, variations in the supply
of money relative to demand must work towards an equilibrium
between the demand for and supply of money with a corresponding
equilibration of the balance of payments. Deficits and surpluses rep-
resent phases of stock adjustment in the money market , and not
eq uilibrium flows. Movements in th e term s of trade , the level of income
and other real factors are unimportant as causal explanations of pay-
ments disequilibrium , or as methods of adjustment , unless they affect
the balance between the supply of and demand for money . Balance-of-
payments deficits are a monetary phenomenon associated with an
excess supply of money . In the framework of the monetary approach
the level of reserves is the only variable that an excess supply of money
can affect because long-run equilibrium in the goods market is assumed ,
the interest rate is given and the demand for money is stable."
Extending the mod el to more than one asset , the fundamental
equations of the monetary approach are easily modified . For example,
introducing bonds into equation (5 .2) would give

(5.4)

where X b is the excess demand for bonds. If Xg = X m = 0 , then

B = X b : that is, a balance-of-payments surplus implies an excess de-
mand for bonds, and a deficit implies an excess supply of bonds."
In this case a surplus will be 'financed' by capital outflows and there
will be no accumulation of reserves , and a deficit will be financed by
The Monetary Approach to the Balance ofPayments 113

capital inflows and there wiIl be no loss of international reserves. In

both cases adjustment takes place without a change in money balances.
Having briefly stated the essence of the monetary approach in
formal terms, let us now evaluate the approach. From the fundamental
equations which form the basis of the monetary approach, it is clear
that balance-of-payments disequilibrium cannot be interpreted solely
as a monetary phenomenon in the causal sense of an excess supply of
or demand for money in the mone y market. Consider equation (5 .1) .
As we said in considering the absorption approach to the balance of
payments , it is quite possible for expenditure (absorption) and income
to be in balance ex ante and yet for the balance of payments to be in
deficit because for one reason or another domestic producers cannot
sell what they have left over after satisfying home buyers. Thus it is
possible that X g - B = A - Y = 0, with B < O. If this is so , balance-
of-payments disequilibrium is consistent with ex ante equilibrium in the
money market. It might be argued that a small country in a large world
economy can sell all the goods it wants at the ruling world price, but
this is not necessarily the case . The law of one price in competitive
export markets does not mean that demand is unlimited . Ex post, of
course, income would fall, so that A > Y , which would show up in
equation (5 .2) as X m < 0 , or an excess supply of money . The excess
supply of money, however , would be a symptom of the payment
deficit, not the cause.
In the two (or many) asset model a monetary explanation of
balance-of-payments disequilibrium would need even closer scrutiny .
As equation (5.4) shows, the balance-of-payments disequilibrium may
reflect disequilibrium in the market for capital assets and be wholly
consistent with ex ant e equilibrium in the money market. The selling
of bonds may lead to a deficit which is then financed by capital inflows
as interest rates rise. In fact the attempt to spend in excess of income
could result from the selling of a whole range of assets - bonds, shares,
land, buildings , etc . - and have nothing to do with excess expansion of
the money supply or dishoarding .
Let us turn now to some of the assumptions of the monetary
approach. A basic assumption is that the demand for money is stable. If
it is not stable, the balance of payments cannot be predicted from
changes in the money supply alone. Increases in the money supply may
induce increases in the demand for money , and decreases in the money
supply may induce dishoarding. Most of the factors that may cause the
demand for money to change as the supply changes , such as income and
114 Balance-of-Payments Theory and U.K. Experience

interest rates, are ruled out by assumption, but these are matters to be
proved rather than assumed for the everyday conduct of balance-of-
payments policy . To understand what is happening to the balance of
payments, and whether disequilibrium may be self-correcting or not,
the long-run characteristics of steady-state equilibrium are hardly
relevant. In practice, over long periods relevant to analysis and policy ,
the level of income and interest rates may vary significantly with
variations in the quantity of money, making it impossible to predict
movements in the balance of payments from changes in the money
supply alone . There is not a one-to-one relation between changes in
the domestic money supply and the level of reserves. Von Whitman
(I975) concludes her survey of 'global monetarism' by saying that most
of the more revolutionary policy implications of the monetary
approach to the balance of payments miss the boat for applicability
to current problems. To assume that real output - even in the long
run - is determined exogenously, and that money does not affect real
variables in an economy , is nonsense . There is far too much emphasis,
she argues, on the long-run static equilibrium state which is never
reached, while ignoring the short and medium term, which are important
for forecasting and for normative policy prescription . The concen-
tration on the long run assumes away all the problems that make
the balance of payments a 'problem ' .
Tsiang (I977) also raises the important question of the specification
of the demand for money function . In the basic monetary approach the
demand for money appears to be specified primarily as a function of
the transactions demand for money, being made a stable function of
income . Yet an increase in the transactions demand for money cannot
imply a withholding of money from spending because by definition it
is meant to be spent. Therefore, an increase in the demand for money
for transactions purposes cannot lead to a balance-of-payments surplus ;
nor can a reduced demand for money for transactions purposes lead to
a deficit if the money is not to be spent. Tsiang argues that only the
demand for money for asset purposes should be included in the demand
for money function , for only variations in the asset demand will affect
the spending on goods .
A serious question mark must also be raised over the basic assump-
tion of the monetary approach that the effect of changes in inter-
national reserves on the domestic money supply cannot be sterilised,
so that balance-of-payments deficits must represent phases of stock
disequilibrium which will ultimately be self-correcting. The monetary
The Monetary Approach to the Balance ofPayments 115

authorities can and do indulge in open-market operations to neutralise

the effect of changes in international reserves on the domestic money
supply, so that a balance-of-payments deficit associated with an excess
supply of money is not necessarily rectified by a loss of reserves and a
reduction in the money supply . Reserves will fall, but the excess supply
of money will remain .
On the question of policy we have the proposition that balance-of-
payments adjustment policies cannot be successful in rectifying pay-
ments disequilibrium except to the extent that they equilibrate the
money market. Thus tariffs, devaluation and expenditure-reducing
policies can only rectify a deficit if they reduce the supply of money
relative to the demand, or raise the demand for money relative to the
supply. Moreover, the success of expenditure-switching policies such as
devaluation can only be temporary . A number of explanations can be
invoked in support of this view. First, devaluation cannot change the
relative price of traded goods because domestic prices rise by' the extent
of the devaluation . Second , because of the high substitutability
between traded goods, all countries are price-takers and each country's
price level is determined by the level of world prices, Hence devaluation
cannot affect relative prices measured in a common currency . Note ,
however, that even accepting the 'law of one price' there could still
be devaluation effects on trade arising from shifts between the non-
traded-and traded-goods sectors of the economy, unless it is assumed
that the price of non-traded goods changes in line with traded goods.
Third, and most in keeping with the spirit of the 'domestic' monetary
approach (as opposed to the 'international' monetary approach to
balance-of-payments problems), it is argued that, starting from stock
equilibrium, devaluation cannot improve the balance of payments
permanently because any increase in the demand for money as a result
of devaluation (see below) will be matched by an equal rise in the
supply of money th!ough an accumulation of reserves. Prices and the
stock of money rise in proportion, leaving the balance of payments
unchanged.
The scepticism of the monetary approach concerning the effective-
ness of traditional balance-of-payments adjustment policies may be
shared by those who do not accept the outlook and assumptions of the
monetary approach to balance-of-payments analysis. While few would
agree that balance-of-payments disequilibrium is necessarily a transient
phenomenon, many would share the view, though not necessarily for
the same reasons, that tariffs, devaluation and other expenditure-
116 Balance-of-Payments Theory and U.K. Experience

switching policies can only have transient effects, and that the main
mechanism through which they work is a real-balance effect. This was
originally stressed by the absorption approach to the balance of pay-
ments in contrast to the elasticity approach with its stress on the sub-
stitution of home for foreign goods.
So far the monetary approach to the balance of paymen ts has been
discussed on the assumption of pegged exchange rates. If exchange
rates are allowed to float, the assumptions are that the balance of pay-
ments looks after itself, there is no movement in the reserves, and the
money supply is entirely under the control of the domestic monetary
authorities. Because there are no reserve movements , and consequent
changes in the money supply, countries are insulated from disturbances
in other countries. In practice , of course , countries are not indifferent
to the exchange rate , and exchange rates are not allowed to fluctuate
freely to produce balance-of-payments equilibrium in the 'total currency
flow sense. Thus even under floating, if it is managed floating, reserve
movements will take place. Even if they did not, however, one country
is not necessarily insulated from monetary disturbances in another
country because in a world of capital mobility interest-rate changes
will cause capital to move from one country to another , thus affecting
the conditions for equilibrium in the money market.
We can end this chapter by saying that the monetary approach to
the balance of payments is useful in stressing that balance-of-payments
disequilibria have monetary consequences, and in helping to understand
the process of balance-of-payments adjustment , but it is not very
useful, and almost certainly misleading, for the understanding of the
causes of balance-of-payments difficulties. A balance-of-payments
deficit may have nothing to do with an excess supply of money in
a causal sense, and its treatment as a monetary phenomenon may
therefore be totally misplaced. As we saw earlier , the money market
may be in equilibrium and yet a deficit may emerge on the balance of
payments owing to real exogenous changes at home and abroad. Such
a deficit could not be relied on to correct itself, and monetary cor-
rection would not be the appropriate policy instrument. Likewise, in an
economy with assets other than money , disequilibrium in the balance
of payments may arise through disequilibrium in the markets for
capital , and monetary correction would not be appropriate unless the
authorities are indifferent to the state of the current account.
Johnson (1958) once argued that the fundamental contribution of
the absorption approach to the balance of payments was to point to the
The Monetary Approach to the Balance ofPayments 117
need to combine expenditure-reducing policies with expenditure-
switching policies at full employment. The question might now be
asked , how will we think of balance-of-payments policy differently in
the light of the new monetary approach? Not very differently, I suspect.
The absorption approach first introduced the notion of a self-correcting
stock deficit and inferred that balance-of-payments difficulties may be
due to an excess supply of money. Also within the absorption
approach , the real-balance effect assumes importance as the main
mechanism of balance-of-payments adjustment resulting from expen -
diture-switching policies. The absorption approach does not stress quite
so strongly as the monetary approach the transient nature of balance-
of-payments adjustment policies, but this simply means that adherents
of the monetary approach must be less sanguine about the possibility
of achieving simultaneous internal and external balance with the
traditional mix of policy instruments. The main contribution of the
monetary approach is undoubtedly to stress that surpluses and deficits
will induce changes in the money stock, at least in the short run, and
that this will affect economic behaviour. But to argue that by these
changes the balance of payments will tend to zero is to assert what
needs to be proved theoretically and empirically . In neither the absorp-
tion approach nor the monetary approach can the authorities afford to
regard balance-of-payments deficits as transient whether the deficits
have monetary or real causes. If the deficits are monetary in origin,
there is not necessarily a one-to-one relationship between changes in
reserves and changes in the money supply ; and changes in the money
supply cannot be guaranteed to eliminate the disequilibrium in the
money market if the demand for money is unstable . If the deficit is
real in origin, monetary correction, if it did occur, would be at the
expense of full employment. The underlying disequilibrium would
remain . The monetary approach is right to warn , of course , that the
monetary authorities should not exacerbate or perpetuate a deficit by
excessive monetary expansion, but that is common sense which does
not need a new theoretical approach for its understanding.
6
Simultaneous Internal and
External Balance
The achievement of simultaneous internal and external balance is
the prime task of macroeconomic policy in most countries. In some
countries there may not be a conflict between the achievement of
internal balance and balance-of-payments equilibrium . Circumstances
may be such that the goals of full employment and a satisfactory
growth rate can be achieved without a balance-of-payments deficit
emerging . In many countries, however , the potential conflict between
internal and external balance, which was discussed in Chapter 2 ,
materialises frequently. In the United Kingdom the conflict between
balance-of-payments equilibrium and faster growth has been acute
since the Second World War, and in recent years the achievement of full
employment has also become problematical.
From the standpoint of theory there are three environments that
need to be considered . The first is the case where external balance is
defined excluding the capital account, with the possibility of exchange-
rate flexibility . The second is the case where external balance includes
the capital account and the exchange rate is fixed . The third is the case
where external balance includes the capital account and the exchange
rate is flexible . A fourth possible case was the situation described in
Chapter 2 where balance-of-payments equilibrium is defined in terms
of the current account and the exchange rate is fixed. This case presents
a severe dilemma for policy-makers because in this situation there
would seem to be only the one instrument of monetary and fiscal
policy to achieve the two independent objectives of internal and
external balance. Meade's (1951) pioneering work considered the first
Simultaneous Internal and External Balance 119
case outlined of the reconciliation of internal and external balance
under flexible exchange rates with external balance defined in terms
of the current account. Meade showed that the confli ct between
internal and external balance could be reconciled if monetary and
fiscal policy were used for internal balance and the exchange rate used
for external balance . But suppose there are limitations on the use of the
exchange rate as a policy weapon , as there seemed to be in the 1950s
and 1960s . According to Meade's analysis this would leave monetary
and fiscal policy to reconcile the two objectives. It was this problem
that Mundell (I 962 , 1963) attempted to explore , investigating the
possibility of using monetary and fiscal policy as separate instruments
by redefining balance-of-payments equilibrium to include the capital
account. Since the capital account of the balance of payments is
sensitive to interest-rate differentials between countries, Mundell was
able to show that the conflict between internal and external balance
could be reconciled by using monetary policy to achieve external
balance and fiscal policy to achieve internal balance. Following the
earlier argument in Chapter I, however , about how balance-of-
payments equilibrium ought to be defined for the conduct of economic
policy, there is a clear difference between the approaches of Meade and
Mundell . Meade's analysis is concerned with adjusting the balance of
payments to conditions prevailing in the real economy . By contrast , the
use of monetary policy to improve the capital account is essentially a
policy of financing balance-of-payments disequilibrium which postpones
the necessary adjustment and possibly stores up difficulties for the
future. The models of Meade and Mundell will be considered below
together with a third category of model in which the definition of
external balance includes the capital account, and the exchange rate
is also variable. The. fundamental question in each of the models is
which policy to assign to which obje ctive. This is called the assignment
problem . As we shall see, the appropriate allocation of policy instru-
ments to objectives becomes very much an empirical matter depending
on the relative sensitivity of targets to instruments, or on what Mundell
has called the principle of effective market classification. Assume two
objectives or targets (Y l and Y 2 ) and two policies or instruments (Xl
and X 2 ) and that each target is responsive to both instruments . There
are thus two structural relationships :
120 Balance-of-Payments Theory and U'K, Experience

and

The principle of effective market classification says assign instrument

X. or X 2 to Y. according to whether a I I /a2 I ~ a12 /a2 2. The instru-
ment not assigned to Y I is then assigned to Y 2 . Alternatively , we
could say, instrument XI should be assigned to Y I or Y 2 according to
whether a. da2 I ~ aI2/a22 ' the remaining instrument being assigned
to the policy not paired with X I .
In the models outlined below we shall be dealing with a single
country with no foreign repercussions .

INTERNAL AND EXTERNAL BALANCE UNDER FLEXIBLE EXCHANGE

RATES WITH NO CAPITAL MOVEMENTS

Consider Figure 6.1 . On the vertical axis internal balance is measured

(arbitrarily) in terms of the state of employment, and on the horizontal
axis external balance is measured by the current account of the balance
of payments. Drawing a vertical line to represent external balance , and
a horizontal line to represent internal balance, gives four quadrants or
zones indicating combinations of internal and external circumstances
from which the appropriate combinations of policies to reconcile
internal and external balance can be deduced. Simultaneous internal
and external balance is achieved at P. In zones II and IV internal and
external balance are not in conflict . In zone II the expansion of demand
will both reduce unemployment and the balance-of-payments surplus.
In zone IVa reduction in demand will both reduce 'overheating' in the
economy and the balance-of-payments deficit. The two objectives of
internal and external balance are not independent and therefore one
policy instrument will suffice in both cases. In practice, of course, it
is highly unlikely that expansion in zone II or deflation in zone IV
will move an economy smoothly and directly to the point of simul-
taneous balance at P. The probability is that the one policy or the
other will move the economy into zone I or III before P is reached,
in which case the problem of conflict will arise.
The conditions prevailing in zone I would be unusual. It would be
a very 'competitive' economy that combined a balance-of-payments
surplus with the fullest possible employment. If this situation did
Simultaneous Internal and External Balance 121

IV
'Over-full'
employment
{
r
Deflate
App reciate
Deflate
1
Internal _ f - - - - - - - - ' f - - - - - - - -
J
l
Balance Inflate P Inflate

Unemployment
i Depreciate :J
II III

.
'====::===~===:::::::===-
Surplus l Defic it
External
Balance

FIGURE 6.1

prevail, however, the natural thing to do would seem to be to use

monetary and fiscal policy to deflate the economy to ach ieve int ernal
balan ce and to appreciate the currency to achieve external balance . To
expand demand in order to reduce the surplus would overheat the
economy even more . If currency appreciation is successful in reducing
the surplus, however, this in itself will be deflationary , and deflation by
monetary and fiscal policy may not be needed . Which combination
and sequence of policies is pursued depends, therefore, on the exact
nature of the internal and external inbalance .
To be in zone III is a much more common occurrence . In this case
the natural thing to do would seem to be to use demand expansion
to produce full employment and to depreciate the currency to rectify
the deficit. To deflate in order to rectify the deficit would simply cause
more unemployment. If currency depreciation is successful in reducing
the deficit, however , this in itself will be expansionary, and further
expansion using monetary and fiscal policy may cause 'ove rheat ing' . As
in zone I, therefore, which combination and sequence of policies is
pursued depends on the nature of the disequilibrium .
To appreciate the problem more clearly we use in Figure 6 .2 the
famous Swan (I963) diagram which measures the ratio of international
prices to home prices on the vertical axis and real expenditure on the
horizontal axis . The AA curves represent internal-balance schedules
which are combinations of relative costs and real expenditure that can
sustain a given level of employment. Suppose A 2A 2 represents internal
balance defined as full employment.' According to Figure 6.2 this can
be achieved either with a relatively low level of domestic expenditure
....
IV
IV

A3
Ratio of 8 2 (external balance)
international
to home prices 83

A3
8,
A 2 (internal balance)
82
A,
Real expenditure

FIGURE 6.2
Simultaneous Internal and External Balance 123

and a high ratio of international to home prices (with exports relatively

cheap and imports relatively dear), or with a higher level of expenditure
and a lower ratio of international to home prices. Curves above and to
the right of A 2A 2 represent given states of 'overfull' employment, and
curves below and to the left of A 2A 2 represent given states of un·
employment. The BB curves show combinations of real expenditure
and relative prices that can sustain a given balance of payments. Let
B 2B2 represent external balance. The upward slope of B 2B2 shows
that this can be achieved either with a relatively low level of expen-
diture and a low ratio of international to home prices , or a higher level
of expenditure combined with a higher ratio of international to home
prices. BB curves above and to the left of B 2B2 represent surplus and
BB curves below and to the right of B 2B2 represent deficit. Only
where A 2A 2 and B 2B2 cross is there simultaneous internal and
external equilibrium . As before , four zones of economic conditions are
defined lying between the two curves. In zone I there is overfull
employment and balance-of-payments surplus ; in zone II unemploy-
ment and balance-of-payments surplus; in zone III unemployment and
balance-of-payments deficit ; and in zone IV overfull employment and
balance-of-payments deficit. Note , however , that while the zone
indicates the state of employment and the balance of payments that a
country is experiencing, this information is not enough for deciding the
direction of policy . What quadrant (given by the intersection of the two
dashed lines) the country is in must also be known because the causes
of disequilibrium differ between the two halves of each zone . For
example, in zone III, which represents balance-of-payments deficit
and unemployment, the nature of the conflict differs between the left
and right side of the vertical dashed line. To the left a slight increase
in expenditure, and then a rise in the ratio of international to home
prices brought about by currency depreciation, are required to bring
about simultaneous internal and external balance. To the right an
increase in expenditure would be inappropriate. What is required is a
reduction in expenditure so that depreciation to rectify the deficit
does not lead to excessive demand expansion. Similar arguments would
apply in other zones.
Care must also be taken in formulating policy to ascertain whether
the observed situation is permanent or temporary. For example, pro-
ductivity improvement will shift all AA curves upwards and all BB
curves downwards, but if the productivity improvement is only
temporary a country may think it is in zone IV when really it is in
124 Balance-of-Payments Theory and U'K , Experience

zone II. Policies appropriate to zone IV would be inappropriate in

relation to the underlying long-term condition of the economy.

INTERNAL AND EXTERNAL BALANCE UNDER FIXED EXCHANGE RATES

WITH CAPITAL MOVEMENTS

According to the model presented above , if exchange rates are fixed ,

there is no solution to the conflict between internal and external'
balance. If external balance is redefined to include the capital account ,
however , Mundell (J 96 2) has shown that internal and external balance
is possible under fixed exchange rates , using interest-rate (monetary)
policy for external balance and fiscal policy for internal balance. We
first present the basic Mundell model , and then consider simultaneous
internal and external balance within the standard macroeconomic
framework of IS-LM curve analysis. Consider Figure 6.3 . BB is the
external-balance schedule with traces the locus of combinations of the
rate of interest and the budget surplus (at full employment) along
which the balance of payments (including the capital account) is in

B
X
Budget I
surplus I
'I"
I
1\
I
1 III II
!\
A I

IV

A (internal balance)

B (external balance)

Rate of
interest

FIGURE 6.3
Simultaneous Internal and External Balance 125

equilibrium. The schedule has a negative slope because rising interest

rates will attract capital inflows , and for the balance of payments to
stay in equilibrium the budget surplus must fall. Any point above and
to the right of BB represents a balance-of-payments surplus, and any
point below and to the left of BB represents a balance -of-payments
deficit.
AA is the internal-balance schedule which traces the locus of
combinations of the rate of interest and the budget surplus along which
there is full employment. This schedule must also have a negative slope
because tighter (laxer) monetary policy must be offset by laxer
(tighter) fiscal policy to stay on the 'schedule. Any point below and to
the left of AA represents overfull employment, and any point above
and to the right of AA represents unemployment. Internal and external
balance will be achieved simultaneously where the two curves cross,
with the appropriate rate of interest and budget surplus . Again, four
zones can be distinguished between the two curves: I = overfull
employment and surplus ; II = unemployment and surplus ; III = un-
employment and deficit ; and IV = overfull employment and deficit.
The relative steepness of the two curves is crucial to the question of
the assignment of policies to the two objectives of internal and external
balance. It can be shown that if BB is steeper than AA , monetary policy
must be assigned to external balance and fiscal policy to internal
balance for the achievement of both objectives. I Let us first consider
the slopes of the curves. It can be demonstrated that BB must be
steeper than AA as long as capital inflows are sensitive to variations
in the interest rate , and that the difference in the steepness of the
curves depends on the degree of respons iveness of capital flows to the
rate of interest and on the marginal propensity to import . The proof
is as follows . The absolute slope of AA is the ratio between the re-
sponsiveness of domestic expenditure to the rate of interest and the
responsiveness of domestic expenditure to the budget surplus . Now if
capital flows are constant , the balance of payments depends only on
expenditure (via the propensity to import), and hence the BB schedule
also measures the ratio between the responsiveness of domestic expen-
diture to changes in the. rate of interest and to changes in the budget
surplus . If, however , net capital inflows are positively related to the
rate of interest, the budget surplus must be changed by more than in
their absence if balance-of-payments equilibrium is to be maintained.
Thus BB must be steeper than AA .
126 Balance-of-Payments Theory and U.K. Experience

It remains to show that if BB is steeper than AA , the system will

only be stable if monetary policy is used for external balance and fiscal
policy is used for internal balance . In zones II and IV there is no
assignment problem because there is no clash of objectives, and
monetary and fiscal policy can work in the same direction. However,
suppose in Figure 6.3 the economy is at w, with internal balance
combined with a balance-of-payments deficit, and consider what would
happen if fiscal policy is used to achieve external balance and monetary
policy is used to achieve internal balance . To correct the balance-of-
payments deficit by fiscal policy requires moving to V, but this will
upset internal balance , causing unemployment. To use monetary policy
to cure the recession requires moving to u, which would cause the
balance of payments to go into deficit again. Fiscal policy to remedy
the deficit would require moving to X, but this causes even more un-
employment. The system is unstable . Suppose instead, however, that
monetary policy is used to correct the balance-of-payments deficit and
fiscal policy to maintain full employment. The economy would then
move from w to b to c, towards simultaneous equilibrium, and the
system would be stable . The same argument applies in zone I, where a
conflict of objectives also exists , but not such a compelling one for the
authorities to reconcile.
The assignment rule is based (as stated above) on what Mundell calls
the principle of effective market classification, which says that policies
should be paired with the objectives on which they have the most
influence . The principle is an extension of the Tinbergen rule that there
must be as many policy instruments as there are independent objectives
(targets), except· that in the latter case the rule does not say which
instrument should be paired with which objective. Pursuing Mundell's
rule, the use of monetary policy for external balance and the use of
fiscal policy for internal balance accords with the principle of effective
market classification because the ratio of the effect of monetary policy
on external balance to its effect on internal balance is greater than the
ratio of the effect of fiscal policy on the balance of payments to its
effect on internal stability ."
Before going on to consider the simultaneous achievement of
internal and external balance in the more familiar macroeconomic
framework of IS-LM curve analysis, there are certain limitations and
criticisms of Mundell's model that must be mentioned. We leave until
last the assumption of fixed exchange rates. The first point to make
is that, strictly speaking, it is not only the interest rate of the country
Simultaneous Internal and External Balance 127
concerned that is relevant to the analysis but the interest rate relative
to that of other countries. The analysis must implicitly assume that
interest rates elsewhere remain fixed . If they do not remain unchanged ,
internal and external balance may be impossible to achieve . Either
international co-operation would have to be sought, or internal balance
would have to be sacrificed for external balance - monetary policy
having been rendered ineffective . The same limitation applies, of
course, to the Meade-Swan model , where exchange depreciation in
one country may be thwarted by competitive exchange depreciation
in other countries. Second , Mundell's model is sometimes criticised
on the grounds that it assumes that international capital movements
are flows, whereas they represent stock adjustments to internal stock
disequilibrium between the demand for and supply of money. If
capital movements do reflect stock adjustments , an ever-increasing
interest rate (or interest-rate differential) would be required to secure
external balance. This criticism is not a strong one , however. In a
growing economy capital movements have two components: a stock -
adjustment component and a flow component. A flow component
exists because, as wealth increases, it has to be distributed among
competing assets . Thus a policy of maintaining a constant interest-rate
differential between itself and the rest of the world may be sufficient
for a country to achieve external balance by attracting the flow com-
ponent from additions to wealth . A third criticism of Mundell is that
the assignment of monetary policy to external balance will produce
resource misallocation because it distorts the allocation of resources
between consumption and investment at home, and because the balance
between home and foreign investment is decided on the basis of the
health of the balance of payments, not on productivity criteria . All
policies of balance-of-payments correction , however , have resource-
allocation effects. Demand-contraction policies would be wasteful of
resources; exchange-rate depreciat ion may have unfavourable internal
repercussions which then lead to policies such as wage and price
control which distort the allocation of resources . The question is not
which balance-of-payments policy has no resource -allocation effects
but which has the least unfavourable . A fourth criticism is that it is
assumed that policy-makers have full knowledge of the slopes of the
curves and that policy adjustment takes place in a smooth and co-
ordinated way . If there is not fuJI knowledge, and policy changes are
not smooth and co-ordinated, the assignment of policies could lead
to divergence from the path towards simultaneous internal and external
 128 Balance-of-Payments Theory and U'K. Experience

equilibrium and to over-shooting . A fifth criticism , and the most

serious , is that using interest rates for external balance is not a balance-
of-payments adjustment policy but a method of financing balance-of-
payments disequilibrium. The state of the balance of payments in
relation to the functioning of the real economy will either remain the
same or probably worsen, and there is a limit to the extent to which
interest rates can be raised to attract capital to compensate for this
' real' disequilibrium. Mundell's policy prescription for simultaneous
balance gives temporary respite but no more . Williamson (1971) has
argued that there are no circumstances in which it would be desirable
to pursue Mundell's strategy rather than adjust the current account
because the interest burden will grow through time, reducing invest-
ment , further worsening the current account , and so on .
Finally, Mundell's model only gives rules for assignment under
fixed exchange rates . Under flexible exchange rates the assignment
problem is much more complex because interest-rate policy may now
have relatively stronger effects on internal balance than fiscal policy.
For example, a policy of monetary contraction and high interest rates ,
by inducing capital inflows, would tend to cause the exchange rate to
appreciate, leading to a further reduction in demand in the economy
greater than the reduction in demand brought about by an equivalent
fiscal contraction which would tend to reduce interest rates, depreciate
the currency , and counter the contraction. Ukew ise, a policy of
monetary expansion, leading to a lowering of interest rates and capital
outflows, would tend to depreciate the currency, leading to a further
expansion of demand greater than the expansion of demand brought
about by an equivalent fiscal expansion which would tend to raise
interest rates, appreciate the currency by inducing capital inflows , and
counter the expansion. Fleming (1962) was one of the first to show
that the expansionary (contractionary) effect of a given increase
(decrease) in the money supply will always be greater under flexible
exchange rates than under fixed rates , and that the expansionary
(contractionary) effect of an increase (decrease) in the money supply
will be greater than a budgetary stimulus (contraction) of the same
amount under flexible exchange rates (except in extreme cases). Thus
it would seem that, according to the Mundell rule, monetary policy
should be used for internal balance, with the achievement of external
'balance left to the exchange rate. But this may be a hasty judgement.
Exchange speculation, if it is equilibrating, will tend to lessen the
difference in effectiveness between monetary and fiscal policy, though
Simultaneous Internal and External Balance 129
the assignment of policies may still be reversed. More important, if the
reserve effects of surpluses and deficits are not sterilised by the
monetary authorities, a surplus due to monetary contraction will
increase the money supply and a deficit due to monetary expansion will
contract the money supply, and in the long run monetary policy will
have no effect on income. However, this qualification to the reversal
of the assignment rule assumes that exchange-rate variations by them-
selves are not sufficient, or work too slowly, to maintain external
balance. Since they do work slowly , the qualification is a serious one
if changes in foreign-currency holdings are allowed to affect the domes-
tic money supply ."

INTERNAL AND EXTERNAL BALANCE IN AN IS-LMCURVE FRAMEWORK

The traditional macroeconomic framework for considering internal

equilibrium, and which has useful pedagogic virtues, is IS-LM curve
analysis , which shows the conditions for simultaneous equilibrium in
the money market and the goods market. This model can be extended
to the open economy, where the slope of the IS curve is affected by the
level of imports, and the position of the curve is partly dependent on
the exchange rate . An external-balance schedule can then be super-
imposed, and policies examined for internal and external equilibrium
under fixed and flexible exchange rates . The model has the advantage
over Mundell's in that the goods and money markets are integrated so
that fiscal policy now has explicit interest-rate effects. Using this model
fiscal policy is defined as shifts in the IS curve and monetary policy as
shifts in the LM curve .
On a diagram (see Figure 6.4) which measures the interest rate on
the vertical axis and the income level on the horizontal axis , the IS
curve is the locus of points at which savings equals investment at
different levels of income given the rate of interest. It slopes down-
wards from left to right, giving a higher level of income as the interest
rate falls. Since imports will rise as income rises, the IS curve will be
steeper in an open economy than in a closed economy since the multi-
plier effect from a given fall in the interest rate will be less. Any auton-
omous increase in demand, including government fiscal policy , will
shift the curve up and to the right. The curve is defined for any given
price level, p . The LM curve is the locus of points at which the supply
of money equals the demand for money at different levels of the rate
130 Balance-of-Payments Theory and U.K. Experience

A (internal balance)
Rate of I
interest I
I

(external
e, » > balance)
....

IV

A Real inco m e

FIGURE 6.4

of interest , given the level of income. It slopes upwards from left to

right showing the interest rate rising as the level of income rises, which
results from an increased transactions demand for money with the
supply of money fixed . In practice the LM curve will be horizontal over
part of its range because there exists a floor to the rate of interest, and
it will become vertical at high levels of income if there is a limit to the
velocity of circulation of money (i.e . a limit to the extent to which
income can be financed with a fixed supply of money). For exposition,
however , the curve is drawn linear in Figure 6.4, in keeping with the
linear structural models that generate these relations." Increases in the
supply of money will shift the curve outwards to the right , and de-
creases in the money supply will shift it inwards to the left. There will
be a different LM curve for each price level p. Where the IS and LM
curves cross , the money market and the goods market are both in
equilibrium . For the sake of argument , let us suppose that this point
of intersection also corresponds to full employment so that it can be
used to locate the internal-balance schedule AA . Points to the left of
AA represent unemployment and points to the right of AA represent
overheating of the economy.
An external-balance schedule (inclusive of the capital account) can
Simultaneous Internal and External Balance 131
now be superimposed on the diagram. drawn through the locus of the
combinations of the interest rate and the level of income which would
maintain external balance . The curve (BB) must slope upwards from
left to right in the interest-rate-income plane because as income rises,
increasing the level of imports , the interest rate must rise to attract
compensating capital flows to maintain balance." Points above BB
represent a balance-of-payments surplus and points below BB represent
a balance-of-payments deficit. The crossing of the AA and BB
schedules again defines four zones : I = overfull employment and
surplus ; II = unemployment and surplus; III = unemployment and
deficit ; and IV = overfull employment and deficit. Zones I and III are
the conflict zones, with zone III the most typical and problematical.
Since all the curves are drawn on the assumption of a given exchange
rate, we can consider in this framework both policies for internal and
external balance under fixed exchange rates (as in Mundell's model) and
also policies under flexible exchange rates which will shift the position
of all three curves. A depreciation of the exchange rate will shift the
BB curve downwards because external balance will now be achieved (if
the conditions for a successful devaluation are met) at a lower rate of
interest, given the level of income , or at a higher level of income, given
the rate of interest. The IS curve will shift upwards to the right if
depreciation is expansionary , and the LM curve will shift upwards to
the left because as prices rise the real value of the money supply will
decline . For comparative static equilibrium analysis the BB curve needs
to be drawn steeper than the LM curve, otherwise a devaluation cannot
be shown to produce a new equilibrium at a higher level of income . Out
of equilibrium there is no reason why the BB curve should not be
flatter than the LM curve .

THE USE OF MONETARY AND FISCAL POLICY UNDER ALTERNATIVE

EXCHANGE-RATE REGIMES

We can consider again the Mundell assignment problem. It was origin -

ally argued that because the ratio of the effect of monetary policy on
external balance to its effect on internal balance is greater than the
ratio of the effect of fiscal policy on the balance of payments to its
effect on internal stability, monetary policy ought to be assigned the
task of external balance and fiscal policy the task of internal balance .
We now show in more detail how the question of which policies to
132 Balance-of-Payments Theory and U.K. Experience

pursue, and thus the mix, depends on the initial conditions, the slopes
of the curves, and the exchange-rate regime in operation. Under fixed
exchange rates, suppose an economy is in zone III, where the crossing
of the IS and LM curves gives recession and a balance-of-payments
deficit. If the BB curve is steeper than the LM curve (case I in Figure
6.4) , the Mundell rule holds whatever the initial conditions within the
zone because expansionary fiscal policy would lead to an even greater
balance-of-payments deficit. If the BB curve is flatter than the LM
curve (cases 2 and 3 in Figure 6.4), it does not matter which policy is
pursued for external balance because expansionary fiscal policy im-
proves the balance of payments." In case 3, however, monetary policy
would have to be contractionary to maintain internal balance, while in
case 2 both monetary and fiscal policy must be expansionary for
internal and external balance. If the BB curve is negatively sloped
because capital movements are income-sensitive, there would be a
strong case for using monetary policy for internal balance and fiscal
policy for external balance.
Under flexible exchange rates the situation is even more complex.
If the economy is not on the BB curve, the exchange rate will alter
and shift all the curves simultaneously. Both internal and external
equilibrium will be affected, and simultaneous internal and external
balance will not be achieved until all curves intersect at the point of
intersection of AA with BB. As suggested earlier , if under flexible
exchange rates monetary policy is relatively more powerful than fiscal
policy in its impact on the domestic economy, monetary policy must
be assigned the task of internal equilibrium. Again, everything depends
on the initial starting-point , the slopes of the curves and how much
each curve shifts as a result of exchange-rate changes.

THE ASSIGNMENT OF POLICIES IN THE UNITED KINGDOM

Fausten (1975) has made a bold attempt to assess the theoretical

validity of the Mundell decision rule for the assignment of economic
policies under fixed exchange rates and to examine the consistency of
policies in the U.K. context. Fausten concludes , like others, that the
assignment rule of monetary policy for external balance and fiscal
policy for internal balance is not logically valid as a universal propo-
sition , and that the policy-mix approach, and particular rules, were
unsuitable for U.K. requirements in the 1960s. Fausten argues his case
Simultaneous Internal and External Balance 133

on three "main grounds. First, the policy -mix approach is inadequate

because it is a short -term approach and disregards the causes of balance -
of-payments disequilibrium. It contains no adjustment mechanism to
reduce the need for further capital inflows . On the contrary, continual
financing of external deficits is required , probably at higher and higher
interest rates if the mobility of international capital represents stock
adjustments rather than flows. Second, the approach assumes that only
the current account is affected by income changes (i.e . capital flows are
insensitive to income) and that only the capital account is affected by
changes in the rate of interest. Further, it is assumed that increases in
income always affect the balance of payments adversely, while increases
in the rate of interest affect it favourably . In practice a wide range of
behavioural responses is possible. Fausten shows that if capital flows
are income-sensitive and exports are interest-sensitive , it is not possible
to specify a unique assignment rule . Everything depends on the relative
parameter values. Third, using estimated relationships for the U.K.
economy, Fausten questions the appropriateness of assigning monetary
policy to the external-balance target. To the extent that the authorities
pursued the assignment rule, he contends that they went astray . In fact ,
however, he finds a discrepancy between the actual and intended thrust
of monetary policy . The restrictiveness of monetary policy for external
reasons was invariably weakened through the generous extension of
finance to the public sector and the desire for a stable gilt-edged market.
7
A History of the U.K.
Balance of Payments
Although the primary purpose of this chapter is to describe the balance -
of-payments experience of the United Kingdom since 1950, it will be
helpful as an introduction to place the experience in historical perspec-
tive, relying heavily on the statistical tables compiled by Mitchell and
Deane (1962). Starting from 1854, data will be given on imports,
exports, the balance of trade, the over-all balance on current account ,
indices of the volume of imports and exports, and the terms of trade.
Throughout the chapter emphasis will be given to discussion of the
statistics rather than to events and policies, and the focus will be on
the real sector of the balance-of-payments accounts rather than on
international monetary developments. The chapter has two appendices
which illustrate that my scepticism of currency depreciation is of quite
long standing .

1854 TO 1939

Table 7.1 provides data for the items referred to above for the period
1854 to 1939. The statistics largely speak for themselves, but certain
points may be noted. The trade balance over the period was always in
deficit (in contrast to post-I 950 , when in three years it was in surplus).
Up to the First World War the average trade deficit was just over £100
million per annum , with a slight trend deterioration through time . The
current -account balance was in surplus , however, by approximately
£70 million per annum , owing to a healthy surplus on overseas invest-
ment earnings and other invisible trade. Between the First and Second
 TABLE 7.1

Balance-of-payments statistics (£m.)

Imports Expo rts Merchand ise Balance Balance of Bullion Balance on Index of Inde x of Terms of
c.i.f', (including trade of overseas all other and spe cie curren t impo rt export trade
re-exports) balance investment invisible account volume volume (1880 = 100\
f'.o .b , earnings trad e (1880 = (1880 =
100) 100 )
(I ) (2) (3) (4) (5) (6) (7) (8) (9) (10)

1854 152.4 115 .8 - 36.6 + 12.6 + 33 .4 - 3.6 +5.8 33 40 94.6

1855 143.5 116 .7 -25.9 + 12.9 + 34 .7 - 7. 8 +13 .9 31 40 89 .4
1856 172.5 139.2 - 32.1 +14 .9 +40.9 - 1.9 +2 \.8 36 48 91.6
1857 187 .8 146 .2 - 40.4 +16 .2 +44 .8 +6.5 + 27.1 36 49 87 .1
1858 164.6 139.8 - 23.8 +15 .9 +40 .2 - 9 .9 + 22.4 37 48 98 .0
1859 179 .2 155 .7 - 22.6 + 16.9 +4 3.2 - 1.4 +3 6.1 39 52 98 .2
1860 210 .5 164 .5 - 45.5 + 18.7 +48.0 +2 .5 + 23 .7 45 55 94 .9
1861 217 .5 159.6 - 57.6 + 19.9 +50 .0 +2.1 + 14.4 47 51 98 .1
1862 225 .7 166 .2 - 58.8 +20 .7 +5 1.9 - 2.3 +11 .5 48 48 105.8
1863 248 .9 196 .9 - 51.4 +21.3 +60.1 - 3.5 +2 6.5 49 51 107. 2
1864 275 .0 212 .6 ·- 6 1.5 +22 .9 +66.0 - 4 .6 + 22.8 49 51 104.7
1865 271.1 218 .8 - 51.2 +24.1 +68 .4 - 6 .4 +34.9 52 55 107.0
1866 295.3 238.9 - 55.2 +26.4 +74.5 - 12.7 +33 .0 56 61 110.0
1867 275.2 225.8 - 48.6 +28 .2 +72.1 - 9 .5 +42.2 56 62 107.8
1868 294 .7 227 .8 - 65.5 +3 1.1 +75.5 -4.6 + 36 .5 60 66 100 .3
..... 1869 295 .5 237. 1 - 57.5 +3 1.1 +75.2 - 4 .1 +46 .7 62 70 103.1
IJ"J
VI 1870 303.3 244. 1 - 57.5 + 35.3 + 76.8 - 10.5 +44 .1 65 76 102.3
....
.....
0\ TABLE 7.1 (cont)

Balance-of-payments statsitcs (£m.)

Imports Exports Mer chandise Balance Balance of Bullion Balance on Ind ex of Ind ex of Terms of
c.i.f. (including trade of overseas all other and specie current import export trade
re-exports) balance invest men t invisible account volum e volume (1880= 100)
f .o .b. earnings trade (1880 = (1880 =
100) 100)
(I) (2) (3) (4) (5) (6) (7) (8) (9) (10)
-- - - - - -- - - ---_._-- - - -
1871 331.0' 283.6' -46.0 +39 .5 +82.4 -4.4 +71 .3 75 85 109.4
1872 354.7 314 .6 - 36.8 +44.3 +89 .8 +0 .7 +98 .0 75 88 113.0
1873 371.3 311 .0 -56 .3 +51.7 +90 .6 - 4.7 +8 1.3 78 85 117 .2
1874 370 .1 297 .7 - 69. 1 +56.6 +90 .9 -7 .5 +70 .9 80 84 113.2
1875 373.9 281 .6 - 90.5 +57.8 +89 .6 -5 .6 +51.3 85 84 111 .6
1876 375 .2 256.7 -117 .8 +57.5 +91.1 -7 .6 + 23.3 87 81 105.4
1877 394 .4 252.4 -14 1.5 +55 .5 +96 .5 +2 .6 +13 .1 90 84 98 .5
1878 368 .8 245 .5 -121.8 +55 .1 +89.1 -5.7 +16 .9 90 84 102.4
1879 363.0 248 .8 - 11 1.8 +55 .9 +88 .0 +4 .4 +35 .5 93 89 101.7
1880 411 .2 286 .5 -121.1 +57 .7 +96.4 +2 .6 +35 .6 100 100 100 .0
1881 397.0 297 .1 -94 .5 +59 .5 +95 .0 +5 .6 +65 .7 98 110 96 .7
1882 413 .0 306 .7 -100.0 +62 .8 +97 .5 -2 .6 +5 8.7 103 III 99.6
1883 426 .9 305.4 --116.9 +64.4 +10 2.1 -0 .8 +48 .8 109 114 98.5
1884 390 .0 295 .9 - 91. ' +66 .8 +95 .0 +1.6 +72 .3 105 115 99.9
1885 371.0 271.5 -98 .5 +70.3 +90 .7 - 0 .2 +62 .3 106 109 102 .5
1886 349 .9 268 .9 -79 .5 +74 .0 +83 .8 +0.6 +78 .9 107 114 104.4
1887 362.2 281 .2 - 78.5 +79 .5 +87 .3 - 0 .6 +87.7 112 119 106 .4
1888 387 .6 29 8.5 -85 .9 +84.5 +92 .7 +0 .6 +91 .9 118 127 102 .3
 1889 427 .6 315.6 -105 .0 +8 8.5 +99 .1 - 2.0 +80.9 128 132 103 .0
1890 420.7 328 .2 -86 .3 +94 .0 +99 .6 -8 .8 +98.5 127 134 109 .1
1891 435 .4 309.1 - 122 .1 +94.3 +99 .6 - 2.4 +69.4 131 127 107.4
1892 423 .8 291.6 - 128.9 +94 .7 +96 .7 -· 3.4 +59.1 133 122 107 .0
1893 404 .7 277 .2 - 124 .6 +94 .7 +85 .6 - 3.7 +53 .0 130 117 109 .3
1894 408 .3 273 .8 - 131.5 +92 .6 +88 .4 --10 .8 + 38 .7 140 122 114 .4
1895 416.7 285 .8 - 126.5 +9 3.6 +87 .8 -- 14 .9 +40 .0 148 13 3 110 .8
1896 441.8 296 .3 - 137.9 +96 .0 +92 .3 +6.4 +56 .8 155 140 110 .8
1897 451 .0 294 .2 -153.9 +9 7.0 +94 .7 +0 .8 +41.6 159 138 110 .0
1898 470 .5 294 .1 - 168.9 +101.2 +96 .8 -6.2 +22.9 164 137 109 .3
1899 485 .0 329 .5 2 - 153.7 +10 3.2 +102 .7 - 9.8 +4 2 .4 166 146 112.2
1900 5 23 .1 354.4 - 167.0 +103 .6 + 109.1 - 7.5 +37.9 166 140 120.0
1901 522 .0 347 .8 -173 .1 + 106.5 +106.7 - 6 .2 +33 .9 171 141 118.1
1902 528 .4 349 .2 - 178.4 + 109.1 +107.9 - 5.3 +33 .3 175 150 114.1
1903 542.6 360.4 - 181.3 +112.2 + 113 .6 +0 .3 +44 .8 177 154 112.4
1904 551.0 371.0 -179.1 +113 .4 +11 5.5 +0 .7 +51.7 179 157 113.3
1905 565 .0 407 .6 - 155.9 +123.5 +120 .1 -6 .2 +81.5 182 173 112.6
1906 607 .9 460.7 - 146.0 +134 .3 +131.0 - 1.8 +117 .5 187 186 114.4
1907 645 .8 517 .9 -126 .8 + 143.8 +142.4 -5 .3 +154 .1 190 201 114.9
1908 593.0 456 .7 - 135.6 + 151.0 +1 32 .5 +6 .8 +154 .7 182 185 114 .7
1909 624 .7 469 .5 - 154 .2 + 158.0 +138 .3 - 6 .5 + 135 .6 189 193 109.4
1910 678.3 534 .2 - 142.7 +170.0 +146.7 - 6.7 + 167.3 193 210 107 .9
1911 680.2 556 .9 -121.2 + 177.3 +146 .8 - 6.0 +196 .9 199 218 112.6
1912 744 .6 598 .9 - 14 3.8 + 186.9 + 158.6 -4 .6 + 197.1 214 230 112.5
1913 768 .7 634 .8 -131.6 + 199.6 + 168.2 - 11.9 + 224.3 220 239 116.2
1914 696.6 526 .2
1915 851.9 484 .0
....
v.. 1916 948 .5 603 .9
'-l 1917 1064 .2 596 .8
....
~ TABLE 7.1 (cont)

Balance-of-payments statsitcs [Em.]

Imports Exports Merchandise Balance Balance of Bullion Balance on Index of Index of Terms of
c.i.f. (including trade of overseas all other and specie current import export trade
re-exports) balance investment invisible account volume volume (1880 = 100)
f.o .b. earnings trade (1880 = (1880 =
100) 100)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

1918 1316.2 532.3

1919 1626.2 963.3
1920 1932 .6 1557.3 -386 +200 +395 +43 +252 195 170 92
1921 1085.5 810.3
1922 1003 .1 723.2 -183 +175 +150 +13 +155 187 164 89
1923 1096 .2 3 885 .83 -210 +200 +148 +16 +153 205 178 91
1924 1277.4 940 .0 -337 +220 +190 +12 +86 228 182 94
1925 1320 .7 927 .4 -393 +250 +188 +10 +54 236 179 97
1926 1241.4 778.5 -463 +250 +199 -12 --26 244 161 95
1927 1218 .3 832 .1 -386 +250 +219 -3 +79 252 184 96
1928 1195 .6 843 .9 -352 +250 +225 -7 +117 244 188 99
1929 1220 .8 839.0 -382 +250 +233 +16 +117 257 193 98
1930 1044 .0 657 .6 -386 +220 +194 -5 +25 249 157 91
1931 861.3 454 .5 - 407 +170 +134 +33 -70 254 120 82
1932 701.7 416 .0 -286 +150 +86 -17 -67 224 120 82
1933 675.0 417.0 -258 +160 +103 -201 -196 224 121 80
1934 731.4 447.2 - 284 +170 +117 --142 -139 234 129 82
1935 756 .0 481.1 -275 +185 +108 -56 -38 237 139 83
 1936 847.8 SOIA -346 +200 +127 -227 - 246 252 142 85
1937 1027.8 596 .5 - 431 +210 +176 -99 - 144 267 154 90
1938 919.5 532 .3 - 388 +200 + 122 +74 +8 254 136 82
1939 885 .5 485.6 - 250

1 Denotes that up to 1870 the values of imports and re-exports were computed by the Board of Trade . All other values were declared
by shippers.
2Denotes that the value of new ships sold abroad was included in exports for the first time in 1899 , when it came to £9.2 million.
3Southern Ireland was treated as foreign from 1 April 1923.
Sources: Columns 1 and 2 are from Mitchell and Deane (1962, pp. 283-4) . Column 3 is from Mitchell and Deane (1962 , pp. 333-5).
Up to 1913 the data are based on lmlah (1958) and from 1920 are taken from the Board of Trade Journal. (Note that because of
the different sources of data, col. 3 does not equal precisely the difference between col. ') and col. 1. It is col. 3, however, which is
used to derive the over-all balance on current account in col. 7.) Columns 4to 7 are from Mitchell and Deane (1962, pp. 333-5).
Columns 8 and 9 are from Mitchell and Deane (1962, pp . 328--9). The indices to 1913 are from lmlah (1958), and from 1920 were
provided by the London and Cambridge Economic Service. The present author put the latter indices on the same basis as the Imlah
indices using base 1880 = 100. Column 10 is from Mitchell and Deane (1962, pp . 331-2). The index to 1913 is from Imlah (1958),
and from 1920 from the Board of Trade Journal (4 August 1951) . The present author put both series on the same basis using base
1880= 100 .

.....
I..u
'0
140 Balance-ofPayments Theory and UK. Experience

World Wars the average trade deficit was £340 million per annum; and
although the surplus on overseas investment earnings and other invisible
trade increased , the improvement was not enough to offset the trade
deficit. Over the whole period there was a small annual average deficit
on current account, with the deficit years occurring in the 1930s.
Up to 1914 the balance of payments was helped by a slight trend
improvement in the terms of trade , the terms-of-trade index rising by
some 20 per cent over the sixty-year period from 1854 . By contrast
there was a marked deterioration in the terms of trade after 1913, the
index falling from 116in 1913t082in 1938(1880= 100) .
The growth in the volume of imports over the period 1854 to
1914 was slightly faster than the growth in the volume of exports,
offsetting the improvement in the terms of trade . The import-volume
index rose 5 .6 times , from 33 in 1854 to 220 in 1913, while the export
volume index rose fivefold, from 40 in 1854 to 239 in 1913 . There was
a noticeable surge in the growth of exports immediately preceding the
First World War. In the interwar period the deterioration in the trade
balance and current account was not only a function of the worsening
terms of trade. More significant was the absolute fall in the volume
of exports while the volume of imports continued to grow . By 1938
export volume was 20 per cent below the level of 1920 and 43 per cent
below the level of 1913 .

1940 TO 1950

For obvious reasons the balance of payments during the war period
went into serious deficit, as shown in Table 7 .2. The combined deficit
on current and long-term capital accounts averaged £750 million per
annum.
In the immediate post-war years, the balance of payments made a
dramatic recovery, while there was a slight deterioration in the terms of
trade . Export volume increased by 75 per cent between 1946 and 1950,
while import volume increased much more slowly by only 25 per cent,
largely because of import controls. The merchandise trade balance
improved in successive years, which, combined with a surplus on
invisible account, turned the balance of payments on current account
from deficit into surplus over the three years 1948 to 1950 . Despite
the improvement , the pound was devalued from ~4.03 to ~2.80 in
1949. It is doubtful, however , whether devaluation contributed to the
 TABLE 7.2
Balance-of-payments statistics 1940-50 (£m .)

Imports Exports Merchandise Invisible Balance on Balance of Index of Index of Net barter
c.i.f. (including trade balance balance current account current and import export terms of
re-exports) long-term volume volume trade
f.o.b. capital (1958=100) (1958=100) (1958=100)
(1) (2) (3) (4) (5) (6) (7) (8) (9)
- -- - -
1940 1082 393 -804 78 37 91
1941 986 324 -816 64 26 95
1942 997 271 - 663 60 19 102
1943 1234 234 - 680 64 15 94
1944 1309 266 - 659 69 16 102
1945 1104 339 - 875 56 24 99
(f.o .b.)
1946 1063 960 -103 - 127 - 230 +18 62 52 99
1947 1541 1180 - 361 -20 -381 - 84 70 57 92
1948 1790 1639 -151 +177 +26 +305 72 72 89
1949 2000 1863 - 137 +136 -1 - 48 78 80 91
1950 2312 2261 -51 +358 +307 +363 78 91 90

Sources : Columns 1 and 2, data for 1940 to 1945 from London and Cambridge Economic Service, Key Statistics 1900-1966, table
K, p. 14; data for 1946 to 1950 from Mitchell and Deane (1962, p. 142). Columns 3 to 6, data from Mitchell and Deane (1962,
p. 142). (The figure in column 6 for the period 1946 to 1950 include the balancing item and are therefore not comparable with
"- later years .) Columns 7 to 9, data from London and Cambridge Economic Service, Key Statistics 1900-1966, table K, p. 14.
-l::..
"-
142 Balance-of-Payments Theory and U'K. Experience
improvement of the balance in 1949 and 1950. Flanders (1963) con-
cludes her study of the 1949 devaluation by saying :

the weakening of Britain's competitive position vis-a-vis other

devaluing countries indicates that exchange rate adjustment was
not a sufficient condition for improving her foreign trade position ,
though it may have been necessary and it is at least probable that
the position would have deteriorated more if there had been no
devaluation .

Flanders examines the value share of U.K. exports in the four com-
modity groups of iron and steel , power-generating machinery, textiles
and vehicles over the period 1949 to 1955 in relation to (i) world
exports, (ii) exports of other devaluing countries, and (iii) exports of
non-devaluing countries. Except for exports of iron and steel and
power-generating machinery to non-devaluing countries, the share of
U.K. exports in all other markets continued to fall. While the results
by themselves would suggest a perverse elasticity of substitution for
U.K. exports in most markets, Flanders believes it is more likely that
other adverse factors were to blame, offsetting a normal negative
elasticity of substitution - factors such as: the Korean war; the recovery
of the Japanese and West German economies; relaxation of trade
controls and regulations ; and inelasticity of U.K. export supply .
What is really relevant, however, is not what happens to export shares
as a result of a devaluation but whether devaluation raises permanently
the trend rate of growth of exports relative to imports. The United
Kingdom's share of world exports was declining before the war and has
declined since. So, for that matter, has the United Kingdom's share of
world production and imports. These facts, however , need not neces-
sarily prevent a country enjoying a balance-of-payments surplus , as the
United Kingdom was later to do in the 1950s. In fact there is no
evidence to suggest that the devaluation of 1949 did raise the rate of
growth of exports, or lower the rate of growth of imports, permanently.
Indeed, economic theory would suggest that the best that devaluation
can do is to exert a once-and-for-all impact on the levels of exports and
imports, leaving the growth rates unchanged (see Chapters 9, 10 and 11).
The question might also be raised of whether devaluation would
have been resorted to had the balance-of-payments statistics at the time
been accurate. When the decision to devalue was taken it was thought
that the aggregate balance-of-payments deficit for the years 1946 to
A History of the UK, Balance ofPayments 143

1948 was over £1000 million. It is now known that it was much less as
Table 7.2 shows ; and in 1948 there was actually a surplus on the current
account and on the basic balance . Other instances will be mentioned
later of possibly precipitate action being taken on the balance of pay-
ments on the basis of figures which have exaggerated the deficit.

1951 TO 1959

Although the emphasis here is on describing the performance of the

balance of payments on current account, the history of the balance of
payments and of balance-of-payments policy cannot be understood
without brief mention of the United Kingdom 's role as an international
banker, and sterling's role as a reserve currency, which have made the
gold and foreign-currency reserve position of the United Kingdom
an element of such vital importance. Because of the large holdings of
sterling by foreigners and the persistent threat of their removal from
London, the exchange rate was also continually under threat in the
1950s and j 960s, notwithstanding surpluses earned on the current
accoun t. It was eventually weakness on capital account (both short and
long term), unrelated to any serious weakening on the current account,
that precipitated devaluation in 1967, and which led sterling to de-
preciate so dramatically from 1972 to 1978 . For reference purposes
Table 7.3 gives the value of official reserves and liabilities of the United
Kingdom since 1950. It can be seen that throughout the 1950s and
1960s the value of external liabilities, as a claim on the reserves, were
four to five times higher than the value of reserves. The ratio worsened
in the last half of the 1970s, largely as a result of the financing of the oil
deficit, but in 1977 there was a dramatic recovery as confidence in
sterling improved and the authorities attempted to hold the exchange
rate down .
The balance-of-payments statistics for the period 1951 to 1959 ,
as well as up to the present, are contained in Table 7.4. The improve-
ment in the balance of payments that was apparent in the immediate
post-war years continued into the 1950s. The large trade deficit in
1951 is attributable to the Korean war and the sharp rise in commodity
prices, but from 1952 there is a gradual improvement punctuated only
by deterioration in periods of excessive overheating of the economy.
The improvement in the merchandise balance compared with the inter-
war period and before was against a background of a commitment to
144 Balance-of-Payments Theory and U'K. Experience

TABLE 7.3
U.K. official reserves and external liabilities of the public sector since
1950

Gold and convertible External liabilities of

currency reserves (£m .) the public sector (£m .)

1950 1178 4060

1951 834 4143
1952 659 3786
1953 899 4004
1954 986 4179
1955 757 4045
1956 799 4091
1957 812 3918
1958 1096 3976
1959 977 4212
1960 1154 4432
1961 1185 4504
1962 1002 4106
1963 949 3737
1964 827 3861
1965 1073 4273
1966 1107 4844
1967 1123 5145
1968 1109 5428
1969 1053 5842
1970 1178 5419
1971 2526 5646
1972 2405 5799
1973 2795 6615
1974 2955 8998
1975 2700 10633
1976 2485 14625
1977 10975 18092
1978 10380 15593
1979 13220 15185
1980 13275 14901
Sources : London and Cambridge Economic Service, Key Statistics 1900 -1966;
and the United Kingdom Balance of Payments: 1981 Edition (London, H.M.S.D.).

full employment which was largely achieved, which makes the balance-
of-payments record even more remarkable. In 1956 and 1958 there
were actually slight surpluses recorded on the trade balance, a phenom-
enon achieved on only three previous occasions since 1800 - in 1816 ,
1821 and 1822 . The performance of the trade balance , combined with
a surplus on invisible account (albeit deteriorating) produced a surplus
A History of the UiK, Balance ofPayments 145
on current account in seven out of the nine years between 1951 and
1959, the exceptions being 1951 and 1955 . Excluding the abnormal
(war) year, 1951, the average surplus on current account was £154
million per annum . Apparently the Treasury's target surplus was
£450 million in order to allow room for net overseas investment and to
build up reserves. It is not clear , however, whether this is the criterion
by which the performance of the balance of payments ought to be
judged . Overseas investment and reserve accumulation are not macro-
economic goals that directly affect the living standards of people
(except to reduce them in the short run), and the figure itself is entirely
arbitrary . The only real standard by which the balance of payments
may be judged unsatisfactory in this period is the low rate of growth of
output compared with other industrialised countries. A higher growth
rate, assuming that would have been physically possible , would un-
doubtedly have raised the rate of growth of imports, which would have
worsened the balance of payments without a corresponding increase
in the rate of growth of exports. If there was a balance-of-payments
problem in the 1950s, it was the inability of the rate of growth of
exports to finance a higher rate of growth of real output. It was the
attempt to achieve faster growth by simply expanding demand, without
first an expansion of exports, which produced such difficulties for the
U.K. economy in the early 1960s and early 1970s, and from which it
has never fully recovered. In Chapters 10 and 11 attention will be
devoted to the balance-of-payments constraint as a source of inter -
national growth-rate differences and to the importance of export-led
growth in an open economy .
Over the period 1951 to 1959 there was a slight tendency for the
volume of imports to grow faster than the volume of exports, but
nothing to suggest the emergence of a fundamental disequilibrium in
the balance of payments at the prevailing full-employment growth rate .
Import volume grew 23 per cent between 1951 and 1959 while export
volume grew 17 per cent, but as far as the balance of payments is con-
cerned this adverse tendency was offset by a 25 per cent improvement
in the terms of trade .
The two major so-called balance-of-payments crises in the 1950s
which invoked domestic deflation were in 1955 when the economy was
clearly overheated and there was also rumour of full sterling con-
vertibility, and the speculative attack on sterling in 1957 when Bank
Rate was raised to the record high level of 7 per cent to protect the
capital account of the balance of payments. The Radcliffe Committee
...... TABL E 7.4
~
0-
Balance-of-payments statistics since 1951 (Em .]
Imports Ex po rts Mer chand ise Invisible Balance o n Balance of Balance for Inde x of Index of Terms of
I.o.b. (including tr ade balan ce ba lance current current and official import expo rt trade
re-exports) accoun t long-term finan cing" volume volume (1970= 100)
f .o .b. capital U970=100) (1970= 100)
account
(basic balance)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

1951 3424 2735 - 689 +320 - 369 n a. - 334 48 53 72

1952 3048 2769 - 279 +442 +163 +29 - 175 44 50 77
1953 2927 2683 - 244 +389 +145 - 49 +296 48 51 82
1954 2989 2785 - 204 +321 +117 -74 +126 49 53 82
1955 3386 3073 - 313 +158 - 155 - 277 - 229 55 57 81
1956 3324 3377 +53 +155 +208 +21 - 159 55 59 83
1957 3538 3509 - 29 +262 +233 +127 +13 56 60 85
1958 3377 3406 +29 +317 +346 +148 +290 57 58 91
1959 3642 3527 - 115 +287 +172 -112 +18 61 61 91
1960 4138 3737 -401 +173 - 228 -457 +325 68 63 92
1961 4043 3903 - 140 +187 +47 +64 - 339 67 65 95
1962 4103 4003 - 100 +255 +155 +14 +192 69 67 97
1963 4450 4331 - 119 +244 +125 - 35 - 58 72 70 96
1964 5111 4568 - 543 +185 - 358 -749 - 695 75 72 94
1965 5173 4913 - 260 +230 -30 - 274 - 353 76 77 97
1966 5384 5276 - 108 +238 +130 - 63 - 547 78 79 98
1967 5840 5441 - 599 +330 - 269 -451 - 671 84 82 100
1968 7145 6433 - 712 +468 - 244 - 422 -1410 93 84 97
 1969 7478 7269 -209 +714 +505 +385 +687 94 97 96
1970 8184 8150 -34 +875 +823 - +1287 100 100 100
1971 8853 9043 +190 +934 +1124 - +3146 104 107 101
1972 10185 9437 -748 +995 +247 - -1265 117 107 101
1973 14523 11937 -2586 +1605 -981 - -771 134 122 90
1974 21745 16394 -5351 +2078 -3273 - -1646 136 132 75
1975 22663 19330 -3333 +1812 -1521 - -1465 126 126 81
1976 29120 25191 -3929 +3048 -881 - -3629 134 137 80
1977 34012 31728 -2284 +2243 -41 - +7361 142 148 82
1978 36605 35063 -1542 +2481 +939 - -1126 149 152 87
1979 44136 40678 -3458 +2595 -863 - +1905 166 158 87
1980 46211 47389 +1178 +2028 +3206 - +1372 157 161 85

*Figures from 1978 refer to total official financing (see note 2 to Chapter 1).

Sources : Columns 1 to 5, data for 1951-5 are taken from Mitchell and Deane (1962, p. 142), from 1956 the data are from UK.
Balance of Payments, various editions, H.M.S.a. Column 6, data from Bank of England Historical A bstract, no. 1, 1970. Column 7,
ibid. for years 1951-5, and for later years UK. Balance of Payments, various editions. (Balance for official financing is synonymous
with total currency flow) . Columns 8 to 10, data for 1951 to 1966 are taken from London and Cambridge Economic Service, Key
Statistics 1900-1966, p. 14, and for later years from NationalInstitute Economic Review, and UK. BalanceofPayments, 1981 Edition.

......
~
'1
148 Balance-of-Payments Theory and U.K. Experience

(1959) remarked (and interestingly, in the light of the history of the

1960s and the belief of many that the U.K. balance of payments has
been in fundamental disequilibrium since the Second World War) that
'the repeated exchange crises have not been due ... to any failure on
the part of the U.K. to pay her way but to the volatility of various
elements in the balance of payments and to the lack of reserves ade-
quate to withstand the resulting pressure on them'.

1960 TO 1969

There seems to be a general consensus that while the balance of pay-

ments was reasonably healthy before 1959, from that date it moved
into a state of underlying disequilibrium, in the sense of an inability of
the economy to achieve the full employment of resources without a
payments deficit emerging. The emergence of such a classic conflict
was the cue for many economists , though by no means all, to advocate
either devaluation , or the abandonment of fixed exchange rates , as the
solution to the dilemma . The advocacy of exchange-rate adjustment
was rarely accompanied, however, by any detailed analysis of the nature
of the difficulties in the early years of the 1960s, or a thorough con-
sideration of the consequences that a fall in the exchange rate would
have on the domestic economy. A consensus on a matter, moreover,
does not mean that the view is necessarily correct. But if a disequili-
brium in the balance of payments did emerge in the sense defined
above, exchange-rate adjustment is only appropriate if the cause 0 f the
disequilibrium is a loss of price competitiveness, or possibly to give
breathing space to tackle the true causes, and then only if the dele-
terious inflat ionary repercussions on the internal economy can be
avoided.
The emergence of a deficit in 1960 from a surplus in 1959 cannot be
attributed to a sudden deterioration in price competitiveness . In fact
the current-account deficit in 1960, against which restrictive domestic
action was taken in 1961 , turned out to be exaggerated . Instead of an
assumed deficit of £344 million, and the forecast of another sub-
stantial deficit, in 1961, the true deficit for 1960 turned out to be £228
million, and there was a surplus of £47 million in 1961. There was a
larger deficit on the basic balance in 1960 caused by an adverse move-
ment in the long-term capital account, but this cannot be attributed to
adverse relative price movements either. In 1962 and 1963 the current
A History of the U'K. Balance ofPayments 149
balance moved further into surplus, with if anything the level of exports
above trend, which is hardly evidence of a strongly deteriorating
competitive position. The fact of the matter is that international price
competitiveness, measured by relative export prices or relative unit
labour costs, did not deteriorate over this crucial period leading up to
the balance-of-payments crisis of 1964. Ray (1966) shows that over the
period 1958 to 1964 total wage costs per unit of output in the United
Kingdom rose II per cent compared with 12 per cent in Japan, 12 per
cent in Italy, 22 per cent in West Germany, 6 per cent in France, and
-5 per cent in the United States . MacGeehan (1968), in her survey of
competitiveness, confirms that between 1960 and 1965 unit labour
costs in the United Kingdom grew more slowly than in Europe . (For a
discussion of measures of competitiveness, see Chapter 9.) Import
volume grew by 20 per cent between 1959 and 1963 and export volume
by 15 per cent - a ratio no different from that of the 1950s.
Despite the facts Coppock (1965) argues that the poor performance
of the current-account balance of payments in the early 1960s merely
reflects the general uncompetitiveness of the U.K. economy in world
trade in manufactures, and that all the evidence adds up to a convincing
case for the existence of fundamental disequilibrium for which the
appropriate remedy is devaluation . He says:

the persistence of [balance-of-payments] crises in conjunction with

the underlying trends in the U.K. foreign trade and the balance of
payments may be held to justify the view that there has been a
cumulative tendency towards fundamental disequilibrium and that
devaluation may be an appropriate or even an essential part of a
package deal to restore equilibrium.

It is true that he qualifies his judgement by saying:

devaluation cannot be expected to remove the deep-seated defic-

iences of British business in productivity growth and export prom -
otion or the institutional bias towards inflation which are generally
accepted as the causes of our troubles. A permanent solution to the
British problem requires the correction of these deficiencies [but , he
goes on] . .. the immediate situation calls for short term relief to
improve the balance of payments position and to enable us to
recover lost ground.
150 Balance-of-Payments Theory and U.K. Experience

In presenting his case he includes the year 1964 in calculating the

average annual balance of payments on current account for the early
years of the 1960s, but surely no one can claim that the deficit that
emerged in 1964 was the result of relative price deterioration. The
worsening of the trade deficit from £ 119 million in 1963 to £543 million
in 1964 is attributable entirely to the massive upsurge in the value of
imports from £4450 million to £5111 million resulting from the very
rapid rate of expansion of demand . The worsening of the trade deficit
caused the current account to move into deficit from a surplus in
1963 ; and an even larger deficit emerged on the long-term capital
account to give a basic balance -of-payments deficit of £749 million,
the largest (until then) in peace-time history. The difficulties of 1964
were cyclical and were on the capital account; they were not the result
of forces to be rectified by exchange-rate depreciation.
The Labour government opposed devaluation in 1964. According to
Stewart (1977) the opposition was on three main grounds : first, the
fear that the Labour Party would be branded as the Party of devalu-
ation ; second, because (to quote Harold Wilson, the Prime Minister) it
may well have started off an orgy of competitive beggar-my-neighbour
currency devaluations which would have plunged the world into mone-
tary anarchy ; and third , a deep -seated distrust of the price mechanism .
It would be misleading to give the impression, however , that there was
an academic consensus in favour of devaluation opposed only by
obstinate politicians on the left. In the early 1960s Sir Roy Harrod and
Thomas Balogh (who became personal advisor to the Prime Minister,
Harold Wilson, in 1964) came out strongly against devaluation , and in
the second half of 1966 three more eminent economists of very differ-
ent political persuasions all expressed doubts . Professor Hicks (1966)
remarked : 'It is widely supposed that the only reason why we do not
devalue is the obstinacy of Mr Wilson. But economists may be grateful
for the obstinacy of Mr Wilson and for the good advice on this matter
we may be sure he is receiving .. . There is really no alternative while
our position is so weak but to cling to the existing parity.' Professor
Joan Robinson (1966) expressed scepticism about the wisdom of
devaluation in conditions of full employment :

In a situation where there is some unemployment and unused

capacity in many lines, a devaluation increases activity and improves
the balance of trade at one stroke. But when there is near full-
employment already! it is liable merely to increase the pressure of
A History of the U'K, Balance ofPayments 151

demand for labour, while the rise in price of imports increases the
pressure for higher money wage rates, so that before long the com-
petitive advantage of lower home costs is completely lost.

Then Lord Robbins declared in the House of Lords on 28 July 1966 : 'I
welcome greatly the declaration . . . expressing the firm decision of the
government not to resort to this expedient [i.e. devaluation] . I do not
think indeed that in present circumstances this view that we should
have devalued has much practical applicability.' Later, in April 1967 ,
Professor Ball (1967) expressed strong opposition to devaluation on the
grounds of elasticity pessimism and noted that the advocation of de-
valuation seemed to be based more on despair than on any positive
merits. When devaluation eventually took place in November 1967 ,
several economists opposed it, and there was wide disagreement on the
measures necessary to make devaluation effective (see Hutchison ,
1977) . Harrod (1968a , 1968b) continued to attack its appropriateness
and wisdom, and Hawtrey (1969a , 1969b) continued to argue, as he
had for many years , that the source of the United Kingdom 's difficulties
was not an overvalued pound , but an undervalued pound causing excess
demand at home .
The response to the 1964 balance-of-payments deficit was the trad -
itional one of demand deflation, coupled with the imposition in
October 1964 of a 15 per cent import surcharge on all impor ts except
food and some raw materials . The surcharge was reduced to 10 per
cent in April 1965 and removed entirely in November 1966 . There have
been several different approaches to the estimation of the effects of the
surcharge . One has been to use import demand equations (see Chapter
9) to forecast imports over the surcharge period and to attribute to
the surcharge the difference between actual and forecast imports.
Johnston and Henderson (1967) adopted this approach using quarterly
data and estimated a reduction in imports due to the surcharge of
between £115 million and £165 million at current prices. Using this
method Black et al. (1967) of the National Institute of Economic and
Social Research estimated a surcharge effect of £200 million . The
weakness of this approach is that it makes no allowance for residual
errors which occur in the normal course of forecast ing, which in the
case of imports could be as high as plus or minus £100 million. This
difficulty can be overcome by incorporating the surcharge into the
import equations as an independent variable and estimating the effect
directly . Black et al. (1967) also adopted this approach and estimated
152 Balance-of-Payments Theory and U'K. Experience

a reduction in imports due to the surcharge of £380 million at current

prices. A third method of approach used by Barker (1970b) is to
include the surcharge as part of the price paid by the importer, using
this price to explain the behaviour of imports, and then to recalculate
import prices excluding the surcharge and use the equations to estimate
imports in the absence of the surcharge . The effect of the surcharge is
calculated to be a reduction in imports of £462 million.! Barker
suggests that perhaps this estimate is closer to the true figure than the
lower estimates given, particularly in view of the large unanticipated
rise in manufactured imports in 1967 after the surcharge was removed .
Between 1966 and 1967 there was a 13 per cent rise in the volume
of manufactured and semi-manufactured good imports, and in 1968
imports exceeded forecasts following devaluation in November 1967.
The events from 1964 leading up to the devaluation of 1967 are
outlined more fully in Appendix 7.1 later in this chapter, which also
contains a detailed analysis of the origins of the 1964 deficit which
started the devaluation bandwagon . Briefly , it is clear from the figures
that the balance of payments responded to the deflationary (and other)
measures taken. Import growth fell in 1965 and 1966, while export
volume grew at close to 7 per cent per annum. The visible deficit
shrunk, and the balance on current account moved into surplus in
1966 .3 The basic balance was in deficit by less than £100 million,
though the balance for official financing was much larger, at over
£500 million . The question is, why did the situation deteriorate so
markedly in 1967 as to make devaluation unavoidable, when the
pressure of demand was falling, and early in the year both the Bank of
England and the National Institute of Economic and Social Research
were forecasting an over-all surplus? The initial source of difficulty
was the slow-down in the growth of world trade which reduced the
volume of exports , while imports grew at an annual rate of 7 per cent.
Even so , the balance of payments was still in surplus in the middle of
the year and there was little pressure on sterling in the foreign-exchange
markets . The steady decline in fortunes started with the Middle-East
crisis in June and the subsequent closure of the Suez Canal, which
prompted the heavy selling of sterling . Short-term interest rates in the
United States and in the Euro-dollar market began to rise, and this also
exacerbated the pressure on sterling. The June trade figures announced
in July were disappointing, and discussion of devaluation began to
appear in the press, all of which added further to the strain in the
foreign -exchange market. In the third quarter of 1967 a balance-of-
A History of the UiK, Balance of Payments 153

payments deficit was registered, and a general lack of confidence caused

further losses of foreign exchange. By September shipments of exports
began to be affected by strikes in the liverpool and London docks.
This was the final straw which confirmed in the minds of foreigners the
apparent precariousness of the U.K. position. The consensus that
measures would have to be taken to protect the balance of payments
and the reserves, which might include devaluation , itself made worse
the outflow of funds. The speculative pressure against sterling became
so intense that what was predicted occurred: the pound was devalued
from $2.80 to $2.40 on 18 November. The devaluation was forced by
a speculative crisis, rather than by any crisis on trading account. The
bad trade figures in the third quarter of 1967 were largely the result
of seasonal factors; and abstracting from the slow-down in the growth
of world trade, the accounts even reveal some underlying improvement
(see Appendix 7.1) . It is also now known that because of a mistake in
the recording of exports there was only a small over-all cumulative
deficit on the current account of the balance of payments between
1964 and devaluation day , contrary to what was thought at the time.
i.e. that the current account was chronically unhealthy. One might
add that had the expansion of demand in 1964 not been so irrespons-
ible, the current account over the period would have recorded a sub-
stantial surplus , and the country may have comfortably ridden the
crisis of confidence in 1967 precipitated largely by external events.
According to the second volume of the Crossman diaries (1976)
most members of the Cabinet were in favour of the pound floating in
1967, as opposed to either devaluation or import controls. Only
Harold Wilson and James Callaghan were against. Crossman records a
dinner on 1 November 1967 attended by himself, Mrs Barbara Castle,
Thomas Balogh, Mr Peter Shore and Mr Tony Benn : 'There is no
disagreement among us ... we all know that we should float the pound
and not devalue to a fixed point. We all agree that import controls will
be quite ineffective.' Crossman also describes how in the Steering
Committee on Economic Policy, on 8 November, Wilson isolated
Callaghan as the only anti-depreciationist , After Callaghan had des-
cribed devaluation as a catastrophe, Wilson is supposed to have replied :
'I am politically open on the subject. 1 am prepared to think there
could be some merit in a free decision, and in that case we must of
course decide whether we float or devalue by a certain amount. My
mind is not closed.' Crossman's account of the lead up to devaluation
is rather different from Wilson's own account (1971). Wilson says:
154 Balance-of-Payments Theory and U.K. Experience

though a contingency plan for devaluation of sterling had long been

in the Treasury files ... change in parity was not in contemplation.
The Governor [of the Bank of England] met with the Chancellor
and me at no . 10 [late October 1967]. I asked him specifically
whether a recommendation to devalue was in his mind. He said
flatly that it was not . .. As we entered November sterling was still
under pressure, but unlike July 1966, there was hardly a serious
commentator pressing for devaluation.

DEVALUATION, 1967

In this section the effects of devaluation are considered . The devalu-

ation of the pound sterling from $2 .80 to $2 .40 reduced the dollar
price of pounds by 14.3 per cent and raised the sterling price of dollars
by 16.7 per cent. Other things remaining the same this would allow
U.K. exporters to reduce prices in foreign currency in foreign markets
by 14.3 per cent, and we should expect the sterling price of imports to
rise by 16.7 per cent. Other things do not always remain the same,
however. Exporters may alter the sterling price of exports and
foreigners may alter the foreign price of domestic imports. What
happens to sterling export prices depends on the pricing policies of
manufacturers and the degree of inflation induced by rising import
prices. The foreign price of imports depends on the elasticity of supply
and the pricing policies of suppliers. The evidence is that export prices
measured in sterling rose by about 8 per cent, and fell by about 6 per
cent measured in foreign currency . Exporters apparently reaped the
benefits of devaluation through a combination of expanding sales and a
higher rate of profit per unit of sales. Import prices measured in sterling
rose by 12 per cent in 1967-8 and by 5 per cent in 1968-9 . Assuming
no change in foreign prices, this would mean that the effect of devalu-
ation on import prices was spread over a minimum of two years.
We shall consider here four major studies of the effect of the 1967
devaluation on the U.K. trade balance. Worswick (1971) examines the
effect on exports and imports separately. On the export side he com-
pares actual exports with predicted exports on the basis of pre-1967
behaviour. The United Kingdom's share of world trade in manufactures
declined by 0.7 per cent per annum prior to 1967. Applying
this figure to 1968 and 1969, Worswick calculates that exports should
have been 14 per cent higher in 1969 than in 1967 in the absence of
 A History of the U.K. Balance ofPayments 155
devaluation . In fact actual exports were 25 per cent higher. He attrib-
utes the 11 per cent difference to devaluation . With an 8 per cent
increase in sterling export prices and an 11 per cent increase in the
volume of exports, the effect of devaluation on the sterling value of
exports is calculated to have been £1240 million. On the import side
the volume of imports rose in 1968 and 1969, above the level predicted
on the basis of past experience , apparently indicating a perverse price
elasticity of demand for imports. Making the less extreme assumption,
however, that devaluation had no effect on the volume of imports,
Worswick calculates that the rise in the sterling price of imports raised
the sterling import bill by £1060 million . The net effect on the trade
balance is therefore estimated to have been +£180 million over two
years . The effect of devaluation on the invisible account is estimated
at +£330 million, giving a total positive contribution to the current
account of £510 million over two years . In a later paper the staff of the
National Institute of Economic and Social Research (1972) 'extend the
analysis to 1970 and revise the estimates downwards, calculating that
by 1970 the improvement in the trade balance was only £130 million,
and in the current balance £425 million.
A study by Artus (1975) for the I.M.F. puts the estimate of the
beneficial effect of devaluation on the trade balance much higher than
do Worswick and the National Institute . This is for two reasons. First,
higher imports due to the higher pressure of demand induced by
devaluation are subtracted ." Second, Artus assumes a cumulative
relative price effect on the volume of imports of -3.5 per cent, whereas
Worswick's study assumes no import-volume response. Artus's study
gives an eventual improvement in the trade balance of £940 million by
the end of 1971.
An interesting study by Masera (1974), who takes broadly the same
approach as Worswick, is also concerned with lags in the response of
import and export volumes to relative price changes as well as with the
ultimate quantitative effect of devaluation on the trade balance . Masera
estimates a J -curve effect (see Chapter 3) lasting until the third quarter
of 1968, after which the trade balance shows a positive improvement
compared with what it would otherwise have been. He estimates,
however, that it took another three or four quarters for the initial losses
to be made good, but ultimately there was a net improvement in the
trade balance of £520 million up to the end of 1969. His estimate is
higher than Worswick's because he argues that in 19675 there was a
shift in the import propensity, which led to an underprediction of
156 Balance-ofPayments Theory and U.K. Experience

imports based on past trends and therefore to an underestimation of

the import response to devaluation .
Figures of the order of magnitude cited above may suggest that
devaluation was successful in relation to the level of the prevailing
deficit. Such a judgement needs to be tempered with caution. First ,
the positive benefits last only so long as the competitive advantage
gained by devaluation is not eroded by competition in world markets
and/or domestic price inflation . Second, even if the competitive ad-
vantage is maintained, the positive balance-of-payments response does
not put the economy on a higher growth path. The conflict between
faster growth and balance-of-payments equilibrium remains , unless
there is continual devaluation. Third, it is not clear that any of the
studies takes adequate account of the rapid growth of world trade in
1968 which expanded in volume by 12 per cent compared with 1967.
Some of the assumed devaluation effect on exports may be due to this
dramatic rise. Care must also be exercised in assuming that what may
have been a successful exercise in the past can be repeated in the
future. The 1967 devaluation was assisted by two important factors ,
in addition to the rapid expansion of world trade. First , there was still
in 1967 widespread money illusion in the economy. Price inflation had
been mild up to then, and productivity growth and an improving terms
of trade combined to raise real living standards by close to 3 per cent
per annum. When the devaluation came, wages were slow to respond
to the price increases, and the competitive advantage of devaluation
was slow to be eroded by domestic price inflation . Inflation did ac-
celerate, however, money illusion disappeared, and it is no coincidence
that from that date the celebrated Phillips curve, found to be stable
and negatively sloped over a long period of history, started to go awry.
Devaluation, or currency depreciation, in the absence of money illusion
in the labour market (or controls to produce the same result) , cannot be
successful.
The second factor which greatly assisted devaluation in 1967 was
the contraction of domestic demand. By 1969 virtually the whole of
economic policy was orientated towards the balance of payments.
Tight monetary and fiscal controls were imposed in November 1968
and in the budget of 1969, giving a total fiscal contraction of over
£1000 million, and in its Letter of Intent to the I.M.F . on 22 May 1969,
the government committed itself for the first time to a target for
Domestic Credit Expansion :" 'The government's objectives and policies
imply a domestic credit expansion for the private and public sectors in
 A History of the U'K, Balance ofPayments 157

the year ending 31st March 1970 of not more than £400 million.' At
the same time the government declared that it 'attaches the greatest
importance to monetary policy, which provides an essential support to
fiscal policies' . An import deposit scheme was also introduced to
restrict imports and reduce liquidity simultaneously. Importers of
manufactured goods were required to deposit 50 per cent of the value
of imports with the government before the goods could be cleared from
Customs and Excise. The required figure was reduced to 40 per cent
in November 1969 and to 30 per cent in April 1970 , and the scheme
was fmally ended in December 1970. With no interest payable on
deposits , and deposits not repayable for 180 days, the scheme was
equivalent to an import surcharge of about 2.5 per cent.
The outcome of this formidable array of policies to improve the
balance of payments was that the current account moved into surplus
in 1969 and reached a record level in 1971 of over £1000 million. The
trade balance was still in deficit in 1969 and 1970, but so it had been
since 1822 except for two years in the 1950s. So severe were the
deflationary measures, however, and so rapid the growth of export
volume (largely as a result of the rapid expansion of world trade) , that
the trade balance itself also moved into surplus in 1971, the surplus of
£190 million being the largest in recorded history (until then). The
'full' employment trade balance would no doubt have been in deficit,
but the current account would still have shown a surplus .
The declared aim of the Treasury in the 1950s and 1960s was to run
a current-account surplus of £450 million to permit long-term overseas
investment and to build up foreign-exchange reserves. By the end of the
1960s the objective had been achieved, though at some cost to employ.
ment. A policy of 'steady as she goes' would probably have consoli-
dated the surplus and allowed the country to retain some of the bene-
fits of devaluation for longer. As it was the surplus was frittered away
by an irresponsible expansion of internal demand, reminiscent of what
happened in 1964 and which led to the troubles and stagnation of the
1960s .

1970 TO 1980

The history of the balance of payments in the first years of the 1970s,
before the pound was allowed to float in June 1972, is told in
Appendix 7.2 to this chapter. Before considering the balance of pay-
158 Balance-of-Payments Theory and U.K. Experience

ments since the par value for sterling was abandoned, a reminder will be
useful of the international events surrounding the U.K. decision. In
August 1971, following a concerted speculative attack against the
dollar, the convertibility of the dollar into gold at $35 per ounce was
suspended and the world's major currencies began to float against each
other for the first time since the Bretton Woods agreement of 1944 .
The expectation was, however, that a new set of parities would be
established with the dollar devalued against other major currencies , and
this was duly accomplished by the Smithsonian agreement of December
1971 . The dollar was devalued by 8 per cent by raising the price of gold
from $35 to $38 per ounce , and the exchange margins allowed around
the new parities were widened from 1 per cent to 2.25 per cent. The
new parities could not be maintained for long, with large volumes of
capital switching from one financial centre to another, and they finally
broke down . All major currencies have been floating since 1974 .
The first currency to break was sterling in June 1972, though the
current account of the balance of payments was still in good surplus.
Switzerland followed in January 1973, and Italy later. Belgium,
Denmark, West Germany, France and the Netherlands operated a
joint float against the dollar for some time , keeping their own curren-
cies within a 2.25 per cent margin of each other. In June 1973 , how-
ever, West Germany was 'forced' to let the mark appreciate , and in
January 1974 France left the joint float and allowed the franc to float
freely.
The expectations held out for floating exchange rates were that
they would lend stability to exchange rates , free domestic economic
policy from external payments constraints, economise on the use of
international reserves, and make management of the international
monetary systerri easier. Alas, the reality has been far removed from the
expectations. First, exchange-rate fluctuations have been quite violent
by any standard, and possibly more disruptive than the periodic devalu-
ations under the Bretton Woods system . Some evidence for the early
years, presented by Hirsch and Higham (J 974), is shown in Table 7.5 .
The exchange-rate movements since 1974 have been just as turbulent.
Hirsch and Higham conclude that the sizeable fluctuations experienced
cannot be attributed to destabilising official intervention, as some
proponents of free floating have tried to argue. If floating had been
completely free , and not managed, the fluctuations would have been
even greater .
Second , domestic policy has not been freed from considerations of
 TABLE 7.5
Fluctuations in exchange rates for sterling, 1 April 1973-31 January 1974 (daily rates)

Range" between lowest and highest rates within

Weekly periods Monthly periods

Sterling against Average Maximum Average Maximum

U.S. dollar 0.8 3.1 2.6 5.9

Deutschmark 1.4 6.9 4.6 9.4
French franc 1.2 6.7 3.9 7.1
Swiss franc 1.5 9.3 4.0 9.3
lira l.l 2.9 4.6 8.1
Yen 0.9 3.5 2.5 6.2

*Expressed as a percentage of the average of the highest and lowest daily rates in the weeks or months involved.
Source : Hirsch and Higham (1974) .

....
V,
'0
160 Balance-of-Payments Theory and U.K. Experience

the balance of payments because floating has not lessened speculative

pressure, and has failed to reduce trade imbalances (see Kaldor, 1978) .
Currency depreciation has led to speculative outflows and often wors-
ened payments imbalance, while currency appreciation has led to specu-
lative inflows and mounting payments surpluses. Floating exchange rates
have tended to weaken the weak and strengthen the strong out of all
proportion to differences in the real economic performance of countries ,
and this has necessitated internal policies as draconian as under fixed
exchange rates (even in conditions of unemployment). Interest-rate
policy has continued to be dominated by consideration of the balance of
payments, and in the deficit countries there has been a growing demand
for trade restrictions . There could be no stronger indictment of the inade-
quacy of exchange depreciation to rectify payments imbalance than the
call for protectionism to become respectable . The U.K. Treasury itself has
undoubtedly become more and more sceptical that a downward float -
ing pound provides the answer to improving the country's balance of
payments and removing the balance-of-payments constraint on faster
growth. Work at the Treasury by Odling-Smee and Hartley (1978),
using the Treasury macroeconomic model, suggests that over 50 per
cent of the competitive advantage of exchange-rate changes is eroded
within two years even on the weak assumption that it takes three years
for wages to fully compensate for price increases. Since wages, in the
absence of controls, tend in practice to adjust much more quickly,
little competitive advantage from exchange depreciation can be ex-
pected at all. Odling-Smee and Hartley conclude that a permanent gain
in competitiveness would require continual depreciation in excess of
the resulting cost inflation, and could thus be achieved only at the
cost of accelerating inflation. It should also be explicitly recognised
that even if a competitive advantage could be maintained, this is not
sufficient to allow the growth rate to be raised permanently. This
would require a continually growing competitive advantage .
Because speculation has exacerbated exchange-rate instability, and
failed to eliminate imbalances, there has been an even greater need, not
less, for international reserves to 'ride out' crises, and an even greater
need for international co-operation to compensate for the lack of inter-
national reserves and to manage the international monetary system .
In the United Kingdom the immediate effect of floating was for the
exchange rate to depreciate and for the balance of payments to worsen .
By the end of 1972 sterling had depreciated against the dollar , and
against a trade -weighted average of twenty-one currencies (including the
A History of the U.K. Balance ofPayments 161

dollar), by 10 per cent. The trade-weighted average of these currencies

is called the effective exchange rate.' By the end of 1973 the de-
preciation against the dollar was 11.2 per cent, and the depreciation of
the effective exchange rate was 17.4 per cent, compared with the
Smithsonian parities of December 1971 . The course of the sterling-
dollar exchange rate and the effective exchange rate since 1971 are
shown in Table 7 .6.
The rapid depreciation of sterling in 1972 and 1973 raised the
sterling payments for imports by much more than the increase in the
sterling receipts from exports, and this helped to produce a merchan-
dise trade deficit of over £2000 million in 1973, the largest in recorded
history . The other major contributor to the enormous deficit was
the steep rise in commodity prices. The import bill between 1971
and 1973 grew by nearly £6000 million. The balance on current
account moved from a small surplus in 1972 to a deficit of nearly
£1000 million in 1973 .
None of these adverse trends had anything to do with the rise in the
price of oil, which did not take place until December 1973 . The quad-
rupling of oil prices led to severe difficulties in 1974, however. Not
only did the U.K. oil bill rise by approximately £2500 million but
exports also suffered as the world was plunged into recession through
the inability of the OPEC countries to spend their surpluses . In 1974
the trade deficit rose to over £5000 million and the current-account
deficit to over £3000 million. Against the background of a worsening
payments imbalance, and the country working a three-day week as a
result of the miners' dispute, with rumours also of a General Election,
sterling came under increasing speculative attack in the early part of
1974 . There was some recovery of sterling in the middle of the year,
but the strong downward pressure continued at the end of the year,
and at least $500 million was used from the reserves in December alone
to support the exchange rate . There was heavy selling of sterling by the
Arabian American Oil Company (Aramco), the world's largest oil
consortium, after a request from Saudi Arabia that future oil royalties
should be paid exclusively in dollars . Earlier in the year , to attract
Arab oil revenue and to induce other sterling holders to maintain their
balances in London, the Treasury announced a new exchange-rate
guarantee scheme for official overseas sterling funds . The new guarantee
was expressed in terms of the effective exchange rate instead of the
dollar (which had been used for previous guarantee schemes) . In return
for the exchange-rate guarantee overseas holders of sterling are required
....
~

TABLE 7.6
Exchange-rate movements since 1971-

The $ price of The effective exchange-rate index

sterling (1971 = 100)

1971 Smithsonian parity 2.606 100.0

1972 2.502 96 .7
1973 2.452 87.5
1974 2.339 84.8
1975 2.222 78.3
1976 1.806 66.3
1977 1.746 63 .0
1978 1.920 63.2
1979 2.123 67.7
1980 2.328 76 .9

-Based on an average of daily spot rates .

Source: Bank of England Quarterly Bulletins.
 A History of the U.K. Balance ofPayments 163

to maintain between 60 and 70 per cent of their national reserves in

London. On April Fools day 1974 it was announced that the Bank of
England would publish, like some medieval ritual, the effective ex-
change rate daily at noon.
By the beginning of 1975 the effective exchange rate had depreci-
ated by 22 per cent against the Smithsonian parities and the dollar price
of the pound was down to $2.38. The Bank of England intervened to
halt the slide of the exchange rate throughout most of 1975, but the
downward trend continued. The internal domestic recession brought
a virtual halt to the growth of imports, and both the trade deficit and
the current-account deficit were cut by some £2000 million compared
with 1974. The payments imbalance remained huge, however, despite
depreciation .
The year 1976 was one of crisis. On 5 March (Black Friday) the
pound fell below the $2 level, heralding in the SOp dollar . Once the
psychological barrier had been broken, there were visions of the pound
falling to $1 .50, even to a dollar a pound . The immediate cause of
sterling's demise was the rum our that Nigeria might withdraw its
sterling funds from London following a diplomatic row with the
United Kingdom in which the High Commissioner in Lagos was ex·
pelled because of his unacceptability to the Nigerian authorities . The
depreciation of the effective exchange rate for sterling reached 32 per
cent. In May the pound fell below $1.80 and the effective rate of ex-
change had fallen by nearly 40 per cent since 1971 . The depreciation
from January to May alone was 12.5 per cent despite moderating in-
flation and an improving balance of payments on current account. The
Bank of England was forced to raise the Minimum Lending Rate to a
record high of 11.5 per cent. In a B.B.C. interview , Mr Healey, the
Chancellor of the Exchequer, declared that 'the value of the pound
had gone down further than was justified by underlying economic
trends and would soon bounce back again'. It might well be asked by
what objective standards did the comparative industrial and commercial
performance of the United Kingdom deteriorate by 40 per cent in
relation to other countries during the previous three years . The truth
is, of course, that there was no justification based on real factors for
such a depreciation. The volatility of floating exchange rates in general,
and the demise of sterling in particular, is to be explained by the whims
and sentiment of speculators, and by the political activities of creditors.
This is the price that a country pays for short-term borrowing and for a
central role in international economic affairs . The price for the U.K.
164 Balance-of-Payments Theory and U.K. Experience

people after 1972 was high and protracted , and probably higher than
if there had been a quick real adjustment to the rise in commodity
prices in 1973 and 1974 with the exchange rate stabilised - by, if
necessary , strict controls on capital movements. The pound continued
its relentless slide into June. The City called for public expenditure
cuts, government backbenchers called for an inquiry into the fall, and
the Opposition in Parliament called for a halt to the slide. The
Chancellor declared that the United Kingdom must keep its nerve
and not panic , but against a background of rumours and speculations
the slide continued . On 7 June a $5300 million loan was announced
from the Group of Ten nat ions , available for six months . The pound
rallied slightly and was also helped by the miners' vote in favour of
the government's second stage of pay policy. On 28 September the
foreign exchanges experienced their worst day since June 1972. The
Chancellor , who was on his way to Hong Kong for a meeting of finance
ministers , returned to London because of the heavy selling of sterling,
which caused the rate to fall by over 4 cents during the day , closing at
$1.637. The effective exchange rate fell to 44 per cent below the
Smithsonian parities. The Bank of England had, in fact , stopped sup-
porting the pound out of the reserves from 9 September . So much for
free floating avoiding the panic and disruption of fixed rates! Then ,
25 October claimed the record as the worst day on the foreign ex-
changes since June 1972 . The pound plunged 5 cents in the day to
$1.595 , with dealers blaming the fall on a report in the Sunday Times
that an agreement had been reached to let sterling fall to $1.50 as a
condition for an I.M.F. stand-by credit. The depreciation of the effect-
ive exchange rate reached 47 .6 per cent. The year ended with a smaller
current-account deficit than in 1975 but a much larger balance for
official financing because of the heavy outflow of short-term capital .
In 1977 pressure in the exchange market was reversed and the
Bank of England had difficulty in preventing the exchange rate from
rising. It is estimated that, during 1977 , £10 ,000 million was sold to
buy up foreign currency to keep the pound from rising in value. By the
end of October the reserves had risen to $20,000 million , compared
with $4000 million at the end of 1976, giving the third largest reserves
in the world next to West Germany and Saudi Arabia . A wiser policy
would have been to let the exchange rate rise immediately. This would
have reduced the import bill, probably raised export receipts, and
contributed to the battle against inflation which might further have
improved the balance of payments. On 31 October the Bank of England
A History of the U.K. Balance of Payments 165
did end its policy of holding down the exchange rate for fear that the
excessive accumulation of reserves would jeopardise control of the
money supply . By the end of 1977 the pound had risen to over $1 .90
and the depreciation of the effective exchange rate stood at 35 per cent.
In December measures were taken to ease exchange controls over
direct investment into the European Economic Community ; and the
requirement applicable to all overseas portfolio investment that 25 per
cent of the sales proceeds must be surrendered at the official exchange
rate (the so-called 'surrender rule') was abolished.
In 1978 the current account moved into substantial surplus from a
slight deficit in 1977 . Throughout the year the American dollar came
under considerable pressure and the pound continued to rise against the
dollar , going above $2 on 20 October for the first time since March 1976 .
The effective rate of exchange against European currencies depreciated
slightly , however, ending the year at 62 per cent of its 1971 level. In
July there was the Bremen initiative to achieve greater stability of
European currencies, but the United Kingdom's endorsement of the
plans was lukewarm . There was a considerable increase in the outflow
of private investment during the year , and this was reflected in a loss of
foreign-exchange reserves despite the current-account surplus.
In 1979 the Conservative Party came to power and altered the course
of economic policy in a strongly monetarist direction . The exchange
rate continued to climb, and by the end of July had risen against the
$U .S. by 21 per cent in a year to $2.32, the highest level for four years .
The effective rate of exchange also appreciated, standing at 73 per cent
of its 1971 level. The rise was attributable partly to continued weakness
of the American dollar and partly to relatively high U.K. short-term
interest rates . The rise in the exchange rate was regarded with equa-
nimity by the government in the international monetarist belief that a
strong pound would mean a lower rate of domestic inflation . There
began in 1979 the final dismantling of all exchange controls after forty
years of restrictions . They were first eased in the Budget and then were
substantially relaxed from 19 July . Currency became available without
limit at the official exchange rate for all outward direct investment.
U.K. residents became free to invest at the official exchange rate in
most securities denominated and payable in the currencies of other
EEC countries. Similar relaxation applied to foreign currency securities
issued by international organisations of which the United Kingdom was
a member. The final remaining controls were removed from 25 October.
The level of capital outflows continued to increase in 1979 . Total U.K.
private investment overseas rose from £2.3 billion in 1977 to £6.5 billion
166 Balance-ofPayments Theory and U.K. Experience

in 1979. There was also a considerable inflow of both long- and short-
term capital, giving a net surplus on capital account of over £2 billion,
half of which was added to the reserves. The year 1979 also witnessed
the biggest rise in the price of oil since 1973, which helped to worsen
the visible trade deficit to over £3 billion - more than twice the visible
deficit for 1978. There was also a 12 per cent rise in import volume
compared with a much smaller rise in export volume of 4.5 per cent.
In 1980, amidst gold fever which pushed the price of gold close to
$1000 per ounce, the pound continued to appreciate on the foreign
exchange market. High interest rates in London and the strengthening
of the current account of the balance of payments were important
contributory factors. No doubt the United Kingdom's position as an oil
producer also played a psychological part in giving confidence to over-
seas investors in sterling . In strictly economic terms, however, there is
no reason why one part of the balance of payments should be singled
out as contributing to the strength of a currency. The large current-
account surplus of £3 billion came about largely through economic
recession at home and the compression of imports. If the full employ-
ment of resources had been maintained, the current account would
have been in substantial deficit , and then what would the price of
sterling have been, notwithstanding Britain being an oil producer? By
the end of October sterling had reached its highest level for seven years
of $2 .45. Despite the high exchange rate, export volume rose. The visible
account was in surplus for only the seventh year in recorded history.
Again there was a huge outflow of private investment overseas of close
to £7 billion, but this was more than matched by inflows of short- and
long-term capital, giving a net increase in the reserves during the year,
pushing them to a record level of over £ 13 billion. Superficially, the
balance of payments looked healthy, but not viewed against a back-
ground of deepening recession, growing import penetration, and an
underlying deterioration in the non-oil account of the balance of pay-
ments .

APPENDIX 7.1
AN ECONOMIC ANALYSIS OF THE UNITED KINGDOM'S PAYMENTS
DIFFICULTIES IN THE 1960s - THE PRELUDE TO DEVALUATION, 1967 8

The history of the UK. balance of payments on international account

since the Second World War makes dismal reading. There have been five
major foreign-exchange crises - in 1951,1955,1959, 1964 and 1967-
and two devaluations, in 1949 and 1967. In 1964 the deficit on current
A History of the U.K. Balance of Payments 167

and long-term capital account was the largest in recorded history . In

1965 and 1966 the payments position improved but in 1967 there was
a sudden deterioration and in November of the same year the pound
was devalued. In retrospect it is sometimes argued that the pound ought
to have been devalued in 1964 . Indeed, at the time of the 1964 crisis it-
self there was a small devaluation lobby, and rumour has it that the
Conservative Party had prepared a contingency plan for devaluation in
the event of re-election to office at the General Election in November.
As it happened the Conservative Party suffered defeat, the Labour
Party inherited a deficit of £744 million , and rejected devaluation as a
method of balance-of-payments adjustment.
Whether or not devaluation was appropriate in 1964, and whether
or not the Labour government was wise in rejecting devaluation as a
solution to the payments deficit, while resorting to it in 1967, requires
a thorough analysis of the origins of payments disequilibrium in these
two years, and a comparison of these years with previous years -
especially 1960. It will be suggested that the 'underlying' balance of
payments on trade and current account remained virtually unchanged
over the period 1960 to 1967, that the larger deficit on current account
in 1964 than in 1960 can largely be accounted for by the rate at which
demand expanded in 1964, and that in general the United Kingdom's
payments troubles have had little to do with growing price uncom-
petitiveness in world markets.
The U.K. payments position on visible account, current account and
long-term capital account is summarised in Table 7.7 for the period
1955 to 1967 . The official balance is defined as the balance on current
and long-term capital account ; it is these items within the accounting
balance which, together with the balancing item, give the balance of
monetary movements or accommodating transfers . The balance on
long-term capital account is normally excluded from econometric work
on the relation between the workings of the domestic economy and the
balance of payments on the grounds that it is largely independent of
such factors as the pressure of domestic demand and movements in the
terms of trade. The balance on long-term capital account clearly has a
bearing, however, on the balance of supply and demand in the foreign-
exchange market and could exert a decisive influence on what form a
balance-of-payments adjustment ultimately takes . Equally important
in this respect are autonomous short-term capital movements . These are
excluded from the official definition of balance-of-payments equili-
brium in many countries (including the United Kingdom)" but will
influence the extent of official action required to stabilise the exchange
rate in the event of disequilibrium between other payments.
168 Balance-of-Payments Theory and U.K. Experience

Whether the current balance , or the current plus the long-term

capital balance , is taken for analysis, it has been noticed that over the
years U.K. payments deficits in successive crises have worsened (see
Table 7 .7). In 1955, for example , the visible deficit was £313 million,
the current deficit £155 million and the official deficit £277 million . In
1960 the respective deficits were £406m ., £265m. and £457m . and in
1964 , £537m ., £381m . and £744m . Since the pressure of demand was
roughly the same in 1964 as in 1960, and was higher in 1955 than in
either of the other two years , the level of demand pressure alone cannot
be used as an explanation of the phenomenon df deteriorating deficits .
Three plausible alternative hypotheses may be put forward . One is
that the economy has been run at too high a pressure of demand in
relation to productive potential at certain periods, causing a rapid
upsurge in imports. The economy has been subsequently slowed down
to maintain foreign-exchange reserves, and to preserve the foreign-
exchange value of sterling, and this has impaired future competitiveness
by discouraging investment and raising unit costs of production in
periods of stagnation . A second hypothesis, related to the first, and
predicting the same result in the long run, is that there has been a
gradual erosion of competitiveness over time by defining 'full' employ-
ment too ambitiously, causing costs and prices to rise faster than in
other industrial countries. A third hypothesis is that the size of pay-
ments deficits is related to the speed of demand expansion from
troughs to peaks of the trade cycle, in addition to the level of demand
at the peak, and that in successive upswings in 1955, 1960 and 1964
the speed of expansion has been faster than in the previous upswing .
The argument is that the more rapid the expansion, the less time pro-
ducers have to react to changes in demand, and the greater will be the
extent of bottlenecks in the economy for imports to fill. Moreover,
when conditions return to normal, imports will not fall to their trend
level because during the boom, with imports high , manufacturers and
consumers are likely to acquire a taste for imports which they will
be reluctant to relinquish when home supplies become available again
(Brechling and Wolfe, 1965) .
Support for the first hypothesis would be given by upsurges in
imports in particular boom years together with a significant negative
trend factor relating the balance of trade (or balance of payments on
current account) to the pressure of domestic demand. The second
hypothesis could be supported by an appeal to the facts on the price
of our exports relative to those of our important customers and com-
petitors, coupled again with a significant negative trend term relating
the balance of trade to time . The third hypothesis would be supported
 TABLE 7.7
The U.K. balance ofpayments 1955-67 ([m .)

Visible balance* Current-account Balance on current

balancet and long-term
capital account t

1955 - 313 -155 - 277

1956 +53 +208 +21
1957 - 29 +233 + 127
1958 +29 +344 + 148
1959 - 117 + 143 - 112
1960 -406 -265 - 457
1961 - 152 -4 +64
1962 - 102 +112 + 14
1963 - 80 + 114 - ·35
1964 - 537 -381 - 744
1965 - 265 -50 - 252
1966 - 103 +64 - 48
1967 ·- 548 -283 -417
Note : The statistics here are not the same as those reported in the main body of the chapter because of revisions.
·Not including payments for U.S. military aircraft and excluding allowance for unrecorded exports. Imports and
exports are f.o .b.
....
0\ tIncluding payments for U.S. military aircraft and allowance for unrecorded exports.
'0
Source : United Kingdom Balance a/Paym ents 1969 (Pink Book) (London: H.M.S.a., 1970) .
170 Balance-of-Payments Theory and U.K. Experience
by finding that the speed of domestic expansion and the rate of ex -
pansion of imports had been greater in successive booms, that imports are
more sensitive to upswings than downswings, and that no significant
negative time trend exists relating the balance of payments to the
pressure of demand .
In recent years these hypotheses have been examined by a number
of investigators with varying conclusions. Brechling and Wolfe present
and substantiate the speed-of-expansion hypothesis . Oppenheimer
(1965) rejects Brechling and Wolfe's explanation and accuses them of
looking in the wrong place for an explanation of the United Kingdom 's
balance-of-payments weakness . Oppenheimer (1965 , p. 183) concludes:

the relationship between business cycles, inflation and the balance

of payments, orthodox or unorthodox , is of quite secondary im-
portance. When a country in Britain 's situation suffers a marked
deterioration in its trade balance not due to any increase in the
pressure of demand ... orthodox theory diagnoses uncompetitive-
ness. And uncompetitiveness is not a cyclical problem.

Smyth (1968) rebutts Oppenheimer. In testing the Brechling-Wolfe

hypothesis with respect to exports of manufactures , Smyth (1968,
p. 32) concludes that 'the assertion of Oppenheimer that the U.K.'s
poor export performance is due to lack of competitiveness and not ,
to any significant extent, to bottleneck influences , is seen to be invalid'.
Eltis (1967) , on the other hand , puts forward a simpler explanation.
He points to the fact that while the deficits and surpluses on current
account have become larger and smaller respectively with successive
booms and slumps , the current account does not show continuous
deterioration if net government invisible imports are excluded . There
was a steady growth of net government invisible imports from £55
million in 1953 to £472 million in 1966 . Paish (1968 , p. 12) also
comes to this conclusion, at least for the period up to 1961 , when he
says:
if we abstract from cyclical fluctuations by calculating the yearly
average for whole cycles we find that . . . the average balance of pay-
ments on current account deteriorated progressively from a surplus
of £119 million in 1952-57 to one of £61 million in 1958-61 and
to a deficit of £66 million in 1962-66 . The worsening between the
first and second cycles was due almost entirely to a greater rise in
public sector payments abroad on invisible account, but between
the second and third cycles the worsening of the trade balance
was greater than on invisible account.
A History of the U.K. Balance-of-Payments 171
While the observations of Eltis and Paish are correct, it is hardly
legitimate to isolate one single item from the balance of payments as
the villain of the piece; in fact government expenditure abroad has been
offset by increases in receipts from other invisible items and the in-
visible balance as a whole has shown a fair degree of stability over the
period under review.
To go over the ground again but in a slightly different way we shall
derive some very simple statistical relationships between two measures
of the pressure of domestic demand and the balance of trade (B T ) and
the balance on current account (Be), respectively, taking quarterly
data for the period 1958 to 1966. With the aid of the estimated
relationships we shall compare first 1960 and 1964, and then 1964 and
1967, together with the intervening years. The two indexes of demand
pressure taken are Paish's (1969) quarterly index of unutilised capacity
(Ue), and a quarterly index of the percentage level of unemployment
(U) . On the assumption that imports are increasingly responsive to
demand pressure, the greater the pressure of demand, non-linear
equations are fitted to the data of the form y = a + b I (X)-I, and
estimated by ordinary least-squares techniques. A time trend is then
added, giving equations of the form y = a + b , (X)-I + b 2 (t), where
t is time in quarters. We thus derive one set of four equations where
y is the balance of trade, and another set of four equations where y
is the balance of payments on current account. The estimated equations
are given in Table 7.8, together with other relevant information. With
thirty-six observations, all the equations are significant at the 95 per
cent confidence level and above. The implied 'trade-off" curves are
drawn in Figure 7.1 , using equations (7.1), (7.1a) , (7.2) and (7.2a).
From the equations, or the 'trade-off" curves, it can be seen what
one might have expected the balance of trade or balance of payments
on current account to have been in the various years given the pre-
vailing level of demand pressure. The difference between the actual
and predicted results for the years in question (the residuals) can then
be examined. While it seems that no more than 50 per cent of the
variance in the balance of payments is related to variations in the
measures of demand pressure, the regression coefficients in equations
(7.1), (7.2), (7.1a) and (7.2a) are highly significant and can therefore
be used with some confidence for the purpose at hand. The residuals
plotted on the basis of equat ions (7.1), (7.2), (7.1 a) and (7 .2a) are
highly cyclical. If there has been a significant trend factor at work
leading to a long-run deterioration in the 'full' employment balance
of payments, one should observe a systematic tendency over time for
the positive residuals to become smaller, and the negative residuals
172 Balance-of-Payments Theory and U.K. Experience

Deficits (£m)
-150
-140
-130
-120
-110
-100
-90
-80
- 70
-60
-50
-40
-30
-20
-10 v isible
balance
Unutilised 6.0 5.0 4.0 3.0 2.0 1.0 o 1.0 2.0 3.0
capac ity (%) / +10 \ Unemployment ('Yo)
current balance +20
+30 current balance

Surpluses (£m)

FIGURE 7.1
Trade-off curves between the balance ofpayments
and the pressure ofdemand

to become larger, at the same pressure of demand . But there is little

evidence of this. The peaks of the positive residuals all seem to be at
roughly the same level and the negative residuals show no consistent
downward trend .
When a time trend was added to the equations (see equations (7.3) ,
(7.4), (7.3a) and (7.4a) in Table 7.8), it emerged as statistically signifi-
cant in both sets of equations using unutilised capacity as the demand
variable, but insignificant in magnitude . To account for the larger
deficit in 1964 and 1960 it does seem we are left with the rate of
expansion hypothesis, but let us look at the two years in more detail .
A History of the U.K. Balance ofPayments 173

A Comparison of 1960 and 1964

In 1960 the actual deficit on the balance of trade was £406 million
and the deficit on current account was £265 million. In 1964 the visible
trade deficit was £537 million and the current-account deficit was £381
million . Since the balance on invisible account (which makes up the
difference between the balance of trade and the balance on current
account) was almost exactly the same in 1964 as 1960, a separate
analysis of the balance of trade and the current balance is not neces-
sary.!? We shall therefore concentrate simply on the balance of trade .
The question is, was the deterioration in the size of the trade deficit
of £131 million between 1960 and 1964 the result of a worsening
trend (due to growing price uncompetitiveness, for example), a greater
pressure of demand , the speed of expansion in 1964 compared with
1960, or other factors?
The pressure of demand in 1964, measured by unemployment and
unutilised capacity, was slightly less than in 1960. We would predict
from equations (7.1) and (7.2), therefore, on the basis of demand
pressure alone, a somewhat lower deficit in 1964 than in 1960.
Had the pressure of demand been the same in 1964 as in 1960 we
should have expected the deterioration in the deficit to be about £10
million greater than it actually was on the basis of the level of un-
employment, and £50 million greater than it was on the basis of the
level of unutilised capacity.
If we now turn to equations (7.3) and (7.4) , which include a time
trend, we find that the trend term has a negative sign in both cases and
is significant at the 95 per cent confidence level in equation (7.4). The
evidence suggests a gradual worsening of the visible trade balance over
the period 1958 to 1966, independent of changes in demand pressure,
of between £1 to £2 million per quarter. I I Supposing the trend term is
taken as -£1.5 million per quarter, this amounts to a deterioration
over four years of £24 million . Now if we take the deterioration in the
balance of trade at a constant level of demand to have been between
£140 million and £180 million (instead of the actual £131 million), and
allow for the trend factor, we are left with a deterioration of between
£116 million and £156 million to account for .
Due to a change in the regulations governing export registration,
exports have apparently been under-recorded since 1964. The balance-
of-payments Pink book gives £20 million as the value of unrecorded
exports in 1964. Making allowance for this we are left with a deteri-
oration of between £96 million and £136 still tobe 'explained'.
The noticeable difference between 1960 and 1964 does seem to be
.....
~

TABL E 7.8

The balance of trade and the pressure ofdemand (quarterly data 1958-66)

(7.1) B T = + 53.652 - 187.432(U;-' r 2 = 0 .297

(30.575) (50 .981) D.W = 0 .865

(7 .2) B T = + 5.957 - 171.895 tuci: r2 = 0.375

(15 .503) (39 .156) D.W = 1.673

(7.3)B T = + 49.025 - 160.400 (U;- 1 - 0.638 (t ) r 2 = 0.309

(31.43 2) (63.138) (0.867) D.W = 0 .877

(7.4) B T = + 25.373 - 156.148 tucr' - 1.433 (t) r 2 = 0 .464

(16.913) (37 .49 1) (0.631) D.W = 1.814
 The balance ofpayments on current account and the pressure ofdemand (quarterly data. 1958-66)

(7.1a) Be = + 128.261 - 230.750 (Ur t r2 = 0.321

(35.576) (59.319) D.W = 0.824
t
(7 .2a) Be = + 74.711 - 225.987 ( UCr r2 = 0.463
(17.020) (42 .987) D.W = 1.719

(7.3a) Be == + 121.933 - 193.783 tur: - 0.872 (t) r2 = 0.337

(36.451 ) (73.219) (1.006) D.W = 0 .842

(7 .4a) Be = + 98.793 - 206.456 (UCr - 1.777 (t) r2 = 0.560

(18.137) (40.204) (0.677) D.W = 1.995

Notes: standard errors are in parentheses ; D.Wis the Durbin-Watson statistic.

....
;:;l
176 Balance-ofPayments Theory and UK. Experience

the faster expansion of imports in the boom of 1964 and the slow-
down of the growth of exports compared with 1960 . These were
sudden changes which can hardly be related to unfavourable move-
ments in relative prices. Unless the elasticity of imports with respect to
income altered between 1960 and 1964, the greater growth of imports
in 1964 must have been related to the speed of expansion itself. The
absolute rise in the value of the physical increase in stocks and work in
progress was roughly the same in the two years. The percentage increase
in consumer expenditure , however , was 6.79 in 1964 compared with
4.98 in 1960, and the rise in gross national product 7.91 per cent and
6 .45 per cent respectively . Coupled with these figures the trade deficit
rose to £406 million in 1960 from £117 million in 1959, with export
growth at 6 per cent and import growth at 13.7 per cent. In 1964 the
deficit deteriorated to £537 million from £80 million in 1963 , with
export growth at 4.3 per cent and import growth at 15.4 per cent. It
can be seen from Table 7.9 that had imports grown no faster in 1964
than in 1960 , the import bill would have been some £75 million less,
and had exports maintained their trend rate of growth of 6 per cent
per annum , export earnings would have been £70 million higher than
they actually were. Differences in the rate of growth of imports and
exports in 1960 and 1964 , related to the speed of domestic expansion ,
seem sufficient to account for the deterioration in the balance of
trade between 1960 and 1964 which is 'unexplained' by the under-
recording of exports and an unfavourable trend .
When quarterly changes in the pressure of demand, as measured by
un utilised capacity, were explicitly considered as an additional explan -
atory variable in the regression analysis, the variable C:l. (UC)-1 emerged
as highly significant. 1 2 The two equations reported below are ex-
tensions of (7.4) and (7.4a) in Table 7.8. The first relates to the balance
of trade (B T ) , and the second to the current balance (Be> :

B T=32.4 - 196.4 (UC)-1 + 156.7 C:l.(UC)-1 -1.1 (t),,2 = 0.58

(16.2) (36.7) (51.8) (0 .6)
(7.5)

B e= 105.6 - 249.0 (UC)-1 + 167.2C:l.(UC)-1 -1.4(t),,2 = 0.65

(17.4) (39.7) (55.7) (0.7)
(7 .5a)
What is most interesting is that the residuals plotted from the above
equations hardly differ from zero in the second, third and fourth
quarters of 1964 . In the case of the trade balance the differences
 TABLE 7.9
The growth ofexports and imports

Exports Export Imports Import

and re-exports growth (f.o.b.) growth
(f.o.b.) (%) (£m .) (%)

1955 3073 10.34 3386 13 .28

1956 3377 9.89 3324 -1.83
1957 3509 3.90 3538 6.43
1958 3406 - 2.93 3377 -4.55
1959 3522 3.40 3639 7.75
1960 3732 5.96 4138 13 .71
1961 3891 4.26 4043 -2.34
1962 3993 2.62 4095 1.28
1963 4282 7.23 4362 6.52
1964 4466 4.29 5003 15.38
1965 4777 6.96 5042 0.17
1966 5108 6.92 5211 3.35
1967 5026 -· 1.60 5574 6.96

.... Source : United Kingdom Balance of Payments 1969 (Pink Book) (London : H.M.S.a., 1970) .
'I
'I
178 Balance-of-Payments Theory and U.K. Experience

between the actual and predicted values are +3 .3, -6.6 and -17.4,
and in the case of the current balance the equivalent values are -7.1,
-0.1 and -18.2. It appears that equations (7.5) and (7.5a) 'explain'
almost perfectly the U.K. balance-of-payments difficulties on trade
and current account in 1964.
The unimportance of the trend term and the overriding importance
of the upsurge in imports in accounting for the bigger deficit in 1964,
at roughly the same pressure of demand as in 1960, should not surprise
us. As far as competitiveness is concerned, the price of U.K. exports of
manufactures rose only 7 per cent between 1960 and 1964, compared
with 7 per cent for West Germany , 4 per cent for France, I per cent
for the United States and 3 per cent for all countries in the world. And ,
needless to say, declines in competitiveness cannot possibly be used
to explain deteriorations in the balance of trade in a single year of
approximately £450 million, as occurred between 1963 and 1964.
The difference in the size of the over-all deficit on current and long-
term capital account between 1960 and 1964 was much larger than the
£131 million difference on trade account. Since the invisible balance
was the same in 1964 as in 1960, the remaining difference must have
been due to a difference in the balance on long-term capital account. In
1960 there was a deficit balance of £192 million and in 1964 a deficit
balance of £363 million. The difference can be attributed partly to a
greater volume of private overseas investment abroad in 1964 and
partly to a lesser volume of foreign investment in the United Kingdom.
In answer to our original question , therefore, of whether the pound
should have been devalued in 1964, we reach the conclusion that the
United Kingdom's balance-of-payments trouble had little to do with
factors traditionally associated with the advocation of expenditure-
switching policies. We confirm Brechling and Wolfe's hypothesis that
the worse over-all deficit in 1964 compared with 1960 at the same
pressure of demand can be attributed almost exclusively to the speed
of expansion in 1964 and an unfavourable change in the balance of
long-term capital movements . Given the uncertainties of curing trade
deficits by devaluation, and the impact of devaluation on the internal
price level, it is to be hoped that no country is ever 'forced' to devalue
its currency through a change in the whim of private investors or
because the government of the day over-expands the economy through
incompetence or for electoral advantage. On the evidence the govern-
ment acted quite responsibly in attempting to control demand before
contemplating devaluation . But if the Labour government were not
unwise in rejecting devaluation in 1964, what was the justification in
A History of the U.K. Balance ofPayments 179

1967? What happened in the intervening years to make devaluation

inevitable?

The Balance of Payments, 1965-7

There was a marked improvement in the balance of payments in 1965

and 1966. The visible deficit fell from £537 million in 1964 to £265
million in 1965 and £103 million in 1966. 1 3 The current balance was
in deficit by only £50 million in 1965 and was in surplus in 1966 to the
extent of £64 million. The basic balance on current and long-term capital
account fell from a deficit of £744 million in 1964 to £252 million in
1965 and £48 million in 1966 . The improvement was the result of some
deflation of the internal economy, but for the years in question, and on
the basis of equations (7.1) and (7.2). the balance on visibleaccount was
much healthier than one would have predicted given the level of demand
pressure. The visible deficit in 1966 was £103 million and the pressure
of demand was higher as measured by the level of unemployment than
in 1959 and 1962, when the trade deficit was of roughly the same
magnitude . If we subtract from the deficit the under-recording of
exports, amounting to £60 million, the 'true' deficit was apparently
no more than £50 million.
The better trade figures (than one would have predicted on the basis
of demand) were partly due to the imposition of the temporary import
surcharge of 15 per cent on 26 October 1964 . This is certainly true of
the first quarter of 1965, when the difference between the actual trade
balance and that predicted on the basis of the level of unutilised cap-
acity amounted to +£75 million. The over-all effect of the import
surcharge from the fourth quarter of 1964 to the last quarter of 1965
is estimated to have been a saving of imports of between £100 and
£150 million, or 1 to 2 per cent of the import bill (see Johnston and
Henderson, 1967) . Import growth was very much slower in 1965, with
a change over 1964 of only 0.17 per cent, and in 1966 it was below its
trend rate of growth at 3.4 per cent. Exports, on the other hand, rose
above their trend rate in 1965 and 1966, at 6.96 per cent and 6.92 per
cent, respectively.
In early 1967 it seemed possible that the United Kingdom would
produce an external surplus large enough to repay the accumulation of
external debts. The pressure of demand was still falling, but unemploy-
ment was not excessive, and the first six months of 1967 were good for
sterling. The current balance was + £41 million in the first quarter and
-£87 million in the second quarter, and the over-all balance was + £51
million and -£105 million in the first and second quarters, respectively .
180 Balance-of-Payments Theory and U.K. Experience
The first quarter's trade balance and current balance were better than
might have been predicted given the pressure of demand, but in the
second quarter the situation undoubtedly worsened . The sterling value
of exports fell, the trade deficit rose to £115 million, and the current
deficit rose to £87 million . The setback was temporary, however, and in
the third quarter the situation improved again. The visible deficit fell to
£78 million and the current account was in surplus by £18 million. The
pressure of demand was still falling up to the third quarter, though it is
probable that the unemployment figures exaggerated the slackness of
demand. It was in 1967 that the relation between unemployment and
unfilled vacancies began to change, with both series rising together. The
percentage level of unemployment in relation to the demand for labour
as measured by unfilled vacancies was probably exaggerated by 0.3 per
cent compared with previous years. Taking a revised unemployment
figure for 1967, the difference in demand pressure between the first
three quarters of 1967 and the first three quarters of 1964 was insigni-
ficant. The actual trade deficit of £222 million in the first three
quarters of 1967 compares with £140 million for the equivalent
quarters of 1964 and £289 million for the same period in 1960, when
the level of demand was also roughly the same. Even if no allowance is
made for the changing structure of unemployment, there is no evidence
of a seriously deteriorating <Underlying' balance of trade between 1960
and 1967.
The figures for the third quarter of 1967 exceeded expectations in
many respects. The growth of world trade had already begun to slow
down, and exports were also beginning to be affected by the closure
of the Suez Canal following the Middle-East crisis in June . The average
monthly cost to export receipts of the Suez Canal closure is estimated
to have been about £20 million. But even in August of 1967 the
National Institute of Economic and Social Research were forecasting
in their Economic Review that the balance of payments - including
estimates for the effect of the war in the Middle-East - would show
in 1967 and 1968 a small surplus of the order of £25 to £50 million
(including the 'normal' positive balancing item of £50 million a year).
The Institute went on to say that 'this , however , conceals some under-
lying improvement', which is in line with the conclusions reached here.
The straw that broke the camel's back was unquestionably the
Liverpool and London dock strikes in September. It is true that the bad
second quarter trade figures and the events in the Middle-East had
caused nervousness in the foreign-exchange markets in July, and that
the possibility of devaluation had begun to be discussed in academic
circles and the press in the summer, but it was the dock strikes more
A History of the U'K. Balance of Payments 181
than anything else which made the balance of payments appear so
dangerously vulnerable to exogenous influences. Between the third
and fourth quarters of 1967 there was a rise in the visible trade deficit
from £78 million to £326 million, £130 million of which was due to a
fall in the sterling value of exports, mainly resulting from delays in
export shipments.
The decision to devalue was taken on 18 November. It was taken
against the background of a lack of confidence, not so much in the
United Kingdom 's ability to pay her way through an underlying wors-
ening in her trading position, but in her ability to cope with the sorts
of emergencies and contingencies such as were suffered in the bleak
months of July to September. It was a foreign-exchange crisis un-
related to any major crisis on trading account - at least compared with
1964 or 1960 . Had the dock strikes been averted the United Kingdom
may well have weathered the Suez Canal closure , and , in line with the
National Institute 's predictions, might have broken even in 1967 . As
it was, the United Kingdom was 'blown off course', and in the face
of pressure in the foreign-exchange market had little choice but to
alter the par value of the pound . However, the basic deficit against
which the pound was devalued was considerably less than in 1964 .
It was the 'market' balance not the 'accounting' balance of payments
which proved decisive.

APPENDIX 7.2
14
A POLEMIC ON FLOATING THE POUND, 1972

When the United Kingdom maintained a fixed rate of exchange for the
pound, a more flexible system allowing the exchange rate to fluctuate
more freely was, according to some, the only solution to the continual
problems of balance-of-payments disequilibrium and slow growth.
Greater exchange-rate flexibility, it was argued, would free the domes-
tic economy from the constraints imposed by the necessity to maintain
a balance on international trading account and would reduce the need
for international reserves and foreign borrowing to defend the foreign -
exchange value of sterling . A fluctuating pound would see the end of
'stop-go' policies and allow the economy to be run at any desired
pressure of demand.
In June 1972, faced with a massive speculative attack on sterling ,
the decision was taken to abandon pegged exchange rates in favour of a
system of managed floating - just one of many forms that greater
exchange-rate flexib ility may take. By the end of 1973 the pound had
182 Balance-of-Payments Theory and U.K. Experience

depreciated 20 per cent against a weighted average of other major

currencies . Meanwhile the economic situation deteriorated rapidly .
The years 1972 and 1973 witnessed unprecedented rates of inflation ,
and the 1973 balance-of-payments deficit was the largest in recorded
history . Since the rise in oil prices did not take effect until the very
end of the year, the origin of the difficulties clearly lay elsewhere. The
question may well be asked whether exchange depreciation itself con-
tributed to the growing economic difficulties, and whether exchange-
rate flexibility is beneficial if it leads to a currency depreciating so
much in such a short space of time.
The purpose here is to argue three points. First, the advocacy of
greater exchange-rate flexibility should not be extended with approval
to a system which permits such excessive depreciation so quickly .
Second, the drastic sterling depreciation that was allowed to occur
exacerbated the United Kingdom's economic difficulties. Third, while
greater exchange-rate flexibility may be desirable, exchange deprecia-
tion itself is largely irrelevant as a policy weapon for tackling the long-
run payments problem , as this has its origin not so much in growing
price uncompetitiveness as in a much higher U.K. income elasticity
of demand for imports than the 'world' income elasticity of demand for
U.K. exports. For this type of problem, common to less-developed
countries, exchange depreciation can at best provide a temporary
palliative. Indeed , it may make the ultimate cure more difficult .

The Casefor Exchange-rate Flexibility

The general case for exchange-rate flexibility is well known and need
be touched on only briefly here . Exchange-rate flexibility gives a
central bank greater room for manoeuvre and avoids the necessity for
a country to hold large reserves or to borrow abroad to maintain the
exchange value of its currency if there is speculation against it. Indeed,
the theoretical expectation is that, in a system in which the exchange
rate is free to vary, speculation will on balance stabilise the currency.
as any initial depreciation will induce the private speculator to move
funds from the appreciating to the depreciating currency. This mech-
anism should serve to limit speculation against a currency, to arrest the
depreciation, and to finance the deficit on the balance of payments
caused by the relative price changes, until the quantities of imports and
exports have had time to adjust to the change in the exchange rate . The
more speculators are willing to support a currency which has depreci-
ated , the more reserves are saved.
For speculation to be stabilising, there must be a consensus in the
A History of the U.K. Balance of Payments 183
market about what a particular exchange rate should be and whether
the existing rate is 'too high' or 'too low'. Since , however, this con-
dition may not be fulfilled, as it is often difficult to know what the
correct rate is, floating rates may lead to a great deal of instabilit y. The
instability in currency markets in recent months illustrates the point.
Because of uncertainty exchange rates have tended to move hectically
and erratically from one over-reaction to the next in a generally de-
stabilising way. As far as sterling is concerned , there is still a great deal
of uncertainty as to what the correct rate is, due in no small measure
to the failure of the balance of payments to respond to the initial
exchange depreciation in the way expected from economic theory.
A second main argument for exchange-rate flexibility is that it
would avoid the necessity of domestic demand deflation - and unem -
ployment - to correct a balance-of-payments deficit. To be sure, if
exchange depreciation occurs in conditions of full employment, demand
contraction is required in order to avoid excess demand. However ,
provided resources can be switched easily and quickly from the domestic
to the export sector, there is no reason why this transition should lead to
higher unemployment. Flexible exchange rates may therefore be a
device for achieving simultaneous internal and external equilibrium if
combined with the appropriate domestic policy . In July 1972 The
Economist went so far as to argue that depreciation of the pound (plus
wage restraint) would mean that ' Britain would probably break through
to something like an economic miracle '.

Problems Stemming from Exchange-rate Flexibility

Managed floating , which the United Kingdom adopted in June 1972.

is only one of many forms of exchange-rate flexibility . There was
nothing in the discarded parity system to prevent exchange rates from
being adjusted . Many would still hold the view that the old adjustable
peg system would have been perfectly workable if only governments
had been willing to make use of the provisions for adjustment. Other
forms of exchange-rate flexibility widely canvassed consist of fixed
parities with wider bands , sliding parities with and without wide bands,
and completely free floating. Although greater exchange-rate flexibility
may be desirable, it does not follow that a system of freely floating
exchange rates is best. There are many considerations, which the
proponents of floating exchange rates tend to ignore , which make one
sceptical of floating as a means of achieving simultaneous internal
and external equilibrium.
The first major consideration frequently ignored is the adverse
184 Balance-of-Payments Theory and U.K. Experience

repercussions that exchange-rate depreciation has initially on the

balance of payments . These may be so severe and so prolonged as to
cause destabilising speculation . A second is the risk that the impact
of exchange depreciation on the domestic price of imports may severely
affect, both directly and indirectly, the rate of domestic inflation in an
open economy. In a popular article Professor H. G. Johnson once
argued that 'the basic argument for floating exchange rates is so simple
that most people have considerable difficulty in understanding it ' , and
because a floating exchange rate keeps the foreign-exchange market
cleared 'a floating rate would save a country from having to reverse its
full-employment policies because they lead to inflation and deficit'
(The Times, 9 December 1968). But there is no mention that exchange
depreciation may generate further domestic inflation and widen the
deficit. Full employment is not the only goal of economic policy and
exchange depreciation will not necessarily remedy a balance-of-pay-
ments deficit.

Nature of the Balance ofPayments

In discussing the effects of floating exchange rates, it is of crucial

importance to draw a distinction between balance in the foreign-
exchange market (the 'market' balance of payments) and the balance
of payments as traditionally defined (i.e. the balance of expenditures
and receipts on current and long-term capital account , or what is some-
times called the 'basic' balance of payments). Since an exchange rate is
the price of one currency relative to another, it is obvious that a freely
floating exchange rate will give balance in the foreign-exchange market,
in the sense that, assuming no market intervention by central banks , the
amounts bought and sold by other operators in the market must be the
same . But the price of a currency which results in a balance in the
foreign-exchange market does not guarantee a balance between ex-
penditure and receipts on the current and long-term capital account of
the balance of payments . A fall in the exchange rate will produce a
market balance and basic balance only if the resulting change in
relative prices between domestic and foreign goods is not offset by
domestic price changes, and if export and import volumes respond
by the correct amounts. If the volumes of exports and imports are
reasonably responsive to changes in relative prices, then according to
economic theory there will be an exchange rate which gives balance in
both senses. However, the time lag between the exchange rate initially
falling and the rate reaching a level which guarantees balance in both
senses may be long; and what may happen to the domestic economy in
A History of the U.K. Balance of Payments 185

the meantime may be extremely serious. Indeed , most of the economic

theory of exchange -rate movements has been concerned with com-
paring two different equilibrium situations. It does not predict the path
between these equilibria. Other dynamic theories in which the economy
would never reach the second equilibrium situation because the re-
sponse of the economy to the initial change in the exchange rate is
perverse could be formulated . For example , one could postulate a
scenario in which an initial depreciation of an exchange rate led to
increased domestic inflation , which in turn reduced domestic confi-
dence , which led to a fall in investment , and so on.

The U.K. Situation

The U.K. experience over the last two years makes a cautionary tale .
A weak balance of payments, in the sense of severe speculation against
the currency, caused the exchange rate to fall. Because imports and
exports were relatively price inelastic in the short run , the balance of
payments deteriorated, speculators were uncertain, and the exchange
rate continued to fall. Sterling import prices rose , pushing up home
costs and prices, and this stimulated the demand for higher wages.
The competitive gains from currency depreciation were eroded and a
spiral of domestic inflation and currency depreciation was set in
motion, forcing the authorities to intervene, necessitating the use of
reserves and foreign borrowing as under the old system of fixed ex-
change rates .
Let us consider in more detail the relationship between the depreci-
ation of the pound and the United Kingdom's worsening economic
difficulties up to the end of 1973. The severe depreciation of the pound
may be said to have aggravated the United Kingdom 's domestic econ-
omic difficulties in three main ways. First , it exacerbated the rate of
domestic. inflation. Second, it diverted attention from the essential
need to maintain the domestic economy in balance. Third, it caused
such deterioration in the balance of payments as to necessitate the use
of restrictive demand-management policies at a time when the economy
was slowing down. It is surely ironical that an exchange -rate policy
designed to free the economy from the constraint of the balance of
payments should have been followed eighteen months later, when there
were still unemployed resources in the economy, by the biggest ever
reduction in govemmen t expenditure, namely the cuts of £1200
million announced in the mini-budget of December 1973. Had the
186 Balance-of-Payments Theory and U.K. Experience

exchange rate been given more support to maintain some par value, we
should almost certainly have been forced to restrain demand earlier but
less drastically. This seems to be yet another example of an inability
to conduct economic affa irs with gradualism and moderation.
The sinking pound , by raising the sterling price of imports by
more than would otherwise have been the case, exacerbated inflation
directly and indirectly . The direct effect arises from the fact that over
20 per cent of total expenditure in the United Kingdom is on imports,
so that for every 10 per cent rise in the sterling price of imports the
price index may be expected to rise by at least 2 per cent. During 1973
the retail price index rose by approximately 10 per cent. Since the
import price index rose by 27 per cent, of which 20 per cent was the
direct result of sterling depreciation , we can say that the total con -
tribution of import price increases to the domestic inflation was 5.4 per
cent , of which 4 per cent can be attributed to the depreciation of
sterling . Indirectly , exchange depreciation influences domestic inflation
by its effect on the price of finished products, especially food , and
hence on wage claims. Wage claims are extremely sensitive to the price
of food, and the United Kingdom imports roughly 50 per cent of her
food requirements. A fall in the exchange rate of 20 per cent, there -
fore , would be likely to raise the food price index by at least 10 per
cent. In fact food prices in the United Kingdom in 1973 rose on average
by 20 per cent , so that roughly half of that rise was the result of the
effective depreciation of sterling , and was within the control of U.K.
economic policy; the remaining half was the result of a rise in world
food prices outside of U.K. control. Some may concur with Professor
Viner (1964), who once remarked that 'the fixed exchange rate -
cult , myth, rigidity, illogicality though it may be - is in many countries
the sole surviving barrier to almost unrestrained inflation' . Even
Professor Friedman (1953) , a leading advocate of floating exchange
rates , has conceded that the inflationary repercussions of exchange
depreciation may be an objection to floating exchange rates in a par-
ticular country at a particular time . Could that country and time have
been the United Kingdom in 1972, when world commodity prices were
already rising and domestic inflation accelerating?
A second way in which the floating exchange rate aggravated the
United Kingdom's difficulties was by relieving pressure on the foreign-
exchange market, and thus giving a false sense of security with regard
to the state of the economy. Whereas under a system of pegged ex-
change rates the necessity to control the domestic economy would have
A History of the U.K. Balance of Payments 187

been apparent from the necessity to support the par value of sterling ,
with the pound floating it took an oil crisis, bottlenecks in particular
sectors of the economy, inflation of over 10 per cent per annum, and
a balance-of-payments deficit running at an annual rate of over £1000
million , for action to be taken to control the rate of expansion of the
economy . It has been a perpetual failing of post-war governments to
expand the economy too fast in the belief that an expansion of demand
by itself can raise the long-run rate of growth of real supply . It
happened in 1964 , under fixed exchange rates, and it happened again
under floating exchange rates, which the government of the day appar-
ently believed would allow the country to expand while the balance
of payments looked after itself.
As noted already , the balance of payments did not look after itself;
it worsened dramatically, and this further aggravated the United
Kingdom's economic difficulties . The increase in sterling payments
for imports as a result of depreciation seems to have far exceeded the
increase in sterling receipts from exports. From a surplus of £1061
million in 1971, the current account moved to a deficit in 1973 to the
tune of £1470 million - a turn-round of approximately £2500 million.
The possibility of a sharp deterioration in the current balance is of
course recognised by the advocates of floating exchange rates, but the
deterioration is supposed to be only short-lived until the quantities
of imports and exports have had time to adjust to relative price
changes . How long the deterioration lasts will, however, depend on
many factors which need careful consideration before a policy of
exchange-rate depreciation is embarked upon . It depends , first, on the
domestic demand for imports and the foreign demand for exports
being sensitive to relative prices. Second, there must be the possibility
of increasing the supply of exports. Third, there must be some means of
managing the exchange market pending the response of imports and
exports. If speculators are not willing to finance the initial deficits
that arise from exchange depreciation, the consequences of floating,
as far as foreign indebtedness and reserve losses are concerned , can be
as severe for a country as the consequences of having to defend the par
value of a currency. In 1972 the United Kingdom suffered a reserve
loss, and had to undertake net official borrowing of over £1000 million
to fmance a deficit on the current and long-term capital accounts
(plus a large negative balancing item). In 1973, despite the size of the
current-account deficit, the situation was eased by large inflows of
long-term private investment and by public-sector borrowing on the
188 Balance-of-Payments Theory and U.K. Experience

Euro-dollar market. There was no official borrowing in 1973 and no

net loss of reserves. Policy for 1974, when the current-account deficit
is predicted to be double that of 1973, largely as a result of the rise in
oil prices, is not yet clear. We know already, however, from the budget
speeches , that foreign borrowing commitments of nearly $4 .0 billion
have already been entered into, to avoid, in Mr Healey's words, 'massive
deflation' .
Over all, there can be no escape from the conclusion that floating
the pound has been debilitating for the balance of payments, and for
the economy at large. It is certainly questionable whether the conse-
quences of maintaining fixed parities would have been any worse than
the consequences of floating . A different mode of exchange-rate flexi-
bility, or a 'dirtier' float, would almost certainly have been wiser.

Past Policy Towards the Exchange Rate

During the period of the last but one Labour government many econo-
rnists.l" including myself, defended, on a variety of grounds, the
government's attempt to preserve the par value of sterling between
1964 and 1967. Sir Roy Harrod (l967a) was strongly opposed to the
1967 devaluation and has described devaluation as 'the most potent
known instrument of domestic price inflation which has such sorry
effects on human misery' . Professor Ball (1967) also expressed doubts
about the wisdom of devaluation and, in particular, scepticism over the
role of prices in determining export performance :

I believe that the non-price factors and the approach to marketing

as a whole, together with the capacity problem, lie at the heart of
the long-term trend, and that the problem of price is by no means
the most significant factor that enters into the balance-of-payments
equation . . .. Even if the once-for-all effect of a single change did
provide a surplus on current account rapidly and of the right order
of magnitude, we have done nothing to ensure that when long-run
growth and expansion is resumed our position will not be further
eroded. The result could be the need for yet another devaluation.

My own defence (Thirlwall, 1970), partly based on econometric evi-

dence, was that the 1964 balance-of-payments deficit had little to do
with the price uncompetitiveness of U.K. goods, and that the conse-
quences of exchange-rate depreciation at that time would have been
A History of the U.K. Balance ofPayments 189

worse than the consequences of maintaining exchange stability .

Some of us may feel we have been vindicated by subsequent events .
The domestic cost inflation of the period 1967-74 almost certainly
stems from the devaluation of the pound in 1967; and the subsequent
floating of the pound merely perpetuated cost inflation without notice-
ably aiding the underlying balance of payments.
The circumstances in which the decision to float the pound was
taken in 1972 bear an uncanny resemblance to the events preceding
the devaluation in 1967. In 1967 the straw that broke the camel's
back was unquestionably the Liverpool and London dock strikes in
September. Speculation fmally breached sterling's defences when a
Labour back-bencher, Mr Robert Sheldon, asked a question in the
House of Commons about the size of a loan agreement negotiated with
foreign banks . At the time devaluation took place, however, the trade
deficit was improving. The recorded deficit on current account was
only £330 million in 1967 - within the margin of accounting error.
Indeed, there was an error. In 1969 it was discovered by Board of Trade
statisticians that because of a change in regulations governing export
registration exports had been systematically under-recorded since 1964
and that there was no over-all deficit on current account between
1964 and 1967. Incidentally this is not the first time that concern with
the balance of payments has been based on faulty statistics. When the
pound was devalued in 1949 it was thought that the aggregate balance-
of-payments deficit for the years 1946-8 was more than £1000 million .
We now know from revised figures that it was only £737 million and
that there had in fact been a slight surplus in 1948. Would the pound
have been devalued by so much in 1949 had the correct figures been
known? Would devaluation have been forced upon us in 1967 if the
figures had been correct and the dockers had not gone on strike?
It was the prospect of a national dock strike, arising this time from
a complex dispute over containerisation, which again prompted specu-
lation against sterling in 1972. Further selling of sterling was precipi-
tated by Mr Healey, then Shadow Chancellor, who announced in the
House of Commons that he expected the pound to be devalued in July
or August. In the three days prior to the pound being floated, over
£1000 million of support for sterling was required . The balance of
payments was still in surplus, however, and the reserves stood at
$7100 million. Mr Barber, the Chancellor of the Exchequer, remarked
in his budget speech that 'there was nothing in the objective facts of
our balance-of-payments position or the level of reserves to justify
190 Balance-of-Payments Theory and U.K. Experience

these [speculative] movements. The trade balance has deteriorated

in recent months but this was from an exceptionally and unusually
favourable position .' The main mot ive for abandoning the pegged rate
was clearly to avoid borrowing to support it. The question remains ,
however , whether it was necessary to allow the pound to depreciate
by 20 per cent within a year, with the attendant inflationary reper-
cussions - or could the same end have been achieved with a more
orderly system of automatic exchange -rate flexibility which would have
contained the amount of depreciation within anyone year - as pro-
vided , for example, by a system of sliding parities?

Disappointing Results of Floating

The floating pound has not lived up to its expectations - at least to

the expectation that it would give the government more freedom to
pursue domestic goals while the balance of payments would look after
itself. It could be argued, of course, that the floating pound has not
been allowed to work , because the float was 'dirty', and because the
economy was so over-extended that the resources were not available
to satisfy export markets . The former argument is not convincing ,
since free floating would have led to even greater depreciation of the
exchange rate . If the latter argument is accepted , the point is conceded
that floating exchange rates by themselves are an inadequate weapon
of economic policy , for what would seem to be implied is that if a
country wishes to expand total demand faster than the rate of growth
of productive potential, it will suffer an ever-deteriorating exchange
rate , or at least deterioration for a long period of time , the conse-
quences of which for the domestic economy could be very serious.
On the other hand, if a country does not wish to expand demand in
excess of productive potential , there is little need for a floating ex-
change rate, and the uncertainties it breeds , when a more orderly system
of exchange-rate flexibility could be adopted. The same argument can
be applied to floating exchange rates as Professor Paish (I969) used to
apply to incomes policy , namely that in conditions of domestic stabil-
ity they are not necessary and in conditions of excess demand they are
unworkable (or at least lead to consequences worse than the disease to
be cured) . A counter-argument would be that floating exchange rates
are necessary to accommodate cost inflation. But this argument begs
two questions: Is it desirable for countries to pursue policies which
allow for cost inflation and which make cost inflation a permanent
A History of the U.K. Balance ofPayments 191

feature of economic life with no apparent benefit to the community

at large? And is it the case that cost inflation has been in the past, and
is likely to be in the future, a major source of balance-of-payments
disequilibrium? The evidence up to 1967 , when exchange-rate adjust-
ment itself initiated a serious cost inflation, is that there was very little
deterioration in the balance of payments attributable to growing price
uncompetitiveness; and almost the lowest rate of increase in labour
costs per unit of output in the Western world in 1973 did not prevent
the United Kingdom from experiencing its largest deficit in recorded
history.

Causes of Poor U'K. Export Performance

There were two major sources of the United Kingdom's payments

troubles over the twenty-five years up to 1974: the periodic attempts
to expand demand faster than the rate of growth of productive poten-
tial ; and the long-term tendency for imports to grow faster than exports
because the U.K. income elasticity of demand for imports exceeded the
'world' income elasticity of demand for U.K. exports. Floating the
exchange rate was not the appropriate solution to either problem. The
solution to the first (self-imposed) source of trouble is to keep domestic
demand in line with productive potential and to desist from rushes for
growth which in the past have involved expansions of monetary
demand of up to 20 per cent in one year. The remedy for the second
source of trouble is more complex, but since the origin of the problem
lies not in price differences between countries but in the characteristics
of the goods produced and sold , it would be surprising if the solution
lay in exchange depreciation . On the contrary, there are good grounds
for supposing that exchange depreciation may aggravate such a struc-
tural problem by ossifying the industrial structure which is the source
of the trouble.
From an important and authoritative study by Houthakker and
Magee (1969), which allows for the effects of relative price changes, it
appears that the 'world' income elasticity of demand for U.K. exports
is only 0.9 compared with a U.K. income elasticity of demand for
imports of 1.7. This means that for a growth rate of world income of,
say,S per cent , our exports may be expected to grow at 5.0 x 0 .9 = 4.5
per cent per annum, while the full-employment rate of growth of
imports would be 3.0 x 1.7 = 5.1 per cent per annum (assuming the
U.K. rate of growth of capacity to be 3 .0 per cent per annum) . The fact
192 Balance-of-Payments Theory and UiK, Experience

that the United Kingdom 's full-employment rate is less than that of
other countries already pro vides a partial adjustment mechanism , but
it is clear from the figures that if the rate of growth of imports is to
be kept in line with the rate of growth of exports , the U.K. growth rate
would have to be slowed to 2.6 per cent. 1 6 But the United Kingdom 's
major problem does not appear to be on the side of imports. The
import elasticity is not so different from that of other countries. It is the
export elasticity which is low compared with other countries. The U.K.
export elasticity of 0.9 compares , for example, with 3.5 for Japan ,
3 .0 for Italy , and 2.4 for West Germany . Only the United States, with
an elasticity of 1.0, appears to be in a comparatively weak position.
The basic payments problem therefore lies in the character of the
goods produced and the markets in which they are sold . A favourable
change in relative prices only makes the United Kingdom more com -
petitive in the wrong range of goods . This type of problem is not
untypical of the developing countries which produce and export mainly
primary commodities the income elasticity of demand for which is low ,
and which import goods the income elasticity of demand for which
is high . These countries have recognised that the correct approach to
their difficulties is through stru ctural change , and in particular by the
encouragement , through protection and subsidisation , of the production
of goods with high income elasticities of demand in world markets. A
change in relative prices, if successful, can only affect the level of
exports and imports; it will not necessarily raise the rate of growth
of exports and lower the rate of growth of imports permanently, unless
of course the change in relative prices is continuous, i.e. unless there is
continual exchange depreciation . This is not feasible or desirable , for
the reasons already discussed.
The major task of economic policy , if the United Kingdom is to
tackle the balance-of-payments problem at its root, is not to look to
exchange depreciation as the panacea , but to take measures at the
micro level to encourage the production and export of goods which
have a high income elasticity of demand in world markets (and which
at the same time reduce the sensitivity of imports to increases in
domestic income). Exchange depreciation merely gives indiscriminate
benefit to all exporters, whereas the aim must be to boost exports in
a discriminating way by the encouragement of certain types of activity.
If it is accepted that the long-run average annual rate of growth of
imports in relation to full-employment income is just over 5 per cent,
and the 'world ' growth rate is approximately 5 per cent , the elasticity
A History of the U'K, Balance of Payments 193

of demand for U.K. exports needs to be increased only from 0 .9 to

just over 1.0 for balance-of-payments equilibrium to be achieved. One
solution may lie in Kaldor's suggestion (New Statesman, 1 March 1974)
of a general subsidy to manufacturing industry which would shift the
balance of resources towards the manufacturing sector where the
income elasticity of demand for goods may be expected to be higher.
A subsidy would also be advantageous if it encouraged a revaluation
of the pound . This would contribute to price stability at home ; and
sterling payments for imports would fall .

Need for Flexibility with Safeguards

Nothing that has been said above detracts from the case for a more
automatic system of exchange-rate flexibility . The argument has been
concerned with the wisdom of allowing the pound to depreciate so
markedly in such a short space of time , and with questioning the
relevance of exchange-rate depreciation as a solution to the United
Kingdom's underlying payments problem . There is still the need for a
flexible system of exchange-rate adjustment, with safeguards against
speculation. What is required is a system which strikes a balance between
minimising the degree of anti-social speculation and protecting the
economy against imported inflation . A system of sliding parities, with
bands around the parities, might achieve this. The flexibility of the
slide and the band would remove the one -way option for speculators,
and if the slide of the parity were confined to, say, 2 per cent per
annum, exchange depreciation would be limited. But the causes of the
underlying payments disequilibrium would still need to be tackled if
the currency were not to depreciate continually.
Whatever long-term policies are pursued , an essential precondition
for internal and external stability is that the exchange rate be stabilised,
and preferably upvalued. Furthermore , let us banish the idea once and
for all that floating exchange rates are a painless recipe for the achieve-
ment of simultaneous internal and external equilibrium . The only
equilibrium that floating exchange rates guarantees is in the foreign-
exchange market . But equilibrium in the foreign-exchange market is not
only consistent with, but may also lead to , severe disequilibrium in the
internal economy and in the basic balance of payments.
8
Import Functions
THE DETERMINANTS OF IMPORTS

Econom ic theory would suggest three major factors determining a

country's demand for imports. First , the capacity of the country to
produce and supply the goods itself. Some imports are not competitive
with domestic goods because the country does not have the physical
capacity to produce them ; others will be competitive , and the demand
for them will partly depend on the ability of domestic producers to
supply the substitutes. Second , the price of imports relative to the
price of domestic substitutes will affect import demand . Third, the level
of expenditure will affect the demand for imports. The composition of
expenditure will also be important to the extent that the import con-
tent of different components of expenditure differs .' For simplicity,
however, and for the purposes of later chapters, it is assumed here that
the import content of different items of expenditure is the same, so
that income can be used as a proxy for expenditure in the import
demand function. We stress again, however, that for income-determin-
ation analysis it is very important to recognise that there may be
different import coefficients attached to different components of
expenditure, and to relate imports to expenditure in deriving the
foreign-trade multiplier (see Chapter 2) . In addition to these three main
factors, stock-building will also affect the demand for imports, as well
as a host of non-quantifiable factors . If the non-quantifiable factors are
correlated in any way with the quantifiable factors , they will affect the
parameter estimates of the estimated import demand function . Chang-
ing tastes, and factors associated with tastes, as income changes, for
example, may influence considerably the relationship between import
demand and income .
Import Functions 195

The form of the import demand function may be assumed to be

multiplicative (linear in the logarithms), implying a constant elasticity
of demand for imports with respect to the independent variables
specified in the function . Thus let

(8.1)

where M is the quantity of imports, C is a measure of capacity utilis-

ation in the importing country, Pt/Pd is the price of imports relative
to the price of domestic goods measured in a common currency, Y is
real income, A is the elasticity of demand for imports with respect to
capacity, tJJ is the price elasticity of demand for imports, and 11' is the
income elasticity of demand for imports .
If the function is not homogenous of degree zero in money income
and prices (i.e . there is money illusion), the import function is

M t = aCtA Prf Pd~ Y~t (8 .2)

where Y m is money income, and it is also recognised that the own price
elasticity of demand for imports may differ from the cross elasticity
of demand , i.e. tJJ *- a.
It has also been suggested (Leamer and Stern, 1970) that if the
import price deflator is unreliable, it may be better to specify imports
in money terms. There is no one correct a priori specification of the
import demand function . Investigators must proceed on a trial-and-
error basis according to the data available , their reliability and the
purpose of the study.
In some import studies investigators use a stock-adjustment model
on the hypothesis that it takes time for importers to adjust imports to
the desired level. From this model short- and long-run elasticities of
demand for imports can then be derived. Consider the following model,
which is a logarithmic transformation of equation (8.1):

t
10gM*t = loga + A10g(Ct ) + tJJ log (P ) + 11' log (Yd
Pd t (8.3)

where M* is the desired level of imports. Now assume that it takes time
for imports to adjust to the desired level, an equilibrium having been
196 Balance-ofPayments Theory and U'K . Experience

disturbed . The change in import demand between two periods will

equal some fraction of the difference between the desired level and the
level of imports in the previous period.

10gMr-logMr _ 1 = 8(1ogM*r-logMr_I)' where 0 < 8 < I

(8.4)

or 10gMr = 8 10gM*r + (I - 8) 10gMr_ 1 (8.5)

Substituting equation (8.3) into (8.5) gives:

10gMr = 8 loga + 8 A10g(Cr) + 81/J log (PI)

P t
+ 81Tlog(Yr)
d
+(1 -8)logMr_ 1 (8.6)

The short-run elasticities are 8A, 81/J and 81T . The long-run elasticities,
derived by setting log M, = log M, - I (which is the long-run equilibrium
condition) , are:

8A 81/J 81T
--- = A = 1/J and = 1T
1-(1 -8) , I - (I - 8) , I -(I -8)

If 0 < 8 < I , the short-run elasticities will be lower than the long-run
elasticities. If adjustment to the desired level of imports is instant-
aneous, then of course 8 = I, and there is no difference between the
short- and long-run elasticities.'

IMPORTS AND CAPACITY

Strictly speaking, a capacity or supply variable belongs to the import

demand function , in addition to a relative price variable, only if excess
demand at home is not eliminated by a change in the domestic price
level. Prices may not change for a number of reasons: first, the demand
curve for domestic goods may be ' very elastic, so that a shortage of
supply does not lead to a rise in price; second, for a variety of reasons,
manufacturers may resist raising prices; and third, the supply of imports
may be so elastic that no price change is observed for domestic goods.
The above points can be illustrated diagrammatically. Assume import
Import Functions 197

demand to be the difference between domestic demand and supply at

the ruling market price (see also Chapter I , p.24) . Suppose in Figure
8.1 that there is an equilibrium at price Pd , and then the supply curve
shifts to SISI, creating excess demand MIM. If demand is less than
perfectly elastic with respect to price , the domestic price in a free
market should rise from Pd to Pd l , and there should be no change in
import demand to explain in terms of supply deficiency , only in terms
of a change in the relative price of domestic and imported goods. If
prices do not change , however, the excess domestic demand will be
filled by imports, which will have to be explained by some pressure of
demand or capacity-utilisation variable . The explicit inclusion of a
capacity variable as an independent argument in the import equation
is to recognise that price may not act as a balance between supply and
demand in the domestic market.
It is important to separate out the two influences of demand and
supply on imports because if the latter exist , a given increase in income
or expenditure will produce a larger increase in imports, the shorter
the period over which the increase occurs. Import equations which
ignore the supply or capacity effects will then tend to give exaggerated
estimates of the income or expenditure elasticity of demand for im-
ports. In some studies discussed below this point is met by distinguish-
ing between the secular and the cyclical income elasticity of demand
for imports, with the expectation that if supply effects are important

Domestic
pr ice
o

s
M, M Domestic supply
and demand
FIGURE 8.1
198 Balance-ofPayments Theory and U.K. Experience

the latter elasticity will be higher than the former . In functions fitted
by Rees and Layard (1972) for the United Kingdom a distinct capacity
variable is introduced and is measured by the proportion of manu-
facturers reported to be working at capacity. The variable is found to
be highly significant for semi-finished capital goods and for consumer
goods but not important in explaining imports of basic materials.
The estimated capacity elasticities, holding expenditure constant, are
2 .1 for semi-manufactured capital goods, 2 .6 for finished capital goods,
and 2.0 for consumer goods .
If a measure of capacity utilisation is not taken , capacity and income
effects may be distinguished by making a distinction between the
cyclical and trend income elasticities of demand for imports. These
elasticities can be measured in the following way . Let

10gMt = s, + B 1 log (I))

Pd t
+ B 2 log Yt (8 .7)

where M* and y* are the trend levels of M and Y, respectively . And let

(IogM t -logM~) = 'Y (log Yt -log Yj) (8 .8)

where log M, - log Mf and log Yt - log Yt represent proportionate

deviations of imports and income , respectively , from their trend levels.
Substituting (8 .7) into (8 .8) gives:

10gM t = e, + B 1 log (PI) + 'Y log Y+ (Bt 2 - 'Y) log Y~

Pd t (8 .9)
where 'Y is a measure of the cyclical elasticity and (B 2 - 'Y) is a measure
of the trend (or secular) elasticity. When Khan and Ross (1975)
followed this procedure for a number of countries they obtained estirn-
ates for the U.K. economy of 0.936 for the cyclical elasticity and 0 .889
for the trend elasticity. Marston (1971) , in a similar but disaggregated
study of U.K. imports finds a cyclical (capacity effect) elasticity of
3.63 for semi-manufactures and 2..15 for finished manufactures.
Attempts to expand the U.K. economy , even from.a position of less
than full employment , have typically led to large upsurges in imports
greatly in excess of the normal incremental increase in demand for
imports that one would expect as income and output expand . This
rapid expansion of imports has partly represented stock-building ,
Import Functions 199
partly bottlenecks in particular sectors, but mainly the inability of
domestic producers to meet the output demands made upon them in a
short space of time. The faster the rate of expansion of demand, the
greater the upsurge in imports has been . Moreover, there has been a
tendency for imports to remain at a relatively high level once the rate
of expansion has subsided and output has returned to its trend level -
the so-called 'ratchet' effect.
The greater the speed of expansion , the greater the importance that
capacity effects will assume. Brechling and Wolfe (1965) noticed that
in the three booms of 1955, 1960 and 1964 the trade gap was wider in
1964 than in 1960 , and wider in 1960 than in 1955, while the mini-
mum unemployment rate had risen . They explain the phenomenon by
the relative speed of expansion in successive booms and the difficulty
of firms adjusting output to demand . They also show a ratchet effect
operating, worsening the 'full' employment balance of payments
through time. The ratio of imports to G.D.P. rises in the booms but
does not fall back to the pre-boom level when income returns to
normal. Eltis (1967) also points to the fact that successive deficits on
the balance of payments became larger in successive booms, and the
surpluses smaller in successive slumps . His explanation , however , is the
steady growth of net government invisible imports from under £100
million in 1953 to nearly £500 million in 1966 . Eltis is clear, however,
that the main factor accounting for the growth of imports in periods
of expansion has been a lack of capacity and supply response . Paish
(1968) makes the same point as Eltis, i.e. that the trend deterioration
in the current account was mainly due, at least up to 1961 , to a deteri-
oration in the invisible account, not in the trade account. On the other
hand, over the cycle 1962-6 the worsening of the trade balance ,
owing to the speed of expansion, was more serious than on invisible
account. In passing we might add that since government spending
overseas is not a function of relative price movements, it cannot be
claimed on this evidence that the deterioration in the balance of pay-
ments was the result of growing price uncompetitiveness .
The effect of capacity constraints on imports is frequently discussed
in official government reports, White Papers and the financial press . The
National Economic Development Council Report on Imported Manu-
factures (1965) endorsed the widely held view that a shortage of
capacity had caused sharp increases in imports, particularly in boom
periods. Having talked with firms in the chemical, paper and board,
textile and clothing, and iron and steel industries , the Report con -
cluded :
 200 Balance-of-Payments Theory and U.K. Experience

The largest annual increases in imports have occurred in years in

which there has been a sharp upsurge of production and of stock-
building, coupled with shortages of domestic productive capacity .
. . . These increases have generally not been reversed in subsequent
years. It may therefore be that the general upward trend of imports
has been accentuated by the effects of temporary shortages of
capacity, giving overseas producers an easy entry to the British
market which they manage to retain and develop .

The government White Paper on Industrial Strategy (resulting from the

Chequers agreement of November 1975) talks in a similar vein: 'Our
manufacturing industry has not done as well as its competitors. In
particular it has not responded adequately to changes in the pattern
of world trade and suffers from structural rigidities which show them-
selves particularly in bottlenecks both of manpower and components
in the early stages of economic upturns .' It was reported in The Times
(1 7 January 1975), for example , that iron and steel imports in 1974
reached £700 million owing to the inability of domestic producers to
meet booming demand .

IMPORTS AND LABOUR-MARKET BOTTLENECKS

Not one of the studies which have focused on the effect of capacity on
imports is very specific about the nature of the capacity constraint.
Both economic theory and casual empiricism would suggest that one
of the most formidable constraints on the ability of domestic manu-
facturers to increase production is the availability of labour. A shortage
of capital stock can be partly overcome by working capital more
intensively - up to twenty-four hours a day if labour is available to
work shift systems. A shortage of raw-material inputs can be met by
importing. A shortage of labour, however, cannot be overcome so
easily . There can be poaching by individual industries but at full
employment this will cause shortages elsewhere. Hours worked can be
increased but not to the same extent as for capital. There can also be
a certain amount of substitution by other factors of production, but
in the short run the supply of labour will be the binding constraint on
output. When domestic output is limited by the availability of labour
supply, and the domestic demand for output exceeds the supply ,
imports will be sucked in. Imports will be increasingly sensitive to the
Import Functions 201

pressure of demand in the labour market until in the limit all addit ional
increases in demand for output will be met by imports. In this section
an attempt is made to quantify the extent to which structural disequili-
brium in the labour market , with shortages in some sectors and excess
supply in others, has raised the annual U.K. import bill at the level of
demand prevailing and at a hypothetical full-employment level. (The
discussion and results draw heavily on a more technical and detailed
paper by Hughes and Thirlwall (I979) which uses data on imports,
vacancies and unemployment classified by 113 Minimum List Headings
(M.L.H.s) of the Standard Industrial Classification (SJ.C .) since 1963 .)
The first question is to decide on a measure of labour-market bottle-
necks . There is no one ideal measure . Given the labour-market data that
are available on an industry basis, the vacancy-unemployment ratio is
taken as the best measure of labour-market tightness. Unfortunately it
is not possible to distinguish between the demand for and supply of
skilled and unskilled labour . Having obtained a time-series measure
of the strength of demand in the labour market industry by industry,
the second question is how to measure the effect on aggregate imports
of bottlenecks in particular sectors. There are two main approaches to
this question, and these are outlined in Hughes and Thirlwall (I979).
The first is to identify the industries in which a labour-market bottle-
neck effect has existed in the past and then to calculate how much less
imports would have been in these industries had the pressure of demand
in them been no greater than the national average . The hypothetical
amount of imports saved industry by industry can be summed to obtain
an aggregate figure . The second approach is the same as that which has
been used in considering whether the dispersion of demand between
markets worsens the conflict between price stability and full employ-
ment. A variable representing the dispersion of labour demand across
industries is included in an import equation and from this it is possible
to calculate what the level of imports would have been if there had
been no dispersion of demand - that is, if there had been no bottle-
necks in some sectors while excess supply existed in others. Both
approaches are very fruitful and yield similar results .
Of the 113 industries examined, fifty-four show a significant rel-
ationship between imports and the pressure of demand in the labour
market over the period 1963 to 1973, holding income and certain other
variables constant , and they account for 61 per cent of total imports.
The industries are listed in Table 8.1 . It would be wrong, of course ,
to conclude firmly from this evidence that it is definitely labour-market
 TABLE 8 .1

Industries showing a significant relationship between imports

and the pressure of demand in the labour market, 1963-73

Minimum
List Heading Industry

001 Agriculture
002 Forestry
211 Grain milling
214 Bacon-curing, meat and fish products
217 Cocoa , chocolate and sugar confectionery
218 Fruit and vegetable products
219 Animal and poultry foods
239 Other drink industries
262 Mineral oil-refining
263 Lubrica ting oils and greases
271 Chemicals and dyes
272 Pharmaceutical and toilet preparations
274 Paint and printing ink
276 Synthetic resins and plastics rna terial
311 Iron and steel
312 Steel tubes
322 Copper, brass and other base metals
331 Agricultural machinery
332 Metal-working machine-tools
334 Industrial engines
335 Textile machining and accessories
336 Contractors' plant and quarrying machinery
339 Other machinery
349 Other mechanical engineering not elsewhere specified
351 Scientific, surgical and photographic instruments
364 Radio and other electronic apparatus
369 Other electrical goods
391 Tools and implements
392 Cutlery
393 Bolts, nuts, screws, rivets, etc.
394 Wire and wire manufacturing
395 Cans and metal boxes
396 Jewellery, plate and refining of precious metals
399 Metal industries not elsewhere specified
412 Spinning and doubling of cotton, flax and man-made fibres
414 Woollen and worsted
419 Carpets
429 Other textile industries
431 Leather (tanning and dressing) and fellmongery

202
TABLE 8.1 (contd)
432 Leather goods
433 Fur
441 Weatherproof outerwear
445 Dresses, lingerie, infants ' wear, etc .
449 Dress industries not elsewhere specified
464 Cement
469 Abrasives and building materials not elsewhere specified
471 Timber
472 Furniture and upholstery
479 Miscellaneous wood and cork manufactures
482 Cardboard boxes, cartons and fibre-board packing cases
491 Rubber
494 Toys, games and sports equipment
495 Miscellaneous stationers' goods
496 Plastics moulding and fabricating

constraints that are the source of the problem without knowing the
correlation between the pressure of demand in the labour market in
these industries and other capacity constraints. But the evidence is
suggestive, especially as in forty-one of the fifty-four industries the
vacancy-unemployment (v/u) ratio was higher than the national average
of 0.42. If viu in these forty-one industries had been no higher than the
national average, it is estimated that total imports would have been on
average £326 million per annum lower (at 1970 prices) .
Using the second approach gives a range of estimates depending on
the specification of the import equation and the measure of dispersion
used. The central estimates, however, come out at between £368/£422
million per annum, holding the average level of income and the average
pressure of demand constant. These estimates are somewhat higher than
that given by the first approach but are close enough to suggest that
there is a bottleneck effect at work , the severity of which has increased
the magnitude of the import bill by an average of between £300 and
£500 million per annum over the period 1963 to 1973 . At full employ -
ment, when sectoral bottlenecks are most acute, the cost of bottlenecks
to the balance of payments is estimated to be £700 million. In the light
of the balance-of-payments deficits in the I970s, these figures may
seem trivial for an individual year , but aggregated over a ten-year period
it is a considerable biIJ to finance by exports and/or foreign borrowing.
It must also be remembered that in the 1960s a deficit of £400 million

203
204 Balance-or-Payments Theory and U.K. Experience

would have been regarded as a severe balance-of-payments problem.

The pound was devalued on a lesser figure in 1967 .
As far as the over-all results for all 113 industries are concerned, the
industries in which imports seem to be the most sensitive to the pres-
sure of demand in the labour market belong to the engineering , elec-
trical and metal groups , while the least sensitive belong to the food,
drink and tobacco , mining , vehicles, and paper groups. In other broad
industrial groupings the results are mixed , no doubt reflecting either
the absence of capacity constraints or the difference in behaviour
between competitive and non-competitive imports. The industries
listed in Table 8.1 would seem to be the major industries on which
attention must be focused if bottleneck effects on imports are to be
avoided in the future . The results have obvious policy implications
for the N.E.D.C.'s Industrial Strategy in particular, and for active
manpower policy in general .

IMPORTS AND INCOME (ESTIMA TES OF INCOME ELASTICITIES)

A second important determinant of imports is expenditure , to be

proxied here by income . A central estimate of the U.K. income elas-
ticity of demand for aggregate imports in the 1950s and 1960s would
be 1.6 (see Houthakker and Magee, 1969). In the 1970s it was probably
closer to 2 : that is, for every I per cent rise in national income, imports
into the U.K. rose by 2 per cent. Disaggregated studies show income
elasticities much higher than this for some groups of commodities,
particularly finished manufactured goods. An elasticity above unity
will increase the ratio of imports to the measure of income through
time .
If one looks at the history of manufactured imports, they started
to increase rapidly in the 1950s , but not until 1960 did the share of
manufactures in total imports reach its pre -war peak of 31 per cent.
Thus , part of the post-war growth of manufactured imports can be
considered as a catching-up process following pre-war, wartime and
post-war trade controls. Since 1960 , however, the growth has been
even faster .' Panic (1968) finds it difficult to blame increases in the
relative price of domestic goods for the upsurge of imports because
from 1958 to 1967 the ratio of wholesale prices of domestic products
to the unit value index of imports remained unchanged . He also believes
that capacity constraints cannot be the whole story either because
imports have continued to rise inexorably even in years of spare capac-
Import Functions 205

ity . This has been even more true of the 1970s than of the 1960s.
Panic's list of possible explanations include : the poor quality , design
and technological performance of domestic products; the slow growth
of the economy and low investment, which has affected adversely the
speed of process and product innovation ; poor marketing techniques
compared with foreign competitors; the growth of subsidiaries of
foreign companies; and investment grants, which may have increased
the quantity of imported capital goods .
In an international comparison of O.E.C.D. countries Taplin (1973)
finds that the United Kingdom had the highest income elasticity of
demand for total manufactured-good imports over the period 1953-4
to 1969-70. His estimates are as follows : the United Kingdom 2 .61 ;
Japan 1.77 ; West Germany 1.5; Italy 1.66 ; the Netherlands 1.25 ;
Canada 1.42; and France 1.89 .
In a comparative study of the United Kingdom , West Germany and
France, taking the five main commodity groupings distinguished by the
Standard Industrial Classification, Panic (1975) estimates the income
elasticities of demand for imports shown in Table 8 .2.

TABLE 8 .2

International estimates of income elasticities ofdemand for imports

United Kingdom West Germany France

Food , beverages and tobacco 0 .35 0 .86 0 .84

Basic rnaterials 0 .66 1.22 0 .70
Fuels 2.47 2.66 1.26
Manufactured goods 3.09 2.14 2.19
Semi-finished 2.37 2.06 na ,
Finished 4.30 3.52 n.a .

Total imports 1.82 1.31 1.63

Source: Panic (1975) .

Panic draws attention to the large discrepancy in the income elasticity

of demand for manufactured goods between the United Kingdom on
the one hand and West Germany and France on the other. He attempts
to explain the difference in terms of shifts in demand away from types
of goods and services traditionally produced and the sluggishness
l\..l TABLE 8.3
<:::>
0,
Ordered money income elasticities of demand for imports

Minimum Industry U.K. income elasticity

List Heading of demand for imports

381 Motor -vehicles manufact uring 3.680

492 Linoleum,leather cloth, etc . 3.454
442 Men's and boys ' tailored outerwear 3.208
364 Radio and other electronic apparatus 2.934
395 Cans and metal bo xes 2.934
382 Motor-cycle, three-wheel vehicle and pedal-cycle manufacturing 2.718
496 Plastics moulding and fabricating 2.617
365 Domestic electric appliances 2.585
362 Insulated wires and cables 2.580
313 Iron castings, etc. 2.578
472 Furniture and upholstery 2.516
240 Tobacco 2.465
473 Bedding, etc. 2.425
444 Overalls and men's shirts, underwear, etc . 2.414
429 Other textile industries 2.377
272 Pharmaceutical and toilet preparations 2.325
443 Women's and girls' tailored outerwear 2.308
441 Weatherproof outerwear 2.255
499 Miscellaneous manufacturing industries 2.248
399 Metal industries not elsewhere specified 2.228
482 Cardboard boxes, cartons and fibre-board packing cases 2.215
 445 Dresses,lingerie, infants' wear, etc . 2.181
363 Telegraph and telephone apparatus 2.174
411 Production of man-made fibres 2.144
239 Other drink industries 2.115
463 Glass 2.104
394 Wire and wire manufactures 2.085
495 Miscellaneous stationers' goods 2.081
213 Biscuits 2.081
369 Other electrical goods 2.046
337 Mechanical handling equipment 1.994
331 Agricultural machinery (except tractors) 1.963
491 Rubber 1.957
432 Leather goods 1.938
393 Bolts, nuts, screws, rivets, etc . 1.932
311 Iron and steel (general) 1.908
312 Steel tubes 1.904
392 Cutlery 1.863
450 Footwear 1.849
370.1 } Shipbuilding and ship-repairmg
1.830
370.2 Marine engineering
417 Hosiery and other knitted goods 1.822
494 Toys, games and sports equipment 1.806
339 Other machinery 1.791
389 Perambulators, hand-trucks, etc. 1.786
277 Polishes, gelatine, adhesives, etc. 1.763
483 .Manufactures of paper and board not elsewhere specified U50
419 Carpets 1.742
334 Industrial engines 1.727
~
'I 391 Tools and implements 1.707
l\.l TABLE 8.3 (contd)
C
00
Minimum Industry U.K. income elasticit y
List Heading of demand for imp orts

231 Brewing and malting 1.706

274 Paint and printing ink 1.691
352 Watches and clocks 1.678
414 Woollen and worsted 1.675
486 Printing, publishing of newspapers and periodicals 1.658
351 Scientific, surgical and photographic instruments, etc. 1.623
361 Electrical machinery 1.595
349 Other mechanical engineering not elsewhere specified 1.594
276 Synthetic resins and plastics materials 1.593
333 Engineers' small tools and guages 1.585
21i Bread and 110ur confectionery 1.578
341 Industrial plant and steelwork 1.546
469 Abrasives and building materials, etc ., not elsewhere specified 1.540
335 Textile machinery and accessories 1.522
421 Narrow fabric s 1.461
413 Weaving of cotton ,linen and man-made fibres 1.424
431 Leather (t anning and dressing) and fellmongery 1.421
271 Chemicals and dyes 10411
416 Rope, twine and net 1.388
336 Contractors' plant and quarrying machinery 1.381
489 Other printing, publishing. bookbinding, engraving , etc . 1.320
219 Animal and poultry foods 1.312
396 Jewellery . plate and refining of precious metals 1.306
418 Lace 1.268
 461 Bricks, fireclay and refractory goods 1.253
217 Cocoa , chocolate and sugar confectionery 1.238
109 Other mining and quarrying 1.231
449 Dress industries not elsewhere specified 1.208
471 Timber 1.192
422 Made-up textiles 1.192
481 Paper and board 1.163
214 Bacon-curing, meat and fish products 1.122
218 Fruit and vegetable products 1.115
412 Spinning and doubling of cotton, flax and man -made fibres 1.109
338 Office machinery 1.037
446 Hats, caps and millinery 0.958
003 Fishing 0.932
433 Fur 0.904
002 Forestry 0.893
332 Metal-working machine -tools 0.883
479 Miscellaneous wood and cork manufactures 0.883
275 Vegetable and animal oils, fats, soap and detergents 0.873
322 Copper , brass and other base metals 0.868
102 Stone and slate quarrying and mining 0.843
216 Sugar 0.803
493 Brushes and brooms 0.778
262 Mineral oil-refining 0.752
001 Agricult ure and horticulture 0.671
229 Food industries not elsewhere specified 0.489
263 Lubr icating oils and greases 0.390

~
<:::>
'0
210 Balance-ofPayments Theory and U'K, Experience

of U.K. producers in adjusting supply to changes in demand . The

United Kingdom's incapacity to grow quickly must have a bearing, he
thinks, on the ability to adjust because of the link between growth,
investment and the speed of technical change . It will be argued in
Chapters 10 and II that the major constraint on growth in open
economies is the balance of payments and that the slow growth rate
of the United Kingdom is primarily a function of the slow rate of
growth of exports, attributable no doubt to factors which also make
the income elasticity of demand for imports relatively high . The syn-
drome of slow growth and a high income elasticity of demand for
imports is part of a vicious circle .
Another comparative study by Humphrey (I976) gives similar
results to those of Panic, with an estimated income elasticity of demand
for imports of total manufactures of over 3, and for finished manu-
factures alone of over 4 . Both estimates are significantly higher than
those for West Germany and France . Humphrey attributes the high
U.K. income elasticity of demand for imports to the fact that 'U .K.
manufacturers are simply less able to satisfy marginal demand for
finished manufactures than is the case in other countries'. He supports
his case by pointing to the low world income elasticity of demand for
U.K. exports (to be discussed in the next chapter), and presumably
believes that both phenomena are the result of a common cause,
namely supply deficiencies in a capacity or quality sense.
With the publication of import (and export) data classified by Mini-
mum List Headings of the Standard Industrial Classification (referred to
earlier) it has become possible to estimate income elasticities of demand
for imports at a highly disaggregated level. Estimates for the period
1963 to 1973 3 are presented in Table 8.3 . The estimates are made using
the money values of imports and income because of the unavailability
of import price deflators at such a disaggregated level. It can be said
that if import prices have risen relative to domestic prices, the elasticity
estimates will be underestimates of the real income elasticities if the
demand for imports is price elastic , and overestimates if the demand
for imports is price inelastic. If import prices have fallen relative to
domestic prices , the converse of the above will be true.

IMPORTS AND RELATIVE PRICES (ESTIMATES OF PRICE ELASTIC'ITIES)

In estimating the influence of relative prices on import demand the

relative price index should not be measured taking all domestic prices
Import Functions 211

relative to import prices because many goods that enter the aggregate
price index are not traded or substitutable for imports. The domestic
wholesale price index would be better, or, ideally, a special index
should be constructed. There is a discussion of measures of price
competitiveness in international trade in the next chapter. Whatever
measure is taken, however, there does not appear to have been any
deterioration in relative prices over the crucial period when the U.K.
balance of payments started to become noticeably weaker. According
to Panic (1968) , there was no deterioration from 1958 to 1967 in the
U.K. wholesale price index relative to the unit value import index, and
Morgan and Martin (1975), who constructed a special index of domes-
tic prices comparable with import prices, find that the ratio of the two
indices was remarkably static until the 1967 devaluation.
The evidence on the importance of relative prices in determining
imports is very mixed . Some studies for the United Kingdom find
imports insensitive to price, other studies find some sensitivity. The
Treasury model of the U.K. economy estimates the price elasticity
of demand for imported goods to be of the order of 0.65, and for
services 0 .2. Rees and Layard (1971) estimate a long run price elasticity
for finished manufactures of 0 .8 for the period 1959- 69, and Whitley
(1979) estimates an equivalent price elasticity of 0.5 for the period
1963-76. At a disaggregated level work by Barker (I 970a) gives the fol-
lowing estimates : food, drink and tobacco , 0 .767 ; basic materials ,
0.288 (perverse sign); fuels, 0.692; semi-manufactures, 1.427, and fin-
ished manufactures , 1.101 . Another disaggregated study of consumer-
TABLE 8.4
Import price elasticities
Country Project Houthakker Taplin (1973)
link and Magee (1969)

Belgium -0.20 - 1.02 -0 .65

Canada -1.98 -1 .46 -1.59
France n .a. +0.17 -0 .39
Italy -0.95 -0.13 -1.03
Japan -0.17 -0 .72 -0.81
Netherlands -0.41 +0 .23 -0 .02
Sweden 0.00 -0.79 -0 .76
United Kingdom 0.00 +0 .22 -0.22
United States -0.52 -0.54 -1.05
West Germany -1.21 -0 .24 -0 .61

Source: Magee (974).

212 Balance-of-Payments Theory and U'K. Experience

good imports, capital goods and semi-manufactures by Hibberd and

Wren-Lewis (1978) gives significant elasticities between 0.5 and 2.0.
Several surveys of the international evidence have been conducted ,
one of the most comprehensive being by Magee (1974) . A selection of
the long-run aggregate estimates for various countries from different
studies is given in Table 8.4.
The elasticities are sometimes perverse but in general they are of the
right sign and lie between zero and unity . The sensitivity of imports to
relative prices seems to be the lowest of all in the United Kingdom
taking an average of the estimates. In fact the average gives a zero
elasticity!

IMPORT PENETRATION

An income (or expenditure) elasticity of demand for imports in excess

of unity leads to the increasing penetration of imports into the domes-
tic market. Import penetration will also occur through upward shifts in
the whole import function through trade liberalisation and increased
international specialisation. For some products the penetration of
imports into the United Kingdom has now reached alarming proportions.
There are two main ways of measuring import penetration . One is to
take the ratio of imports to sales of the product in the domestic mar-
ket, where sales are equal to domestic production plus imports minus
exports. The other is to take the ratio of imports of a product to its
domestic production. The second measure (the import ratio) is to be
preferred since the first measure is affected by export performance,
and in any analysis of import penetration in relation to the function-
ing of the domestic economy this could mislead. For example, starting
from a point of balance between exports and imports, a faster growth
of exports than imports of a product will raise import penetration
according to the first measure cited above, and this should not be a
source of concern if great importance is attached to raising the rate of
growth of exports. To analyse import penetration in relation to the
functioning of the domestic economy it is better to exclude exports
from consideration and to take the ratio of imports to domestic pro-
duction . This point is discussed more fully by Hughes and Thirlwall
(1977), who examine import penetration in the U.K. manufacturing
sector over the period 1963 to 1974, using import data classified by
M.L.H.s of the S.Le . and output data from the Census of Production
 TABLE 8.5
Import penetration ratios, 1970 and 1979

1970 1979 % change

III Food , drink and tobacco 0.1627 0.1414 - 13.1

2 11 Grain milling 0.0262 0.0460

212 Bread and flour confec tionary 0.00 20 0.0033
213 Biscuits 0.0015 0.0222
214 Bacon curing, meat and fish produ cts 0.6761 0.3005
215 Milk and milk produ ct s 0.1906 0.1629
216 Sugar 0.4657 0.4677
217 Cocoa , chocolate and sugar confectionary 0.0795 0.1204
218 Fruit and vegetabl e produ cts 0.3478 0.2859
219 Animal and poultry food s 0.0584 0.0505
221 Vegetabl e and animal oils and fats 0.725 2 0.3881
229 Food indust ries n.e.s. 0.4171 0.2584
231 Brewing and malting 0.0 131 0.0128
232 Soft drink s 0.0276 0.09 73
239 Other drink industries 0.0864 0.2090
240 Tobacco 0.0038 0.0112

IV Coaland petroleum products 0.1933 0.15 76 - 18.5

261 Coke ovens and manu facturing fuel 0.0008 0.0213

~ 262 Mineral oil refining 0.2242 0.1505
....
I..u 263 Lubricating oils and greases 0.1624
".l
.....
-t..
TABL E 8.5 (conte/)

V Chemicals and allied industries 0.1483 0. 2260 +52.4

271 General chemicals 0.2272 0.2710

272 Pharmaceutical chemicals and preparations 0.0777 0.1370
273 Toilet preparat ions 0.0315 0.1359
274 Paint 0.0260 0.046 2
275 Soap and dete rgent s 0.0433 0.06 37
276 Synthetic resins et c. 0.2 279 0.4264
277 Dyestuffs and pigment s 0.1481 0.1679
278 Fertilis ers 0.095 3 0.1010
279 Other chemical industries 0.0849 0.20 39

VI Metal manufacture 0.1736 0.2424 +39. 6

311 Iron and steel (general) 0.0748 0.1789

312 Steel tub es 0.0706 0.18 32
313 Iron castings, etc . 0.0149 0.0248
321 Aluminium, et c. 0.2275 0.2218
322 Copp er , brass, et c. 0.4679 0.4432
323 Oth er base metals 0.4249 0.5409
 1970 1979 % change

VII MecJumical engineering 0.1261 0.2081 +65.0

331 Agricultural machinery (except tractors) 0.1505 0.313 2

332 Metal working machine tools 0.2207 0.4449
333 Pumps , valves and compressors 0.1326 0.1864
334 Industrial engines 0.1154 0.1465
335 Textile machin ery. etc. 0.2455 0.3619
336 Constructi on and earth-moving equipment 0.2182 0.2735
337 Mechanical handling equipment 0.0875 0.218 2
338 Offi ce machinery 0.4274 0.7006
339 Oth er machinery 0.1517 0.2685
34 1 Indu strial plant and steel work 0.02 39 0.0702
349 Oth er mech anical engines n.e.s. 0.0599 0.1048

VIII Instrument engineering 0.2342 0.4 738 +102.3

351 Photographi c and do cument

copying equipment 0.3628 1.1011
352 Watches and clocks 0.5699 1.4761
353 Surgical inst ruments and appliances 0.1419 0.3166
354 Scientific and indus trial instrum ents 0.2041 0.3323

\\,)
.....
v,
....0-tv
TABLE 8.5 (contd)

IX Electrical engineering 0.1322 0.3008 +12Z5

361 Electrical machinery 0.0656 0.1544

362 Insulated wires and cables 0.0138 0.0455
363 Telephone and telegraph apparatus
and equipment 0.0420 0.0779
364 Radio and electronic components 0.2783 0.5212
365 Broadcast receiving and sound
reproducing equipment 0.0983 0.5678
366 Electronic computers 0.8759 0.8395
367 Radio, radar and electronic capital goods 0.0958 0.1463
368 Electrical (domestic) appliances 0.0939 0.2896
369 Other electrical goods 0.0767 0.1883

X Shipbuilding and marine engineering 0.0739 0.1028 +39.1

370 0.0739 0.1028

Xl Vehicles 0.0732 0.3625 +395.2

380 Wheeled tractor manufacturing 0.0887 0.2664

381 Motor-vehicle manufacturing 0.0469 0.3455
382 Motor-cycle, tricycle and pedal cycles 0.1278 1.0720
383 Aerospace equipment 0.1686 0.4729
384 Locomotives, railway track equipment, }
0.0130 0.0282
385 railway carriages, wagons and trains
 1970 1979 % change
XII Metal goods n.e.s. 0.1941 0.2193 +13.0

390 Engineers' small tools and gauges 0.1116 0.1412

391 Hand tools and implements 0.1638 0.3650
392 Cutlery, spoons, forks , etc. 0.1449 0.2476
393 Bolts, nuts, screws, rivets 0.0558 0.1809
394 Wire and wire manufacturing 0.0282 0.0833
395 Cans and metal boxes 0.0090 0.0174
396 Jewellery (excluding diamonds) 0.4138 0.8121
399 Metal industries n.e.s. 0.0371 0.1042

XIII Text iles 0.1252 0.2844 +12Z2

411 Production of man-made fibres 0.1382 0.2892

412 Spinning and doubling on cotton
flax system 0.1875 0.3228
413 Weaving of cotton, linen and manmade fibres 0.3739 0.9684
414 Woollen and worsted 0.0569 0.1622
415 Jute 0.3449 0.1432
416 Rope, twine and net 0.0792 0.1902
417 Hosiery and other knitted goods 0.0848 0.2766
418 Lace 0.1733 0.6601
419 Carpets 0.0771 0.1988
421 Narrow fabrics 0.0576 0.1517
N 422 Make-up textiles 0.1350 0.2544
.....
'-J 429 Other textile industries 0.0329 0.1342
l\,)
....... TABLE 8.5 (contd)
00

XIV Leather 0.1790 0.4355 +143.3

431 Leather (tanning) 0.1725 0.3426

432 Leather goods 0.1203 0.5427
433 Fur 0.3295 0.7005

XV Oothing and footwear 0.1169 0.2992 +155.9

441 Weatherproof outerwear 0.1298 0.3932

442 Men's and Boy's tailored outerwear 0.0726 0.3681
443 Women's and girl's tailored outerwear 0.0623 0.1976
444 Overalls and men's shirts, underwear, etc. 0.2264 0.4132
445 Dresses, lingerie, infants' wear, etc. 0.0695 0.1975
446 Hats , caps and millinery 0.1010 0.1867
449 Dress industries n.e.s. 0.2185 0.2301
450 Footwear 0.1394 0.3908

XVI Bricks, pottery, cement, etc. 0.0498 0.0859 +725

461 Bricks, flreclay and refractory goods 0.0348 0.0277

462 Pottery 0.0612 0.1270
463 Glass 0.1076 0.1962
464 Cement 0.0064 0.0076
469 Miscellaneous build ing materials 0.0314 0.0585
 1970 1979 % change
X VIl Timb er, furniture etc. 0.260 8 0. 286 7 + Z9
47 1 Tim ber 0.516 0 0.5668
47 2 Furnit ure and upholstery 0.0 285 0.1177
473 Bedding, etc. 0.0045 0.0208
474 Shop and o ffice fitting - 0.00 53
475 Wooden containers and basket s 0.1069 0.1375
479 Miscellaneou s wood and cork manu factur ing 0. 1032 0.1461
XVlII Paper, printing, publishing 0.1 635 0.1724 +5.4
481 Paper and board 0.7594 0.9 288
482 Packaging products 0.0062 0.0 160
483 Manufactur ed stationary 0.0167 0.0380
484 Manufacture of pap er and board n.e.s. 0.0 257 0.0773
485 Print ing and publishing of
} 0.0092 0.0080
486 newspap ers and period icals
489 General print ing and publishing 0.0475 0.065 5
X/X Oth er manufacturing industries 0.0 760 0.1654 +l1 Z6
491 Rubber 0.0476 0.14 25
492 Linoleu m, etc . 0.1247 0.2537
493 Brushes and brooms 0.1769 0.1800
494 Toys, games , etc. 0.151 2 0.3059
495 Miscellaneous stationers' goods 0.11 33 0.2891
496 Plasti cs product s n.e.s, 0.05 35 0.11 32
499 Miscellaneou s manu factu ring 0.1852 0.3282
N Total manufacturing 0.147 0.223 +51.7
....
'0
220 Import Functions

Reports. The extension of this analysis comparing 1970 and 1979 is

given in Table 8.5. We merely highlight here some of the main features
of the factual analysis. The import ratio was 12 per cent in 1963, rising
to 14.7 per cent in 1970 and to 22.3 per cent in 1979 . This represents
an average annual rate of increase of 3.2 per cent between 1963 and
1970 and 5.74 per cent between 1970 and 1979. Import penetration
appears to be accelerating . For individual industries, the experience
differs considerably. There was a positive trend rate of change in the
import ratio in fifteen of the seventeen S.Le. Orders, and in 106 of the
120 M.L.H. industries . The handful of industries where import pen-
etration fell appear to be mainly food and mineral oil refining. The
vehicle industry experienced the highest trend rate of growth in import
penetration, with an average annual rate of 44 per cent per annum . This
has been followed in severity by trends in the clothing and footwear
industry; leather; textiles; electrical engineering and instrument engin-
eering.
If the trend in import penetration continues at the same rate over
the period 1980 to 1989 as it did over the period 1970 to 1979, the
import bill for manufactures will be equivalent to 34 per cent of final
domestic manufacturing output in 1989. There is already evidence
that the trend is accelerating and the ratio could be higher than this .
If the rising trend which comes from upward shifts in the import func-
tion is not compensated by an autonomous improvement in export per-
formance relative to output," growing import penetration poses a severe
threat to employment and the manufacturing base. For an industry in
which imports grow relative to output at a rate of 10 per cent per
annum, the import penetration ratio would double in roughly seven
years, which could pose a serious threat to the industry's survival. In a
world of increasing specialisation it is almost inevitable that the level of
import penetration will continue to rise. What is important is that it
should not rise faster than in other countries, or faster than the per-
formance of exports relative to the full-employment rate of growth of
output.
9
Export Functions
THE DETERMINANTS OF EXPORTS

The factors affecting the quantity of goods exported by a country will

be the same as those affecting a country 's demand for imports, though
they may be expected to differ in relative importance . First , there is the
ability and willingness of domestic producers to supply, which depends
partly on capacity and partly on the domestic pressure of demand,
which may divert goods away from foreign markets to the home market.
Second, the price of exports compared with prices charged by foreign
competitors in different export markets may be expected to exert some
influence . Third, the level of income and expenditure in foreign mark-
ets will affect the quantity of exports sold. In addition a host of non-
price factors may be expected to affect the demand for exports-such
as the quality, design and reliability of goods, and the time it takes for
their delivery. Some of these factors , more difficult to quantify , are
discussed in Chapter 12.
like the import function, the export function is usually assumed to
be multiplicative, implying a constant elasticity of exports with respect
to each of the explanatory variables specified in the export function .
Hence the export function may be written :

(9 .1)

where X, is the quantity of exports, C is a measure of capacity utilis-

ation in the home country, PdlPf is the price of exports relative to
competitors' prices in export markets measured in a common currency,
Z is the level of 'world' income, r is the elasticity of exports with
222 Balance-ofPayments Theory and U.K. Experience

respect to changes in capacity utilisation (or pressure of demand) , T/ is

the price elasticity of demand for exports, and e is the world income
elasticity of demand for exports.
If the function is not homogeneous of degree zero in money income
and prices (i.e. there is money illusion), the function may be written :

(9.2)

where Zm is world income measured in money terms , and it is also

recognised that the 'own' price elasticity of demand for exports may
differ from the cross elasticity (i .e. T/ *" 0) .
If equations (9.1) and (9 .2) are thought of as demand equations,
the capacity variable, strictly speaking, does not belong except to the
extent that prices do not adjust as the supply of exports varies. In
periods of excess supply prices will not fall if the demand for exports
is perfectly elastic in world markets; similarly, the price of exports
may not be raised as a rationing device when domestic demand pressure
is diverting exports to the home market. If export behaviour is to be
explained in these circumstances, a supply variable must be included
in the export function as an independent argument. Let us consider
each of the factors mentioned above , and. examine the empirical
evidence relating to the United Kingdom .

EXPORTS AND CAPACITY

When the United Kingdom enjoyed relatively full employment in the

1950s and 1960s it used to be claimed that a part of the reason for her
relatively poor export performance was that the internal pressure of
demand was excessive, leaving insufficient capacity to supply exports
that would be readily absorbed by the world at the ruling market price .
Export performance was thought to be negatively related to the inter-
nal pressure of demand and the level of domestic capacity utilisation .
What is the theoretical prediction, and what is the evidence? The
theoretical prediction about the division of output between home and
foreign sales depends on the assumptions made about firm behaviour,
and on the shape of the demand curve facing firms in foreign markets.'
For example, assume a single-product profit-maximising firm which is
a price-maker in the home market and a price-taker in foreign markets,
so that the demand curve facing the firm is downward-sloping at home
and horizontal abroad . The allocation of the firm's output between the
Export Functions 223
Price

s
Me

PF ~-------="~----~----71~---­
ARF=MRF

o '----------'---------'-------=---
Home sales QH Foreign sales Q Quantity

FIGURE 9.1

two markets is shown in Figure 9.1. H stands for the home market
and F for the foreign market.
The total output o f the firm is Q, where the producer maximises
profits by equating marginal cost (M C) with marginal revenue in both
markets (MR H and MR F). Total output is divided between OQH for
the home market and QHQ for the foreign market , giving a price in
the domestic market of PH and a price in the foreign market of PF .
Now the total marginal revenue curve is the horizontal sum of the
marginal revenue curves in each market, which equals STU in Figure
9.1 . If the Me curve cuts the total MR curve to the right of T , the
equilibrium level of output does not change with an increase or de -
crease in the domestic demand for output. Thus in this particular
model it can be said that if the domestic demand for output increases,
the quantity available for export will fall; and if domestic demand
falls, the quantity available for export increases. The quantity of exports
will vary inversely with the internal pressure of demand .
Now suppose the firm is a price-maker in world markets so that
the demand curve facing the firm slopes downwards in both markets.
The total marginal revenue curve will now no longer have a kink in it.
Thus when demand in the home market changes , the equilibrium level
224 Balance-of-Payments Theory and U.K. Experience

of output will also change. What happens to the quantity of output

available for export depends on whether output increases or decreases
more or less than the increase or decrease in domestic demand . This in
turn will depend on the shape of the MC curve . If MC is increasing, an
increase in home demand will cut exports because total output will
increase by less than domestic demand . If MC is decreasing, an increase
in domestic demand will increase exports because total output will
increase by more than domestic demand. If MC is constant, the
quantity of exports will remain unaffected. These propositions are
illustrated in Figure 9.2.

Price

MC,

1---+-;-'--'r-----------'::::;,l,"'=:--t-r----MC3

\
\
\
MR ,
I
IMR
I

oL...---...l...-----'---------......L..--:l-...L.-L,-----
Q

FIGURE 9 .2

The total marginal revenue curve (MR) in Figure 9 .2 is the sum of

MR H and MR F . The equilibrium level of output is Q, divided between
OQH for the home market and OQF for export. Now suppose there is
an increase in domestic demand which shifts MRH to MR H 1 ' and which
at the same time shifts MR to MR 1 by the same amount. The new
equilibrium level of home sales is at QH1 , and whether this affects
the availability of output for exports depends on the response of
total output. If MC is increasing (MC.), it is clear that output rises
less than the increase in domestic demand, to Ql , and export volume
Export Functions 225

will fall. If MC is decreasing (MC2 ) , output rises to Q2, which is more

than the increase in domestic demand , and the supply available for
export increases . Finally, if MC is constant (MC3 ) , output rises to Q3,
which exactly equals the increase in domestic demand, and exports
remain the same.
Needless to say, if the firm is not a profit-maximiser and there is a
disturbance of domestic demand, it is difficult to know, and impossible
to predict a priori. what the relation will be between changes in the
internal pressure of demand and export performance .
In the formal theoretical models outlined above , an increase in the
pressure of domestic demand makes production for the home-market
relatively more profitable than exporting in very competitive markets,
and this is presumably part of the mechanism which produces an inverse
relation between domestic demand pressure and export performance.
Some models of export behaviour (see Winters, 1974) take explicit
account of changes in the profitability of exporting, hypothesising that
for any given degree of demand pressure on capacity the extent to which
exports are cut back will be directly related to the profitability of home
sales relative to exports. Winters allows for this by specifying the
model :

(9.3)

where Trt is the relative profitability of exports measured by the ratio of

the domestic price index for manufactures to the export unit value
index for manufactures. (We later discuss different measures of compet-
itiveness and profitability.) Winters fmds that profitability affects export
performance , in addition to the level of capacity utilisation itself,
though not necessarily independently of it.
Before turning to the empirical evidence of the relationship between
export performance and the internal pressure of demand , a distinction
needs to be made between models which are designed to explain the
level of exports and models which are designed to explain export
performance relative to other countries, i.e. a country's share of total
world exports. While the level of export performance depends on
capacity, competitiveness and the level of 'world' income, export
performance relative to other countries depends on capacity, competit-
iveness and the geographic and commodity composition of trade. The
level of 'world' income becomes largely irrelevant in explaining shares,
since the variable is more or less common to all but very large countries.
226 Balance-of-Payments Theory and U.K. Experience

It is now the 'direction' of trade that is important. The slower the

growth in demand for the commodities that a country exports , and the
slower the growth of the markets to which it exports, the more unfav-
ourable a country's share of total exports will become, other things
remaining the same . Some studies of export performance examine both
the level of exports and variations in the export share in relation to the
internal pressure of demand , and where this is so both findings will be
reported. (For a detailed analysis of the apparent unimportance of the
direction of trade as an explanation of the United Kingdom's declining
share of total world exports, see the final section of this chapter.)
One of the earliest studies of U.K. export performance and the
internal pressure of demand is that by Ball, Eaton and Steuer (1966)
for the period 1954 to 1964 . Working first with a model to explain
the level of exports, only world demand emerged as an important
explanatory variable. At a time when most economists still believed
in the importance of relative prices , however , the authors were moved
to remark that 'such an interpretation is open to strong a priori and
statistical objections. It is difficult to believe that price and non-price
competition have not played an important role in determining the
downward trend in the U.K. share [of total exports J over the period
studied .' They seem to attribute their poor and unexpected results
to multicol1inearity between variables. This leads them to divide the
level of U.K. exports by the level of world trade and to work with the
U.K. share of total exports as the dependent variable in the analysis .
When this is done their relative price variable becomes statistically
significant, but the internal pressure of demand variable as measured
by an index of the ratio of industrial output to its trend level remains
insignificant. The authors then include in their estimating equation
the U.K. share of world output as a measure of non-price factors
affecting the United Kingdom 's declining share of world trade. When
this is done the internal pressure of demand variable becomes signifi-
cant. The authors also include a time trend in the analys is, which is
another way of picking up the effect of non-quantifiable adverse
influences. Again they find the internal pressure of demand variable
statistically significant. Ball, Eaton and Steuer conclude over all that
if the average pressure of demand in the United Kingdom over the
period had been lower at 2.5 per cent unemployment (instead of
approximately 1.5 per cent), the level of exports would have been
some £80 million to £180 million per annum higher , depending on
the precise year taken.
Export Functions 227
Henry (1970) undertakes a similar study to the one above, but at
a disaggregated industry level, taking different countries. He examines
five U.K. industries over the period 1953 to 1965 , and in addition
thirteen Belgian and eight U.S. industries , testing the basic hypothesis
that domestic sales are preferred, and exports discriminated against,
when the domestic pressure of demand is high . Demand pressure is
measured by the ratio of the actual value of the industrial production
index to its trend value, and the effect of periods of high and low
pressure of demand are distinguished using dummy variables . There
seem to be a number of industries where at a high pressure of demand
exports are sensitive to domestic demand and not to world demand,
and where at a low pressure of demand exports are sensitive to world
demand but not to domestic demand . His evidence supports the capac-
ity hypothesis , but there appears to be no particular industry or
industry characteristic associated with the phenomenon . The adverse
impact of demand pressure on exports appears to be quite pervasive.
Smyth (1968) examines the pressure-of-demand hypothesis but
relates the U.K. share of total exports to the rate of change of demand
rather than to the absolute level of demand pressure. He also tests
for a ratchet effect to see whether the export share falls more when the
rate of change of demand is positive than it rises when the rate of
change of demand is negative . The idea that a ratchet mechanism may
operate originates from the work of Brechling and Wolfe (1965),
mentioned in the last chapter, who observed successive' deteriorations
in the trade balance in successive booms. Smyth finds that comparing
the end-boom years 1955 , 1960 and 1964 , the percentage decline in the
share of exports was greater in 1964 than in 1960 and greater in 1960
than in 1955, while the average pressure of demand was lower in 1964
than in 1960 and lower in 1960 than in 1955 . This is casual empirical
support for the Brechling-Wolfe hypothesis applied to exports alone,
and is supported by statistical regression analysis . Smyth finds no
support for the Oppenheimer (1965) assertion, made in response to
Brechling and Wolfe, that the United Kingdom's poor export perform-
ance can be attributed to a lack of competitiveness . Changes in relative
prices in Smyth's study seem to have no impact on the U.K. share of
total exports.
One manifestation of a high internal pressure of demand is that if
the market is not cleared through price adjustment , the waiting-time
for export orders to be dispatched may lengthen . If the waiting-time
for existing orders lengthens, this may affect the quantity of new
228 Balance-of-Payments Theory and U.K. Experience
export orders and hence the quantity of future deliveries. Steuer ,
Ball and Eaton (1966) have examined the relationship between waiting-
time and export orders for machine-tools using quarterly data for the
United Kingdom, the United States and West Germany over the period
1956 to 1962 . The authors find that the flow of foreign orders received
by the U.K. machine-tool industry was influenced more by long
delivery delays than by relative prices . West German waiting-time
(though not U.S. waiting-time) also affected orders for U.K. machine-
tools. West Germany and the United Kingdom are apparently stronger
competitors than the United States and the United Kingdom. In quanti-
tative terms, a one-month rise in U.K. waiting-time reduces orders by
about 10 per cent , and a one-month fall in West German waiting -time
reduces U.K. orders by about 20 per cent.
In a similar study Artus (1973) investigates the export production
behaviour of machinery industries (including machine-tools) in the
United Kingdom, the United States and West Germany , using quarterly
data from 1956 to 197 J. A 10 per cent increase in demand resulted
ultimately in an increase in export delivery time of twenty days for
U.K. machinery (an increase of 6.7 per cent), five days for U.S. ma-
chinery (4.7 per cent), and eighteen days for West German mach inery
(4.5 per cent). For machine-tools, the lengthening of the delivery time
was eight days for the United Kingdom (2 .6 per cent) and ten days for
the United States (4.6 per cent) . Using the work of Steuer et al., Artus
calculates that a 10 per cent increase in U.K. demand will ultimately
reduce the United Kingdom's export of machine-tools by 3 per cent.
Finally, the study of Panic and Seward (1966) may be mentioned
as an introduction to the rest of the chapter. They are concerned with
explaining the United Kingdom's declining export share , and consider
three hypotheses: declining competitiveness; unfavourable commodity
structure; and unfavourable market structure. They contend that the
price argument is not convincing because, at least over the period 1959
to - 1964 , export prices rose no faster than in other countries. They also
dismiss the unfavourable commodity structure argument on the
grounds that the U.K. share of total exports in expanding and declin ing
commodities in world trade is similar to other countries. The unfavour-
able market structure argument is dismissed as well since they note
that the area pattern of trade has shifted towards the E.E.C. countries,
which is one of the most rapidly expanding markets . Panic and Seward
favour a supply-side explanation of the United Kingdom's declining
share of world trade . This could either mean capacity difficulties in
Export Functions 229
a supply sense, or the production of goods with unfavourable charact-
eristics giving them a lower world income elasticity of demand than
similar goods of other countries.
Before turning to the evidence on the relation between exports and
world income , and between exports and relative prices , a word of
warning is in order in interpreting an inverse relation between a
country's share of total world exports and the internal pressure of
demand . If the elasticity of demand for a country 's exports with
respect to the growth of world trade is less than unity, the country's
share of exports is bound to fall as world trade grows and to rise as
world trade slumps . If cycles of expansion and contraction within
the domestic economy broadly coincide with cycles in world trade ,
the finding of an inverse relation between export share and the internal
pressure of demand may be a reflection not simply of capacity con-
straints but of many factors, including the characteristics of goods ,
causing the elasticity of demand for exports with respect to world trade
to be less than unity. The U.K. Treasury's estimate of this elasticity for
U.K. exports is approximately 0 .6.

EXPORTS AND WORLD INCOME (ESTIMATES OF INCOME ELASTICITIES)

All the evidence we have indicates that it is the growth of world income
and world trade that is the major determinant of the growth of a
country's exports, with the United Kingdom being no exception . In
all export demand functions the variable standing for world income
or world trade is always significant, and in most cases is the only
variable that appears to matter. The United Kingdom's problem, which
we shall elaborate upon in Chapters 10, 11 and 12 , is that the world
income elasticity of demand for her exports is much lower than for
other industrialised countries, with the exception perhaps of the United
States, which makes the rate of growth of U.K. exports relatively low.
Houthakker and Magee (1969) make the following international
estimates of the income elasticity of demand for exports over the
period 1951 to 1966 : United States 0.99 ; Canada 1.41 ; United
Kingdom 0.86 ; Japan 3.55 ; West Germany 2.44; Italy 2.96; Netherlands
1.88; France 1.53; Belgium 1.74 ; Sweden 1.76; Australia 1.18; Switzer-
land 1.47; Denmark 1.69 ; and Norway 1.58 . Disaggregated work,
using export data classified by M.L.H.s of the Standard Industrial
Classification, gives the money income elasticities of demand for
230 Balance-of-Payments Theory and U.K. Experience

U.K. exports shown in Table 9.1. The money income elasticities will
be overestimates of the real income elasticities if U.K. export prices
have risen less than world prices and the demand for exports is price
elastic, and underestimates if the demand for exports is price inelastic .
If U.K. export prices have risen more than world prices, the converse of
the above will be true . The estimates of the income elasticities are
invariably lower than the income elasticities of demand for imports
in the equivalent industries, reported in Table 8.3 above . The weighted
average of the estimated income elasticities of demand for exports
by industry is approximately unity, which supports the aggregate
estimates of approximately unity by Houthakker and Magee and other
investigators. The growth implications for a country in which the
world income elasticity of demand for exports is lower than the domes-
tic income elasticity of demand for imports were outlined in Chapter I
and will be elaborated upon in Chapter 10.

EXPORTS AND RELATIVE PRICES (ESTIMATES OF PRICE ELASTICITIES)

There are a number of factors which affect a country's price competit-

iveness in world markets : the domestic rate of inflation ; production
costs per unit of output ; and changes in the exchange rate . This makes
it difficult to define one unique measure of price competitiveness.
Rather , there are a number of complementary measures, each with
certain advantages and disadvantages? The appropriateness of the
various measures will depend partly on the nature of the market being
analysed . In very competitive markets , for example , where a virtually
identical good is being sold, relative prices can hardly change and an
index of relative prices is unlikely, therefore , to be a good predictor
of sales. In this case some measure of profitability, or relative costs,
would be more useful.
The basic measure of export price competitiveness used in most
studies is the ratio of export prices to a weighted average of the export
prices of major competitors expressed in a common currency. The
measure suffers from a number of limitations, however. First, as men-
tioned above, it is not appropriate for use if the market is very compet-
itive. Second, the index is based on unit values which do not make
allowance for changes in the composition of exports. Third, the index
measures competitiveness only in relation to the exports of competitors
and leaves out of account competitiveness in relation to domestic
 TABLE 9.1
Ordered money income elasticities of demand for exports
Minimum Industry Income elasticity
List Heading of demand for exports
002 Forestry 2.753
261 Coke ovens and manufactured fuel 2.718
214 Bacon-curing, meat and fish products 2.557
471 Timber 2.493
003 Fishing 2.382
419 Carpets 2.224
472 Furniture and upholstery 2.200
102 Stone and slate quarrying and mining 2.133
216 Sugar 2.108
444 Overalls and men 's shirts, underwear, etc . 2.014
262 Mineral oil refining 2.009
229 Food industries not elsewhere specified 1.966
433 Fur 1.832
364 Radio and other electronic apparatus 1.803
499 Miscellaneous manufacturing industries 1.792
109 Other mining and quarrying 1.756
496 Plastics moulding and fabricating 1.740
352 Watches and clocks 1.727
001 Agriculture and horticulture 1.716
464 Cement 1.709
442 Men's and boys ' tailored outerwear 1.694
461 Bricks, fireclay and refractory goods 1.691
~ 215 Milk products 1.679
...... 337 Mechanical handling equipment 1.653
 TABLE 9.1 (contd)
~ 417 Hosiery and other knitted goods 1.622
271 Chemicals and dyes 1.600
483 Manufactures of paper and board not elsewhere specified 1.552
396 Jewellery, plate and refining of precious metals 1.549
219 Animal and poultry foods 1.517
231 Brewing and malting 1.513
351 Scientific, surgical and photographic instruments, etc . 1.469
218 Fruit and vegetable products 1.452
412 Spinning and doubling of cotton, flax and man-made fibres 1.444
481 Paper and board 1.444
272 Pharmaceutical and toilet preparations 1.435
493 Brushes and brooms 1.431
276 Synthetic resins and plastics materials 1.341
411 Production of man-made fibres 1.324
393 Bolts, nuts, screws, rivets, etc . 1.323
482 Cardboard boxes, cartons and fibre-board packing cases 1.315
312 Steel tubes 1.285
383 Aircraft manufacturing and repairing 1.279
369 Other electrical goods 1.279
445 Dresses, lingerie, infants' wear, etc. 1.278
273 Explosives and fireworks 1.276
263 Lubrica ting oils and greases 1.276
277 Polishes, gelatine, adhesives, etc. 1.272
469 Abrasives and building materials, etc ., not elsewhere specified 1.268
313 Iron castings, etc . 1.268
213 Biscuits 1.260
321 Light metals 1.255
217 Cocoa, chocolate and sugar confectionery 1.254
336 Contractors' plant and quarrying machinery 1.246
 361 Electrical machinery 1.237
275 Vegetable and animal oils, fats, soap and detergents 1.233
339 Other machinery 1.213
211 Grain-milling 1.206
349 Other mechanical engineering not elsewhere specified 1.202
479 Miscellaneous wood and cork manufactures 1.194
491 Rubber 1.189
274 Paint and printing ink l.180
399 Metal industries not elsewhere specified 1.179
492 Linoleum , leather cloth, etc . 1.172
463 Glass 1.160
462 Pottery 1.145
432 Leather goods 1.138
389 Perambulators, hand-trucks, etc . 1.121
341 Industrial plant and steelwork 1.118
494 Toys, games and sports equipment 1.117
240 Tobacco 1.116
365 Domestic electrical appliances 1.1 07
495 Miscellaneous stationers' goods 1.089
391 Tools and implements 1.082
429 Other textile industries 1.082
333 Engineers' small tools and guages 1.082
103 Chalk, clay, sand and gravel extraction 1.076
385 Railway carriages and wagons and trams 1.059
443 Women's and girls' tailored outerwear 1.049
239 Other drink industries 1.043
322 Copper, brass and other base metals 1.035
~ 362 Insulated wires and cables 1.017
 TABLE 9.1 (contd)
~ 489 Other printing, publishing, bookbinding, engraving, etc . 1.017
363 Telegraph and telephone apparatus 1.009
449 Dress industries not elsewhere specified 1.008
413 Weaving of cotton , linen and man-made fibres 1.000
431 Leather (tanning and dressing) and fellmongery 0.976
335 Textile machinery and accessories 0.962
381 Motor-vehicle manufacturing 0.939
382 Motor-cycle, three-wheel vehicle and pedal cycle 0.937
331 Agricultural machinery (except tractors) 0.924
332 Metal-working machine-tools 0.889
422 Made-up textiles 0.881
486 Printing, publishing of newspapers and periodicals 0.850
394 Wire and wire manufactures 0.849
338 Office machinery 0.819
334 Industrial engines 0.819
311 Iron and steel (general) 0.776
475 Wooden containers and baskets 0.763
441 Weatherproof outerwear 0.718
414 Woollen and worsted 0.709
450 Footwear 0 .705
421 Narrow fabrics 0.676
392 Cutlery 0.658
395 Cans and metal boxes 0.633
415 Jute 0.627
473 Bedding, etc . 0.582
446 Hats, caps and mininery 0.514
370 .1} Shipbuilding and ship-repairing
370.2 Marine engineering 0.488
Export Functions 235
producers in the various export markets. This weakness could in
principle be overcome by calculating the ratio of export prices to
some weighted combination of competitors' export prices and domestic
producers' wholesale prices in the export markets . A fourth limitation
is that the index measures only delivery prices and not quotations,
and therefore only reflects trade that actually takes place rather than
underlying competitive conditions. Strictly speaking, any price at which
business is actually transacted is by definition competitive. Moreover,
as Posner and Steer (1979) have noted , if costs rise in a marginally
competitive export industry, forcing the industry's product out of the
export market, the aggregate index may register an improvement in
competitiveness! The basic problem is that the index may not give a
good indication of the ability of the export sector to produce as effic-
iently as foreign competitors, or the willingness of exporters to supply
(which depends on profitability). Efficiency and cost changes will
be reflected in profitability in competitive markets where the export
price is determined by world prices. To measure the relative profit-
ability of exports, the ratio of export prices to wholesale prices may be
taken, as in Table 9.2 .
The measure of competitiveness preferred by the U.K. Treasury,
which is used by the I.M.F., and which best explains export trends in
the work of Enoch (1978) at the Bank of England is, however, relative
unit labour costs normalised for cyclical variations in output so that
labour costs are related to potential output rather than to actual out-
put. This measure is claimed to have several advantages. First , it covers
potential as well as actual exporters. Second, it measures, in effect ,
a combination of both price competitiveness and profitability . Third ,
it overcomes in principle the problem of the export price index incorp-
orating only successful export quotations. The problems with the index
are that labour costs are not the only costs of production, and that
producers in one country may work on a lower profit margin than in
others. Which index to use, however, is essentially an empirical
question, and if the normalised unit labour cost index relative to
other countries performs the best in explaining export trends then it
should be preferred not only for forecasting but as a guide to policy.
The different measures of competitiveness discussed above are
shown in Table 9.2 for the United Kingdom over the period 1970
to 1980 . There is no presumption, of course, that in the base year
U.K. producers were necessarily competitive in any absolute sense.
Lower values of the indices represent a more competitive position .
236 Balance-of-Payments Theory and U.K. Experience

TABLE 9.2
Measure of competitiveness of U.K. manufactured goods (1970 = 100)

Relative export Relative normal unit Relative profitability

prices labour costs of exports

1970 100.0 100.0 100.0

1971 102.2 103.1 95.6
1972 102 .6 100.8 95 .2
1973 94 .1 91.8 98.0
1974 93.1 97.9 98.1
1975 96.5 99.2 100.0
1976 94 .8 90.8 101.2
1977 99.3 85.8 102 .7
1978 105.1 97 .1 102.9
1979 112.1 112.1 99.8
1980 124.7 137.8 97 .3
Source : Econom ic Trends, H.M.S.O. (various editions, particularly
1980) .

Sometimes the indices move in the same direction , at other times

not. Some interesting features are apparent, however. First , despite the
depreciation of the pound from 1972 to 1977, which reduced relative
export prices init ially , the index in 1977 was back to its 1970 level with
the advantage of depreciation apparently eroded by domestic inflation.
Second , relative normal unit labour costs declined over the period from
1972 to 1977 and the relative profitability of exporting improved .
From this evidence of an over-all improvement in competitiveness one
might have expected a considearble export-volume response were it not
for the overriding importance of world income and world trading con-
ditions in the determination of exports, both of which were depressed
by the oil price rises in 1973-4. From 1977 to 1980 competitiveness
and the relative profitability of exporting deteriorated , but export vol-
ume continued to rise. How much different export performance would
have been had there not been changes in competitiveness, however, is
an open question , depending on one 's judgement of the export price
elasticity, which we now consider .
The empirical evidence available for the United Kingdom shows that
export volume is responsive to a change in price competitiveness,
though the response lag for the price elasticity of demand to exceed
unity may exceed two years. Some studies conclude that the price
elasticity never exceeds unity. Some representative estimates of the
price elasticity of demand for U.K. exports, using different measures
of competitiveness and different sample data , are as follows: Hutton
Export Functions 237

and Minford (1975), 1.5; Duffy and Renton (1970), 2.5 ; Houthakker
and Magee (1969),0 .44; Junz and Rhomberg (1965),1.86 (short run),
2.27 (long run) ; and Batchelor and Bowe (1974); 1.13 to 2.80 . Posner
and Steer (1979) have summarised the estimates of export price
elasticities implied by the major U.K. macroeconomic models (see
Table 9.3).

TABLE 9.3

Empirical estimates of export price elasticities for the United Kingdom

One Two Long Period of

year years run estimation

London Business School 0.37 0.39 0.99 1951-76

Cambridge Economic Policy
Group 0 .35 1.23 2.36 1960-76
National Institute of
Economic and Social Research 0.46 1.13 1.61 1967-75
Treasury 0.26 0.65 1.30 1966-76

There would seem to be a general consensus among investigators

that the short-run elasticity is very low, but after two years the elas-
ticity is probably of the order of between 1.5 and 2.0, so that the
Marshall-Lerner condition for a successful devaluation would be ful-
filled if all the other necessary conditions are met. We repeat, however,
that there is no guarantee that an induced change in competitiveness
through exchange depreciation can be retained for any length of time,
so that any balance-of-payments gain is temporary. Second , an improve-
ment in competitiveness may not be enough to counter other adverse
factors affecting export performance. Third, as we shall see in the next
chapter, a once-for-all change in competitiveness cannot raise the
balance-of-payments equilibrium growth rate permanently. A continual
improvement in competitiveness would be required .
Studies which relate a country's export share to relative prices are
estimating the elasticity of substitution of one country's exports for
another. Thus the estimates are a combination of the 'own' price
elasticity and the cross elasticity of demand in both sets of countries.
Special assumptions would have to be made to deduce the 'own' price
238 Balance-ofPayments Theory and U'K. Experience

elasticity of demand for exports for anyone of the countries . Junz and
Rhomberg (1965, 1973) find a relation between relative price changes
and market shares, but only after a lag of three years, with the full
effect coming after five years . The share elasticity for the United
Kingdom is estimated to be approximately 1.7 in their 1973 study and
3 in their 1965 study.

THE UNITED KINGDOM'S DECLINING SHARE OF WORLD EXPORTS

The U.K. share of world exports has fallen from over 20 per cent in
1950 to under 10 per cent in 1978. Some observers have taken this
fact as a measure of the weakness of the UK . balance of payments.
It is difficult to subscribe to this view without some qualification.
Indeed, Harrod (1967a) described the view as 'surely the most absurd
ever perpetrated in a diagnosis', adding that 'the British share of world
exports has declined, is declining and will continue to decline, hope-
fully at an accelerated pace'. Without wishing to concur with the last
sentiment, Harrod is right that the decline should occasion no surprise,
and there is nothing incompatible between a declining share of total
world exports and a healthy balance of payments at a satisfactory
growth rate . The health of the balance of payments depends on the
relation between the absolute level of exports and imports associated
with a given growth rate, not on the growth of exports relative to the
growth of total world exports. While the UK . export share has fallen,
the U.K. share of world imports has also fallen. The decline in both
shares has a common explanation, namely that the U.K. share of the
world's manufacturing population and output has declined as resources
have shifted from agricultural to industrial activities throughout the
world. The United Kingdom's legitimate worry is that some of her
European neighbours, even in the face of these changing world con-
ditions, have managed to retain their share of world exports. It is not
without interest , therefore, to analyse this declining share. Is it that
the United Kingdom is exporting commodities the demand for which
is growing slower than for all commodities? Is it that the United King-
dom is exporting commodities to markets where demand is rising
more slowly than in other markets? Or is the United Kingdom losing
its share in all commodity and geographic markets regardless of whether
they are fast- or slow-growing? A classificatory device is available to
help answer these questions which divides a country's changing share
Export Functions 239
of world trade between any two periods into three components: (i)
that part due to the commodity composition of trade, which may be
called the differential product growth effect ; (ii) that part due to the
destination of trade , which may be called the differential market
growth effect ; and (iii) that part due to the decreasing share of all
individual commodity and geographic markets because of growing
price uncompetitiveness, non-price factors, supply constraints, low
income elasticities of demand, and so on . This third component is
obtained as a residual and is sometimes, misleadingly, called a 'compet-
itiveness' effect . As we shall see from the studies that have been done ,
the United Kingdom's declining share of world exports cannot be
accounted for by the commodity composition or destination of trade
and must therefore be the result of an over-all declining share of every
market . The classificatory device to be outlined is taken from Magee
(1974), but the methodology is a standard one widely used in regional
analysis for analysing interregional differences (see Dixon and Thirlwall,
1975b).
An individual country is denoted by the subscript k , the world by
subscript w, the base year by 0 , and the terminal year by t , A country's
base-year share of world exports can therefore be expressed as X ok/
X ow' and its terminal-year share as Xtk/X tw' The relationship between
the two rat ios is given by (I + Kk)/(I + Kw ) = R , where K stands for
the rate of growth of exports and R can be called the relative growth
factor . Thus :

(9.4)

Another way of writing R is:

R=LJ~}
11 + s;
i (9.5)

where the is stand for individual commodities. Although the terms in

the denominator contain no i superscripts they are kept to the right
of the summation sign for reasons which will become clear shortly.
The numerator of (9 .5) can be expanded to :

(9.6)
240 Balance-of-Payments Theory and U.K. Experience

where the js refer to individual markets (destinations) . Equation (9.5)

can thus be written:

(9.7)

Multiplying top and bottom of equation (9 .7) by (l + g~) xJk gives:

Multiplying top and bottom of equation (9 .8) by (I + g!b gives:

~(l +g~) X~",,(l +gp) (1 +g!J)X~jk
R=LJ--- ---~-- . . _ ..- . - -.-
i (1 + gw) X ok j (I + g:J) (l + g~) X~k (9 .9)

(a) (b) (c) (d) (e)

We now have the relationship between a country's share of exports in

the base and terminal year as the product of three factors : 3
(i) (a) x (b) : that is, a country's export share may change because
the country specialises in commodities which grow relatively faster/
slower than the average for all commodities. This is the differential
product growth effect.
(ii) (d) x (e) : that is, a country's export share may change because
the country concentrates its products in markets which grow relatively
faster/slower than the average for all markets. This is the differential
market growth effect.
(iii) (c) : that is, a country's export share may change because the
country's share in every market is increasing/decreasing due to a host of
factors connected with competitiveness and the characteristics of the
products exported.
It is obvious that even if a country maintains its share of every
product in every market «c) = I), its share of total exports may still
go down because its exports are orientated towards relatively slowly
growing products or slowly growing markets.
To estimate the actual magnitude by which exports have suffered
Export Functions 241
as a result of a declining share of world trade due to the three 'explan-
ations ' above, the same methodology can be applied but in a slightly
different way. First, define the difference between actual exports and
what they would have been had they grown at the world rate, thereby
maintaining the share constant :

(9.1 0)

The effect of product composition difference is:

~ (I +gJ)Xoik - ( I + gw) X o k (9.1 I)

i
The differential market growth effect is:

(9 .12)

Thus

X tk - (1 + gw) X Ok = (9.1 I) + (9.12) + a residual

where the residual

Studies which have applied the methodology for analysing changes

in the U.K. share of world exports include Major (1968) , National
Economic Development Council (1963), and Panic and Seward (1966).
Major (1968) and Kreinin (1967) attempt to estimate the change in the
absolute magnitude of exports resulting from changing shares. Panic
and Rajan (1979) have investigated the distribution of exports between
fast- and slow-growing product categories in world trade in the United
Kingdom , France , West Germany and Japan . The broad commodity
and geographic composition of U.K. exports (and imports) since 1967
is given in Tables 9.4 and 9.5 in the appendix to the chapter.
Major (1968) takes nine areas and seven commodities (sixty-three
'markets'in all) over the period 1954 to 1966, during which the UK.
share of world exports of manufactures fell from 20.9 per cent to
l3 .1 per cent. He finds that only 9 per cent (or 2 percentage points)
of the fall in share can be accounted for by the area/commodity comp -
osition of trade . The major part of the declining share is attributed to
242 Balance-of-Payments Theory and U.K. Experience

a falling share in every market. In fact, of the sixty-three markets, in

only eleven did the share not fall. The cumulative value of lost exports
(compared with the maintenance of a constant share) is estimated at
~31.7 billion . Since biases in trade towards slowly expanding com-
modities and markets cannot account for the major part of the United
Kingdom's declining share of total manufacturing exports, Major con-
cludes that the loss must therefore be due to 'declining competitiveness
in the broadest sense' . The National Economic Development Council
(1963) reaches the same conclusions for the period 1954 to 1960, and
gives the following reasons for the declining share : the ending of
discrimination in favour of the United Kingdom in sterling area
countries; the relatively slow rate of expansion of output and produc-
tivity in the United Kingdom ; the faster increase in costs and prices in
the United Kingdom compared with other countries; and poor design,
quality, delivery and salesmanship .
Panic and Seward (1966) examine the period 1959 to 1964, during
which the decline in the U.K. share of total exports was from 16.7
per cent to 13.3 per cent. The results of their study were given earlier
in the chapter (p .229). They dismiss trade bias and price uncornpetitive-
ness as explanations of the declining share. and argue that it is impos-
sible to divorce the problem of the slow growth of exports from that of
slow economic growth in general . If growth and change are slow, the
supply of exports and their quality are likely to be inferior to those of
goods available from other parts of the world .
Panic and Rajan (1979), in their international comparison of the
structure of trade according to the growth performance of industries
over the period 1955 to 1973, find that the United Kingdom's disad -
vantage (compared with France, West Germany and Japan) lies not in
the broad product structure of her exports but in the poor relative
performance in all markets.
The upshot of this extensive research on the United Kingdom's
declining share of world exports is that the low elasticity of demand
for U.K. exports with respect to the growth of world trade is not the
result of a heavy weighting of goods towards slow-growing markets or
slow-growing products; rather, that in the same markets and in the
same goods (broadly defined) as other countries, U.K. products sell
less well. In other words the problem lies with the supply and character-
istics of the goods themselves, not the markets in which they are sold
or the range of goods produced. It could be, of course, that within
the broad categories of goods defined, the United Kingdom is produc-
ing a product the market for which is growing relatively slowly, and
Export Functions 243
that a more disaggregated analysis would show the commodity bias of
trade to be an important explanation of the United Kingdom's declin-
ing share of world exports . Barna (1963) addressed himself to this
question by examining forty fast- and slow-growing products within
the industries of machinery, transport equipment and chemicals. He
found that the UK. share of fast-growing industries in world trade,
even at a disaggregated level, is only marginally below the average for
other major countries studied . Having earlier expressed doubts about
the conclusions of more aggregative studies (The Times, April 1963) ,
Barna supports their conclusions by finding that the United Kingdom's
slow export growth has been the result of a slower increase in the sales
of both fast- and slow-growing products. The initial bias in trade is
apparently not important.
Accepting this fundamental point , on which all the studies seem to
agree, the deeper question is still why UK . products of a similar type
to those of other countries are treated as inferior. The explanation
would seem to lie with the view expressed by Panic and Seward (1966),
and those who have advanced a nee-technical theory of trade (e .g.
Kravis, 1956) , that the slow growth of the UK. economy and its lack
of investment and technical progressiveness have made the country
weak in the production of the most recently developed and technically
advanced products. It is the newness and technical sophistication of
products, together with their marketing, which make them income
elastic in world markets, ensuring that shares rise in every market. The
importance of the income elasticity of demand for exports in world
markets, in relation to the domestic income elasticity of demand for
imports, in determining a country's growth rate is discussed in the next
chapter. The virtuous circle that may be set up between higher export
growth and higher output growth is discussed in Chapter 11, and some
of the non-price factors that determine the growth of exports are
briefly discussed in Chapter 12. The view taken here is that while the
slow growth of exports may well be a reflection of the slow growth of
the economy, the growth of exports itself must be raised before the
rate of growth of the economy can be improved, unless the income
elasticity of demand for imports can be drastically reduced. Otherwise
there will be an intolerable balance-of-payments strain in the transition
from expanding demand to raise the growth rate and the benefits from
this feeding through to export performance. To raise the rate of growth
of exports requires raising the income elasticity of demand for exports
in world markets . Exchange depreciation will not work, and any
improvement in internal price competitiveness would have to be on a
continuous basis to raise the rate of growth of exports permanently.
 APPENDIX
t COMMODITY AND GEOGRAPHIC COMPOSITION OF U.K. EXPORTS AND IMPORTS

TABLE 9.4

Commodity composition of visible trade [Em.]

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

EXPORTS
Food, beverages and tobacco 505 569 638 849 1034 1388 1654 2181 2867 2905 3241
Basic materials 269 279 317 420 583 556 807 963 1036 1290 1467
Mineral fuels and lubricants 210 239 242 374 781 827 1265 2092 2375 4324 6418
Semi-manufactured goods 2782 2958 3058 4013 5630 5851 7996 10122 10889 12623 14152
Finished manufactured goods 4100 4734 4900 5910 7764 9987 12654 15467 16852 18332 20727
Commodities and transactions
not classified according to kind 284 264 282 371 602 721 815 903 1044 1204 1384

Total 8150 9043 9437 11937 16394 19330 25191 31728 35063 40678 47389
 TABLE 9.4 (contd)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

IMPORTS
Food, beverages and tobacco 1846 1969 2136 2832 3475 4089 4617 5483 5659 5955 5619
Basic materials 1178 1097 1128 1692 2213 1967 3022 3427 3201 3600 3413
Mineral fuels and lubricants 691 952 949 1320 4145 3912 5241 4905 4511 5407 6589
Semi-manufactured goods 2323 2276 2601 3731 5682 5355 7014 8492 9568 11889 12523
Finished manufactured goods 1997 2401 3212 4704 5742 6746 8719 11211 13107 16376 16862
Commodities and transactions
not classified according to kind 149 158 159 244 488 594 507 494 559 909 1205

Total 8184 8853 10185 14523 21745 22663 29120 34012 36605 44136 46211

~
~

TABLE 9.4 (contd)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

VISIBLE BALANCE
Food, beverages and tobacco -1341 -1400 - 1498 - 1983 - 244 1 -2701 - 2963 -3302 -2792 -3050 - 2378
Basic materials -909 - 818 - 811 - 1272 - 1630 - 14 11 - 2215 - 2464 -2165 -2310 - 1946
Mineral fuels and lubricants - 481 -713 - 707 - 946 -3364 -3085 - 3976 - 2813 - 2136 -1083 - 171
Semi-manufactured goods +459 +682 +457 +282 - 52 +496 +982 +1630 +1321 +734 +1629
Finished manufactured goods +2103 +2333 +1688 +1206 +2022 +3241 +3935 +4256 +3745 +1956 +3865
Commodities and transactions
not classified according to kind +135 +106 +123 +127 +114 +127 +308 +409 +485 +295 +179

Total -34 +190 - 74 8 - 2586 -5351 - 3333 - 3929 - 2284 -1542 - 3458 +1178

Source: Central Statistical Office, U.K. Balance of Payments 1981 Edition (London : H.M.S.O., 1981).
 TABLE 9.5

Geographic composition of visible trade (£m.).

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

VISIBLE TRADE ON A BALANCE.QF·PAYMENTS BASIS:

EXPORTS
European Community" 2416 2536 2849 3851 5546 6227 8936 11674 13348 17306 20422
Other Western Europe 1330 1450 1551 1951 2623 2972 3862 4615 4385 5661 6838
North America 1226 1422 1559 1879 2278 2316 3065 3773 4219 4786 5285
Other developed countries 981 1085 940 1209 1719 1835 1932 2075 2302 2485 2657
Oil exporting countries 477 584 641 794 1225 2275 3172 4335 4680 3667 4822
Rest of world 1720 1966 1897 2253 3003 3705 4224 5256 6129 6773 7365

Total - all areas 8150 9043 9437 11937 16394 19330 25191 31728 35063 40678 47389

~
"I
~

TABLE 9.5 (contd)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

IMPORTS
European Community" 2325 2720 3441 5178 7680 8734 11194 13605 15863 19935 19713
Other Western Europe 1295 1470 1733 2462 3219 3240 4146 4795 5224 6908 6907
North America 1706 1585 1633 2153 2964 2955 3887 4585 4953 5853 6946
Other developed countries 822 851 1083 1410 1615 1852 2040 2627 2714 2771 2896
Oil exporting countries 597 802 775 1122 3393 2948 3854 3421 3033 2963 3980
Rest of world 1439 1425 1520 2198 2874 2934 3999 4978 4818 5706 5769

Total - all areas 8184 8853 10185 14523 21745 22663 29120 34012 36605 44136 46211
 TABL E 9.5 (contd)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

VISIBLE BALANCE
European Community" +91 - 184 - 592 -1327 -2134 -2507 - 2258 -1932 - 25 15 - 26 29 +709
Other Western Europe +35 - 20 - 182 - 5 11 - 596 -268 -284 - 180 - 839 - 1247 - 69
North America - 480 - 16 3 - 74 -274 - 686 -639 -822 - 812 - 734 - 1067 - 166 1
Other developed countries +159 +234 - 143 - 20 1 +104 - 17 - 108 - 552 - 4 12 - 286 - 239
Oil exporting coun tries - 120 -218 - 134 -328 - 2168 - 673 - 68 2 +914 +1647 +704 +842
Rest of world +281 +541 +377 +55 +129 +771 +225 +278 +1311 +1067 +1596

Total - all areas - 34 +190 - 748 -2586 - 5351 -3333 - 392 9 - 2284 - 154 2 - 3458 +1178

• Figures for all years relate to the nine countries.

Source : Central Statistical Office, U.K. Balance of Payments 1981 Edition (London: H.M.S.O., 1981) .

~
'0
10
The Balance-of-Payments
Equilibrium Growth Rate
In the long run a country cannot grow faster than the rate of growth
of output consistent with balance-of-payments equilibrium on current
account. 1 We call this growth rate the balance-of-payments equilibrium
growth rate 0'B)' In this chapter the import and export functions
specified in Chapters 8 and 9 are used to determine the balance-of-
payments equilibrium growth rate and to highlight its major determin-
ants . An attempt is then made to estimate the balance-of-payments
equilibrium growth rate for a variety of countries, including the United
Kingdom, using some simplifying, though not unrealistic , assumptions.
It is shown how closely the actual growth experience of countries
approximates to the rate of growth of exports divided by the income
elasticity of demand for imports (i .e. x/rr), except in the case of Japan,
which has grown much slower than its balance-of-payments equilibrium
rate and built up a massivebalance-of-payments surplus. The findings
of this chapter underline the importance of raising the rate of growth
of exports to improve the balance of payments permanently, and lend
support to export-led growth models (which are considered in the next
chapter).
The importance of a healthy balance of payments for growth can be
stated quite succinctly. If a country gets into balance-of-payments
difficulties as it expands demand before the short-term capacity growth
rate is reached, then demand must be curtailed, supply is never fully
utilised, ihvestrnent is discouraged, technological progress is slowed
down, and a country's goods compared with foreign goods become less
desirable, so worsening the balance of payments still further, and so on.
A vicious circle is initiated. By contrast, if a country is able to expand
The Balance-ofPayments Equilibrium Growth Rate 251

demand up to the level of existing productive capacity, without balance -

of-payments difficulties arising, the pressure of demand upon capacity
may well raise the capacity growth rate. There are a number of possible
mechanisms through which this may happen : the encouragement to
investment, which would augment the cap ital stock and bring with it
technological progress; the supply of labour may increase by the entry
into the work-force of people previously outside or from abroad ; the
movement of factors of production from low-productivity to high-
productivity sectors ; and the ability to import more may increase
capacity by making domestic resources more productive . It is this
argument that lies behind the advocacy of export-led growth, because
it is only through the expansion of exports that the growth rate can be
raised without the balance of payments deteriorating at the same time .
Believers in export-led growth are really postulating a balance -of-
payments constraint theory of why growth rates differ . It should be
stressed, however , that the same rate of export growth in different
countries will not necessarily permit the same rate of growth of output
because the import requirements associated with growth will differ
between countries, and thus some countries will have to constrain
demand sooner than others for balance-of-payments equilibrium . The
relation between a country's growth rate and its rate of growth of
imports is the income elasticity of demand for imports.

THE DETERMINATION OF THE BALANCE.()F-PAYMENTS EQUILIBRIUM

GROWTH RATE

Balance-of-payments equilibrium on current account measured in

units of the home currency may be expressed as

(10.1 )

where X is the quantity of exports, Pd is the price of exports in home

currency, M is the quantity of imports, Pf is the price of imports in
foreign currency, E is the exchange rate (i.e. the home price of foreign
currency), and t is time. In a growing economy, starting from equili-
brium, the condition for balance-of-payments equilibrium through time
is that the rate of growth of the value of exports equals the rate of
growth of the value of imports, i.e ,
252 Balance-ofPayments Theory and U.K. Experience
(10 .2)

where lower-case letters represent rates of change of the variables.

Following Chapter 8, but omitting capacity as a variable (for simpli-
city), the quantity of imports demanded may be specified as a multi-
plicative function of the relative price of imports and domestic substi-
tutes (measured in units of the home currency in order to incorporate
the effect of exchange-rate changes) and of domestic income. Thus :

(10.3)

where 1/1 is the price elasticity of demand for imports (1/1 < 0), Y
is domestic income, and 1r is the income elasticity of demand for
imports (1r > 0). The rate of growth of imports may be written:

(lOA)

where lower-case letters again represent rates of change of the variables.

Following Chapter 9 , again omitting the capacity variable (for
simplicity), the quantity of exports demanded may also be expressed
as a multiplicative function , in which the arguments in the demand
function are the price of domestic goods relative to foreign goods
(measured in units of foreign currency to capture the effect of ex-
change-rate changes) and the level of world income. Thus:

(10 .5)

where X, is the quantity of exports, Pdt is the domestic price of

exports, PIt is the foreign price of goods competitive with exports,
Zt is the level of world income, l/Et is the foreign price of home
currencyv n is the price elasticity of demand for exports (11 < 0), e is
the world income elasticity of demand for exports (e > 0), and t is
time.
The rate of growth of exports may be written :

(10.6)
The Balance-ofPayments Equilibrium Growth Rate 253
Substituting equations (lOA) and (10 .6) into (10 .2) we can solve for
the rate of growth of domestic income consistent with' balance-of-
payments equilibrium (YSt) :

YS t = ~~l _ (Pdt - Ptt - et) + € (Zt)

n (10.7)

Remembering the signs of the parameters (1'/ < 0 , t/J < 0, e > 0 , and rr
> 0) , equation (10 .7) expresses several fundamental economic propo-
sitions:
(i) Domestic prices rising faster than foreign prices will worsen the
balance-of-payments equilibrium growth rate if the sum of the price
elasticities of demand for exports and imports is greater than unity
in absolute value [i .e. ifl1'/ + t/J,I> 1).
(Ii) Devaluation or currency depreciation, I.e. a rise in the home
price of foreign currency (e, > 0) will improve the balance-of-payments
equilibrium growth rate provided the sum of the price elasticities of
demand for imports and exports exceeds unity in absolute value , which
is the so-called Marshall-Lerner condition (i .e. if I 1'/ + t/J I> 1).2 Notice,
however, the important point that a once-for-all depreciation of the
currency cannot raise the balance-of-payments equilibrium growth
rate permanently. After the initial dep reciation, et = 0 , and the growth
rate would revert to its former level. To raise the balance-of-payments
equilibrium growth rate permanently would require continual depreci-
ation , i.e . et > 0 in successive periods.
(iii) A faster growth of world income will raise the balance-of-
payments equilibrium growth rate.
(iv) The higher the income elasticity of demand for imports (n),
the lower the balance-of-payments equilibrium growth rate .

The interesting question is, how well does equation (10 .7) predict
the actual growth experience of countries, and can we derive any simple
rule or 'law'? First of all, it must be noted that there may be an asym-
metry in the system. While a country cannot grow faster than its
balance-of-payments equilibrium growth rate for very long , unless it
can fmance an ever-growing deficit, there is little to stop a country
growing slower and accumulating large surpluses. In particular this may
occur where the balance-of-payments equilibrium growth rate is so high
that a country simply does not have the physical capacity to grow at
254 Balance-of-Payments Theory and U.K. Experience

that rate . This typifies many oil-producing countries, and would also
seem to typify the experience of Japan, as we shall see below.
To calculate the balance-of-payments equilibrium growth rate from
equation (10 .7) for a number of countries requires a substantial amount
of data and estimates of parameters and variables which are not readily
available. If the assumption is made, however, that there can be no
long-run change in relative prices measured in a common currency,
l,e. (Pdt - Pft - et ) = 0 , equation (10.7) reduces to:
e (Zt)
YBt=
rr

or, from equation (10.6) on the same assumption,

xt
YBt= -
rr (10 .9)

Such a simple formula from a complicated expression certainly makes

the empirical task easier. How realistic is the assumption used to derive
the simple formula? Many models (see Ball, Burns and Laury, 1977 ;
Wilson, 1976) suggest that over the long period there can be little
movement in relative international prices measured in a common
currency , either because of arbitrage (the law of one price), or because
exchange depreciation forces up domestic prices equiproportionately.
In the expression (Pdt - Pft - et ) a rise in et either causes Pft to fall by
the same amount (the law of one price) or causes Pdt to rise by the
same amount, or a combination of both, so that Pdt - Pft - et = O.

A FUNDAMENTAL LAW OF GROWTH

Applying equation (10.9) to international data gives a remarkable

approximation to the growth experience of many countries over the
last twenty years . It might almost be stated as a fundamental law that,
except where the balance-of-payments equilibrium growth rate exceeds
the maximum feasible capacity growth rate, the rate of growth of a
country will approximate to the ratio of its rate of growth of exports
and its income elasticity of demand for imports. The approximation
itself would seem to vindicate the assumption used to arrive at the
simple rule in equation (10 .9). The hypothesis is tested on two sets of
data on the growth of output and exports: one for the period 1953 to
 TAB L E 10 .1

Calculations of the growth rate consistent with balance-of-payments equilibrium, 1953-76

Country % change of % change in Income elasticity Balance-of-payments

real G.N .P. export volume of demand for equilibrium growth
(y) [x } imports (1T) rate from applying
equation 00 .9)

United States 3.23 5.88 1.51 3.89

Canada 4.81 6.02 1.20 5 .02
West Germany 4.96 9 .99 1.89 5.29
Netherlands 4.99 9.38 1.82 5 .15
Sweden 3.67 7.16 1.76 4 .07
France 4.95 8.78 1.62 5.42
Denmark 3.58 6.77 1.31 5.17
Australia 4 .95 6.98 0 .90 7.76
Italy 4.96 12.09 2.25 5.37
Switzerland 3.56 7.20 1.90 3.79
Norway 4.18 7 .70 1.40 5.50
Belgium 4.07 9.24 1.94 4.76
Japan 8.55 16.18 1.2 3 13.15
Austria 5.17 11.1 2 na .
United Kingdom 2.71 4.46 1.51 2.95
South Africa 4.97 6.57 0.85 7.73
Spain 5.94 11.10 n.a,
Finland 4.55 6 .63 na .
~
Source: Kern (1978) and Houthakker and Magee (1969) .
 TABLE 10.2
~
Calculations of the growth rate consistent with balance-of-payments equilibrium, 1951-73,
using data given by Cornwall (1977)

Country % change in % change in Income elasticity Balance-of-payments

G.D .P. exports of demand for equilibrium growth
(y) (x) imports rate from applying
tn) equation (10.9)

Austria 5.1* 10.7 na .

Belgium 4.4* 9.4 1.94 4.84
Canada 4.6 6.9 1.20 5.75
Denmark 4.2t 6.1 1.31 4 .65
France 5 .0 8.1 1.62 5.00
Italy 5.1 11.7 2.25 5 .20
Japan 9.5 15.4 1.23 12.52
Netherlands 5.0 10.1 1.82 5 .55
Norway 4.2 7.2 1.40 5.14
United Kingdom 2 .7 4.1 1.51 2.71
United States 3.7 5.1 1.51 3.38
West Germany 5.7 10.8 1.89 5.71

*1955-73.
t 1954-73.
Source: Cornwall (1977 . p. 162) .
The Balance-of-Payments Equilibrium Growth Rate 257

1976 (Kern, 1978) , and the other from Cornwall (1977) for the period
1951 to 1973. On the income elasticity of demand for imports,
Houthakker and Magee's (1969) estimates have been taken as applying
to the whole of these periods even though they were only estimated
for the period 1951 to 1966. They are the best consistently estimated
international estimates available, but are probably now on the low side.
The data, and the results of applying equation (10 .9), are presented in
Tables 10.1 and 10.2. In both tables there is a general tendency for the
estimates of the balance-of-payments equilibrium growth rate to be
higher than the actual growth rate, which, if true, would produce a
balance-of-payments surplus . For countries which have built up sur-
pluses the estimates are consistent with the empirical evidence. Japan is
a striking example of a country where the gap between its actual
growth rate and its balance-of-payments equilibrium growth rate has
resulted in the build-up of a huge payments surplus . Presumably Japan
could not grow faster than it did because of an ultimate capacity
ceiling. But Japan still grew considerably faster than other countries
because demand was unconstrained and induced its own supply of
factors of production . For countries which have moved into deficit
over the period, the estimate of their balance-of-payments equilibrium
growth rate must be too high. As suggested above, this may be because
the assumed income elasticity of demand for imports is an under-
estimate for the period stretching into the late 1960s and 1970s. Also,
adverse relative price movements combined with various price elasticity
conditions cannot be entirely ruled out as determinants of the balance
of payments even though they may be of minor significance compared
with income movements and income elasticities of demand for imports
and exports.
The estimate of the UK . balance-of-payments constrained growth
rate over the 1950s and 1960s and into the 1970s is approximately 2.7
to 2.9 per cent, the lowest of any of the countries in the sample. The
simple policy conclusion for most countries, including the United
Kingdom, is that if they wish to grow faster they must first raise the
balance-of-payments constraint on demand . To raise the rate of growth
of productive capacity (by improving productivity, for example),
without being able to raise the rate of growth of demand because of the
balance of payments, will merely lead to unemployment. If the balance-
of-payments equilibrium growth rate can be raised, however, by making
exports more attractive and by reducing the income elasticity of
demand for imports, demand can be expanded without producing
258 Balance-of-Payments Theory and U.K. Experience

balance-of-payments difficulties; and , within limits , demand can gen-

erate its own supply by encouraging investment, absorbing under-
employment, raising productivity growth, and so on . Thus the explan-
ation of growth-rate differences must lie primarily in differences in the
rate of growth of demand, and the major constraint on the rate of
growth of demand in most countries is the balance of payments. The
model and the empirical evidence lend strong support to the advocates
of export-led growth.
The deeper question lies in why the balance-of-payments equilibrium
growth rate differs between countries and why it is so low in the United
Kingdom. The answer must be primarily associated with the character-
istics of goods produced, which determine the income elastiticy of
demand for a country's exports, and a country's propensity to import.
For countries like the United Kingdom with a slow rate of growth of
exports, combined with a relatively high income elasticity of demand
for imports, the message is plain : the goods produced by the country
are relatively unattractive at both home and abroad .
11
Export-led Growth
THE IMPORTANCE OF EXPORT -LED GROWTH

In the last chapter the importance of raising the rate of growth of

exports was emphasised for the maintenance of balance-of-payments
equilibrium at a permanently higher growth rate. If a continuous
improvement in relative price competitiveness is ruled out as being
unattainable, the major determinant of the growth of exports is the
income elasticity of demand for exports, for any given growth of world
income .
The export-led growth models to be found in the literature on
applied growth theory , which will be elaborated on later, concentrate
more on export-led growth setting up a virtuous circle of faster growth
and faster exports than on simply relieving a country of the balance-
of-payments constraint on demand . There is a problem with these
models of the virtuous-circle type, however , which is that they do not
contain a balance-of-payments equilibrium condition or constraint.
The implicit assumption seems to be that provided it is export growth
that is the engine of growth, as distinct from domestic autonomous
demand, the balance of payments will look after itself. Indeed, it is
assumed in some models that the initial export growth and trade
surplus generates such favourable responses in the economy that the
balance-of-payments surplus actually grows. No consideration is given
to the possibility that the rate of growth of income determined by the
structure of the model may generate a rate of growth of imports greater
than the rate of growth of exports, thereby imposing a constraint on
the export-led growth rate if balance-of-payments equilibrium must be
preserved . The import side in export-led growth models seems to have
been neglected. This, of course, is not to pour cold water on export-led
260 Balance-of-Payments Theory and U.K. Experience

growth models. If the balance of payments is constraining the actual

growth rate before the capacity rate is reached , then growth led by
export demand, as opposed to other elements of demand , will raise the
constraint ; and if the actual growth rate can reach the capacity rate , the
capacity rate itself may be raised. But it cannot be taken for granted in
export-led models that there is no constraint on growth at all. If balance -
of-payments equilibrium is a requirement , the equilibrium growth
rate in an export-led growth model must reflect this requirement ,
otherwise the model may be useless for predictive purposes.
Models of export-led growth are typically very unspecific about the
precise relationship between the rate of growth of exports and income ,
and how much income growth might be associated with a given growth
of exports. It is clear from historical and present-day evidence , how -
ever, that exports tend to grow faster than income , which must make
one immediately suspicious of models, for predictive purposes at least,
which set the rate of growth of income equal to the rate of growth of
exports. There is no doubt a variety of explanations as to why the rate
of growth of exports typically exceeds the rate of growth of income
through time, but a balance-of-payments constraint, related to the
characteristics of goods traded, is one powerful explanation. It is
obvious that if the income elasticity of demand for imports is greater
than unity , and there is no continual compensating improvement in
competitiveness, an equality between the rate of growth of exports and
income would generate a higher rate of growth of imports than exports,
and income growth would sooner or later have to be curtailed. Thus,
as long as the income elasticity of demand for imports is greater than
unity , which it is for most countries, the ratio of export growth to
income growth will almost certainly show an historical tendency to
exceed unity. If the growth equation is specified as y, = 'Y (x t ) , with
'Y < I, we would predict from equations (lO .6) and (lO .7) in the
previous chapter that if relative prices measured in a common currency
do not change , 'Y = l/rr.
This weakness of traditional export-led growth models can be
rectified by incorporating the balance-of-payments equilibrium con-
dition specified in the previous chapter . Before doing so, however, it
will be useful to do two things : first , to discuss further the relationship
between the balance-of-payments constrained growth rate (YB) and the
actual and capacity rates of growth (y A and Yc); second , to consider
some of the early (unconstrained) models that have laid stress on
export growth as the key determinant of a virtuous circle of growth.
Export-led Growth 261
The possible relationships between YB' YA and Yc are now outlined :
(i) YB = YA = rc Balance-of-payments equilibrium and full employ -
ment
(ii) YB = YA < Yc Balance-of-payments equilibrium and growing un-
employment
(iii) YB < YA = Yc Increasing balance-of-payments deficit and full
employment
(iv) YB < YA < Y C Increasing balance-of-payments deficit and growing
unemployment
(v) YB > YA = Y C Increasing balance-of-payments surplus and full
employment
(vi) YB > YA < Y C Increasing balance-of-payments surplus and growing
unemployment
It is a fundamental proposition in economics that in the long run, when
all resources are fully utilised, a country's actual growth rate cannot
exceed its capacity rate determined by the rate of growth of the labour
force and the productivity of labour - the Harrodian natural rate of
growth. The U.K. experience has been that the country has rarely had
the opportunity to grow at its capacity rate because the balance-of-
payments equilibrium growth rate has been below it. The United
Kingdom has approximated to situation (iv) above. The lure of export-
led growth lies in the possibility of moving from situation (iv) to at
least (i) , if not (v) (the Japanese case), where the balance-of-payments
equilibrium growth rate lies above the capacity growth rate, allowing
the actual growth rate to equal the capacity growth rate without
balance-of-payments difficulties arising . Indeed, in this situation the
export-led growth advocates argue that the buoyancy of demand at full
employment will raise the capacity growth rate, for the reasons out-
lined in the previous chapter: the encouragement to investment and
technical progress ; the augmentation of other factors of production;
the increased mobility of factors of production from low-productivity
to high-productivity sectors; and the ability to import more may
increase capacity by making domestic resources more productive. In this
spirit Cornwall (J 977) argues persuasively that the major explanation
of growth-rate differences between countries is differences in the
pressure of demand to which supply adjusts, though for reasons we
shall consider later (and question) he does not believe that the balance
of payments has been a constraint on demand in the United Kingdom .
As far as the U.K. economy is concerned, it is difficult to know what a
262 Balance-ofPayments Theory and U'K. Experience

faster rate of growth of exports might do for the capacity growth rate
because export performance in the past has never been good enough
to escape a balance-of-payments constraint before the capacity growth
rate has been reached . It is frequently argued, however, that if only the
United Kingdom did not run into balance-of-payments difficulties
before full employment is reached, demand would not have to be
contracted, investment would remain high, and that these conditions
in the long run would raise the capacity growth rate closer to that of
other countries. While the argument is speculative , it would be surpris-
ing if the U.K. growth record relative to other countries since the
Second World War did not have something to do with the character-
istics of its trading position compared with other countries, particularly
its low rate of growth of exports combined with a relatively high
income elasticity of demand for imports. Certainly part of Japan's
phenomenal success must be related to the fact that, as we showed in
Chapter 10 , its growth rate consistent with balance-of-payments equili-
brium has exceeded by a considerable margin its actual growth rate ,
which has continually pressed on its capacity rate .
Kaldor (1974) has argued that the main cause of unemployment in
the United Kingdom in the last hundred years, barring periods of acute
depression, has not been oversaving but insufficient exports relative to
the level of imports which would be required at full employment :

despite commitment to full employment from 1944 it took a very

long time - in fact until 1967 1 - before it was realised that the
true effect of the new system [i.e. commitment to full employment 1
was simply to transmute the chronic pre-war unemployment prob-
lem into the chronic post-war balance of payments problem.

Kaldor goes on to say that what the United Kingdom has really suffered
from has been the slow growth of exports, a historical fact which can
be explained by the industrialisation of other regions of the world
which constantly narrowed the markets for U.K. goods. Kaldor is a
strong believer in export-led growth, maintaining that the common
feature of all industrial economies is that their economic growth has
been invariably led by a faster growth of exports which has given a
higher rate of growth of industrial productivity . He claims that the
United Kingdom could have grown at 5 per cent per annum had it
achieved export growth of 10 to 15 per cent per annum .
To sum up we can make two propositions. First, that up to the
Export-led Growth 263
capacity growth rate a country's actual growth rate is fundamentally
determined by its rate of growth of exports in relation to the growth
of imports if it is to maintain balance-of-payments equilibrium . Second ,
it is probable that a country's capacity growth rate is also partly deter-
mined by its export performance because of the link between high
demand and the response of factor supplies, and because faster growth
itself generates faster productivity growth . This is the idea of a virtuous
circle of growth led by exports, which we now develop.

MODELS OF EXPORT-LED GROWTH

In the European context, Lamfalussy (1963) was one of the first

economists to propound an export-led growth theory to account for
differences in the growth performance of Western European countries.
In Lamfalussy 's model export-led growth is important for three main
reasons : (i) the rate of growth of exports, as a determinant of demand ,
is likely to be an important determinant of investment; (ii) growth
requires imports, and if exports do not rise as fast as import require-
ments , growth will be constrained by the balance of payments; and
(iii) the smaller the domestic market, the greater the importance of
external demand in enabling economies to reap economies of scale in
production to make enterprises viable that would not otherwise be so.
Lamfalussy envisages a virtuous circle commencing with higher exports
leading to more investment, which in turn leads to a higher rate of
growth of productivity , lower export prices and thus higher exports.
There is, however, no explicit balance-of-payments constraint in his
model.
Beckerman (1962) sees a similar virtuous circle in export-led growth,
but his model also lacks a balance-of-payments equilibrium condition :
demand determines investment and growth; an important component
of demand is exports (and it is only this component of demand that can
help to balance the import requirements at a higher level of demand);
a high level of demand and investment is favourable to growth, which
contributes to greater competitiveness and further export demand .
Beckerman claims that the growth 'o f exports is closely related across
countries to the growth of competitiveness, and that differences in
competitiveness are mainly a function of differences in productivity
growth. As Caves (1970) rightly noted , Beckerman's model in its
original form lacks an equilibrium condition. Also, the export demand
264 Balance-of-Payments Theory and U.K. Experience

function is of a very ad hoc nature, making the rate of growth of

exports a function of the absolute difference between domestic and
foreign prices . Using the more conventional multiplicative export
demand function, which makes the rate of growth of exports depend
on the difference between the rate of growth of domestic and foreign
prices, also gives the model an equilibrium condition. The Beckerman
model, as here modified, runs as follows : export growth is a function of
the difference in the rate of growth of domestic and foreign prices;
faster export growth leads to faster productivity growth; faster produc-
tivity growth contributes to a lower rate of growth of wage costs
per unit of output if wages do not rise in line with productivity; a
lower rate of increase in wage costs per unit of output leads to a lower
rate of domestic price increase; and a lower rate of domestic price
increase leads to a faster rate of growth of exports. The virtuous circle
is complete. If the model is formulated algebraically, using linear
relations , with signs correctly specified in the model , we have :

(11.1)

where x is the rate of growth of exports, Pa is the rate of growth of

domestic prices, PI is the rate of growth of foreign prices, and ao
represents the rate of growth of exports determined by other factors
(e .g. the growth of world income). Also:

( 11.2)

where r is the rate of growth of labour productivity. Also:

(11.3)

where w is the rate of growth of wages. Finally :

Pa = w - r (11.4)

Substituting (11.3) into (1104) gives:

(11.5)

Substituting (11 .2) into (11.5) gives:

Export-led Growth 265
Pa = az + (b z - 1) (al + b l x)

(11.6)

Substituting (11.6) into (11.1) gives:

x = ao - b o [(a2 - a1 + b z ad + (b z b, -bdx] + bO(Pf)

(11 .7)

Solving for the equilibrium growth rate of exports gives:

ao -b o (az -al + b z ad + b o (Pf)

x = -------------
1 + b o b, (b z -1) (I 1.8)

Note that the virtuous circle of export-led growth depends crucially on

the rate of increase in wages being less than the rate of increase in
productivity (i.e . b z < 1). If b z = I, there would be no 'circular'
process : that is, no induced rate of growth of exports from the initial
expansion of exports itself. Balassa (1963) has also argued that if
wages respond to the level of employment, the virtuous circle may be
choked . He is concerned that if b z < 1, Beckerman's model may lead to
diverging country growth rates which are not observed in practice .
The model is easily modified by relating changes in wages to the level
of unemployment, using a Phillips curve relation. This makes for
stability by causing wages and prices to rise faster in regions where
export growth is high than in regions where export growth is sluggish
and unemployment high. Divergence of growth rates may take place
until full employment is reached, but then the virtuous circle would
break down. Beckerman (1963) replies that wage-rate increases between
countries bear little relation to unemployment-rate differences . This is
clearly an empirical matter . A lot would seem to depend on the extent
to which labour supply can adjust to demand across countries. The
recent work of Cornwall (1977) suggests that labour is very flexible
and that demand and growth have not been constrained in Europe in
the post-war period by a lack of factor supplies. Beckerman stresses
an economy obtaining its initial advantage in trade through a favourable
movement in relative prices : that is, through some competitive shock
such as devaluation. It has already been shown in Chapter 10, however,
that a once-for-all competitive shock cannot raise the rate of growth of
266 Balance-of-Payments Theory and U.K. Experience

exports permanently. Continual devaluation (or depreciation) would

be required. By contrast, we stress here the importance of countries
obtaining their initial trading advantage in goods with a high income
elasticity of demand in world markets, which affects the term ao in
equation (II.!).
Before proceeding to develop our own model, incorporating a
balance-of-payments equilibrium condition, a brief examination of
Caves's (1970) comments on the empirical content of the individual
functions making up the export-led growth models of Lamfalussy and
Beckerman may be useful to clarify the argument. He makes two major
points but neither are substantial enough to alter one's view on the
importance of export growth for simultaneous balance-of-payments
equilibrium and a high rate of growth of income . First , Caves asks,
'what is special about the growth of exports compared with the growth
of any other component of aggregate demand of equal size?'
Beckerman himself gives the answer when he argues that if other items
of demand are expanded businessmen may fear that demand growth
will not continue smoothly because of the balance-or-payments impli-
cations of the expansion of other types of demand . Caves seems to
recognise this as the crucial point, but then says: 'but it does attribute
to business enterprises an aversion to demand fluctuations induced by
public authorities that defies easy credibility'. Caves appears to be
questioning the influence of the level of demand on investment. But if
investment is sensitive to the pressure of demand in relation to capacity
output, the point made by the export-led growth school remains valid
if export-led growth raises the rate of growth at which a balance-of-
payments constraint becomes operative. Caves's second point concerns
the relationship between higher export growth and a higher rate of pro-
ductivity growth. He contends that there may be a link between export
and output growth on the one hand and productivity growth on the
other, but the direction of causation is anything but clear. This
question relates to the controversy over the Verdoorn relationship."
Suffice it to say that while productivity growth is obviously a compon-
ent of output growth, there are also good economic reasons why higher
output growth should be a stimulus to productivity growth. The
question is not whether there is a relationship but whether the estirn-
ate of the relationship is biased owing to its two-way nature.
Another recent critic of the idea that growth is constrained by the
balance of payments is Cornwall (1977) , who argues that the forces
that lead to the rapid growth of output, such as technological change
Export-led Growth 267
and entrepreneurial dynamism, also work to relieve a country of a
balance-of-payments constraint. He rightly observes that the developed
countries with recurring balance-of-payments difficulties have also been
the slow growers , but he wants to argue from this that it is the slow
growth that has caused the balance-of-payments difficulties - rather
than the other way round. It is hard to accept this view. It is not the
case that all of the forces that lead to rapid output growth in a country
necessarily improve the balance of payments. Population growth is one
obvious force. Another is rapid productivity growth if it comes about
largely through economising on men rather than through the expansion
of output. Technological change may be expected to improve produc-
tivity at home and also increase the desirability of a country's goods
abroad , but this is only one of many factors behind the growth of out -
put. In fact Cornwall's argument does not fit the U.K. experience,
where, if anything, productivity growth and the capacity growth rate
have risen, and yet the balance-of-payments constraint has not im-
proved because the forces making for the rise in productivity have not
improved the demand characteristics of the goods exported.
As we indicated earlier, Kaldor is a strong devotee of the export-
led growth school and has become one of its major protagonists. He
has presented many models in words, one of which (Kaldor, 1970)
Dixon and Thirlwall (l975a) have attempted to formalise . We shall
use this latter model to incorporate a balance-of-payments equilibrium
condition. Kaldor 's argument is essentially the Hicksian view that it
is the growth of autonomous demand which governs the long-run rate
of growth of output. In open economies export demand is the main
component of autonomous demand , so that the rate of growth of
exports governs the long-run rate of growth of output to which invest-
ment and consumption adjust. Kaldor is not explicit on the form of the
export demand function but seems to be suggesting the conventional
multiplicative function :

xt = (
E
Pdt) "z:t
p
r ft (11.9)

where the variables and parameters are defined as in equation (10.5)

in Chapter 10. The rate of growth of exports may then be written :

(I I.l 0)
268 Balance-ofPayments Theory and U.K. Experience

where lower-case letters represent rates of change of the variables.

The rate of growth of income outside the economy (Zt), the rate of
change of competitors' prices (Pr) and changes in the exchange rate
(e.) are taken as being exogenous . The rate of growth of domestic
(export) prices (Pd) is assumed to be endogenous, however, and is
derived from a mark-up pricing equation of the form :

(11.11)

where W is the level of money wages, R is the average product of

labour, and Tis I + % mark-up on unit labour costs . Thus :

Pdt=Wt-'t+Tt (11.12)

where lower-case letters represent rates of change of the variables.

The third relation in Kaldor's model, which gives a virtuous circle
of export-led growth, is the dependence of the growth oflabour prod-
uctivity on the growth of output, which is Verdoorn's law referred to
earlier:

't = 'at + A(Yt) (11.13)

where 'at is the rate of autonomous productivity growth, and A is

the Verdoorn coefficient. Equation (11.13) makes the model 'circular'
since the higher the rate of growth of output, the faster the rate of
growth of productivity; the faster the rate of growth of productivity,
the lower the rate of increase in prices; and the lower the rate of
increase in prices, the faster the rate of growth of exports and hence
output. It is also the Verdoom relation which gives rise to the
possibility that once an economy obtains a growth advantage it will
keep it. Suppose, for example, that an economy obtains an advantage
in the production of goods with a high income elasticity of demand
in world markets, which raises its growth rate above that of other
countries. Because of the Verdoorn effect, productivity growth will
then be higher, and the economy will retain its competitive advantage
in these goods, making it difficult for other countries to establish an
advantage in the same commodities. The income elasticity of demand
for exports is probably the most important determinant of comparative
export performance; and the income elasticity of demand for imports
must likewise assume key importance in an export-led growth model
Export-led Growth 269

which incorporates a balance-of-payments equilibrium condition.

The lower the income elasticity of demand for imports, the higher the
growth rate consistent with balance-of-payments equilibrium, other
things remaining the same. In models of cumulative causation in which
some economies produce goods which are expanding quickly in demand
while other economies produce goods which are sluggish in demand,
it is the difference between the income elasticity characteristics of
exports and imports which is the essence of the theory of divergence
between 'centre' and 'periphery' and between industrial and agricul-
tural economies . This is also the essence of Kaldor's view that the
opening-up of trade between economies may create growth-rate differ-
ences which are sustained or even widened by the process of trade, for
example the United Kingdom vis-a-vis the European Economic
Community .
Combining equations (J 1.10), (11.12) and (11.13) gives an expres-
sion for the rate of growth of exports:

(11.14)

AN EXPORT-LED GROWTH MODEL WITH A BALANCE-oF·PAYMENTS

CONSTRAINT

Kaldor's model, formalised above , also lacks a balance-of-payments

equilibrium condition , and could therefore overpredict the sustainable
rate of growth depending on the assumed relation between Y r and xr '
Equation (J 1.14) , however , can be combined with a balance-of-pay-
ments equilibrium condition to give an export-led growth model which
expresses the idea that export growth may establish a virtuous circle
of growth, but which also recognises explicitly that what is happening
to imports is also a determinant of growth if balance-of-payments
equilibrium must be maintained in the long run . The balance-of-pay-
ments equilibrium condition is (see equation (10.2) of Chapter 10):

(J 1.15)

Substituting equation (10.4) for mr and equation (J 1.14) for x r gives

the export-led growth rate consistent with balance-of-payments equili-
brium:
270 Balance-of-Payments Theory and U.K. Experience

11
(11.16)

Therefore

( 11.17)

This compares with the equilibrium growth rate derived from the
Kaldor model without a balance-of-payments equilibrium condition,
Ietting v, = r (x.) , of :

'Y[TI (w t - rat + Tt - PIt - e t) + € (Zt)]

Yt = -
I+'YTIX (11.18)

Note that if the price and income elasticities of demand for imports
are both unity (i .e. 1J; = 1 and 11 = 1) and l' = 1,equation (11.17)
collapses to (II.18) because the balance of payments would always
be in equilibrium whatever the rate of growth of income and foreign
prices. In an export-led growth model without a balance-of-payments
constraint, 'Y must take on that value which preserves balance-of-
payments equilibrium as income and the price of imports change.
Export-led growth incorporating the idea of a virtuous circle through
the Verdoorn coefficient (X) will raise the equilibrium growth rate
above what it would otherwise be only if the Marshall-Lerner con-
dition is satisfied, i.e. if I TI + 1/1 I> 1. Comparing equations (10.7)
and (11.17) it can be seen that the Verdoorn coefficient in the denom-
inator of (11.17) will only raise the balance-of-payments equilibrium
gro wrth rate if the bracketed term is negative, which implies that the
two (negative) elasticities must sum to greater than unity. There is no
guarantee, of course, that a country at any point in time will grow
at its export-led growth rate consistent with balance-of-payments
equilibrium. If the country grows faster, however, (assuming it has
the capacity to do so), it will run into deficit; if it grows slower, it
will accumulate surpluses; and in either case the actual growth rate
may be less than the capacity rate, causing growing unemployment.
Note again from equation (11.17) that the exchange rate would have to
Export-led Growth 271

be depreciating continually (e t > 0) for exchange depreciation to raise

the balance-of-payments equilibrium growth rate permanently, and
then only if the Marshall-Lerner condition is satisfied and wages do
not rise.
As far as the United Kingdom is concerned, the export-led model
incorporating an explicit balance-of-payments equilibrium condition is
a far better predictor of the actual growth experience than the uncon -
strained model of equation (11.18) . When the unconstrained model was
applied to data for the period 1951 to 1966 (see Dixon and Thirlwall,
1975a) it predicted a growth rate of 4 per cent on the basis of the
following parameter values: T/ = -1.5, Wt + T t = 0.06, rat = 0.02,
Pft = 0.02, € = 1.0, Zt = 0 .04, A = 0.5, and 'Y = 1. Since the actual
growth rate over the period was 2.8 per cent , it was mentioned at the
time that the model may be overpredicting because of a balance-of-
payments constraint - apart from any constraint on capacity. If 4
per cent were a permissible rate consistent with balance-of-payments
equilibrium, the country should have enjoyed a growing surplus, but
it did not. In order to apply the balance-of-payments equilibrium
model (equation (11.17» estimates are required of the additional
parameters t/I and 1T. It is assumed that imports are insensitive to price,
and we take Houthakker and Magee's (1969) estimate for 1T of 1.51.
Using equation (11.17), with the exchange rate fixed (as it was over
this period), we obtain an estimate of the growth rate consistent with
balance-of-payments equilibrium of 2.4 per cent. This is closer to the
actual growth rate of 2.8 per cent, and the discrepancy between the
two rates is consistent with the balance of payments moving into deficit
over the period ."
Applying the equations to the post-1966 period does not give such
reasonable results, probably because the period has been one of general
economic upheaval, and the model is very sensitive to small variations
(errors) in the variables and in the assumed parameter values. When
the constrained model (equation (11.17» is applied to the data over the
period 1967 to 1976 the estimated growth rate is 5.2 per cent, corn-
pared with the actual growth rate of 1.8 per cent per annum ." Since
the country moved into substantial deficit, the model is clearly over-
predicting. It could be that the U.i<.. income elasticity of demand for
imports is much higher than the assumed value of 1.51, or that the
price elasticity of demand for UK. exports is less than the assumed
1.5. Assuming the lesser value of unity (so that with t/I = 0, the
Marshall-Lerner condition is just satisfied) allows us to apply the
272 Balance-of-Payments Theory and U.K. Experience

simple growth rule enunciated in equation (10 .8), which gives a bal-
ance-of-payments equilibrium growth rate of 0.03 -:- 1.51 = 2 per
cent. This looks much more consistent with the facts , particularly
if the income elasticity of demand for imports has risen. A figure for
n of approximately 2 would give a balance-of-payments equilibrium
growth rate of 1.5 per cent per annum over the period , and this would
be entirely consistent with the actual growth experience and the
balance of payments on current account moving into substantial
deficit. Despite the effort of formulating a fairly sophisticated export -
led growth model , incorporating the idea of a virtuous circle led by
exports, the empirical evidence suggests that the simple growth rules
set out in equations (10.8) and (10.9) of the previous chapter are better
predictors of actual growth performance. This is not to disparage
export-led growth models which incorporate a virtuous circle . It is
rather to suggest that the link between exports and growth via the
Verdoorn effect may not be very important either because relative
prices do not change or because the price elasticities of demand for
exports and imports are not high enough to matter. Again we conclude
that everything would seem to hinge on the income elasticities of
demand for exports and imports. It is in this sense that we refer to the
balance of payments as a structural problem .
12
The Balance of Payments
as a Structural Problem
The U.K. balance of payments has acted as a serious constraint on the
attainment of other macroeconomic goals since the Second World War,
particularly the achievement of faster, investment-led growth. As the
economy has expanded towards its capacity level, bottlenecks in partic-
ular sectors of the economy have raised the aggregate level of imports
in excess of the capacity to export, necessitating contraction of the
economy . Likewise, attempts to grow faster in order to match the
growth rates of other European countries have resulted in imports
growing more rapidly than exports, and experiments of domestic
demand-led growth have had to be abandoned. In the last two chapters
the prime importance of the growth of exports, and the income elastic-
ity of demand for imports, as determinants of domestic economic
performance have been stressed . Nevertheless a strong body of opinion
has traditionally argued that the fundamental problem of the UK.
balance of payments has been a lack of price competitiveness in world
markets, because unit costs of production in the United Kingdom have
risen faster than elsewhere, and that the most appropriate solution to
this problem is to let the currency depreciate. Since 1972 the pound
sterling has in fact depreciated substantially against the dollar and
other major currencies, as shown in Table 7 .6 above (p.l62).
The argument that the U.K. balance-of-payments problem has been
one of price uncompetitiveness, coupled with an inflexible exchange
rate up to 1972, contains a number of explicit and implicit assump-
tions, some of which were questioned in Chapter 3 on the elasticity
approach to balance-of-payments adjustment. It assumes that costs and
prices have had an independent tendency to rise faster in the United
274 Balance-of-Payments Theory and U.K. Experience

Kingdom than elsewhere, that the demand for traded goods is deter-
mined primarily by relative prices , that if relative prices fall the demand
for domestic goods is sufficiently elastic , that manufacturers make the
appropriate pricing response to a change in the foreign value of the
currency and reap the benefit , that the production capacity is available
to meet increased export demand and the increased demand for import
substitutes, that the effects of currency depreciation are not fully
eroded by the domestic inflation which depreciation breeds, and so on .
Many economists, formerly believers in the efficiency of exchange-rate
changes, are now questioning these explicit and implicit assumptions
and the relevance of relative price adjustment as a method of balance-
of-payments adjustment and as a means of relieving the balance-of-
payments constraint on full employment and faster growth in the
United Kingdom . The reasons for scepticism are axiomatic. The balance
of payments has been increasingly problematical since the pound was
devalued in 1967 and then allowed to float down in 1972. The promise
of a new dawn held out by the currency depreciation school has not
materialised . It is true that the balance of payments improved between
1968 and 1971, but that was mainly due to the sharp upturn in world
trade and at the expense of domestic employment. When the economy
was expanded during 1972-3 the record surpluses that had been built
up on current account not only melted away but became record defi-
cits. This was caused largely by a consumption-led boom , but the
deteriorating balance-of-payments situation was also worsened by the
currency depreciation itself which reduced the foreign-currency earn-
ings from exports without noticeably decreasing the foreign-currency
expenditure on imports. The commodity boom and oil price increase
made matters even worse in 1973 and 1974, and the balance of pay-
ments only began to recover in 1977 after a prolonged stagnation of
output and consumption and the increased flow of North Sea oil. It is
difficult to detect any significant raising of the balance-of-payments
equilibrium growth rate from currency depreciation itself. After the
events of the last decade the United Kingdom's weak balance of pay-
ments is increasingly seen as a structural problem related both to
the capacity to produce and to the characteristics of goods produced
and exported, both of which are not directly amenable to exchange-
rate depreciation .
Textbooks dealing with the balance of payments also traditionally
treat balance-of-payments disequilibrium as something that can be
rectified by appropriate relative price adjustments. By contrast, one of
The Balance of Payments as a Structural Problem 275

the purposes of this book is to stres s that many countries' balance-of-

payments difficulties originate from the supply side and in non-price
factors , for which price adju stments are not appropriate . Downward
price adjustment, by man ipulat ing exchange rates, merely makes
countries more competitive in the short run in goods with a given set
of characteristics which were the source of the balance-of-payments
difficulty in the first place. In a growth context continual depreciation
would be required to raise the balance-o f-payments equilibrium growth
rate permanently, and again the industrial structure would be ossified .
Certainly it is difficult to see how the United Kingdom 's long-run
balance of payments could be improved at full employment by contin-
ual exchange depreciation . The more recent monetary approach to the
balance of payments offers nothing by way of prescription either. If
the balance-of-payments constraint on growth is to be alleviated , the
concentration of policy must be on the supply side . More specifically,
following the arguments in Chapters 10 and II, policy must be directed
to raising the rate of growth of exports by non-price means . Given the
growth of world income, the growth of exports is primarily a function
of the world income elasticity of demand for exports ; and the income
elasticity of demand for exports, in turn, is a function of the type of
goods produced within the product range - their quality, reliability ,
marketing, and so on . The world income elasticity of demand for U.K.
exports is very low by international standards , probably less than unity,
and the elasticity of demand for U.K. exports with respect to the
growth of world trade is even lower .' The latter elasticity, which the
Treasury estimates to be about 0.6 , accounts for the United Kingdom's
declining share of world trade. The former elasticity (particularly its
low magnit ude) lies at the root of the United Kingdom's long-run
balance-of-payments difficulties . In this chapter, to conclude the book,
we emphasise the non-price factors affecting the balance of payments,
and the importance of removing the balance-of-payments constraint on
growth if an expanding industrial sector is to be preserved. Improving
the non-price factors should regenerate the industrial sector without a
loss of jobs and help to remove the balance-of-payments .constraint on
growth at the same time. Simply to revitalise industry by increasing
productivity, without removing the balance-of-payments constraint on
growth, will cause employment in the industrial sector to contract and
unemployment to rise .
276 Balance-of-Payments Theory and U.K. Experience

NON·PRICE FACTORS DETERMINING EXPO RTS AND IMPORTS

The National Economic Development Office's Report on non-price

factors in international trade, written by David Stout (1977) , stresses
two broad aspects of non-price competitiveness: first, the act of selling;
second, aspects of the product including design. ease of maintenance ,
quality. reliability, delivery time and after-sales service . To quantify
the importance of these factors is not easy , and the anecdotal evidence
that these factors matter is perhaps stronger than the actual evidence ,
but from the work of the National Economic Development Office and
other organisations, and the statements of industrialists and impartial
observers, no one can doubt that they rank in importance with price ,
and in many cases may be the decisive influence in the decision of
foreigners whether to buy British and in the decision of the British
whether to buy foreign . Here we refer briefly to some representative
work and statements of organisations and individuals under the general
headings of nuzrketing, and quality, design and reliability .
Mr David Orr (l978), ex-Chairman of Unilever, has defined marketing
as

the skill of providing the right product for delivery at the right
time in the right place at the right price and with the right back-up
service - for a profit. [He goes on 1 the most decisive factors today
are concerned with design, styling, packaging, advertising; and for
capital goods - credit terms , delivery dates, reliability, technical
service and so on.

A six-month delivery delay on a large order, when the money or goods

could have been invested profitably , more than offsets any price
advantage over a competitor charging a higher price but who can deliver
immediately . Mr Orr believes that marketing and selling need to be
accorded much higher prestige in the hierarchy of business.
With regard to the act of selling, a study by the British Export Trade
Research Organisation (1975) found that leading U.K. companies had
either no one, or only one permanent export representative, in key
foreign markets compared with an average of eleven for West German
and Japanese companies.
The danger of currency depreciation is that it can lead rnanufact-
The Balance ofPayments as a Structural Problem 277
urers to slide progressively 'down market' in the quality of exports,
supplying the world with exports of goods with low unit values. This
danger of depreciation is particularly stressed in the work of the
National Economic Development Office (Stout, 1977), which makes
the distinction between eliminating a competitive disadvantage through
currency depreciation, which is a static process , and the need for
dynamic change in a growth context (if the balance of payments is to
be improved permanently), which the act of depreciation may thwart
because it perpetuates the production of the same commodities, pre-
sumably of inferior design and style to those of other countries.
Certainly the profitability of exporting must be maintained but not at
the expense of neglecting what the market is demanding .
Strong evidence is emerging from the work of the National Econo-
mic Development Office that the United Kingdom tends to export
relatively low-valued products within a given product range, and to
import relatively high-valued products, compared with West Germany
and France. The evidence for this comes from analysis of industries
covered by thirty-five Sector Working Parties set up as part of the
government's Industrial Strategy.' If true, this would account for the
low world income elasticity of demand for U.K. exports and the higher
income elasticity of demand for imports. This would also account for
the fact that while the United Kingdom exports goods which are
expanding rapidly in demand, and to geographic markets which are
also expanding, none the less the U.K. share of world trade has fallen
more than in the case of other industrialised countries. The average
unit values of West German exports were found to be higher than those
of U.K. exports in twenty-nine out of the thirty-five industry groups,
and in twenty-three out of thirty-four in the case of the comparison
with France. While part of the difference could be accounted for by the
United Kingdom having lower prices for similar goods, probably the
major factor lies in differences in the product mix and differences in
product quality. It is this evidence which suggests that the United
Kingdom is specialising in 'down-market' or inferior products.
The other side of the coin is the continued export success of
countries where the currency has appreciated . Higher unit values for
exports and lower unit values for imports than other countries may be
the result, of course, of a country's costs and prices being too high ,
but not if its exports are growing fast and its share of world trade is
rising. Despite a 30 per cent appreciation of the Deutschmark between
278 Balance-of-Payments Theory and U.K. Experience

1973 and 1977 , West German exports continued to expand rapidly .

The former German Minister of Finance, Mr Hans Apel, explained thus:
'obviously the variety and the quality of German goods fits almost
exactly what the customer wants . Also, customers can rely on the
dates of delivery promised by German suppliers being met' (Artus and
Sosa, 1978).
Sir Frederick Catherwood , Chairman of the British Overseas Trade
Board, stressed in his Annual Report for 1977 the overriding need for
investment because it is now vital to sell exports on design, quality
and reliability and not simply on price .
Mr David Orr, quoted earlier , argues that the United Kingdom is
obsessed with price (presumably he means economists in government
and industry brought up on the wrong textbooks and trained in neo -
classical trade theory!), but that price is not sufficient to offset bad
design, poor quality, an inadequate range of models and unreliability .
Competitiveness does not mean cheapness (just as low-cost labour does
not mean cheap labour if productivity is inferior). Mr Orr reiterates the
point made by the National Economic Development Office , i.e. that
the unit value of U.K. exports tends to be lower than that of imports,
and that it is dangerous to move 'down market' . What is needed is to
add extra value to U.K. products to meet customers' needs better than
competitors, as the Japanese set out , successfully, to do . A number of
other countries are now adopting the Japanese strategy , which poses a
further threat to U.K. industry if it fails to respond in the manner
outlined . Industry must gear its output to the export market and not
simply treat export sales as peripheral to home sales, which often
appears to be the case .
Management cannot escape responsibility for poor quality, inferior
design , lack of reliability, delivery delays and lack of interest in mar-
keting abroad . Some would lay the blame squarely on management's
shoulders, and lift it from the traditional scapegoats, the trade unions
and the British worker : 'The root of Britain 's export problems are
outstandingly bad managers, lack of interest, and terrible communi-
cations' (letter to The Times, 5 December 1977). The writer of the
letter concerned, who runs an importing business in France, refers to
one U.K. company who received his order in January, promised deliv-
ery in April and then decided in August that it was no longer interested
in exporting. Another two companies increased price without notice.
Another company promised delivery which did not arrive, and letters
The Balance of Payments as a Structural Problem 279
requesting an explanation produced no response . Other evidence
suggests that the experience of this one particular foreign buyer is not
atypical.
The Central Policy Review Staff Report on the Future of the
British Car Industry (1975) considered the competitive weakness of the
industry to be due to : inadequate distribution networks; slow delivery;
high costs, low productivity and under-investment ; overmanning ; poor
labour relations; slow work-pace and line-speeds: and product unsuita -
bility relative to market conditions and requirements.
The National Economic Development Office Report on Imported
Manufactures (1965) ascribed import penetration to, among other
things ; higher relative prices and costs , a shortage of capacity in boom
periods coupled with a 'ratchet' effect, poor technical performance
of the home product (particularly in mechanical engineering products,
electronic capital goods and scientific instruments), and a slow market-
ing response by U.K. manufacturers.
The National Economic Development Council survey of Investment
in Machine Tools (1965) asked U.K. machine-tool-users why they
bought foreign machines. Only 5 per cent said price was the main
factor. The more important factors were (percentage of respondents
listing the factor as the main one in parentheses) : technical superiority
of the foreign product (30 per cent); machine specifications not avail-
able in the United Kingdom (21 per cent); quick and reliable delivery
(20 per cent) ; willingness of foreign producers to meet special require-
ments (8 per cent) ; and better after-sales service (5 per cent) . Technical
factors were cited by over one-half of the sample in the decision to buy
foreign goods.
In the current work of the National Economic Development Office
as part of the Industrial Strategy, pertinent factors are emerging on
an industry-by-industry basis. In machine-tools for technologically
advanced products export success seems to depend primarily on deliv-
ery and reliability . The unit values of the exports of machine-tools
appear to be lower than the unit values of imports, which suggests that
the United Kingdom is not competing successfully in the most sophisti-
cated machine-tool product groups . In pumps and valves the technical
quality of U.K. products is good but the industry seems to lack the
organisational support and resources necessary to match the marketing,
delivery and after-sales service of competitors. In electric motors, the
Sector Working Party concludes that U.K. industry is deficient in
establishing overseas marketing and technical networks, and in delivery.
In electronics, consumer goods suffer from a reputation of poor quality
280 Balance-of-Iuyments Theory and U.K. Experience
and apparently have not been designed in the past with the European
market in mind. In clothing, it is concluded that designs and fashions
are often out of line with continental preferences, and that quality and
design are often sacrificed to keep price down. In construction equip-
men t, reliability and after-sales service is of key importance.

NORTH SEA OIL AND THE BALANCE OF PAYMENTS

The above diagnosis of the United Kingdom's underlying balance-of-

payments weakness has implications for the balance-of-payments policy
that ought to be pursued if the country enjoys a surplus from North
Sea oil. It also has implications for the use of tax revenue from North
Sea oil.
The latest Treasury estimates of the balance-of-payments effects
of the North Sea oil and gas programme are shown in Table 12.1. The
cumulative effect on the total balance of payments from 1977 to 1985
is estimated at £35.4 billion at 1977 prices. The actual net contribution
of oil to the current account was £6 billion in 1980.
In addition the various taxes and royalties that the government
collects will give it around 70 per cent of the earnings from North
Sea oil. The government expected to collect £5000 million by 1980,
and £3500 million a year by 1985 after companies have written off
their development costs against tax liability. In discussing future policy
we can distinguish, conceptually at least, between the different uses to
which the balance-of-payments surplus might be put and the different
ways in which the government can use its oil tax revenues. However,
the distinction must not be drawn too finely since any policy the
goverrunent pursues in relation to its oil tax revenues will obviously
have implications for the balance-of-payments surplus. Nevertheless it
will be useful to begin by considering the potential surplus and how it
might be used. The worry is that if the import savings from oil are not
used wisely and productively, the non-oil balance will deteriorate
further, and when North Sea oil runs out the country will be faced with
a current-account deficit of gigantic proportions.
Repayment of official sterling liabilitiesr' and the movement of
the economy back to full employment, will probably absorb most of
the benefit in the early years, but after this there are five main uses to
which a surplus might be put. First, it could be used to bring about
 TABLE 12.1
Balance-of-payments contribution ofNorth Sea oil and gas (lb. at 1977 prices)

1977 1978 1979 1980 1981 1985

1. Oil and gas sold , at

equivalent import value 4.5 5 .0 6.8 8.1 9.7 10.2
2. Imports of goods and services
directly for the programme 1.2 1.1 1.0 0 .9 0 .8 0 .7
3. Interest, profits and dividends
due overseas 0 .4 0.8 1.2 1.5 2 .2 0.9

4 . Net con tribu tion to the current

account (equals line 1 minus
line 2, minus line 3) 2 .9 3.2 4.6 5.7 6 .7 8 .6
5. Net effect on capital account 1.3 1.1 0 .9 0 .4 0.1 - 0 .1

6. Net effect on the balance for

official financing (line 4 plus 4.2 4.3 5 .5 6.1 6.8 8.5
line 5)
N - . .. ._ - - --
00
..... Sour ce: H.M. Treasury, Economic Progress R eport , October 1978.
282 Baiance-of-Payments Theory and U.K. Experience

an immediate increase in real living standards by supporting an increase

in consumption. If this policy is pursued, however , the fundamental
structural weaknesses in the balance of payments would remain, and
would reveal themselves again - probably even more acutely - when the
flow of oil dries up. A consumption bonanza would be a gross extrava-
gance that would have to be paid for dearly later.
Second , the surplus could be used for investment abroad. Since
this would provide an income for the future, it is certainly a more
attractive proposition than spending the surplus to increase consump-
tion , but it presupposes a lack of investment opportunities in the
United Kingdom to provide a similar income flow in the future. While it
is understandable that private capital will seek profitable outlets abroad
if the private net rate of return in the United Kingdom is lower, it is not
necessarily in the national interest that it should do so because such
investment does not create jobs at home, nor does it yield the same tax
revenue to the government . In addition the return flow of investment
income may be little more than a trickle if the investor pursues a policy
of reinvestment overseas.
The third possibility would be to use the surplus to increase the
foreign-exchange reserves. Since the U.K. foreign-exchange reserves
have been traditionally low in relation to the value of liabilities, any
addition to them would be beneficial and would confer greater stability
when the days of surplus are over. Of course, if the sterling liabilities
are not redeemed on schedule, but funded instead , the accumulation of
reserves could later be used for the repayment of liabilities at a more
convenient time . The early redemption of sterling liabilities has some-
thing to recommend it since it would reduce the volume of volatile
money deposited in London and help reduce the vulnerability of the
pound to speculative attacks when the surplus runs out. To accumulate
reserves, and to repay debt, however , implies holding the exchange
rate down.
An alternative option would be to let the exchange rate rise. This
would confer the important benefit on the economy of reducing the
sterling price of imports and reducing upward pressures on the internal
cost structure. There can be little doubt that the conduct of economic
policy in the 1950s and early 1960s was greatly helped by a continually
improving terms of trade, allowing living standards to rise faster than
the rate of productivity advance , which in turn reduced the pressure on
wages to rise equiproportionately with prices, thus avoiding accelerating
inflation. My preference would certainly be to let the exchange rate rise
The Balance of Payments as a Structural Problem 283

and to disregard those who say that the result of such a policy will be
that the United Kingdom will price itself out of foreign markets. On
the contrary, the United Kingdom's competitive position might im-
prove , but in any case factors other than relative prices affect imports
and exports, and , as we argued above, these factors have been, and still
are , of decisive importance in the case of the United Kingdom. Succes-
sive devaluations , while bringing temporary relief, have not solved the
fundamental underlying structural problems. A surplus at least provides
the opportunity for an experimental period of currency appreciation.
The fifth policy alternative is to use a surplus to mount a massive
programme of investment to re-equip and revitalise major sectors of
U.K. industry , with particular orientation towards raising the rate of
growth of exports. This is the most attractive option of all, and (if it
can be realised) the only one which will bring lasting benefits to the
U.K. economy . It will be difficult for the government to use a surplus
directly to purchase imported investment goods, but the oil tax rev-
enues will provide the government with considerable leverage to mobil-
ise the resources for such a programme of investment. Some of the
funds could be used to increase investment directly, through, say, the
National Enterprise Board . The remainder could be used to reduce
taxes on profits in order to stimulate investment in the private sector.
In the light of our earlier analysis investment should be directed prim-
arily to improving the world income elasticity of demand for U.K.
exports and reducing the income elasticity of demand for imports and
the level of import penetration.

THE BALANCE OF PAYMENTS AND DE-INDUSTRIALISATION

Some concern has been expressed in recent years that the United
Kingdom is entering a period of de-industrialisation with severe impli-
cations for the future level of employment and unemployment ,"
The United Kingdom's weak balance of payments, and her slow growth
rate constrained by the balance of payments, provide a ready explan-
ation of the process (see Thirlwall , 1978 ; and Singh, 1977). If the
growth of manufacturing output is constrained by the balance of
payments below the rate of productivity growth in manufacturing,
industry is bound to shed labour.
The relation between the growth of manufacturing output (gm)
and the growth of total income (y) may be expressed as
284 Balance-of-Payments Theory and U.K. Experience

gm = JJ(y)
(l2.l )

where JJ is the elasticity of manufacturing output with respect to

total income, or a measure of the income elasticity of demand for
manufactured goods. Given JJ , and the growth of total income con-
sistent with balance-of-payments equilibrium, the rate of growth of
manufacturing output consistent with balance-of-payments equilibrium
(im) is also determined :

( 12.2)

where Yo is the balance-of-payments equilibrium growth rate that we

defined in Chapter 10. If Yo can be approximated by the rate of growth
of exports (x) divided by the income elasticity of demand for imports
(1T), the maximum long-run rate of growth of manufacturing output is
- JJX
gm= -
1T (12.3)

The only ways to increase the long-run rate of growth of manufacturing

output, given the over-all constraint of the balance of payments, are to
raise JJ by producing goods with a higher income elasticity of demand,
to raise the rate of growth of total exports by making all goods more
desirable abroad, and to lower 1T by making all foreign goods less desir-
able. If g~ is less than the rate of growth of labour productivity in
manufacturing, less labour will be demanded. What are the facts? For
the period 1951 to 1966 we estimate the balance-of-payments con-
strained growth rate to have been about 2.4 per cent. For the period
since then our simple rule predicts an estimate of between 1.5
and 2.0 per cent depending on the assumed income elasticity of
demand for imports. From statistics on the share of manufacturing
output (see Table 12.2) it would appear that JJ is virtually unity, with
the share of manufacturing output to total output remaining constant
over time. This means that the growth of manufacturing output consis-
tent with balance-of-payments equilibrium is exactly the same as the
over-all balance-of-payments constrained growth rate . Now produc-
tivity growth in manufacturing industry over the period 1950 to
1964 was 2.7 per cent per annum, which is approximately equal to the
rate of growth of total output and manufacturing output, and thus is
consistent with the economy retaining labour in the manufacturing
The Balance ofPayments as a Stru ctural Problem 285

TABL E 12 .2

Manufacturing employment and output

in the United Kingdom since 1950
Share of manufacturing Share of manufacturing Level of
employment in output in G.D.P. at manufacturing
total employment constant (1963) employment
(%) prices (%) (000)
1950 34.7 29.3 8519
1955 35.9 30.6 9222
1960 35.8 31.0 8850
1965 35.0 31.1 9028
1970 34.7 31.7 8910
1975 30.9 29.1 7488
1980 30.2 29.5 6807
Sources: Brown and Sheriff in Blackaby (1979) and Economic Trends (various
editions).

sector over the period and the economy moving into payments deficit.
From the mid-1960s to the early 1970s, however, productivity growth
accelerated, but the growth of manufacturing output permitted by the
over-all balance-of-payments constraint did not rise commensurately
and hence manufacturing employment declined. During the rest of the
1970s, productivity growth slowed but output growth slowed even more ,
and was actually negative in 1980, causing further (sharp) falls in manu-
facturing employment, as shown in Table 12.2. The slow growth of the
manufacturing sector does not lie in the lack of factor supplies but in
constraints on demand . The major underlying long-run constraint on
the growth of demand is the balance of payments. The problem lies in
the characteristics of the goods that the United Kingdom produces and
sells at home and abroad . The British treat them as 'inferior' goods and
so does the rest of the world.
It has been argued by some (e.g. Kay and Forsyth, 1980) that the
production and export of oil must inevitably lead to a contraction of
the manufacturing sector because oil can only be exchanged for tradeable
goods which in the United Kingdom consist almost exclusively of manu-
factures. The mechanism through which this contraction is supposed
to take place in through an appreciation of the exchange rate. This indeed
will happen if the economy is static and at full employment , and the
foreign-exchange earnings from oil are neither accumulated at home nor
invested abroad. Then there will be no increase in domestic output, the
exchange rate will rise by the full amount justified by the foreign-exchange
286 Balance-of-Payments Theory and U.K. Experience

gain from exporting oil, and the domestic production of tradeable

(manufactured) goods must fall. Clearly, however, there is nothing
inevitable about the process in practice except in a relative sense. There
are so many ways in which the foreign-exchange earnings from a natural
resource can be dissipated that there need be no rise in the exchange
rate and no contraction of domestic tradeable goods production. Start-
ing from less than full employment, the economy could be expanded,
and this will use foreign exchange and stimulate domestic production at
the same time . Foreign exchange could be accumulated , keeping its
price down, or earnings could be invested overseas. To attribute the
decline in industrial output and employment in 1980 and 1981, which
was the most rapid since 1931, to the fact that the United Kingdom then
became a net oil exporter would be highly misleading and disingenuous .

IMPORT CONTROLS

The apparent inability, and failure up to the present time, of currency

depreciation to reconcile the conflict between full employment and
balance-of-payments equilibrium has led to the call for import controls.
This solution to the conflict has been consistently argued by the
Cambridge Economic Policy Group, a body which still apparently
believes in the efficacy of exchange-rate adjustment but which feels
that , in the U.K. case, the depreciation would have to be so great as to
be impractical. The major weapons of import control are tariffs and
import quotas, which could be applied across all industries or selectively
against specific imported goods. It is not certain, however, whether im-
port controls would be more effective than depreciation in relaxing the
balance -of-payments constraint on demand and growth. There would
certainly be an initial direct cut in imports, and quotas would prob-
ably be less inflationary in a cost and demand sense if capacity is
available to meet the consequent increase in demand for domestic
goods . On the other hand, there are a number of factors that may
mitigate the balance-of-payments gain. If the home market becomes
more profitable, firms may relax their export effort , and exports may
be diverted to the home market to substitute for imports. Protection
may breed inefficiency and worsen in the long run an industry's out-
put and export performance. Other countries may retaliate, and ulti-
mately nullify any balance-of-payments gain. Last but not least, import
controls, like a once-for-all devaluation, cannot raise the rate of growth
of output permanently . The controls would have to become more and
The Balance ofPayments as a Structural Problem 287
more stringent for the country to move to a permanently higher growth
path consistent with balance-of-payments equilibrium. This is clearly
impractical. At best , therefore, import controls could only give some
temporary respite . For such a policy to give lasting benefit , other
policies of a structural kind would have to be pursued at the same time .
These policies must operate to raise the income elasticity of demand for
UK . exports in world markets and to reduce the U.K. income elasticity
of demand for imports. If the variables of price and income which
determine imports and exports cannot be influenced effectively and
permanently, the parameters relating imports and exports to the said
variables must become the targets of policy . There is no way in which
the balance-of-payments equilibrium growth rate of the United
Kingdom can be raised to the level of other industrial countries, and
de-industrialisation halted, without a rise in the world income elasticity
of demand for U.K. exports and/or a fall in the U.K. income elasticity
of demand for imports. My own policy preference would be for selec-
tive subsidies to export activities with growth potential. This would
alter the export mix in favour of faster-growing products, so that for
any given growth of world income and trade , U.K. export growth
would be faster. Neither import controls nor currency depreciation
would help to improve the income elasticity of demand for U.K.
exports in the same way . Subsidies could take the form of corporation-
tax relief, and would constitute one means of using North Sea oil tax
revenue . The fact that export subsidies might contravene certain inter-
national economic agreements should not be allowed to stand in the
way of such a policy . The situation in the United Kingdom has become
far too serious to allow political good manners to decide whether
the country survives as a prosperous industrial nation or whether
it becomes the economic backwater of Europe .
Notes
CHAPTER 1

1. This replaced the 'liquidity balance' concept in 1965 on the

recommendation of the Bernstein Committee . However, in 1976 an
Advisory Committee on the Presentation of Balance of Payments
Statistics recommended doing away with all 'balances' for fear that
they may mislead.

2. The old measure of total currency flow measured by the balance

for official financing excluded the allocation of Special Drawing Rights
and gold subscriptions to the I.M.F.

3. A basic balance-of-payments surplus or deficit will not neces-

sarily lead to an accumulation or decumulation of foreign-currency
reserves, or to a change in the exchange rate, because the surplus or
deficit may be fully offset by private short-term capital movements.

4 . The gold specie system which preceded the gold bullion system
differed in that the domestic coinage itself consisted of gold. However,
the mechanism of international adjustment under the two systems
was the same.

5. The convertibility of the domestic currency into gold was

abandoned by most countries during the First World War.

6. The exception to the proposition would be if the rectification of

an over-valued exchange rate increased the over-all level of productivity
in the economy, thereby raising income at full employment. (See
Chapter 4 for further discussion.)
Notes 289
CHAPTER 2

1. Throughout the chapter and the book, exports and imports are
defined in such a way that the national income equals the national
output. The difference between income and output is net factor pay-
ments abroad .

2. For expositions of the theory, see Cripps and Godley (I 976)

and Rowan (1976).

3. The traditional approach to the foreign-trade multiplier is intro-

duced first in order to dismiss it later.

4 . We assume for simplicity that the import coefficients attached to

autonomous and induced consumption are identical.

5. If so , we have lI.a = (t1Mlt::.1)a and ma = (t1Mlt::.Y)a. where T

is total expenditure. Since t::. Yo = (t::.T - t1M)a . ma = lI.al(I - lI.a) ,
or lI.a = mal(I + ma). Likewise for lI.b . Substituting for lI.a and lI. b in
equation (2 .39) gives equation (2.40).

6. The real-balance effect derives from the assumption that people

desire to hold a constant proportion of their real income in the form
of real-money -balance holdings. If the price level goes up. the real
value of money holdings goes down, and to restore the real value the
demand for nominal money must increase.

CHAPTER 3

I . In terms of domestic currency the Marshall-Lerner condition

becomes more stringent starting from deficit , i.e. Ex (PXIM) + Em > I .
This leads to the possibility that while the trade balance may be im-
proved in terms of foreign currency, it may not be improved in terms of
domestic currency.

2. Divide the top and bottom of the first half of the expression by
Sx and the top and bottom of the second half by Sm '

3. Starting from payments equilibrium the same result can be

290 Notes

derived, measuring changes in the balance of payments in domestic

currency.

4. Some mitigation is provided by the Export Credits Guarantee

which int roduced in 1975 a scheme to provide exporters of capital
goods costing more than £2 million with a manufacturing period of
more than two years with cover against inflation on fixed-price
contracts.

5. If the competitive advantage of devaluation is maintained, the

J-eurve will become a square-root sign (Masera, 1974), and if it dis-
appears the curve will come down to the level at which it started ,
leaving the balance of payments unchanged .

6 . The barter terms of trade is the ratio of export prices to import

prices measured in a common currency, and changes in the barter terms
of trade give a measure of the change in the real purchasing power of
exports over imports.

CHAPTER 4

I . Specifically, this concerns the suggested idea that all balance-of-

payments deficits are caused by an excess supply of money .

2. Except for the possibility that devaluation might increase

productivity if the traded-goods sector has higher productivity than the
non-traded-goods sector. (This will be discussed later.)

3 . Changes in absorption will also have an effect on income. We

incorporate later the interaction between changes in income and
absorption.

4. Assuming, of course, people have the extra money income to

increase their money expenditure in line with prices . If not, the saving
would be 'forced '.

5. This section draws on Thirlwall (1979a) .

6. The same formula in a different form has been presented by

Johnson (1958), but it is notclear how it is derived .
Notes 291
7. Johnson (1958) is guilty of this, intentionally or not.

8. By Walras's Law, which says that the sum of excess demands and
supplies in all markets must sum to zero.

CHAPTER 5

1. An alternative approach suggested by Coppock (I 978 , and in

correspondence) is to distinguish between traded and non-traded goods.
Let Xg = X n + X t , where XII is the excess demand for non-traded
goods and X, is the excess demand for traded goods (= - B). The
budget constraint leads to X n + X, + X m '" O. Then, if X n = 0, X m =
-Xt = B.

2. The growth of reserves (R) can be shown (see Johnson , 1973) to

be inversely related to the growth of domestic credit expansion (D)
according to the following expression :

1 ,- -' ) dD/D
dR/R = ( - -

where r is the ratio of reserves to the money supply. The growth of

reserves is also positively related to the growth of income , through the
increased demand for money , implying that the higher a country's
growth rate (holding the supply of money constant) , the greater its
balance-of-payments surplus. But how can a higher growth rate be
financed without an increase in the money supply? The evidence does
not square with the theory.

3 . Employing the earlier distinction between traded and non-traded

goods we would have X n + X, + X m + X b '" O. If X n = X m = 0 ,
then X b = - X t = B.

CHAPTER 6

1. It is possible (see the 1S-LM curve analysis later) for fiscal

policy to improve the balance of payments by raising interest rates to
such an extent that higher capital inflows exceed the increase in irn-
292 Notes

ports. In this case the BB curve would be positively sloped and the
problem of assignment would not arise. Mundell's model ignores the
interrelation between fiscal policy and interest-rate changes.

2. In terms of the model outlined above, if YI is external balance,

Y2 is internal balance , XI is monetary policy and X 2 is fiscal policy.
then monetary policy should be assigned to external balance and fiscal
policy to internal balance because a I I /a2 I > a12 /a2 2·

3. There are a number of other factors which may increase the

impact of fiscal policy and reduce the impact of monetary policy which
are too involved to discuss at this introductory level. For a useful
survey, see Currie (1978).

4. For a derivation of the slopes of the IS and LM curves in an open

economy, see Stern (1973, pp . 312-16).

5. Assuming that capital movements are not sensitive to income,

in which case the schedule may slope downwards.

6. This is ruled out of Mundell's original model discussed earlier.

CHAPTER 7

I. Unemployment in the middle of 1966 was still at a record low

of 1.1 per cent .

2. The estimated annual magnitude of the surcharge effect is of

the same order (if not higher) as the estimated annual impact of the
1967 devaluation (see later), which suggests that controls on imports
can be at least as effective as devaluation.

3. The figures shown in Table 7.4 are different from those pub-
lished at the time and therefore differ from those given in Appendix
7.1.

4. But this is also implicit in Worswick's procedure.

5. Probably because of the removal of the import surcharge .

Notes 293
6 . See Chapter 4 for an exposition of this concept.

7. A I per cent movement in the rate is intended to be equivalent

to a I per cent change in sterling 's rate against all currencies as far as
the effect on the U.K. trade balance is concerned .

8. Abridged from my paper 'Another Autopsy on Britain's Balance

of Payments. 1958-67', Banca Nazionale del Lavoro Quarterly Review,
September 1970. The balance-of-payments statistics used here are
different from those in the tables in the chapter because of revisions.

9. Up to 1970.

10. The gap in the visible account the two years is not the
same as the gap in the current account since the figures for the current
balance include payments for U.S. military aircraft and an allowance
for the under-recording of exports, which are excluded from the visible-
account figures . Readers are reminded that the figures here are slightly
different from those in the main body of the chapter because of
revisions made .

I I. A simple time trend fitted to the data yielded a significant result

of the balance of trade (B T ) deteriorating at approximately £1.9
million per quarter (t) , i.e.

BT = -22.27 - 1.920 (t)

(15.31) (0 .763)

12. Since no distinction was made between upswings and down-

swings of activity, the possibility of a ratchet effect was not explored .
The ratchet hypothesis is obviously not refuted, but the level of sig-
nificance of the regression coefficients on A(UCr' reported suggests
a fair degree of symmetry and that any ratchet effect is probably
quite weak in relation to the balance of payments as a whole (as
distinct perhaps from individual items in the account, e.g. imports) .
This lends support to our previous finding that the degree of 'under-
lying'.deterioration in the balance of payments for this type of reason ,
and others, has probably been exaggerated .

13. Excluding payments for U.S. military aircraft and under-

recording of exports.
294 Notes

14. Abridged from my paper, 'The Panacea of the Floating Pound' ,

National Westminster Bank Quarterly Review, August 1974 .

15. For a fascinating catalogue of quotes (and academic somer-

saults!) see Hutchison (1977).

16. Hence Houthakker and Magee's conclus ion that 'Britain can
therefore only grow half as fast as the rest of the world in the long run
if it wants to maintain its exchange rate'. It does not follow, however ,
that the appropriate long-run solution is to allow the exchange rate
to depreciate.

CHAPTER 8

1. A 'Koyck' distributed lag function could also be used to estimate

short - and long-run elasticities if it takes time for imports to adjust to
changes in the independent variables ; but for a tractable solution it
would have to be assumed that the distributed lag-adjustment weights
attached to each of the variables were the same.

2. For the experience since 1967, see Table 9.4 in the appendix
to Chapter 9 .

3. The official statistics only go back to 1970 , but unpublished

data are available from the Treasury for the period 1963 to 1970.

4. It is important that the rise in the ratio of exports to output

should reflect a genuine improvement in exports and not a fall in
output induced by the increase in imports via the workings of the
Harrod trade multiplier. For a discussion of this topic , see Kennedy
and Thirlwall (I 979c).

CHAPTER 9

1. For good discussions of this topic, see Ball (1961) and Cooper,
Hartley and Harvey (1970).

2. Good discussions of the various measures are contained in H.M.

Notes 295
Treasury's Economic Progress Report, February 1978, and in Enoch
(1978).

3. The empirical magnitude of these effects will depend on the

order in which the summations occur. Here it is i first and i second.

CHAPTER 10

1. Exceptions to this statement would be countries able to attract

a continual net inflow of long-term or short-term capital despite a
growing current-account deficit.

2. See Chapter 3.

3. This result would also obtain if the Marshall-Lerner condition

was just satisfied .

CHAPTER 11

1. Kaldor, once an advocate of currency depreciation to improve

export performance, has belatedly turned against it because of its
inflationary repercussions domestically.

2. That is, the relationship between the rate of growth of produc-

tivity as the dependent variable and the rate of growth of output as the
independent variable, which when tested empirically gives a coefficient
of about 0.5. The law is not simply an empirical generalisation , how-
ever. In the mathematical appendix of Verdoorn's original article
he also develops a predictive model (see Thirlwall and Thirlwall , 1979) .

3. Our simple rule enunciated in equation (10 .8) gives a balance-

of-payments equilibrium growth rate of 2.6 per cent.

4. The parameter values and estimates of the variables (rates per

annum) used were 1/ = 1.5, wt + T t = 0.26, ra = 0 .01, PI = 0 .28,
€ = 1, Z t = 0 .03, A = 0.5 , and 1T = 1.51 , and the exchange rate fell
by about 4 per cent per annum over the period .
296 Notes

CHAPTER 12

I . The relation between the two elasticities is as follows. The

income elasticity of demand for exports (e) is xjy . The elasticity of
world trade with respect to world income could be expressed as A
= tly . Thus the elasticity of exports with respect to world trade is
el): If A > I, which it is, then the elasticity of demand for U.K. exports
with respect to world trade will be less than the income elasticity
of demand for U.K. exports.

2. The number of Working Parties has since increased from the

original thirty-five .

3. Assuming that the liabilities are not funded. Since writing this
section of the book the prospects for the non-oil deficit have become so
alarming that any discussion of what to do with the balance-of-payments
gains from North Sea oil now seems hypothetical. The full-employment
current-account deficit is probably of the order of £10 billion at the
time of writing (1982), which is almost twice the net annual gain to the
current account from oil. The discussion is left for hypothetical interest!

4. See Blackaby (1979), which contains papers presented to a

conference on the subject organised by the National Institute of
Economic and Social Research.
References
Alexander , S. (1952) 'Effects of a Devaluation on a Trade Balance',
I.MF. StaffPapers, April.
Alexander , S. (1959) 'Effects of a Devaluation : A Simplified Synthesis
of Elasticities and Absorption Approaches', American Economic
Review, March.
Artus , J . R. (1973) 'The Short Run Effects of Domestic Demand
Pressure on Export Delivery Delays for Machinery', Journal of
International Economics, February.
Artus , 1. R. (1975) 'The 1967 Devaluation of the Pound Sterling ',
I.M.F. Staff Papers, November .
Artus, J. R. and Sosa, S. C. (1978) ' Relati ve Price Effects on Export
Performance: The Case of Nonelectrical Machinery' , I.M.F. Staff
Papers. March.
Bacon , R. and Eltis , W. (1978) 'The Non-market Sector and the Balance
of Payments', National Westminster Bank Quarterly Review, May.
'Balance of Payments : Methods of Presentation' (1964) Bank of
England Quarterly Bulletin, December.
Balassa, B. (1963) 'Some Observat ions on Mr Beckerman's " Export -
Propelled" Growth Model' , Economic Journal, December.
Ball, R. J. (1961) 'Credit Restrictions and the Supply of Exports' ,
Manchester School, May.
Ball, R. J. (1967) 'The Case Against Devaluation of the Pound ' ,
Bankers 'Magazine, April.
Ball, R. 1., Bums, T. and Laury , J. S. (1977) 'The Role of Exchange
Rate Changes in Balance of Payments Adjustment : The U.K. Case',
Economic Journal, March.
Ball, R. J. and Drake , P. S. (1962) 'Export Growth and the Balance
of Payments', Manchester School, May.
Ball, R . J., Eaton , J. R. and Steuer, M. D. (1966) 'The Relat ionship
298 References

Between United Kingdom Export Performance in Manufactures

and the Internal Pressure of Demand', Economic Journal, September.
Ball, R. 1. and Marwah, K. (1962) 'The U.S. Demand for Imports
1948-1958', Review of Economics and Statistics, November
Barker, T. S. (1968) Devaluation and the Rise in U.K. Prices' ,
Bulletin of the Oxford Institute of Economics and Statistics , May.
Barker, T . S. (1970a) The Determinants of Britain's Visible Imports
1949-1966, Department of Applied Economics, A Programme for
Growth No. 10.
Barker, T . S. (1970b) 'Aggregation Error and Estimates of the U.K.
Import Demand Function', in The Econometric Study ofthe United
Kingdom , ed. Hilton, K. and Heathfield, D. (London, Macmillan).
Barker, T . S. and Lecomber, 1. R. C. (1970) 'The Import Content of
Final Expenditure for the U.K.', Bulletin of the Oxford Institute
of Economics and Statistics, February.
Barna , T. (1963) 'Fast- and Slow-growing Products in World Trade',
National Institute Economic Review, August.
Batchelor, R. A. and Bowe, C. (1974) 'Forecasting U.K. International
Trade : A General Equilibrium Approach', Applied Economics,
1une.
Beckerman, W. (1962) ' Project ing Europe's Growth',EconomicJournal,
December.
Beckerman, W. (1963) 'Some Observations on Mr Beckerman's "Export-
Propelled" Growth Model: A Reply, Economic Journal, December.
Black, I. G., Kidgell, 1. E. and Ray, G . F. (I967) 'Forecasting Imports :
A Re-examination', National Institute Economic Review, November.
Black, 1. (1959) 'A Savings and Investment Approach to Devaluation',
Economic Journal, June .
Blackaby, F . (1965) Britain's Share in World Trade in Manufactures,
Woolwich Economic Papers No .8 .
Blackaby, F. (ed.) (I979) Deindustrialisation (London, Heinemann).
Branson, W. H. (1972) The Trade Effects of the 1971 Currency Rea-
lignment, Brookings Papers on Economic Activity No .1 .
Brechling, F . and Wolfe, J . N. (1965) 'The End of Stop-Go', Lloyds
Bank Review, January.
British Export Trade Research Organisation (1975) Export Concen-
tration (London) .
Brooman, F. (1970) Macroeconom ics (London, Allen & Unwin).
Brown, A. J. (1942) 'Trade Balances and Exchange Stability', Oxford
Economic Papers, April .
Caves, R. (1970) 'Export-led Growth : The Post-war Industrial Setting',
References 299
in Induction, Trade and Growth, ed . Eltis , W., Scott, M. and Wolfe ,
J . N. (London, Oxford University Press) .
Central Policy Review Staff (1975) The Future of the British Car
Industry (London, H.M.S.O.).
Cooper, R., Hartley, K. and Harvey, C. (1970) export Performance and
the PressureofDemand (London, Allen & Unwin) .
Coppock, D. J. (1965) 'The Alleged Case Against Devaluation',Man-
chester School, September.
Coppock, D. J . (1978) 'So me Thoughts on the Monetary Approach
to Balance of Payments Theory' , Manchester School, September.
Coppock, D. J. and Metcalfe, J . S. (1974) ' Fo reign Trade and the
Balance of Payments', in The U.K. Economy: A Manual ofApplied
Economics, ed . Prest, A. R . and Coppock, D., 5th edn (London,
Weidenfeld & Nicolson) .
Cornwall, J . (1977) Modern Capitalism: Its Growth and Transformation
(London, Martin Robertson) .
Cripps , F. and Godley, W. (1976) 'A Formal Analysis of the Cambridge
Economic Policy Group Model', Economica, November.
Crockett, A. D. and Goldstein, M. (1976) 'Inflation Under Fixed and
Flexible Exchange Rates' ,I.M.F. StaffPapers, November.
Crossman, R . (1976) The Diaries of a Cabinet Minister : vol. 2 Lord
President of the Council and Leader of the House of Commons,
1966-68 (London, Hamilton/Cape).
Currie, D. (1978) Monetary and Fiscal Policy and the Crowding-
out Issue, Discussion Paper No . 53 (London, Queen Mary College).
Deppler , M. C. (1974) 'Some Evidence on the Effects of Exchange Rate
Changes on Trade', I.M.F. Staff Papers, November.
Dixon , R. J . and Thirlwall , A. P. (l975a) 'A Model of Regional Growth
Rate Differences on Kaldorian Lines', Oxford Economic Papers, July .
Dixon, R. J . and Thirlwall, A. P. (l975b) Regional Growth and Unem-
ployment in the United Kingdom (London, Macmillan) .
Duffy, M. and Renton , A. G. (1970) 'A Model of U.K. Exports of
Manufactures to Industrial Countries' , Applied Economics, vol. 2 ,
no.3 .
Eltis, W. (1967) ' Econ omic Growth and the British Balance of Pay-
ments', District Bank Review, December.
Enoch, C. A. (1978) 'Measures of Competitiveness in International
Trade', Bank ofEngland Quarterly Review, June .
Fausten, D . (1975) The Consistency of British Balance of Payments
Policies (London, Macmillan).
300 References

Flanders, M. J. (1963) 'The Effects of Devaluation on Exports : A Case

Study, United Kingdom 1949-1954', Bulletin of the Oxford Instit-
ute ofEconomics and Statistics, August.
Fleming, J . M. (1962) 'Domestic Financial Policies Under Fixed and
Floating Exchange Rates', I.M.F. StaffPapers, November.
Fleming, 1. M. and Tsiang , S. C. (1956) 'Changes in Competitive
Strength and Export Shares of Major Industrial Countries', I.M.F.
StaffPapers, August.
Frenkel, 1. A. and Johnson, H. G. (eds) (1976) The Monetary Approach
to the Balance of Payments (London , Allen & Unwin) .
Friedman, M. (1953) 'The Case for Flexible Exchange Rates ', in Essays
in Positive Economics (Chicago University Press).
Gilbert, R. S. and Major, R. L. (1961) 'Britain's Falling Share of
Sterling Area Imports', National Institute Economic Review, March.
Godley , W. and Shepherd, J. (1965) 'Forecasting Imports ', National
Institute Economic Review, August.
Goldstein, M. (1973) 'Downward Price Inflexibility, Ratchet Effects
and the Inflationary Impact of Import Price Changes: Some
Empirical Evidence', I.M.F. StaffPapers, November.
Goldstein, M. (1974) 'The Effect of Exchange Rate Changes on Wages
and Prices in the U.K.: An Empirical Study', I.M.F. Staff Papers,
November.
Goldstein, M. and Khan, M. S. (1976) 'Large versus Small Price Changes
and the Demand for Imports',I.M.F. StaffPapers, March.
Gregory, R. G. (1971) 'United States Imports and the Internal Pressure
of Demand 1948-1968', American Economic Review, March.
Gribbin, J . (1971) The Profitability of U.K. Exports, Government
Economic Service Occasional Papers No. I (London, H.M.S.O.).
Haberler, G. (1949) 'The Market for Foreign Exchange and the Stability
of the Balance of Payments: A Theoretical Analysis', Kyklos, fasc. 3.
Hague, D. C., Oakeshott, E. and Strain, A. (1974) Devaluation and
Pricing Decisions (London, Allen & Unwin) .
Hahn, F . (1977) 'The Monetary Approach to the Balance of Payments',
Journal of International Economics, August .
Harberger, A. (1950) 'Currency Depreciation, Income and the Balance
of Trade', Joumal ofPolitical Economy, February.
Harberger, A. (1953) 'A Structural Approach to the Problem of Import
Demand', American Economic Review, Papers and Proceedings, May
Harberger, A. (1957) 'Some Evidence of the International Price Mech-
anism', Journal ofPolitical Economy, December .
References 301

Harrod, R. (1947) Are These Hardships Necessary? (London, Hart-

Davis).
Harrod , R. (l967a) 'Assessing the Trade Returns', Economic Journal,
September.
Harrod, R. (1967b) Towards a New Economic Policy (Manchester
University Press).
Harrod, R. (l968a) 'The British Economy and Prospects for Sterling ',
Bankers'Magazine, January.
Harrod, R. (l968b) 'Devaluation - The Year After', Bankers' Magazine,
December.
Hawtrey, R. (1969a) 'The Case for a Floating Pound', Bankers' Maga-
zine, June .
Hawtrey, R. (1969b) 'A Stable Floating Pound', Bankers' Magazine,
December.
Hemming, M. F. W. and Ray, G. (1959) 'Imports and Expansion',
National Institute Economic Review, March.
Henderson , H. D. (1949) 'The Function of Exchange Rates', Oxford
Economic Papers, January .
Henry, G. (1970) 'Domestic Demand Pressure and Short Run Export
Fluctuations' , Yale Economic Essays, Spring.
Hibberd, J . and Wren-Lewis, S. (1978) A Study of LiK. Imports of
Manufactures, Government Economic Service Working Paper No.6
(London, H.M.S.O.).
Hicks, J. R. (1966) After the Boom : Thoughts on the 1966 Economic
Crisis (London, Institute of Economic Affairs) .
Hirsch, F. and Higham, D. (1974) 'Floating Rates - Expectation and
Experience', Three Banks Review, June .
Holmes, P. M. (1978) Industrial Pricing Behaviour and Devaluation
(London, Macmillan).
Houthakker, H. S. and Magee, S. P. (1969) 'Income and Price Elastici-
ties in World Trade', Review of Economics and Statistics, May.
Hovell, P. (1968) 'Export Pricing Policies', District Bank Review,
September.
Hughes, J. J. and Thirlwall, A. P. (1977) 'Trends and Cycles in Import
Penetration in the U.K.: A Disaggregated Study', Bulletin of the
Oxford Institute ofEconomics and Statistics, November .
Hughes, J. J. and Thirlwall, A. P. (1979) 'Imports and Labour Market
Bottlenecks: A Disaggregated Study for the U.K.', Applied Econ-
omics, March.
Humphrey, D. H. (1976) 'Disaggregated Import Functions for the U.K.,
West Germany and France', Bulletin of the Oxford Institute of
Economics and Statistics, November.
302 References

Hutchison, T. W. {l977) Knowledge and Ignorance in Economics

(Oxford, Blackwell).
Hutton, J. B. and Minford, A. P. L. (1975) A Model of VX Manu-
facturing Exports and Export Prices, Government Economic Service
Occasional Paper No. 11 (London, H.M.S.a.).
Imlah, A. H. (1958) Economic Elements in the Pax Britannica (Cam-
bridge , Mass., Harvard University Press).
Johnson, H. G. (1958) 'Towards a General Theory of the Balance of
Payments', in International Trade and Economic Growth (London,
Allen & Unwin).
Johnson, H. G. (1972) 'The Monetary Approach to Balance of Pay-
ments Theory', Journal of Financial and Quantitative Analysis,
March.
Johnson , H. G. (1973) 'The Monetary Approach to Balance of Pay-
ments Theory', in Further Essays in Monetary Economics (London,
Allen & Unwin).
Johnson, H. G. (1977a) 'The Monetary Approach to Balance of Pay-
ments Theory and Policy: Explanation and Policy Implications',
Economica, August.
Johnson, H. G. (l977b) 'The Monetary Approach to the Balance of
Payments' , Journal of International Economics, August.
Johnston, J. and Henderson, M. (1967) 'Assessing the Effects of the
Import Surcharge', Manchester School, May.
Junz, H. and Rhomberg, R. (1965) 'Prices and Export Performance
of Industrial Countries, 1953-63',IM.F. StaffPapers, July .
Junz, H. and Rhomberg, R. (1973) 'Price Competitiveness in Export
Trade Among Industrial Countries', American Economic Review,
Papers and Proceedings, May.
Kaldor, N. (l970) 'The Case for Regional Policies', Scottish Journal
ofPolitical Economy, November.
Kaldor , N. (1974) 'The Road to Recovery' , New Statesman, 1 March.
Kaldor, N. (1979) 'The Effect of Devaluations on Trade in Manufac-
tures', in Further Essays on Applied Economics (London , Duck-
worth).
Kay, J. and Forsyth, P. (1980) 'The Economic Implications of North
Sea Oil Revenues', Fiscal Studies, July.
Kenen, P. (1975) Floats. Guides and Indicators, International Finance
Section (Princeton University Press).
Kennedy, C. and Thirlwall , A. P. (l979a), 'The Foreign Trade Multi-
plier Revisited' , in Proceeding of the A. V.T.E. Conference 1978, ed.
Currie , D. and Peters, W. (Croom Helm).
Kennedy, C. and Thirlwall, A. P. (1979b) 'The Input-Output Formu-
lation of the Foreign Trade Multiplier',Australian Economic Papers.
June .
References 303

Kennedy , C. and Thirlwall, A. P. (1979c) 'Import Penetration , Export

Performance and Harrod 's Trade Multiplier ' , Oxford Economic
Papers, July.
Key Statistics 1900-1966 , London and Cambridge Economic Service,
Khan , M. S. and Ross, K. Z. (1975) 'Cyclical and Secular Income Elas-
ticities of the Demand for Imports' , Re view of Economics and
Statistics, August.
Kindleberger, C. P. (1969) 'Measuring Equilibrium in the Balance of
Payments', journal of Political Economy, November-December.
Klein, T. M. (1964) 'The United Kingdom Balance of Payments
Accounts' , Economic Journal. December .
Kravis, I. (1956) 'Availability and Other Influences on the Commodity
Composition of Trade',Journal ofPolitical Economy, April.
Kreinin , M. E. (1967) ' Price Elasticities in International Trade' , Review
ofEconomics and Statistics. November.
Kreinin , M. E. (1977) 'The Effect of Exchange Rate Changes on the
Prices and Volume of Foreign Trade ' , I.M.F. Staff Papers, July .
Krueger, A. (1969) 'Balance of Payments Theory' ,Journal ofEconomic
Literature, March.
Lamfalussy, A. (1963) The United Kingdom and the Six (London,
Macmillan).
Lamfalussy, A. (1976) 'Beyond Bretton Woods: Floating Exchange
Rates and Capital Movements', in Keynes and International Mon-
etary Relations, ed . Thirlwall , A. P. (London , Macmillan).
Laursen , S. and Metzler, L. (1950) 'Flexible Exchange Rates and the
Theory of Employment ', Review of Economics and Statistics ,
November .
Leamer , E. and Stern , R. M. (1970) Quantitative International Econo-
mics (Boston, Allyn & Bacon).
London and Cambridge Economic Service (1968) Key Statistics 1900-
1966.
MacGeehan , J. M. (1968) 'Competitiveness: A Study of the Recent
Literature ' , Economic Journal, June.
Machlup, F. (1950) 'Three Concepts of the Balance of Payments and
the So-Called Dollar Shortage ', Economic Joumal, March.
Machlup, F. (1955) 'Relative Prices and Aggregate Spending in the
Analysis of Devaluation', American Economic Review. June.
Machlup, F. (1956) 'The Terms of Trade Effects of Devaluation upon
Real Income and the Balance of Trade ', Kyklos, fasc. 4.
Machlup , F. (I973) 'Exchange Rate Flexibility', Banca Nazionale del
Lavoro Quarterly Review, September.
304 References

Magee, S. P. (1974) 'Prices, Income and Foreign Trade' , in International

Trade and Finance: Frontiers for Research, ed . Kenen , P. B. (Cam-
bridge University Press).
Major, R. L. (1960) 'World Trade in Manufactures', National Institute
Economic Review, July .
Major, R. L. (1968) 'Note on Britain's Share of World Trade in Manu-
factures 1954-1966' , National Institute Economic Review, May.
Marston, R. (1971) ' Income Effects and Delivery Lagsin British Import
Demand 1955-1967', Journal of International Economics, November
Masera, R. (1974) 'The J-Curve: U.K. Experience After the 1967
Devaluation' , Metroeconomica , December-January.
Meade, J . E. (1951) The Theory of International Economic Policy:
Vol. 1 The Balance ofPayments (Oxford University Press).
Meade, J . E. (1966) 'Exchange Rate Flexibility', Three Banks Review,
June .
Michaely, M. (1960) 'Relative Prices and Income -Absorption Approa-
ches to Devaluation : A Partial Reconciliation', American Economic
Review, March.
Mitchell, B. R. and Deane, P. (1962) Abstract of British Historical
Statistics (Cambridge University Press).
Morgan, A. D. (1970) 'Income and Price Elasticities in World Trade :
A Comment', Manchester School, December.
Morgan, A. D. (1971) 'Imports of Manufactures into the U.K. and
Other Industrial Countries 1955-69', National Institute Economic
Review, May.
Morgan, A. D. and Martin, A. (1975) 'Tariff Reductions and U.K.
Imports of Manufactures 1955-71', National Institute Economic
Review, May.
Mundell, R. (1960) 'The Monetary Dynamics of International Adjust-
ment Under Fixed and Flexible Exchange Rates', Quarterly Journal
ofEconomics, May.
Mundell, R. (1961) 'Flexible Exchange Rates and Employment Policy',
CanadianJournal ofEconomics, November.
Mundell, R. (1962) 'The Appropriate Use of Monetary and Fiscal
Policies for Internal and External Stability', IM.F. Staff Papers,
March.
Mundell, R. (1963) 'Capital Mobility and Stabilisation Policy Under
Fixed and Flexible Exchange Rates' , Canadian Journal of Econ-
omics, November.
Mussa, M. (1974) 'Monetary Approach to Balance of Payments Analy-
sis',Journal ofMoney, Credit and Banking, August.
 References 305
National Economic Development Council (1963) Export Trends
(London, H.M.S.O.).
National Economic Development Council (1965) Investment in Mach-
ine Tools: A Survey by the Management Consultants' Association
(London, H.M.S.O.).
National Economic Development Office (1965) Imported Manufac-
tures: An Inquiry into Competitiveness (London H.M.S.O.).
National Institute of Economic and Social Research (1972) The
Effects of the Devaluation of 1967 on the Current Balance of
Payments' , Economic Journal, Supplement, March.
Odling-Smee, J. and Hartley, N. (1978) Some Effects of Exchange
Rate Changes, Government Economic Service Working Paper No.2
(London, H.M.S.O.).
Oppenheimer, P. M. (1965) 'Is Britain's Worsening Trade Gap due to
Bad Management of the Business Cycle?', Bulletin of the Oxford
Institute of Economics and Statistics, August.
Orcutt, G. H. (1950) 'Measurement of Price Elasticities in International
Trade', Review of Economics and Statistics, May.
Orr , D. (1978) 'The Customer Rules OK .', Sunday Times, 9 August.
Page, S. A. B. (1975) 'The Effect of Exchange Rates on Export Market
Shares', National Institute Economic Review, November.
Paish, F. W. (1968) ' How the Economy Works', Lloyds Bank Review,
April.
Paish, F. W. (1969) Rise and Fall of Incomes Policies, Hobart Paper
No. 47 (London, Institute of Economic Affairs).
Panic, M. (1968) 'British Imports of Manufactured Goods ' , Business
Ratios, Winter.
Panic, M. (1975) 'Why the U.K.'s Propensity to Import is High', Lloyds
Bank Review, January.
Panic , M. and Rajan, A. H. (1979) Product Changes in Industrial
Countries' Trade, 2nd edn (London , N.E.D.O.).
Panic, M. and Seward, T . (1966) 'The Problem of UK. Exports',
Bulletin of the Oxford Institute of Economics and Statistics,
February.
Pearce, I. F. (1961) 'The Problem of the Balance of Payments', Inter-
national Economic Review , January.
Posner, M. and Steer, A. (1979) 'Price Competitiveness and the Per-
formance of Manufacturing Industry', in Blackaby (1979).
Prais, S. J . (1962 'Econometric Research in International Trade :
A Review', Kyklos, fasc. 3.
306 References

Radcliffe Committee (1959) Report on the Working of the Monetary

System, Cmnd 827 (London, H.M.S.O.).
Ray, G. F. (l966) The Competitiveness of British Industrial Products:
A Round-Up, Woolwich Economic Paper No. 10.
Rees, R. and Layard, P. (I 971) Determinants of U'K, Imports, Govern-
ment Economic Service Occasional Paper No.3 (London, H.M.S.O.).
Renton, G. A. (l967) 'Forecasting British Exports of Manufactures to
Industrial Countries', National Institute Economic Review,
November.
Robinson, J. (1937) 'The Foreign Exchanges', in Essays in the Theory
of Employment (Oxford, Blackwell).
Robinson, 1. (l966) Economics : An Awkward Corner (London, Allen
& Unwin).
Rosendale, P. (1973) <the Short Run Pricing Policies of Some British
Engineering Exporters', National Institute Economic Review, August.
Rowan , D. C. (l976) 'Godley's Law, Godley's Rule and the "New
Cambridge Macro Economics"', Banca Nazionale del Lavoro Quar-
terly Review, June.
Scott , M. (1973) A Study of UK Imports (Cambridge University Press).
Singh, A. (1977) 'U.K. Industry and the World Economy : A Case of
De-industrialisation?', Cambridge Journal of Economics, June .
Smyth, D. 1. (1968) 'Stop-Go and United Kingdom Exports of Manu-
factures' , Bulletin of the Oxford Institute of Economics and Statis-
tics, February.
Smyth, D. J . (l969) The Effect of Internal Demand Pressure on United
Kingdom Imports of Manufactures' , Journal of Economic Studies,
November.
Sohmen, E. (1961) Flexible Exchange Rates: Theory and Controversy
(Chicago University Press).
Stern, R. M. (1973) Balance ofPayments Theory and Economic Policy
(LDndon, Macmillan).
Steuer , M. D., Ball, R. J . and Eaton , J. R. (l966) The Effect of Waiting
Times on Foreign Orders for Machine Tools', Economica, November.
Stewart , M. (1977) The Jekyll and Hyde Years: Politics and Economic
Policy Since 1964 (London, Dent).
Stout, D. (1977) International Price Competitiveness, Non-Price
Factors and Export Performance (London, N.E.D.O.).
Swan, T . (1963) 'Longer Run Problems of the Balance of Payments',
in ' The Australian Economy, ed. Arndt, H. W. and Corden, M.
(Melbourne, Cheshire) .
 References 307

Taplin , G. R. (1973) 'A Model of World Trade', in The Int ernational

Linkage of National Economic Models, ed. Ball, R. J. (Amsterdam ,
North-Holland) .
Thirlwall , A. P. (1970) 'Another Autopsy on Britain 's Balance of
Payments: 1958-67' , Banca NazionaIe del Lavoro Quarterly Review,
September.
Thirlwall , A. P. (1974) 'The Panacea of the Floating Pound' , National
Westminster Bank Quarterly Review, August.
Thirlwall, A. P. (1978) 'The U.K.'s Economic Problem : A Balance of
Payments Constraint', National Westminster Bank Quarterly Review,
February.
Thirlwall, A. P. (1979a) 'The Interaction Between Changes in Income
and Changes in Absorption in the Absorption Approach to the
Balance of Payments' ,Journal ofMacroeconomics.
Thirlwa II, A. P. (1979b) 'The Balance of Payments Constraint as an
Explanation of International Growth Rate Differences ', Banca
Nazionale del Lavoro Quarterly Review, March.
Thirlwall , A. P. and Thirlwall, G. (1979) 'Factors Governing the
Growth of Labour Productivity [translation of P. J. Verdoorn's
article in L 'industria, 1949], Research in Population and Economics,
Autumn .
Thirlwall, A. P. and White, H. (1974) 'U .S. Merchandise Imports and
the Dispersion of Market Demand', Applied Economics, December.
Tsiang , S. G. (1961) 'The Role of Money in Trade Balance Stability :
Synthesis of the Elasticity and Absorption Approaches' , American
Economic Review, December.
Tsiang, S. G . (1977) 'The Monetary Theoretic Foundation of the
Modern Monetary Approach to the Balance of Payments' , Oxford
Economic Papers, November.
Viner, J. (1964) Problems of Monetary Control (Princeton University
Press).
Wells, S. (1964) British Export Performance : A Comparative Study
(Cambridge University Press).
Whitley, J. D. (1979) 'Imports of Finished Manufactures: The Effects
of Prices, Demand and Capacity', Manchester School, December.
Whitman, M. von (1970) Policies for Internal and External Balance.
Special Papers in International Economics No.9 (Princeton Uni-
versity Press).
Whitman, M. von (1975) Global Monetarism and the Monetary
Approach to the Balance of Payments, Brookings Papers on Econ-
omic Activity No.3 .
Williamson, J. H. (1965) The Crawling Peg, International Finance
Section , Princeton University.
Williamson, 1. H. (1971) 'On the Normative Theory of Balance of
308 References

Payments Adjustment', in Monetary Theory and Monetary Policy

in the 1970 's, ed . Clayton, G., Gilbert, J . and Sedgwick , R. (Oxford
University Press) .
Wilson, H. (1971) The Labour Government 1964-1970 (London,
Weidenfeld & Nicolson).
Wilson, T . (1976) 'Effective Devaluation and Inflation', Oxford Econ-
omic Papers, March.
Winters, L. A. (1974) 'United Kingdom Exports and the Pressure of
Demand: A Note', Economic Journal, September .
Worswick, G. N. D. (1971) 'Trade and Payments', in Britain's Economic
Prospects Reconsidered, ed . Cairncross, A. (London, Allen & Unwin).
Yeager, L. B. (1970) 'Absorption and Elasticity : A Fuller Reconcili-
ation', Economica, February.
Name Index
AJexander,S . 68,69,98 ,101,106 Dixon, R.J. xiii, 239, 267, 271
Apel, Hans 278 Duffy , M. 237
Artus, J . R. 83,155,229,278
Eaton, 1. R. 226 ,228
Balassa, B. 265 Eltis, W. 170,199
Ball, R. J . 90,96,151 ,188, 222n9.1, Enoch, C. A. 230n9 .2,235
226,228,254
Balogh, T. 150 ,153 Fausten, D . 132,133
Barber, A. P. L. 189 Flanders, M. J. 142
Barker, T. S. 89,96, 152 ,211 Fleming, J. M. 128
Barna, T. 243 Forsyth, P. O. 285
Batchelor, R. A. 237 Frenkel, 1. A. 110
Beckerman, W. 263,265 ,266 Friedman, M. 31,90,186
Benn, A. 153
Black, I. G. 151 Godley, W. 49n2.2
Blackaby, F. 283n12.4 Goldstein, M. 47,87,89,97
Bowe,C. 237 Goy , J . xiii
Branson, W. H. 87 Gribbin, J. 80
Brechling,F . 170,178,199,227
British Export Trade Research Organi- Hague, D.C. 80 ,82
sation 276 Hahn , F . 110
British Overseas Trade Board 278 Harberger , A. 78
Brooman , F . 63,64 ,65 Harrod, R. 68, ISO, 151, 188 ,
Bums, T. 90,254 221n8.4, 238
Hartley, K. 222n9.1
Callaghan, J . 153 Hartley, N. 160
Cambridge Economic Policy Group Harvey,C. 222n9.1
286 Healey , D. 163 ,188, 189
Castle, B. 153 Henderson, H. D. 90
Catherwood, F . 278 Henry, G. 227
Caves, R. 263, 266 Hibberd,1. 212
Cooper, R. 222n9 .1 Hicks, J . R. 150
Coppock, D. 1. 10, 111n5 .1, 149 Higham, D. 158 , 159
Cornwall, J. 256,257,261,265 ,266, Hirsch, F. 158,159
267 Holmes, P. M. 80 ,83
Craven, J . xiii Houthakker, H. S. 191, 192n7 .16,
Cripps, F. 49n2.2 204, 211 , 229, 230, 237 , 255 , 257,
Crockett, A. D. 47 271
Currie, D. 129n6.3 Hovell, P. 82
Hughes,J .J. 52,201,212
Deane, P. 143, 139n3, 141 , 147 Hume, D. 67
Department of Trade and Industry HumphreY,D.H . 210
xiii Hutchison, T. W. 151, 188n7.15
Disney, R. xiii Hutton, J. B. 236
310 Name Index
Imlah, A. H. 238n3 Oakeshott, E. 80,82
Odling-Smee.T, 160
Oppenheimer . P. M. 170,227
Johnson, H. G. 69,70, 98, 104n4.6, Orcutt, G. H. 93,96 ,97
106n4.7, 107,110, 112n5.2, 116, 184 Orr, D. 276,278
Johnston,J.151
Junz ,H . 83,237,238
Parish,F.W . 170,190,199
Panic, M. 204,205 ,210,211,228,
Kahn, M. S. 97 241,242,243
Kaldor, N. 160, 193, 262nl1.1, 267, Posner, M. 235,237
268,269,270 Prais, S. J . 96
Kay,J. 285
Kenen,P. 42 Radcliffe Committee 145
Kennedy, C. xiii, 49 , 61, 22n8.4
Rajan, A. K. 241,242
Kern, D. 255,257 Ray,G.F. 149,151
Khan, M. S. 198 Rees, R. 198,211
Kidgell, 1. E. 151 Renton, A. G. 237
Kindleberger, C. P. 19 Rhomberg, R. 83, 237, 238
Kravis, I. 243 Robbins, Lord 151
Kreinin, M. E. 241
Robinson , J. 23,50, 67, 150
Rosedale, P. 83
Lamfalussy, A. 33,263,266 Ross, K. Z. 198
Laury ,J.S. 90,254 Rowan, D. C. 49n2 .2
Layard, P. R. 198,211
Leamer, E. 195 Seward,T. 228,241,242,243
London BusinessSchool 90
Sheldon, R. 189
Shore, P. 153
MacGeehan, J. M. 149 Sinclair, T. xiii
Machlup, F. 20,42,101 Singh, A. 283
Magee, S. P. 191, 192n7.16, 204, Smyth, D. J. 170, 227
211,212,229,230,237,239,255 Social Science Research Council xii
257,271 Sosa, S. C. 278
Major, R. L. 241 Steer, A. 235,237
Marston, R. 198 Stern , R. M. 78,92, 130n6.4 , 195
Marwah, K. 96 Steuer , M. D. 226,228
Masera, R. 84, 84n3 .5, 155 Stevens, J. xiii
Meade,J .E. 20,90,118,119 Stewart, M. 150
Metcalfe, J. S. 10 Stout, D. 276, 277
Michaely, M. 106 Strain, A. 80,82
Minford, A. P. L. 237 Swan, T. 121
Mitchell, B. R. 134, 139n3, 141, 147
Mundell, R. 117, 125n6.1, 126n6 .2,
Taplin,G.R. 205,211
132n6.6 Thirlwall, A. P. 49,52,61, 103n4.5,
181n7.14, 188,201,212,
National Economic Development Coun- 221n8.4, 239, 267, 271, 283
cil 199,200,204,241 ,242 Tsiang, S. G. 106, 114
National Economic Development Office
276,277,278 Unilever 276
National Enterprise Board 283
National Institue of Economic and
Social Research 83,89,151 Viner, J. 186
Name Index 311
White, H. 52 Winters, L. A. 225
Whitley, J. D. 211 Wolfe, J. N. 170,178,199,227
Whitman, M. von 114 Worswick, G. N. D. 154, 155,
Williamson, J. H. 42,128 155n7.4
Wilson, H. 150, 153 Wren-Lewis,S. 212
Wilson, T. 254
Subject Index
valuation; disequilibrium 5, 9, 19,
E. = Equation ; F. = Figure; T. =Table 21, 22, 28, 39-41, 44, 71, 106,
110, 112, 113, 128,148,167,181,
absorption approach to balance of pay- 191, 274; equilibrium xi, 2, 5-22,
ments 66, 68-9, 92, 98, 98n4.1, 27-30, 43, 49-53, 66, 125, 167,
99-109, (E.4.1-4.18), 117 167n7.9, 250nl0.l , 251, 263,
Advisory Committee on the Presenta- 266, 269-72, 271nl1.3, 11.4,
sion of Balance of Payments 272, equilibrium growth rate, see
Statistics 8n1.1 balance-of-payments equilibrium
'Another Autopsy on Britain's Balance growth rate; faulty statistics 179,
of Payments: (1958-61)' (Thirlwall), history, see balance of payments
Banco Nazionale del Lavoro Quar- history of United Kingdom, integra-
terly Review 293n7.8 tion into national accounts 48-50;
Arabian American Oil Company Marshall-Lerner condition, see Mar-
(Aramco) 161 shall-Lerner condition ; nature of
Articles of Agreement (I. M. F.) 36 183-4 ; official financing 8, 8nU,
1.2,9-10,17-18 (T.1.4), 19, 1951
balance of payments: accounting xi, -80 146-7 (T .? .4),164; rectifying
1-22, 48-71; accounts 10, 11 imbalance 27, 28, 68, 90, 160;
(T.1.1); adjustment, see balance-of- relationship to functioning of econ-
payments adjustment policy; balanc- omy 49-54; saving 7, 49, 99,
ing items 4, 8, 10, 35; capital 99(E.4.5);simultaneous internal and
account 3, 9, 17-18(T.1.4), 45, external balance, see simultaneous
70, 110, 145-7, 165-7, 177, internal and external balance; statis-
'below the line' 39 ; conflict be- tics 1854-1939 135-9 (T.7.!),
tween equilibrium and economic 1940-50140-1 (T.7.2), 1951-80
policy 50-4; contribution of North 146-7(T.7.4), 152n7.3, 169(T.7.7);
Sea oil and gas 280-3, (T. 12.1); structural problems, see structural ;
corrective action 2, 69, 99, 183; surplus 2, 6, 8, 20, 28, 29, 38, 45,
current account xii, 3-16, 46, 50, 66-7, 111, 112, 125, 143,
(E.1.-5), (T.1.2, 1.3), 23, 70, 145,270,280
110, 'above the line' 39, balance balance-of-payments adjustment policy
6, 8, 9, 12-13 (T.1.2), 99, (E.4.4, xi, 22, 23, 37-8, 40, 65-71,
4.5), 135-9 (T.7.1), 1940-50 72-97,98-109,128,274, absorp-
140-1 (T.7.2), 1951-80 146-8 tion approach 66, 68-9, 92, 98,
(T.7.4), 174 (T.7.8), 176 (E.7.5a), 98n4.1, 99-109, (E.4.1-4.18),
177, 260-73, basic balance 8, 9, 117; biasing factors on price 93-4,
10, 143, 146-7 (T,7.4), 152, 183; 94(F .3.5), 95-6(E.3.34-3.39), 97;
deficit 2, 5-8,19,20,23,27,29, classical price-specie flow mechan-
38, 39, 44,45,46,4~66-7,6~ ism 66-7 , elasticity approach 65,
98n4.1, 106, 106n4.7, 107-8, 111, 67-9,72-97, (F.3.1-3 .5), (E.3.1-
112, 125, 126, 140, 141 (T.7.2), 3.39), 274; expenditure-switching
150, 151, 165, 166, 173, 17n7.3, policies 99, 99n42, 100, 108,
270, 280; and devaluation, see de- 115-17, 178; monetary approach
 Subject Index 313

66, 69-71, 90, 107-8, 110-17; capital assets 1, 7, 31, 113; export of
term of trade and devaluation 3, 31-3; imports of 4; inflows 31,
92-3,100-1 108, 113, 230nl0.1, long-term 4,
balance - of - payments equilibrium 7, 10, short-term 4, 5, 7-10; long-
growth rate 52-3, 66, 237, 250, term 1940-50 140-1 (T.7.2);
251-8, (E.1O.1-10.9), (T.IO.1, movements 5, 6, 23n1.3 31-3
10.2) 41, 44-·5, 127, 131n6.5, 158; out-
balance of payments, history of, in the flows 9, 31, long-term 10, short-
United Kingdom 134-93, (T.7.1- term 10, 164; private inflows 19-
7.9), (E.7.5,7.5a), (F .7.1) 21; reserves 4; transactions analysis
Balance of Payments Pink Book 17-18 (T.1.4)
1967-1977 144,147 Caves model of export-led growth
balance of trade, see trade balance 263,266
Bank of England 152, 163, 164 Census of Production Reports 212,
Bank of England Historical Abstract 220
147 Central Policy Review Staff Report on
Bank of England Quarterly Bulletin Future of the British Car Industry
162 279
bank rate 145 Central Statistical Office 10, 13, 16,
Banker. The 79 18,246,249
barter 21, 92, 92n3.6, 93, 101 classical price-specie flow mechanism
Beckerman model of export-led growth 66-7
263-6, (E.l1.1-11.8) commodity composition of visibletrade
Bernstein Committee 8n1.1 244-6 (T.9.4)
bias: aggregation of price 95-6, commodity exports comparisons 142
(E.3.34-3.39) ; factors on price competition 79; foreign 80; mono-
93-4, 94 (F.3.5), 95-6, (E.3.34- polistic 80; perfect 80; trade
3.39),97 ; of trade 242, 243 170-1
Board of Trade 139nl competitiveness, see exports ; imports ;
Board of Trade Journal 139n3 , 189 price
bond market 108, 108n4.8, 112, 113 competitveness, non-price 275, 276
book-keeping double-entry 1-3 consumption: autonomous 57 (E.2.15),
borrowing 70; foreign 7, 17-18 58(E.2.16), 59, 62n2 .4; disposing
(T.1.4), 45, 49 , 180, 203, govern- of income 48 ; effect of rising prices
ment 19; from International Mone- on 102; induced 59, 62n2.4 ; goods
tary Fund 40 ; public sector 19; 59-60, 62 ; -led bloom 274; North
186-8; short-term 163; vulnerable Sea oil revenues supporting 282 ;
to speculative crisis 7, 10, 19 private, generating income 48
bottlenecks: affecting imports 52, contracts, forward 83
170, 187. 199, 204, 273, economic costs 85; import 85-7, (E.3.12-
200, 273 ; labour market 200-1, 3.22) ; inflation 190-91 ; profits
203 85-7, (E.3.12-3.22) ; structure 42 ;
Bretton Woods (1944) 33, 39,40,41, wage 85-7, (E.3.12-3.22), 88
43,47,158 (E.3.25), 149, 190, 235-6, 264
British Export Trade Research Organi- crawling peg exchange rate 42
sation 276 currency: appreciation 33, 42, 121
bullion and specie 134, 135-9 (T.7.1) (F.6.1) ; buying own with foreign
exchange 1; depreciation xi, 19,
30,33,36 ,42,67, 84, 121(F.6.l),
253, 262, 262nl1.1, 273, 274,
276-7. 285; depreciation to rectify
Cambridge Economic Policy Group balance-of-payments deficit 68,
286 160; devaluation xi, 36,41 ,42,67,
314 Subject Index
currency [contd] 25; price of exports 74, 75, 75
70, 150, 253; flows (total 8-10, (F .3.1); price of imports 75, 75
11(T .1.1) , 22-3, 27, 116, 146-7 (F .3.2); selling forward 35; sterling
(T.7.4); home, in excess demand 1; price 24, 26-30, (F.1.2-1.6), 89,
home, in excess supply 1; main- 161, 162(T.7.6); supply of 23,
taining external value 9; managed 26-30, (F .1.3- 1.6)
floating 42-3, 158, 181-4; par- Domestic Credit Expansion 108-9,
value 33, 40, 41; reserves 42, 43, 112n5.2, 156, 156n7.6
47,70,160,164,182; selling own,
for foreign exchange 1; specula-
tion 22, 30, 31-3, 182; strong
42, 109; supply and demand 22,
31; weak 42, 45 ,109 economic analysis of United Kingdom's
See also foreign currency ; specu- payments difficulties in the 1960s
lation 166-7, 168(T.7.7) , 170-1, 174-5
Customs and Excise 157 (T.7.8), 172, 172(F .7.1), 173-4,
177(T.7.9), 176, 176(E.7.5, 7.5a),
deflation 157 178-81; difficulties of United
de-industrialisation 283, 283nI2.4, Kingdom and depreciation of the
286 pound 184-7, 274 ; policy objec-
delivery delays 83, 227 -8, 276, 278, tive 6, 106; Robinson's (Mrs)
279 'magic quadrilateral ' 50; strength
demand: dollar 23-5, 26(E.1.2) , 27, 5 , structure xii, 108
28(E.1.4), 30(E.1.6 .); elasticity, see Economic ProgressReport (HM . Treas-
exports, imports; pressure 176-8 ury) 230n9.2, 236(T.9.2), 281
depreciation, see currency ; dollar; (T.12.!)
sterling Economist. The 182
depression 21, 262 ; world 39 economy; bottlenecks 52, 170, 187,
devaluation xi, 36, 41, 42 , 165n7 .8, 199, 200, 204, 273 ; capacity level
266; effect on absorption 100, 222-30, 252, 261-3, 273-4,
100n4.3 , 101-3 ; effect on balance closed 7, 48 , 54-7; domestic 22,
of payments 67, 70, 76, 77, 32-3, 37, 47, 80, 148, 158, 180,
77(E.3.2), 78,84,98, 104n4 .6, 140, 183, 184, unfavourable effects on,
142 ; effect on income 100-3, 103 of devaluation 68; expanding 187,
(E.4.9, 4.10); effect on price, see 189, 274 ; international 39; macro
price; effect on terms of trade 92-3, 49 , 50, 118, 124, 145, 160,273 ;
100-1 ; effect on trade balance national 48n2 .1, 48-71; open 7,
154-5, 155n7.4, 156-7 ; Goldstein 48, 49, 54-7, 130n6.4; policy,
model 89-90; and inflation 68, conflict with balance-of-payments
79n3.4, 85-92, 149; opposition to equilibrium 50-4, 191-2; private
149-154 ; and response of rums sector in balance 49 , 49n2 .2; siege
78-81, 81(F .3.3) , 82-4; Wilson 20; world 2, 22, 45 -7 See also
model 85-9(E.3.12-3.25) real economy
disequilibrium on balance of payments, E.E.C. countries 228, 269
see balance of payments; in mone- elasticity approach to balance-of-pay-
tary policy 70-1 ments adjustment policy 65,67-9,
dollar : appreciation 24, 35; converti- 72-97, (F .3.1-3.5), (E.3.1-3.39),
bility into gold 40, 41, 158; 273
demand for 23, 26, 26(E.1.2), 27, employment: constraints on 19, 40,
28, 28(E.1.4), 30, 30(E.1.6); depre- 275 ; full, and balance-of-payments
ciation 24, 35; devaluation 41, 74, equilibrium xi, 44n1.6, 52, 120,
75, 158 ; elasticity of demand 24, 121(F.6.1), 122-3, 125, 126, 145,
Subject Index 315
173, 261, 262, 262nll.l, 280, 285 export-led growth 259-72
full, and devaluation 77, 105, 150 export performance, causes of poor
full, and export output 221, 262 United Kingdom 191-3
level 42, 283, 283n12.4 ; manu- cxports :capacity 222-30,252,261-3 ;
facturing 285, 285(T.12.2); safe- capital 3, 31; changes affecting
guarding full xi, 20,40-1,50,51 national income 5; commodity
equilibrium on balance-of-payments composition 241, 244(T .9.4) ; com-
account, see balance of payments petitiveness 79n3.4, 227, 228,235,
exchange rate: adjustable peg 40-2, 236, 236(T .9.2) ; credit 10; deter-
47; adjustment xi, 40, 142, 148, minants 221-2; dollar price elas-
192, 285 ; alternative regimes 37- ticity 74, 75, 75(F.3 .1) ; effects of
43,131-2, 262n6 .2, 6.3 ; apprecia- devaluation on 154-7 ; elasticity
tion 31 ; changes xi, 10, 23, 27, 29, 53-4, 71-8, (E.3.1-3.7), 80 ,
29, 37, 42, 50, 66-8, 73, 274, 92-3, 222, 275n12 .1; functions
282-3; changes on wage and price 221-49; geographic composition
behaviour, Goldstein model 89-90, 247(T .9.5); gold 38-9 ; growth
(E.3.26-3.28); crawling peg 42; rate, see growth rate ; inelasticity
crises 265 ; depreciation xii, 29- 29; and investment 54-6; invisible
32, 90, 127, 160, 181, 295 ; 2, 4, 6, 134-9 ; marketing 51 ;
destabilising 31, 32; devaluation multiplier 59, 59(E .2.22) ; price
76-83, (E.3.5-3.11), (F.3.3); dis- elasticity 222, 226, 236, 237, 237
equilibrium 28, 30; effective 161, (T.9.3) , 238; profitability 79-81,
161n7.7 , 162(T.7.6) ; equilibrium 81(F .3.3), 82-3,225,235-6,236
21, 22, 29, 30, 32 ; fixed 1, 21, (T.9 .2); relative price 227, 230-6,
22, 23, 32, 33, 37-9, 40-2, 51, 236(T.9.2) , 237 -8 ; statistics (1854
65, 67, 124-9, 132, 180, 273; -1939) 135-39, (T.7.1), (1940-
fixed versus flexible 43 - 7; flexible 50) 141(T.7.2) , (1951-80)
(floating) xi, 1,9 ,21,22,23,27, 146-7, (T.7.4) ; supply and demand
29,31-3,41,43,47,51,67,84, 23, 28, 29, 75, 75(F.3.1), 221 ;
90, 116, 120-4, 132, 158, 181-4, supply and demand , effects of price
185; forward 33-4, 34(£ .1.7), on 26, 28-30,44 ; output, division
35-7; gold bullion standard 37-9; between home and foreign 222-29,
manipulating 178 ; movements since (F .9.1, 9.2), (E.9.3) ; visible 2, 4,
1971 162 ; par-value 40-2, 185; 6; volume (1854-1939) 135-39,
pegged 180, 186, 190; policy (T.7.1) , (1940-50) 141, (T.7.4),
weapon 119, 182 ; speculation, ef- (1951-80) 146-7 (T.7.4), 149,
fect of 31-3; spot 31, 33-7 , 155,157 ,183,255 -6(T.I0.l, 10.2) ;
162; stability 28, 29, 31,32,37, world 238-49
189,193
expenditure : and absorption approach
69, see a/so balance-of-payments fiscal policy 118,119,121,124,125,
adjustment, absorption approach; 129, 156, 157; under alternative
autonomous 54-64 , 64n2.5, 65, exchange-rate regimes 125n6.1,
77, 77n3 .3, 78; multiplier xii, 126n6.2 ,131-2
59-65, (E.2 .23-2.43), 194; real floating pound, see pound
121, 121(F .6.2), 122-3; relation of food, price of 88, 185; price index
imports to 57-61,194 ; response to 186
devaluation 78, 78(E.3 .11); switch- foreign assets 31, 35, 107
ing policies 99, 99n4 .2, 100, 108, foreign currency 4, 21, 33, 73,
115-17 ,178 76n3 .1, 79; buying spot 36; de-
Export Credit Guarantee 79n3.4 valuation effects on exports and
316 Subject Index
foreign currency (contd} generating income 48, 49 ; freedom
imports 76-83, (E.3.5-3 .10) , of, curtailed by capital inflows 7;
(F.3.3), government transactions transactions in foreign currency 33
33; reserves 6, 9, 22, 23, 33; sell- Group of Ten nations 164
ing forward 36 growth rate xi, xii, 50, 53-4, 180,
foreign-exchange market 1, 10. 22-47 , 186, 210, 283 ; and balance-of-
152, 183 ; balance of payments 27; payments constraint 269-72,
equilibrium in 21, 27-30, 193 ; (E.l1.l5-11.l8), 275, 283, 284,
reserves 45, 108, 111, 143; stable 286; and balance-of-payments equi-
27-9 ; supply and demand for inter- librium 52-3 , 66, 250-8 (T.I0.l,
national trading 4, 22-30; supply 10.2), 271, 271nl1.3,11.4, 286,
and demand to achieve desired ex- Beckerman model 263-6,
change rate 43 ; supply and demand (E.ll.l -11.8) ; Caves model 263,
for speculation 31-3; unstable 266; differential market 239-40,
29,30 (E.9.4-9.12) , 240nl1.3, 241 , dif-
foreign-exchange supply and demand ferential product 239-40, (E.9.4 -
4, 21, 22, 23-33, (F .1.1-1.6), 9.12), 240nl1.3, 241; export-led
83 ; crises 165 250, 251 259-72, (E.ll.l-11.8),
foreign markets 222-49 ; see also 273; of exports xii, 66, 80, 140,
world trade 142, 145, 177(T.7.9), 190, 191,
foreign trade: multiplier 49, 50, 192n7 .16, 210, 238, 243, 250-8 ,
55n2.3, 56-61; repercussions 61-5, 273-5 ; fundamental law 254; of
(E.2.30-2.43) imports 140, 142, 145, 152, 174
forward exchange rate 33-7 (T.7.9), 190, 191, 204n8 .2, 243,
Futur e of the British Car Industry 250-8, 273, of income 259-60;
(1975) (Central Policy Review Staff Kaldor model 267 -70, (E.l1.9-
Report) 279 11.14); Lamfalussy model 263 ;
manufacturing output 283-5,
Germany, trade expansion 277 -8 (T.12.2), (E.12 .1-12.3); neurosis
gold: export 38-9 ; par value 40; 52-3 ; world trade 156, 179,
reserves 6, 9, 143, (1950-80) ; 238-49
144(T.7 .3); sale of 4,40
gold bullion standard 37-8, 38n1.4,
39-41, 109; suspended 39, 39n1.5,
65,67 Hahn model 110-13, (E.5.1-5.4)
Goldstein's model of exchange-rate hedging 33, 36
changes on wageand price behaviour HM. Treasury xiii, 157, 160, 161,
89-90, (E.3.26-3.28) 275; Economic Progress Report
goods 1,2,4,31, I11n5.1 , 112n5 .3; 230n9.2, 236(T .9.2), 281(T.12.l)
after-sales service 279-80; demand
for U.K., in U.S. 26 ; design xii,
276, 278 ; domestic demand and
supply in relation to price 24(F.1.1), import controls 152n7.2, 286-7 ;
25(F .1.2) , marketing xii, 276, 278, quotas 99,286
279 ; quality xii, 276, 278, 279, Imported Manufactures (1965) (Nation-
281 ; reliability xii, 276, 278, 279, ~1 Economic Development Office)
280; supply curve, U.S. 23; un- 199-200,279
favourable characteristics 229,242, imports xii, 2, 20, 48, 49, 51; bottle-
258,274,275,285 necks in United Kingdom economy
government : budget deficits 49; ex- affecting 52, 170, 199, 204, 273 ,
penditure abroad 170, 199; expen- capacity 196-200, commodity
diture, cuts in 184 ; expenditure composition 245 (T.9.4); competi-
Subject Index 317
tive 24, 194; credit 10; determin- insurance, foreign-exchange 46
ants 194-6, (£.8 .1-8.6); dollar interest rates 124(F .6.3), 125n6.1,
price elasticity 75, 75(F.3.2); dom- 131n6.5, 152; differences between
estic price 42, 44, 46, 74 ; effects countries 33-4, 126-8; high 19-
of devaluation on 154-7 ; elasticity 20, 128; level 19, 114, 129-31;
24(F .1.1), 25, 53-4, 71-8, rise and fall with gold flow 40-1;
(£.3.1-3.7),80,92,181 ,196n8 .1, rising with inflation 102
275nI2.1 ; functions 194-221, interna tional monetarist school of
geographic composition 247(T .9.5); economists 85, 90
growth rate, see growth rate; in- International Monetary Fund (I.M.F.)
visible 2, 4, 170; multiplier 58-9, 7, 9, 11(T.1.1), 12(T. 1.2) , 18
(E.2.20, 2.21) ; penetration ratios (T.l.4), 19, 39,40, 155, 156, 164;
212-20, 220n8.4, (T.8.5); ratio to Articles of Agreement 36,40
expenditure 49 , 50 ; relative price investment 12-13, (T.1.2), 17-18,
194, 196, 197, 199-212 ; statistics (T.1.4) ; abroad 3, 5, 177; abroad
(1854-1939) 135-39, (T.7.1), of North Sea oil revenue 282 ; dif-
(1940-50 141, (T.7.2), (1951 ferential yield 34(£ .1.7),35,36,39;
-80), 146-7, (T.7.4) , 149, 155, domestic 20; domestic of North
183 ; supply and demand 23, 24, Sea oil revenue 283 ; in I.M.F. 11
52,75 , 75(F.3.2), 97,168,194-7, rr.i.n, 12-13(T.1.2) ; multiplier
(E.8.1-8.6), (F.8 .1); surcharge 54-65 ; need for 278; private gen-
151-2, 152n7.2, 155n7.5; visible erating income 48, 49, 283 ; private
2, 4, 6; volume, see statistics above long-term 187
income: and absorption approach 69, investment earnings overseas (1854-
103, 103n4.5, (E.4.9-4.18) ; adjust- 1939) 135-9, (T.7.1), 140
ment xi, 27; devaluation effect on Investment in Machine Tools (1965)
100-1 ; elasticities 205-9, (T.8.2, (National Economic Development
8.3),231-4 , (T.9.1), 244,255-6, Council) 279
(T.I0.l, 10.2),272,273, 275nI2.1, IS·LM model of simultaneous internal
286; equilibrium 48 , (£ .2.1-2.3), and external balance 125n6.1, 129,
49, 54-7 , (E.2.4-2.14), 62; expen- 130n6.4, (F .6.4),131
diture-generating 48, 57; fluctua-
tions 49 ; growth rate 259-60, 283; Japan 242, 250, 254-7, 261, 262,
methods of disposal 48 , 54 ; multi- 276,278
plier 50, 57-8, (E.2.15-2.19), J-curve 84(F.3.4), 84n3.5
62-5, 68 ; price changes, effect on
105-6; relating imports to 49;
world 225, 229-30, 236, 253
Kaldor model of export-led growth
industries: export waiting time 83,
267-70, (£ .11.9-11.14)
227-8, 276, 278, 279, and im-
Keynesian theory 54, 65, 67, 68, 70,
port penetration ratios 213-19 ,
85
220n8.4, (T.8.5) ; and labour de-
Korean war 142, 143
mand 202-3{T.8.l)
inflation 22,30,41 ,42,47,164,180,
185, 189-90; and devaluation 68,
79n3.4, 85-92, 149; domestic 184, labour costs, see wages
185,187,188; 'Scandinavian' model Labour Government 177, 11111
91(E.3.29-3.31) labour market: bottlenecks 200-1,
Input-Output Tables of the United 203; supply and demand 200-1,
Kingdom (1971) 61 213
318 Subject Index
Labour Party 150 money : reserves 115; stable 113; stock
labour productivity 264, (E.ll.2-11 .5) of 107-8, 110, 112; supply and
lags 155, 184 ; monetary 4, 5, 35; demand 98n4.1, 102n4 .8, 110-17,
physical 83, 84 (E.5.1-5.4)
Lamfalussy model of export-led growth money market 112, 115
263 monopoly 80,81
leads: monetary 4, 5, 35; physical multiplier : and exchange-rate change
83,84 68 ; and exports 59, (E.2.22) ; and
Letter of Intent 156 foreign trade 49, 50, 55n2.3, 56-
liabilities: external, of public sector 61; and imports 58-9, (E.2.20,
1950-80 144(T.7.3); to foreigners 2.21) ; income and expenditure 50,
3,4 ; sterling 280n12.3 54-65, (E.2.15-2.19), (E.2.23-
London and Cambridge Economic Ser- 2.43), 68, 194; investment 54-65
vice, Key Statistics 1900-1967 Mundell mode! of simultaneous internal
139n7.3 ,141, 144, 147 and external balance 119, 124-5,
125n6.1,126n62,127-32,132n6~
management, bad 278, 279
manufacturing: output, growth of National Economic Development Coun-
283-5 (T.12.2) cil 241, 242
market classification 119, 120 National Economic Development Coun-
market conditions 80 cil Report on Industrial Strategy
market structures 89, 228, 276, 279 (1975) 200,204,277,279
Marshall-Lerner condition 29, 67, 68, National Economic Development Office
72-6 , 76n3 .1, 77, 77n3.2, 78-80, Report on Imported Manufactures
83,85,93,101,237 ,253,253nl0.2, (1965) 199-200,279
270,271 National Economic Development Office
Meade model of simultaneous internal Report on non-price factors (1977)
and external balance 119-24, 127 (Stout) 276, 277
mercantilist trade protection 67 National Economic Development Of-
Minimum Lending Rate 163 fice Report on non-price factors
Minimum List Headings (M.L.H.s) (1977) (Stout) 276, 277
201-3, (T.8.1), 206-9 , (T.8.3), National Enterprise Board 283
210, 210n8.3, 212, 220, 229, national income 3, 7, 48n2.1, 99 ; ex-
231-4 port and import changes affecting 5
monetary approach to balance of pay- National Institute Economic Review
ments 66, 69-71, 90, 107-8, 147
110-17 ; Hahn model 110-13, National Institute of Economic and
(E.5.1-5 .4) Social Research 83, 89, 151, 152,
Monetary Approach to the Balance of 155,180
Payments (1976) (Frenkel and North Sea oil 280; contribution to
Johnson) 110 balance of payments 281(T.12.1)
monetary policy 118, 119, 121, 124,
125, 128, 129, 156, 157; under O.E.C.D. countries 205
alternative exchange-rate regimes official financing, see balance of pay-
37-43,131-2, 262n6 .2, 6.3;assign- ments
ment of 132-3; correction xi; oil prices 161, 180, 188, 192-3,
disequilibrium 70-1 ; independent 236,274
44-5 ; international movements 70; oligopoly 79, 80, 81, 83
manipulation xii; mixed 132-3 OPEC countries 161
Subject Index 319
'Panacea of the Floating Pound' profitability 79-82, 225,235-6,237
(1974) (Thirlwall): National West- (T.9.2)
minster Bank Quarterly Review public sector, external liabilities (1950
181n7.14 -80), 144(T.7.3)
parities, sliding 190, 192; Smithsonian
161-3 Radcliffe Committee 145
Phillips curve 156,265 ratchet effect 178n7.12, 199,279
pound : depreciation of, and economic real-balance effect 70n2.6 , 102, 116,
difficulties 185-8, 274; floating 117
181, 181n7 .14, 182-4;
price: adjustment xi, 27,274-5 ; aggre- saving 7, 49, 99(EA .5) ; disposing of
gation bias 95-6 (E.3.34-3.39) ; income 48, 102; forced 102n4.4;
changes affecting income 105-6; and imports 54-6
competitiveness 80, 148-9, 230, 'Scandinavian' model of inflation 91
230n9 .2, 235-7, 243,259 ; effects (E.3.29-3.31)
on output, Wilson model 85-9 , Sector Working Parties 277n12.2, 279
(E.3.12-3.25) ; effects on supply securities, buying and selling 31, 44
and demand 26, 28-30, 44, 187, simultaneous internal and external
196-7; effect on wages 88-92, balance 118-33 ; under fixed ex-
(E.3.24-3.28) ; elasticity in exports change rates with capital movements
and imports 67-8, 74, (E.3.4), 124-9, (F .6.3); under flexible ex-
93-7, (F.3.5), (E.3.34-3.39), 210, change rates with no capital move-
211, 211(T.8.4), 222, 226, 236, ments 120-4, (F.6.1, 6.2); IS-LM
253 ; Goldstein model 89-90, curve model 125n6.I,129, 130n6.4 ,
(E.3.26, 3.28) ; index 92, 186,210- (F.6A), 131; Meade model 119-
21, 230; inflation 68, 86, 88-90, 24, 127; Mundell model 119,
156, 185, 186, 188 ; international 124-5, 125n6.1, 126n6.2,127-32,
agreements 79; levels 45, 46 , 70, 132n6.6 ; Swan model 121-4, 127
87, 115; levels, internal 87-8, Smithsonian Agreement (1971) 41,
(E.3.24, 3.25) Keynesian model 85 ; 158 ; parities 161-3
movements xi, 79; observation speculation 163; attack on sterling
errors 97 ; real relative 70; relative 161; bull 36; capital 31-3 ; cur-
194, 196, 197, 199, 212, 227, rency 22, 30, 31-3, 182; effect
230-8, (T.9.2), 237,238,254 ,274, on spot and forward rates of ex-
283 ; rise and fall with gold flow change 36-7; pressure 42 ; safe-
38-41,66-7; rising to equilibrate guards against 192; stabilising and
supply and demand 52; short-run destabilising exchange market 31-3,
elasticity 96 ; simultaneous equation 47 ,183,184
bias 93-5, (F.3 .5); stable 50, sterling: appreciation 35; depreciation
201 ; Treasury model 211; uncom- 24, 35, 143, 160-2, 164; devalua-
petitiveness 182, 186, 191, 200, tion 37, 79, 80, 140; fluctuations
239; variations due to devaluation in exchange rates (1973-4) 159
79-81, 81(F .3.3), 82-3, 85-90, (T.7.5) , 181-4 ; forward buying
(E.3.12-3.28), 115, 155-6, 178, 34-7; liabilities 280n12 .3;par value
273 ; world market 80,91 ,92,94, 40, 158, 188n7.15; pressure on
95, 115 153-4, 161; price of dollar 24,
productivity 44n1.6, 51, 99n4.2, 251, 26-30, (F .1.2-1.6), 89, 161, 162
261, 263, 266, 266n11.2, 267-69, (T.7.6); reserve 143; selling of
283 161,164,189
320 Subject Index
structural: balance of payments as 222, 222n9 .1, 226, 229-30, 238 -
problem 272, 273-87 ; economic 50, 266, 268, 273, 274, 275nI2.1,
policy xii, 108 286
Standard Industrial Classification See also exports; imports
(Sl.C .) 201, 205, 210, 212, 220, trade balance 6, 76n3.1, (1854-1939)
229 135-9, (T.7.1), 144; (1940-50)
Steering Committee on Economic 141, (T.7.2) , (1951 -80) 146-7 ,
Policy 153 (T.7.4) , 154-7 , 160, 161n7.7,
stock-building 194, 198, 200 168-70, (T .7.8), 173n7 .10, 173
strikes 153,180,181 ,189 n7.11, 176n7 .12, (E.7.5), 178,227
subsidies 99,286 trade-off curve 51 , 52, 52(F .2.1), 172,
Suez Canal crisis 152, 180, 181 172(F .7.1) ,173
supply and demand, see under the trade protection 67, 286
specific subjects Treasury , see H.M. Treasury
Swan model of simultaneous internal Treasury, model and price elasticity
and external balance 121-4, 211
127
unemployment xii, 20, 21, 44, 51,52,
tariffs 70, 115; constraints on 19; 53, 120-3, (F.6.1), 125, 126,
high 20; import 51, 99 150n7.1, 182,201, 262, 265, 283,
tax : disposing of income 48; relief 283n 12.4; and balance-of-payments
286 ; revenue from North Sea oil equilibrium 261 ; and devaluation
280,286 77, 105,150; rise in 53,107,179,
taxation 49 270,275
technical backwardness of United United Kingdom Balance of Payments
Kingdom 243, 267 (pink Book) 169, 177, 246,
terms of trade 41,45,67, 282; effect 249
on, of devaluation 92-3, 100-1 ; United Kingdom's declining share in
statistics (1854-1939) 134-40, world exports xii, 230 , 238 -49,
(T .7.1), (1940-50) 141 (T .7.2), (T.9.4) , (E.9.10 -9.12), 275, 277
(1951-80) 146-7, (T.7.4) United Kingdom Exchange Equalisation
Tinbergen rule 126 Account 37
trade: commodity composition of United Kingdom as international banker
visible 244-6, (T.9.4) ; competition 143
170-2; deficit 134, 140, 161, United Kingdom reserves 1950-80
173-80, 189 ; discouraged by ex- 144(T.7 .3)
change uncertainty 46-7; effect of United States: balance-of-payments
currency depreciation and devalua- deficit caused by war in Vietnam
tion 67, 92-3, 100-1, 154-5 , 46 ; interest-rate influence on United
155n7.4, 156-7 ; geographic com- Kingdom 34-5 ; loss of confidence
position of visible 247-49 (T .9.5) ; in dollar 41
and income equilibrium 48, 48
(E.2.1-2.3), 49; international, see Verdoorn relationship 266 , 266nl1.2,
world, below; invisible (1854-1939) 268,270,272
134-9, (T.7 .1), 140; non-price fac-
tors 275, 276; structure of 242 ; wage(s): claims 88-9, 186; costs
terms of, see terms of trade; United 85-7, (E.3.12-3.22), 88, (E.3.25),
Kingdom declining share 229, 238 149,190,235-6,264-5, (E.l1.1-
-49, (T.9.4), (E.9.10 -9.12), 275, 11.8); effect on prices, Goldstein's
277 ; world 156, 180, 200, 212, model 89-90, (E.3.26-3.28) ;
Subject Index 321
policy 88 ; restraint 183; nsing world economy 2, 22,45-7 ; exports
151, 265, 271; rising and falling 238-49, (E.9.4-9.12), (T.9.4); ex-
with gold flow 38-9 ports , United Kingdom decline, see
See also real wage United Kingdom declining share;
waiting time affecting exports 83, income 225, 229-30, 236, 253 ;
227-8,276,278,279 prices 80, 91 , 92 , 94 , 95, 115;
Walras's Law 108n4 .8, III trade 156, 180, 200, 212, 221,
Wilson model of devaluation and in- 222, 222n9 .I, 226, 229-30, 238-
flation 85-9, (E.3.12-3 .25) 50, 266, 268, 273, 274 , 275nI2.1,
working parties , see sector 286