ECONOMIC SANCTIONS

LIST OF PREVIOUS PUBLICATIONS

Instructor’s Manual: Economics. 2006. Leeds:
Von Allmen and Schming, Addison-Wesley.

Brand damage valuation: Theory and practice. 2005.

Journal of International Wine Marketing. 17: 21–29.

Financial aspects of wine. 2004. With Tony Correia. Wine:

A Global Business, ed. Liz Thach and Tim Matz. 121–136.
Elmsford, NY: Miranda Press.

Direct shipping laws in the wine industry. 2003. Journal of

International Wine Marketing 15: 25–36.

Competing in the U.S. wine market: Australian imports

and tasting scores. 2001. Journal of International
Wine Marketing 13: 32–42.

Imported wine demand and stock market returns. 1999.

With Eric N. Sims. Journal of International
Wine Marketing 11: 64–84.

Consumption effects of economic sanctions. 1998. PhD

Dissertation. Davis, CA: University of California, Davis.
Economic Sanctions
International Policy and
Political Economy at Work

Robert Eyler
ECONOMIC SANCTIONS
Copyright © Robert Eyler, 2007.
All rights reserved. No part of this book may be used or reproduced in any
manner whatsoever without written permission except in the case of brief
quotations embodied in critical articles or reviews.
First published in 2007 by
PALGRAVE MACMILLAN™
175 Fifth Avenue, New York, N.Y. 10010 and
Houndmills, Basingstoke, Hampshire, England RG21 6XS
Companies and representatives throughout the world.
PALGRAVE MACMILLAN is the global academic imprint of the Palgrave
Macmillan division of St. Martin’s Press, LLC and of Palgrave Macmillan Ltd.
Macmillan® is a registered trademark in the United States, United Kingdom
and other countries. Palgrave is a registered trademark in the European
Union and other countries.
ISBN-13: 978–1–4039–7463–1
ISBN-10: 1–4039–7463–2
Library of Congress Cataloging-in-Publication Data
Eyler, Robert.
Economic sanctions : international policy and political economy
at work / Robert Eyler.
p. cm.
Includes bibliographical references and index.
ISBN 1–4039–7463–2 (alk. paper)
1. Economic sanctions. 2. International relations. 3. International
economic relations. I. Title.
HF1413.5.E95 2007
327.117—dc22 2007005284
A catalogue record for this book is available from the British Library.
Design by Newgen Imaging Systems (P) Ltd., Chennai, India.
First edition: December 2007
10 9 8 7 6 5 4 3 2 1
Printed in the United States of America.
To Louis and Rena Albini
This page intentionally left blank
 Contents

List of Figures ix
List of Tables xi
Preface xiii

1 To Sanction or Not to Sanction? 1

2 Basic Sanction Analysis 9
3 Sanction Initiation and Continuance:
Enter Game Theory 35
4 Public Choice Theory and Smart Sanctions 59
5 Open Economy Macroeconomics and Sanctions 85
5A Mathematical Derivations of NOEM
Sanctions Model 105
6 Empirical Analyses of Sanction Effectiveness 125
7 Conclusions and Policy Recommendations 159
Appendix 1 Brief Cases Histories of Selected
Sanction Episodes 169

Notes 207
Bibliography 223
Index 235
This page intentionally left blank
 List of Figures

1.1 Sanction cases involving the United States 3

1.2 Sanction effectiveness continuum 8
1.3 Sanction timeline 8
2.1 The target’s export market/sender’s import market 13
2.2 Import sanction effects 15
2.3 Export sanction effects 18
2.4 The target’s market for loanable funds 21
2.5 Financial sanction effects on the target 21
3.1 A sanction game in extensive form 39
4.1 The sender’s policy market 63
4.2 The target’s policy market 64
4.3 The policy markets’ international connections. 69
5.1 The world trichotomy 90
5A.1 Demands for z and z* in each nation 107
6.1 Humanitarian data distribution
(% of total observations [65 obs]) 145
6.2 Political effectiveness indicator for sixty-five
observations (distribution of values by
percentage of total) 152
This page intentionally left blank
 List of Tables

3.1 A sanction game in normal form 38

5.1 Sanctions effects on utility functions 100
5.2 Sanction effects sensitivity to parameter
changes: sender 102
5.3 Sanction effects sensitivity to parameter
changes: target 103
5A.1 Symbol list for sanctions NOEM model 106
6.1 Cases and basic data for VAR analysis 134
6.2 Initial and cumulative effects of sanction
shock—first month (1 or 1 or 0) versus
last month (1 or 1 or 0) 136
6.3 Variance decompositions (%) for VAR results 138
6.4 Humanitarian data estimates for sanction cases 142
6.5 Humanitarian logistic regression results:
initial effects of sanction shock 147
6.6 Humanitarian logistic regression results:
cumulative effects of sanction shock 148
6.7 Political logistic regression results: initial
effects of economic coercion 154
6.8 Political logistic regression results:
cumulative effects of economic coercion 155
A1.1 Selected cases 170
This page intentionally left blank
 Preface

W hen I was eight years old, I began to read military history. As I

progressed intellectually, I expanded my reading of different conflicts
and began to understand warfare in its historical context more com-
pletely. In sixth grade, my social science text suggested an Arabian
philosopher once wrote that the study of history is simply the study of
war. In high school, I had a dynamic history teacher that set up his
U.S. history curriculum as lessons on one international conflict to the
next, where the interbellum periods were simply movements in a
catastrophic symphony that had no end in the foreseeable future.
As an undergraduate at Chico State University in California, my
fascination with economics began mainly because it is a way of under-
standing human thought at its core. I felt to understand economics
meant to understand people. When I entered graduate school at the
University of California, Davis, I had a dissertation topic already in
mind. I wanted to write about the economics of warfare. At that
point, there was no doubt in my mind that the study of war was an
obvious blend of economics, history, physical sciences, and sociology
such that all the subjects that mattered to me were in one place. I was
advised against such a thesis due to its perceived lack of salability in
the job market, and was directed instead (and happily so) to a disser-
tation on economic sanctions. This text began there, over ten years
ago, as did my interest in using economic statecraft as a substitute for
military conflict.
I presented a smaller version of the new open economy macro
model in chapter five of this book at the American Economic
Association meetings in 2005. I am indebted to the Middle East
Economic Association (MEEA) for allowing me to co-organize a ses-
sion on sanctions there. However, I am truly indebted to my profes-
sors at UC Davis, especially Peter Lindert who helped me to see a
better way to think and understand economics and writing in general.
I also want to thank my colleagues at Sonoma State University in
xiv PREFACE

California, who I was able to use as information centers to make sure I

was not going too far down the wrong road. I also want to thank my
friends and family who allowed me to write this text, sacrificing time
with them to do so; this is especially true for Chèrie, whose patience
was incalculable and whose support was necessary. I also need to
thank an anonymous referee for outstanding comments. Anthony
Wahl, the editor I worked with directly at Palgrave/Macmillan, must
also receive special thanks, as his help was invaluable. Finally, I want to
thank Stanford University’s Economics Department, Clark Reynolds
and Tim Bresnahan in specific, which hosted me for six months while I
completed this text.
This text, in my mind, can be used by political scientists and econ-
omists alike to understand the way an economist thinks about policy.
It is also meant to provide a broader view of these policies in the hope
they can continue their use in more efficient ways. This marriage of
politics and economics may not work in the end, but it is truly the
only substitute for war available.
 Chapter 1

To Sanction or Not to Sanction?

O n June 25, 2006, an Israeli soldier was kidnapped by militants,

known as “Hamas,” after a raid into Israel from positions in the Gaza
Strip. The soldier was held hostage and his captors’ demands included
the freeing of over one thousand jailed Palestinians. As a reaction to
this kidnapping, where one soldier’s life was in question, the Israeli
government ordered troops over the border into Gaza to destroy
bridges and electric transformers and to shut off all utilities to the sus-
pected location of the Hamas kidnappers. The limited invasion sought
to constrain the kidnappers’ activity and find the soldier in question,
not to initiate full-scale conflict against Hamas in Gaza. On July 5,
however, as negotiations closed in on a stalemate, the Israeli army
began to mass on the Gaza border for a full strike.1
On July 12, 2006, an extremist arm of Hamas, called “Hezbollah,”
captured two more Israeli soldiers after crossing the southern Lebanese
border. Lebanon as a nation, accused of providing a home to
Hezbollah and acting as an armory for Hamas, was in the middle of
this conflict. An Israeli naval and air blockade of Beirut, Lebanon’s
capital and largest port, was an act of economic coercion. The damage
to Lebanon and her people was front and center in the international
community’s eyes concerning the human cost of this sanction and par-
allel military action. U.N. Security Council Resolution 1701 effectively
ended the conflict in August 2006 (Sharp 2006, 4).
On July 4, 2006, North Korea conducted missile tests for a long-
range weapon system claimed to be capable of reaching U.S. shores.
Before July, after months of diplomacy, North Korea began to come
around and discuss the delay of such tests and also not pursuing atomic
weapon manufacture. The missile test provoked reactions worldwide,
especially from the United States; the United Nations immediately
began deliberations on imposing economic sanctions against North
2 ECONOMIC SANCTIONS

Korea. The United States urged immediate action and found many
allies. Other countries, such as Russia and China, were initially against
sanctions and wanted more diplomacy. The U.N. Security Council
passed Resolution 1695 as a weak reaction on July 15, 2006.2
In the Israeli case, military force made the threat credible that
actions against Israel were going to be met swiftly and violently. By
closing off the border into Israel and Beirut’s port, the Israeli govern-
ment ostensibly embargoed both the Gaza Strip and Lebanon for the
actions of a relatively small number of people. In the North Korean
case, the entire international community saw the missile tests as objec-
tionable. A North Korean underground nuclear test in October 2006
showed further defiance. On October 14, 2006, the U.N. Security
Council initiated arms sanctions in full with Resolution 1718.
Economic sanctions, wherein economies use trade and financial ties
with other nations as diplomatic instruments, are not always used as
proactive moves or reactions to international deviance and are rela-
tively new as devices of international conflict resolution. When should
sanctions be used? Why should they be used? Will they be effective,
and what role do they play in international relations? American
foreign policy since 1945 has been a strange mix of both economic
coercion and military engagement as second or third steps in diplo-
macy. Many countries have used sanctions as diplomatic instruments;
however, the United States has been by far the leader on these
policies’ use. Figure 1.1 breaks down the sanction cases to date and
American involvement.
This text aims to achieve the following goals. First, this study
attempts to survey not only where the sanctions literature is to date
but also how economists describe various aspects of a sanction
episode. Basic economics is essential to understanding these policy
choices and their rationale. Cartel theory helps understand why insti-
tutional, multilateral, and universal sanctions may succeed or fail.
Public choice theory tells us that policies aimed at specific interest
groups within the deviant economy, those focused on decision
makers, may sway another country’s politics. For this reason, the lit-
erature review is scattered throughout the text, connecting the most
appropriate works to specific topic areas, rather than all lumped into
one chapter. Appendix 1 provides more literature sources and brief
synopses of sanction episodes.
This book’s central theme is to examine sanctions as macroeconomic
policies that seek marginal movements toward diplomatic goals. The
second goal is building an open-economy macroeconomic model
describing welfare redistributions from economic coercion. The
 TO SANCTION OR NOT TO SANCTION? 3

Figure 1.1 Sanction cases involving the United States.

Sources: Hufbauer, Schott, and Elliott (1990) and Drury (1998, 2005).

model in chapter five is the major contribution of this study. This

model of exchange rate movements from policy is described in detail,
derived in such a way that economists and other social scientists can
follow how sanctions act like other macroeconomic policies. Finally,
this text expands the literature’s empirical studies by using economet-
rics to not only estimate exchange rate movements due to sanctions
but also to connect these changes to both humanitarian and political
effectiveness. Because data can be parsimonious and non-credible in
many sanction cases, any empirical studies have their caveats and
pitfalls. However, the empirical analyses in this text provide alternative
ways to measure sanction effectiveness, where economic effects build
marginally toward political goal achievement. The empirical models of
chapter six corroborate other studies’ results and add new conclusions
and dimensions to the literature.

Sanctions and their Political Economy

The study of wealth redistributions from policy decisions and their
economic effects is generally how political economy is defined.
Sanctions have become immediate diplomacy reactions, with the
4 ECONOMIC SANCTIONS

United States leading the charge.3 Whether the sanctions are unilateral
or universal, potential profits attract sanction busting, taking advan-
tage of another country’s higher prices and lost markets. Even with
sanctions being more focused and new legislation to economically
punish countries that aid and abet sanctioned nations, U.S. sanctions
still struggle to achieve their stated goals. Scholars must agree on a
model or set of models to predict and analyze sanction effects. Such a
consensus helps recommend policy choices to sender governments
that balance economic, humanitarian, and political effectiveness
simultaneously.
Sanctions are ultimately macroeconomic phenomena, and their
effects should be analyzed and forecasted as such. Scholars have
become more concerned about the economic harm caused by these
policies, especially the target socioeconomic groups upon whom
potential damage is borne. Sanction effectiveness is a large issue in the
academic literature and popular media, where many claim sanctions are
ineffective. Why continue to impose sanctions? The simple answer lies
in the parsimony of other options: more negotiations, placation, or
war, regardless of whether the military option is initially debated or
not. Sanction use is likely to increase over time; this text attempts to
expand our understanding of these policies’ political economy.

The Sanction Cast and This Text’s Thesis

First is the sanction itself. Many studies define sanctions using an
economic interpretation. An economic sanction is a country’s dis-
criminatory economic restriction of either trade or credit flows with
another country in an attempt to affect or reverse current policy in the
sanctioned nation. The target is the sanctioned country pursuing
political policies deemed deviant by other countries. Senders choose
themselves by initiating economic restrictions against this target
nation. Sanctions involve economic weapons to wage a nonmilitary
campaign, extending the diplomatic process beyond verbal negotia-
tions. Third parties also exist, as the final cast members, and their
involvement level depends on how they react to the sender’s initial
policy. Some help the sender and others help the target, all depending
on the potential profits of those choices. There is much debate about
the importance of third parties and there is no consensus on whether
multilateral sanctions or the existence of sanction busting, third-party
nations make any difference in policy efficacy.
This text’s thesis is as follows. Economic sanctions are diplomatic acts
used to change a foreign government’s political policies, where sanctions
 TO SANCTION OR NOT TO SANCTION? 5

act as if they are macroeconomic policies transmitting coercive economic

effects from senders to targets. Whether we call a sender’s policies embar-
gos, sanctions, coercion, statecraft, or any other term, the rationale and
intended effects have the same genesis ex ante policy. The economic
rationale behind each is close enough that this study uses these terms
interchangeably throughout this text.4

Brief Overview of the Text

Sanction policy is at an enigmatic crossroads. With their effectiveness
in question and human suffering from them now a major concern,
economists must understand and forecast sanction effects better than
before. Economic statecraft aims to move along a policy effectiveness
continuum toward stated political goals, where marginal movements
toward a goal defines relative policy effectiveness. We must, as best we
can, drop the term “success” from the sanction nomenclature.
Chapter two introduces basic economic theory in sanction analysis
using models of international trade. Since sanctions are trade barriers
at their core, models from international economics provide much
needed insight to policy makers. Chapter two posits a basic model that
acts as foundation for how economic coercion drives welfare losses for
both countries from these policies. The importance of recognizing that
sanctions imply costs for both senders and targets is realized in the
macroeconomic model of chapter five. Some of the literature’s major
works are case study analyses that use fundamental economics to draw
conclusions. Hufbauer, Schott, and Elliott (HSE) have produced two
editions of over 100 cases studies (1985, 1990), with a third edition of
over 170 cases on the way. HSE’s second edition (1990) is this litera-
ture’s encyclopedic foundation. Other seminal works are included in
chapter two. Knorr (1975) and Baldwin (1985) are also seminal, and
eloquently summarize all earlier writings on political economy and
sanctions. Drury (2005) is the apex of the latest analyses, especially
empirical work based on and since HSE (1990). Chapter two illus-
trates in simple terms how important the international relationships are
to these policies’ credibility and potency. Appendix 1 provides brief
histories and sources on over 65 cases from 39 countries.
Chapter three focuses on understanding sanction initiation and
continuation. Economists and political scientists alike use game theory
as a tool to explore sanction initiation as diplomacy’s first step. This
chapter first outlines economic and political issues before sanctions
begin. In some historic cases, sanctions are merely a prelude to war; in
other cases, war as an option is never discussed or credible. Nations
6 ECONOMIC SANCTIONS

pick strategic paths, based on perceived payoffs, given the other

nations’ choices. Nash Equilibrium is formed by conditional and
strategic, not tactical, diplomacy choices in theory. The idea of coop-
eration transcends the theoretical issues and becomes a practical ques-
tion when considering why other nations would join with senders.
Cartel theory may help explain why there is a perceived lack of effec-
tiveness in sanctions that have two or more senders. Chapter three
also presents simple lessons from basic game theory and statecraft
applications. Studies such as Eaton and Engers (1992, 1999), Martin
(1992), Bonetti (1997b), and Drezner (1999) use game theory to
explain how payoffs drive both the sender and target nations to
choose policy strategies. In some cases, sanctions possess credible
threats of major economic damage and an inability to continue a
deviant action may force the target’s policies to change. In other cases,
sanctions may end because they lack credibility and have no expectation
of success.
Chapter four expands on Chapter three’s themes and integrates
more specific political economy theory in examining sanction effects.
Public choice models conceive of political markets in each country
that perpetuate the target’s deviant policies and the sender’s sanctions.
Interest groups in each country influence policy actions and reactions.
Economic coercion leads to wealth redistributions that change politi-
cal decision-making, as certain groups in each country pay decision
makers to receive benefits from policy. Public choice models naturally
lead to smart sanction analyses, where sanctions are focused on the
target’s political decision makers. Such sanctions reduce potential
humanitarian crises from general embargoes, increasing the cost of
interest groups paying for continued policy directly if possible.
Examples include arms sanctions, travel restrictions, asset seizures,
and reduced aid packages. Chapter four discusses seminal works in
smart sanctions and their conclusions, such as Cortright and Lopez
(1995, 2002). Sanctions maintaining this focus, while influencing
political change, may begin to look like macroeconomic policies in
their effects, creating a paradox for smart sanction pundits.
Sanctions are macroeconomic phenomena. Chapter five introduces
a macroeconomic policy model where economic coercion acts like
monetary and fiscal policies to transmit economic damage such that
target political change follows. This model uses recent advances in
open economy macroeconomics to derive exchange rate effects from
sanctions. Few studies have examined how exchange rate fluctuations
can measure and elucidate the ways that the sender’s policy transmits
economic costs to targets on purpose. The major contributions here
 TO SANCTION OR NOT TO SANCTION? 7

are as follows. First, the model illustrates how market imperfections

can enhance sanction efficacy under different sanction mixes. Second,
the model in chapter five predicts that financial sanctions, where the
sender reduces its financial flows with the target and freezes asset
payments, are the most economically effective sanctions; the effects of
trade sanctions are ambiguous in this model. Chapter five provides a
way to view sanction policy like any other macroeconomic policies,
where damaging welfare effects are meant to be transmitted from
senders to targets through exchange rate movements. While econo-
mists may find this chapter’s technical material useful, the model’s
algebra is largely relegated to chapter five’s appendix.
Chapter six draws conclusions about sanction effectiveness, from
economic to humanitarian to political. First, the sanction could be
completely ineffective. In this case, the sanction must be seen as pure
political rhetoric. Second, there is economic success. To be effective,
the sanction must cause economic damage to the target. Time series
econometrics is used to estimate the direction and magnitude of
exchange rate shocks from economic coercion, following Sobel
(1998). If target currency values fall as a result of coercion impulses,
policy is deemed economically successful. Next is humanitarian effec-
tiveness, when economic effectiveness is not a harbinger of humani-
tarian problems. If economic damage results, the resultant harm may
or may not be large in terms of human costs. Assessing the sanction’s
ability to circumvent an “innocent” populace in its detrimental effects
measures coercion’s human effectiveness.5 Finally, the political effects
are measured. Economically effective sanctions that force marginal
policy changes are deemed politically effective. Using discrete choice
modeling for both the humanitarian and political estimations tests
how economic coercion leads to these marginal movements, the
essence of political economy. Exchange rate effects and control vari-
ables are used together to determine humanitarian and political
effectiveness, following Drury (2005).
Chapters two through five discuss and theorize about sanctions,
whereas chapter six empirically tests hypotheses concerning sanction
effectiveness (or lack thereof) in sixty-five U.S. cases. Import sanctions
are seen as the most effective coercion measure in terms of human and
political effectiveness, whereas export and financial sanctions have
ambiguous or small effects. Table 6.1 shows that import sanctions are
by far the least used sanction historically, which may be why many
perceive sanctions to be ineffective.
Chapter seven discusses conclusions and policy recommendations.
The first conclusion is that sanctions must act like a cartel, where
8 ECONOMIC SANCTIONS

Ineffective Economic Humanitarian Political

effectiveness effectiveness effectiveness

Figure 1.2 Sanction effectiveness continuum.

Initiation Assessment Adjustment Success/Withdrawal

Figure 1.3 Sanction timeline.

senders monopolize the policy market and impose costs on targets

that are enough to force marginal policy changes. Next, sanctions are
macroeconomic phenomena allowing economic statecraft to be dis-
cussed as if it was monetary or fiscal policy. Also, smart sanctions
struggle to be effective or remain focused in practice, especially if they
are macroeconomic phenomena. Policy recommendations include
forming sanction alliances to credibly impose coercive measures
regionally rather than globally. Regional connections are likely to be
more important to targets funding their policy decisions. Scholars and
policy makers should build and agree upon macroeconomic models to
measure economic statecraft effectiveness and marginal movements
along the Sanction Effectiveness Continuum. Finally, senders must
recognize that all sanctions must have an endpoint, as perpetual
sanctions show little evidence of efficacy.
The Sanction Effectiveness Continuum (figure 1.2) focuses on
how sanctions work and why marginal movements, from left to right,
are important. The Sanction Timeline (figure 1.3) provides a parallel
tool to compare and contrast cases as to how they progressed from
initiation to policy termination.
Without using economic theory, analyzing sanction effects is pure
conjecture. Chapter two begins our journey by discussing the basics.
 Chapter 2

Basic Sanction Analysis

Introduction
To study economic statecraft is to study international economics.
These policies of economic coercion, to affect political change, are
case-specific in their characteristics, involving many nations and
cultures. By changing the way a nation consumes, works, and saves,
theory suggests economic sanctions can levy a tax on the target’s
regime and populace to engage its rulers directly for change. Policy
choices transmitting welfare redistributions among the parties
involved, and also creating externalities that affect other groups indi-
rectly, are the essence of political economy. In this chapter, basic
concepts from international economics are applied to sanctions and
some seminal works in the literature are discussed vis-à-vis these basic
ideas. This chapter’s literature review is selective and not meant to
be comprehensive. Relevant literature is mentioned and discussed
throughout this text when topical.
Sender nations must have economic connections with the target to
sanction effectively. Direct connections and more inelastic choices
lead to target reactions in line with the sender’s goals. Market eco-
nomics shows that international trade free of market barriers begets
economic gains from comparative advantage and specialization.
Political economy is a simple tangent of international economics in
this sense: using trade and financial barriers as a political weapon is
meant to undermine target gains from trade to force political gains for
the sanction sender.
Sanctions are used to exploit one specific similarity between sovereign
nations, regardless of any differences: citizens need to eat. Exploiting the
target’s necessity for certain goods is economic warfare, where trade
restrictions are used in lieu of military weapons. Spindler (1995)
10 ECONOMIC SANCTIONS

provides an outstanding study of how similar sanctions are to trade

barriers. “Economic sanctions can be regarded as instruments of foreign
policy that lie somewhere on a continuum between completely unre-
stricted international exchange and absolute war at the other” (205).
Sanctions are not direct substitutes for war but for other diplomatic
tools, whereas war is the end of diplomacy. Fundamentally, a sanction is
nothing more than a quota or a quantity restriction on the amount of
goods, services, or financing flowing between two nations.

A typical sanction places restrictions (ranging from 0 to 100 percent)

on exchange between the sanction-sender and the sanction-target
nations. A typical protection policy places quantity restrictions, quotas,
restrictive taxes—tariffs, or both tariffs and quotas, on exchange
between the protective nation and other nations. Other protective
policies, such as national preferences laws, may be regarded as having
some tariff or quota analogue. (206)

Looking beyond basic international economic theory to determine

sanction effects is dangerous given what sanctions are at their core.
Foundationally, sanctions are nontariff barriers to trade, import, and
voluntary export restraints (VERs), used for changing another coun-
try’s political choices rather than domestic protectionist purposes.
The politics surrounding a sanction’s imposition immediately puts a
duration and mix onto these policies, which differentiates economic
statecraft from protectionism.1 A seminal study that outlined the
hypothesized effects of sanctions is that by Galtung (1967). This work
focused on sanctions against Rhodesia in the late 1960s, imposed by
the United Nations and the United Kingdom.2 Galtung outlined
seven “dimensions” for classifying sanctions, which have been a driving
force in this literature and are paraphrased here:

1. Sanctions either reward (are positive) or punish (are negative);

2. Sanctions are aimed at individuals or the collective;
3. Sanctions are imposed due to internal or external actions by the
target;
4. Sanctions are unilateral, multilateral or universal;
5. Sanctions are general or selective;
6. Sanctions are total or partial; and
7. Sanctions restrict a mix of the target’s trade, financing, communi-
cations, and diplomatic relations. (381)

Galtung suggested that certain ideal conditions exist for damaging

the target’s economic system without causing much harm to the
 BASIC SANCTION ANALYSIS 11

sender nation (384). But while it may be theoretically possible to

achieve an optimum, this second-best optimum comes with societal
losses, as do all international trade and financial restrictions.3 Because
embargoes are quantity restrictions, deadweight losses result from a
sanction’s imposition. This is a classic problem of trade barriers, tariff
or nontariff, in economic theory.4
Galtung’s case study of Rhodesia was seminal, the framework for
cases in Baldwin (1985) and Hufbauer, Schott, and Elliott (1990),
two leading texts discussed later. Appendix 1 in this text provides case
histories for episodes used in the empirical analyses in chapter six.
A standout case is South Africa and sanctions against apartheid.
Sanctions against South Africa were reactions to apartheid, and the
consensus view is that they contributed highly to apartheid’s end
(Levy 1999). Apartheid was a system of segregation and discrimina-
tion in South Africa against black South Africans, based on race. The
first sanctions against South Africa were in 1962, and were on and off
at various times throughout the next thirty years. This case was a mix
of multilateral policies, with the measures strengthening and weaken-
ing over time. South Africa became more and more economically
isolated as time passed, and apartheid broke down through interna-
tional pressure.5 Case studies show international policy at work, and
help us understand the basic theory, especially when these measures
failed to produce the political result we expected.
This chapter has four parts. First, it provides basic economic theory
to explore how sanctions work to reduce the target nation’s wealth, at
some cost of sender wealth. This section combines many ideas from
the literature and textbook on international trade and finance. Next,
this chapter provides a selective literature review of seminal works that
tie together the basic theory and sanction cases. Two texts stand out:
whereas Baldwin (1985), using Galtung (1967) and Knorr (1975),
summarized and expanded the existing literature, the work by
Hufbauer, Schott, and Elliott (1990) has become the literature’s
foundation, building from Galtung (1967) and Baldwin (1985). The
cases in Appendix 1 are a subset of the total cases provided by
Hufbauer, Schott, and Elliott (1990), a sanction case encyclopedia.6
The third section of this chapter concludes by positing stylized facts
about economic sanctions to be discussed through the book, both in
theory and empirically (in chapters five and six). Do alternative mar-
kets and sender hegemony play a role in sanction effectiveness? Is
policy duration a factor in effective sanctions? How do we measure
effectiveness? This chapter begins building toward the general equi-
librium model of chapter five, concluding with thoughts about
12 ECONOMIC SANCTIONS

economic theory’s limits in both explaining sanction effects and the

literature to date.

Basic Sanction Theory

Economic sanctions are just that: economic. Profit pursuit and subse-
quent market movements drive economic activity in all countries
involved, regardless of political regime. In less capitalist and demo-
cratic societies, that profit may be absorbed by an autocratic amoeba of
bureaucracy and market frictions. Thus, sanction effects may be super-
ficial to governments, harsh on its population, but must ultimately
change the target’s politics to be both credible and effective.
We view these policies from the sender’s perspective generally,
rather than the target economy. Because of the economic connections
between the two nations, and the fact that the sender is initiating the
policies, using this vantage point should minimize confusion. For
example, the target’s export market with the sender (the sender’s
import market with the target) reacts to the sender’s import sanctions.
Analogously, the target’s import market with the sender reacts to the
sender’s export sanctions. We begin with import sanctions, which are
akin to classic trade barriers, restricting sender goods and services
inflows from the target.

Import Sanctions
Theoretically, the target’s export market reflects the worldwide supply
and demand for its goods. We assume that if the target’s exports are
sanctioned, it reduces worldwide demand from the current quantity
demanded at the current world price. If we assume markets are in
equilibrium before sanctions were imposed, basic economics illustrates
the welfare effects of a change in price and quantity on each market.
Figure 2.1 compares the pre-sanction equilibrium in the target’s
export (EX) market to the sender’ import (IM) market.
The original world equilibrium takes place at a world price (PW),
where the line S is the supply curve and line D is the demand curve
in each market. Variables with asterisks are domestic equilibrium
values. The quantities supplied (QS) and demanded (QD) determine
the market surplus (QS  QD) or shortage (QS  QD) at the cur-
rent world price vis-à-vis the equilibrium price without trade (P*).
Since the world price is above equilibrium in the target’s export
market (P*T) the target is a net exporter of these goods, where a
market surplus is sold to foreign buyers. The sender faces the
 BASIC SANCTION ANALYSIS 13

The Target’s Export Market

PEX

Phigh
S

PW

P*

Plow
D
QD Q* QS QEX

CS = ½(Phigh- Pw) × QD; PS = ½(PW– Plow) × QS

PIM The Sender’s Import Market

Phigh

P*

PW

Plow D

QS Q* QD QIM
CS = ½(Phigh – Pw) × QD; PS = ½(PW – Plow) × QS

Figure 2.1 The target’s export market/sender’s import market.

opposite conditions as the world price is below its closed-economy

equilibrium. This basic model assumes the target does not dictate its
own export prices, the world market does instead. A trade sanction is
assumed to have a “ceteris paribus” effect, where all other economic
variables are held constant. In the model in chapter five, import sanc-
tions reduce the target’s wealth through reduced export revenues.
This lost revenue acts like an indirect payment, a “tax” for pursuing
deviant policies.
14 ECONOMIC SANCTIONS

In this simple supply-and-demand analysis, “welfare” is the sum of

consumer and producer surplus derived from market actions. Consumer
surplus is the gain for buyers who purchase a good at an equilibrium
price below what they otherwise would have been willing to pay.
Producer surplus is the gain experienced by producers from selling their
goods at prices greater than the market’s equilibrium price. The sum of
these two surpluses is equal to societal welfare derived from the market.
The triangles CS and PS represent these surpluses in figure 2.1.
Sanctions reduce welfare by creating market frictions, much like
any trade barrier.7 Import sanctions act like import quotas in theory.
Quotas are generally imposed as protectionist measures on a specific
good.8 One major difference is that sanctions do not generate revenue
from selling licenses to foreign producers to sell their goods, as they
would under a quota. The sanction is meant to restrict specific
imports (oil from Iraq, Libya, Syria, and Iran, e.g.) or all imports from
a country (Cuban goods, services, tourism travel, etc.) to decrease the
target’s ability to fund its policies.
The embargo’s resultant welfare effects are shown in figure 2.2,
where the shock represented by the dark, vertical lines has two effects.
First, target export sales fall, reducing target revenues. Instead of
exports equaling QSQD, they now equal QS2QD2, while price falls
to P2. The next effect is from the sender’s import price rising to P2,
where its market shortage is now QD2QS2. In figure 2.2, the sender
economy experiences wealth redistributions in the following ways
from the policy. Area a is a gain to the sender’s producer, the revenue
generated by new production of import-competing products. Areas b,
c, and d are the net welfare reduction from sanctions in the sender
economy; area a is a welfare wash because it also represents a loss to
domestic consumers who now pay higher prices.
In basic trade theory, area c should represent the government’s
revenue from selling licenses under a quota. Since this is a sanction,
we assume no licenses are sold, thus area c is additional welfare lost
and not monopoly rents for the government imposing the barrier.
The target producers experience lost sales in their export market, and
prices fall toward the closed-economy equilibrium, with the producer
losing more than the consumer gains from lower prices. The target
economy also loses wealth in net, where areas e, f, g, and h in figure 2.2
are the consumer gains (area e) and the producer losses (areas e, f, g,
and h in sum). The new market conditions are illustrated graphically
in this figure.
The resultant shift’s magnitude (from D to D) and the domestic
supply and demand elasticities dictate the welfare effects’ magnitudes.
 BASIC SANCTION ANALYSIS 15

The target’s export market

PEX f g h

PW
e
P2
PT*

QD QD2 Q* QS2 QS QEX

The sender’s import market

PIM

c d
PS*
P2

PW

QS QS2 Q* QD2 QD QIM

a b

Figure 2.2 Import sanction effects.

Sanctions are partial in certain cases, focused on specific target goods

rather than on all goods (see Travel Sanctions later), a relatively small
shift of sender demand; in other cases, embargoes are comprehensive
causing larger losses. The sender’s welfare loss in figure 2.2 is increased
as its domestic supply curve becomes flatter (more elastic); this implies
the domestic supplier cannot fulfill domestic demand easily and take
advantage of higher prices. In the target market, the flatter its domestic
demand curve (more elastic), the larger are the welfare losses. Target
producers lose even more revenue because domestic consumers can
easily find substitutes for the good. Relatively lower prices do not
16 ECONOMIC SANCTIONS

increase consumer surplus as much as producers lose revenue. Both

economies experience welfare losses in net, as does the world.
Restricting travel to the target nation is an example of import sanctions.

Travel Sanctions
The theory behind using travel sanctions, especially if the deviant
government derives tax revenue from expenditures, such as hotel stays
and casino gambling, is tantamount to focal or “smart” sanctions as
discussed in chapter four. Many developing countries, especially Latin
America and Caribbean nations, gear themselves toward tourism busi-
ness where gambling and nightlife draw hundreds of thousands of
people annually to their shores. Cuba is a classic example of current
U.S. sanctions on travel. Given Cuba’s proximity to the United States,
traveling to Cuba would otherwise be extremely easy. Flights from
Miami, Houston, and New Orleans would likely leave multiple times
per day, especially during the American winter, feeding the Cuban
economy with foreign reserves of U.S. dollars. While the rest of the
world travels to Cuba, Americans must find circumspect ways to make
that journey, and even then must be careful when returning from
Cuba with souvenirs (especially cigars, of course). These sanctions
cost both countries because trade no longer flows freely.
Travel sanctions are import sanctions. The inability to export travel
and recreation to sender travelers, who import these services while
traveling in the target nation, has three effects. First, it reduces the
target’s income from domestic travel and related industries, ranging
from hotels to transportation to casinos to local food markets. Also,
there are reduced amounts of foreign currency in the target economy.
The lack of foreign reserves may restrict the target’s ability to trade
generally, as foreign currency (specifically U.S. dollars) acts as a vehicle
for foreign transactions.
The analyses presented earlier are simple but illuminating. The
sender manipulates the target market quantities, forcing a reduction
in the target’s welfare. Import sanctions are classic sanctions and have
effects like trade barriers, specifically quotas. The earlier figures show
how a sender may affect a target’s wealth, transmitting effects to lower
overall welfare and funding for the target government’s actions. The
other trade sanction is on a sender’s sales to the target.

Export Sanctions
Export sanctions are analogous to a sender economy imposing a
voluntary export restraint (VER). The target’s import market mirrors
 BASIC SANCTION ANALYSIS 17

the sender’s export market; as earlier, we assume that the worldwide

supply and demand for the sender’s goods dictate the market price to
the target. In export sanctions, the target’s supply of imports is affected.
Export sanctions increase the world price, forcing higher prices and
lower quantities in the target economy for the sender’s goods. Firms in
the target country gain from export sanctions, as sales of the domesti-
cally produced goods rise when sender firms abandon this market.
The sender’s producers face lower revenues for their goods, a cost of
domestic governments initiating export sanctions (see Losman 1998).
This shift’s magnitude and effects depend on the trade and eco-
nomic integration between the sender and target. The sender good’s
elasticity of demand in the target economy dictates how much target
welfare is ultimately lost. Similar to figure 2.2, figure 2.3 shows the
post-sanction markets for the sender and target. The reader can infer
the pre-sanction situation by reversing the graphs in figure 2.1. In this
case, the sender is the exporter and the target the importer.
The net effect on both economies’ welfare is negative, as in import
sanctions. Area a represents target producers gaining revenue while
consumers lose due to reduced sender imports, where areas b, c, and
d represent the additional consumer loss of welfare. The export sanc-
tion mocks the sender imposing an export quota on itself. For the
sender, the areas e, f, g, and h represent the gains (area e to the
consumer) and losses (areas f, g, and h to the producer) from this
sanction’s imposition. Specific goods of strategic importance have
been historically curtailed by senders. These goods include arms sales,
developmental aid, and technology.

Arms Sanctions
These export sanctions are specific to military hardware and weapons
systems, seen as examples of smart sanctions.9 These embargoes
reduce the target government’s ability to make war and force its
citizens into nondemocratic political systems. Arms sanctions have
evolved with geopolitics, ultimately as a way to constrain belligerent
governments from igniting regional conflicts, holding volatile nations
on the brink of military obsolescence. Bondi (2002) discusses reasons
why arms embargoes have recently increased after the end of the Cold
War. “The end of a bipolar world permitted the emergence of a non-
ideological debate over crisis management and conflict prevention
and resolution” (109). Arms embargoes have focused on reducing
terrorist activities since September 11, 2001, especially in countries
that either allegedly house terrorists or sell arms to anyone, where
Syria and Iran are prime examples.
18 ECONOMIC SANCTIONS

The target’s import market

PIM

c d
PS*
P2

PW

QS QS2 Q* QD2 QD QIM

a b

The sender’s export market

PEX f g h

PW
e
P2
PT*

QD QD2 Q* QS2 QS QEX

Figure 2.3 Export sanction effects.

Small arms sales dominate weapons sales worldwide, as weapons

systems and larger ordinance have fewer sources. Peru (1968), Turkey
(1974), Nicaragua (1977, 1981), and El Salvador (1977, 1987) are
case studies where arms sales curtailment is aimed at reducing threats
of regional conflict, civil war, or human rights abuses. Bondi (2002)
suggests that international monitoring of the arms trade is poor.
“[The small arms dealer’s] increasing and recognized prominence has
not yet been accompanied by more stringent controls over their oper-
ations, which remain largely unregulated and consequently almost
invariably go unpunished, even when in breach of international
 BASIC SANCTION ANALYSIS 19

sanctions” (113). Weapon system embargoes are specific examples of

capital and technology sanctions, restrictions that may also come in
the form of developmental aid embargoes.

Developmental Aid, Physical Capital,

and Technology Sanctions
One economic weapon the United States had wielded over many
developing countries is aid reductions. The form of aid restriction is
country-specific. Countries such as India (1965), Zimbabwe (1983),
and Liberia (1992) are examples of American aid embargoes. Unilateral
transfers of food are the least likely to be cut, due to humanitarian
considerations. The 1989 case of the Sudan is one where all but food
aid was cut by the United States (see O’Sullivan 2003).
Export sanctions may also curtail sales of physical goods used in
production. These goods range from tractors to intellectual capital,
computer software to new pharmaceutical factories. The loss of
physical capital from foreign sources reduces net investment in the
target’s macroeconomy. Theoretically, these sanctions threaten to
slow the target’s economic growth, leading to slower savings
rates, reduced GDP, and decreased consumption. Given advances in
telecommunications and computer technology, trade barriers to these
improvements may drive target infrastructure and capital quickly
toward obsolescence.
Reducing these inflows also reduces the target’s ability to export
goods. As a result, foreign asset holders may sell their ownership to
domestic citizens and firms, potentially causing asset price bubbles
without economic growth as support. Some authors believe such
actions helped sanctions against South Africa to defeat apartheid (see
Porter 1979). South African sanctions focused on reducing foreign
direct investment (FDI). The ultimate effect of “disinvestment” in
South Africa, the parallel sale of assets by foreigners along with a ban
on new investment, was to force asset prices down.

Forced disinvestment by foreigners of ownership claims to output

produced in South Africa has the effect of driving down the market
prices of these assets, and thus increasing their expected rates of return.
The effect on market prices will be a function of the size of the disin-
vestment [itself] and the sensitivity of international asset substitutability
relationships. (Kaempfer and Lowenberg 1986, 385)

Alternative markets mitigate welfare damage from these policies.

Sanctions on physical capital act like export sanctions because their
20 ECONOMIC SANCTIONS

focus is on restricting the availability of goods or services from outside

the target. For arms, aid, or capital sanctions, embargoes are on
specific industries or funding. Policy-imposed scarcity in these markets
aims to reduce productive capacity, lower funding available to deviant
governments and decrease returns on capital. Through the balance of
payments, as the model in chapter five makes explicit, trade embar-
goes are tied to capital account balances, as financial sanctions are tied
to current account balances.

Financial Sanctions
Curtailing financing is different than an embargo on capital goods, as
these are sanctions on financial entitlements to income.10 International
trade and finance are intrinsically linked through the balance of
payments. Trade sanctions indirectly affect target financial markets;
financial sanctions also indirectly affect the target’s goods markets.
Mechanically, financial sanctions are similar to trade sanctions. When
senders sanction their “imports” of target lending (capital inflows), the
sender reduces expected interest payments and income for target
investors. When curtailing its domestic sources of financing for the
target economy, the sender reduces funds supply, causing the target’s
financing costs to increase. The target is assumed to not determine its
own interest rates, the outside world does instead.11
Another type of financial sanction reduces or eliminates income
flows from target assets currently held in the sender economy, essen-
tially “freezing” the assets. If the target has used the sender’s financial
markets as a proxy for its own, these sanctions can be quite detrimen-
tal. To restrict asset income flows to foreigners, the sender must have
a certain precondition. A sender cannot credibly curtail target asset
income if it is a net creditor of the target. It is likely that targets, once
cut off from their own asset income flows, would retaliate in kind.
The threat to curtail these flows is only credible if the sender holds
more target assets than the target holds of the sender’s. To use credit
sanctions, where the sender neither provides nor demands financing,
the only precondition is that a financial market exists between the two
countries. A basic model of financial sanctions is shown in figures 2.4
and 2.5, analogous to the export embargo example (figure 2.3).
Figure 2.4 shows the market equilibrium for the target’s loanable
funds, or available credit. Assume that the target has a shortage of
available credit at the current world interest rate. In figure 2.5, the
target’s available credit from abroad is reduced from QDQS to
QD2QS2. This demand contraction forces asset prices down in the
 BASIC SANCTION ANALYSIS 21

R*

RW

QS Q* QD QLF

Figure 2.4 The target’s market for loanable funds.

c d
R*
R2

RW

QS QS2 Q* QD2 QD QLF

a b

Figure 2.5 Financial sanction effects on the target.

target economy, while increasing interest rates. As earlier, the elastic-

ity of demand and supply in each country’s market dictate the final
welfare effects’ magnitudes. The target’s net welfare losses are defined
by the trapezoid formed by areas b, c, and d.
Elliott (2002) suggests that financial sanctions have market forces
behind them that reinforce their effects: banks outside the sender–
target dyad may perceive the target to be more risky due to financial
sanctions. This risk augmentation is stronger when the target is a
country with few alternative sources of financial markets. Market
22 ECONOMIC SANCTIONS

forces are dictated by the elasticity of demand and supply, by the

number of substitute markets available.
If the target government uses asset income and foreign borrowing
derived from outside to fill its treasury with foreign currency, these
sanctions can be very focal. Steil and Litan (2006) discuss these sanc-
tions at length, making two points that underscore the current view of
these policies. First, capital flows affect trade flows in such a way to
shock exchange rates and basic economic conditions in the target
country; also, the mere threat of financial sanctions, especially from
the United States, cause objections from target governments and
beyond who view these measures as extreme and extraterritorial to
the target economy (5). If sanctions cause a currency crisis, directly
or indirectly, such a crisis must affect the average citizen, and may
spill over regionally. Sanctions are structured as if macroeconomic
policies.

Sanction Theory and Political Economy

Reducing a nation’s wealth by policy driving higher prices and inter-
est rates, lower income levels, and reduced productivity are all part of
a sanction’s political economy. However, measuring sanction effects is
difficult. Using historic case studies helps understand different con-
texts and sanction characteristics, especially the multiple ways sanc-
tions are influenced to begin, continue, and end. Sender nations
initiate sanctions because they perceive the benefits of target political
change exceed the costs sanctions guarantee to all parties. A policy’s
efficacy, credibility, and forecasted damage should be tracked by
senders to decide when policy should change or end. The political
economy of economic statecraft is tracked by marginal movements
toward the sanction’s goals. The following studies are examples of
seminal sanction investigations concerned with political economy and
linking basic theory to sanction effectiveness.

Sanctions as Diplomatic Tools

Statecraft is a word used to describe how nations engage in diplomacy.
Using negotiations to settle international disputes is an evolving
legacy for the United Nations. Chapter 7 of the U.N. Charter, Article 41
in particular, calls for a specific approach to any threat to international
peace.12 Article 41 suggests that the United Nations must first curtail
economic connections with rogue nations or a group of belligerents.
Article 42 suggests that after economic sanctions are determined not
 BASIC SANCTION ANALYSIS 23

to work, the need and magnitude of military action will then be

discussed and potentially initiated. One can easily see the subtlety that
ties Articles 41 and 42 together. The economic weapon is to be used
before military force, if military force is debated at all. This makes a
precise and deep examination of sanction effects necessary before
initiating any additional actions. Ultimately, the United Nations
should assess every sanction, as should all senders. The driving force
in the literature, beginning with Galtung (1967), has been identifying
the characteristics and signs of sanction effectiveness qualitatively and
empirically.
The 1966 Rhodesian case, in many ways, sparked this literature.
The UN statecraft policies reacted to human rights violations, inter-
national concerns over a deteriorating and nonrepresentative political
structure, and possible regional conflict. Questions of multilateral
measures and their efficacy, potential leakages from a sanction coali-
tion, and the target political system’s ability to be changed or com-
pletely turned over, all were questioned in this case. One text not only
expanded on the political economy theory and definitions of eco-
nomic coercion, it also examined case studies where the Rhodesian
case was front and center.13
David Baldwin’s study Economic Statecraft (1985) is a pillar in this
literature. Baldwin did not attempt to measure sanction success or
failure empirically. He explained how sanctions worked as political
economy instruments to succeed, defined economic statecraft in alter-
native ways, gave caveats for how trade and aid sanctions are defined
and employed, and presented case studies. His text focused on cases
such as South Africa, Cuba, and Rhodesia, employing conflict resolu-
tion theory, and also discussed sanctions in power relationships, as in
Knorr (1975).
Knorr (1975), in The Power of Nations, examines economic warfare
and the use of national power through international policy. He
defined how power relationships and hegemony can be used in eco-
nomic sanctions, a theme continued by Baldwin (1985), a much-
debated issue since. “Like all power, material economic power can be
used for coercion, that is, for influencing the behavior of the weaker
actor, or for directly achieving a desired effect, whether harmful in an
adversary relationship or supportive between friendly or allied coun-
tries” (Knorr 1975, 138). Knorr’s assessment of collective sanctions
greatly extended Galtung (1967) and connects to sanction cartel
possibilities as discussed in chapter three of this text. “In principle,
collective trade sanctions should be more effective than economic
reprisals by one state or a few, because they would be based on an
24 ECONOMIC SANCTIONS

international cumulative degree of monopolist or monopsonist control

over the world market” (Knorr 1975, 160). Knorr further suggested
military blockade throughout history, taught us that it was possible to
embargo efficiently, and that trade sanctions were the new type of
blockade if used correctly.
Knorr’s thoughts and ideas provided further foundations for
Baldwin. Baldwin expanded on economic sanctions in specific, whereas
Knorr was more general in examining trade policy, aid provision, and
international relations. Baldwin’s case study of 1966 Rhodesian
sanctions provides key ideas for this text. The sanction’s main goal was
for Rhodesia to allow majority rule, while secondary goals included
avoiding force, isolating Rhodesia economically and politically,
encouraging regime opponents, and imposing costs if majority rule
was not instituted (Baldwin 1985, 191). This case explains sanction
effects from many angles; the Rhodesian human rights situation typi-
fied the reasons to use sanctions rather than military intervention.
Baldwin also made the point that the average citizen was affected by
sanctions; as the conditions of black Rhodesians, especially in terms of
unemployment versus their white leadership post-sanctions, were seen
as worse and disproportionate in the sanction’s influence (195–96).
Baldwin’s look at Rhodesia provides two general questions for further
analysis: (i) do sanctions cause change or support political evolution
already underway?; and (ii) in what ways does the populace suffer
from sanctions?
Putting the world on notice that economic coercion may cause
collateral damage is also a seminal insight in Baldwin’s text. Chapter
four of this text discusses smart sanctions, where the policy focus is
minimizing general harm by focusing on target decision makers. Like
other texts, Baldwin’s analysis is at a principles level and provides
background for viewing these policies using conflict resolution tech-
niques. In his conclusions, Baldwin suggests that success is a matter of
degree (371). The concept of marginal effectiveness was born here,
which is in stark contrast to Hufbauer, Schott, and Elliott (1990) and
their contribution and success scores. The theme of marginal move-
ments toward goals continues throughout this text in the open econ-
omy macroeconomic model in chapter five and the empirical work
in chapter six. Baldwin drew from the work of Doxey (1971) and
Renwick (1981) for the Rhodesian case study and questions of
sanction legality (Baldwin 1985, 336–69).
Doxey (1971, 1987) looked at international law and whether
nations had the “right” to impose economic statecraft measures.
Doxey (1971) suggested that sanctions are methods of coercive
 BASIC SANCTION ANALYSIS 25

diplomacy, statements against deviant policies that crossed inter-

national boundaries as well as jurisdictions. Her work laid the
foundation for debates over the importance of cooperative efforts,
special interest group influence, and sanction legality. Doxey (1987)
extended the original study using specific cases in detail, discussing
conditions under which countries should be allowed to sanction. Her
socioeconomic approach was seminal in stating why sanctions may
fail: they are viewed by the target populace as an act of war, and would
be resisted.14
Doxey (1987) made two major points that prevail in the literature
today. First, sanction success must be linked with its stated or unstated
purpose. The Sanction Effectiveness Continuum illustrates the possi-
ble outcomes a sanction can achieve. Economic damage is itself a
“success” of sorts, as reducing target welfare is an implicit statecraft
goal. How that damage exacerbates or creates problems determines if
it is a humanitarian success; if the stated political goal is achieved, it is
a political success. Doxey’s study provided seminal reasoning for
multiple thresholds and for measuring third-party effects.
Robin Renwick (1981) examined specific sanctions in great detail
and identified historic patterns in economic statecraft; Renwick also
concentrated on the Rhodesian case, focusing on decision-making,
where these sanctions were flawed, and why sanctions worked or not.
Using the Rhodesian example, Renwick makes generalizations that
provide foundation for understanding in what circumstances the
UN’s economic muscle should be flexed. In the Rhodesian case, the
military option was threatened but not used. Embargo effectiveness
may be enhanced when military threats act complementary to sanc-
tions, the very action the sanctions are attempting to avoid if follow-
ing Chapter 7 of the UN Charter. Renwick’s work also emphasized
the need for universality in sanction imposition. “Sanctions frequently—
one might also say, generally—are decided in large measure as a
consequence of the lack of feasible alternatives” (85).
Sanctions, as international diplomacy tools, must be prudent and
self-aware, used only if there is international consensus on its size and
scope. Third parties must prepare themselves for sanction effects.
Senders and targets must expect outside parties to react in the pursuit
of profit. In sum, international diplomacy starts with negotiations.
Embargoes should substitute for military options, the first choice
once verbal diplomacy breaks down for any sender. Baldwin (1985)
was the peak of the literature that it followed, as it was comprehensive
in its literature review and provided illustrative cases. We now review
the contributions of this literature’s current foundation.
26 ECONOMIC SANCTIONS

Hufbauer, Schott, and Elliott (1990)

Hufbauer, Schott, and Elliott (HSE; 1990) compiled over 110 cases
of economic sanctions since World War I, analyzing each case by
assigning points. This extended their 1985 study such that the 1990
study is heralded as encyclopedic and comprehensive.15 Each case is
broken down into similar parts for comparison. Their thesis is that
each case, while different, can be assessed as to its success in goal
achievement. After providing a detailed chronology of each episode,
HSE assigned case-specific “scores” for both political and economic
success, each ranging from one to four (one being the least effective
and four being the most). The integer assigned for political success is
based on the level of target capitulation to the sender’s demands. The
way HSE present and assign economic success is through welfare loss
and gain, measured by shocks to GNP attributable to the embargo. If
the target economy experienced a severe drop in GNP, or if the sender
used very little of its GNP relative to the target’s loss, in percentage
terms, the “contribution score” is three or four (success). The eco-
nomic contribution score is multiplied by the political success score to
calculate an overall success score.16 If the sanction episode receives an
overall success score of nine or greater, it is termed a relative success;
if the sanction scored eight or lower, the episode is a relative failure.
Scholars have identified two potential sources of empirical prob-
lems in HSE. First, endogenity bias may exist in the data for economic
success. The economic contribution score and the political success
score are treated as mutually exclusive, but logic suggests they are
dependent on one another. This endogenity problem becomes an
empirical problem because economic policy and its goals are rarely
exogenous of each other (see Sims 1980). Second, a selection bias
problem exists because many sanction cases were excluded by HSE as
mere threats not involving explicit sanctions. Drezsner (2003) focuses
on this problem in specific, and echoed the concerns of Lam (1990),
Drury (2002), and others. The success rate stated by almost every
news article or scholarly work that cites HSE says there is an approxi-
mated 33 percent success rate of sanctions since 1945. Dreszner
(2003) suggests that if the threats were also included, the success rate
would be over 50 percent, depending on the threat’s genesis
(653–54).
HSE’s analysis compares and contrasts the economic environment
before and after sanction imposition in cases that had ended by 1990.
Their data categories (cost to sender, loss of trade for target, etc.) pro-
vide information and logic for the subjective, economic contribution
 BASIC SANCTION ANALYSIS 27

score. As target GNP falls, the economic pressure on the populace

increases, placing the target’s current policies at risk because the
current government is at risk. How well the target’s rulers can govern
their people and coercion’s effects simultaneously helps measure
political success. Hence the endogenity problem.
HSE states the sanction goal and then looks for evidence of how
close the target came to acquiescence. For example, the 1970 sanctions
against Chile were to combat human rights abuses and the Allende
government. These sanctions are seen as a success, as Allende was
assassinated in 1973. HSE gives this case a political success score of
“four,” as the stated goal was achieved. This is in sharp contrast
to Cuban (HSE case 62–3), North Korean (case 50–1), Libyan (case
78–4), and Iraqi (case 90–1) sanction episodes, where the ousting of
each country’s leader was the stated goal, and none of these sanctions
succeeded. Because actual versus stated policy goals can be deceptively
dissimilar, HSE attempted to focus on just stated goals. Regardless of
the caveats, both the economic contribution and political success
scores in HSE provide a strong foundation for further research of each
individual case, but must be seen as ad hoc, subjective assessments.17
Three major characteristics of successful cases empower HSE’s
results. Each characteristic’s importance is debatable, subjects of many
scholarly works. Derived from Galtung (1967), Knorr (1975), and
Baldwin (1985), each characteristic was investigated in depth and
breadth through the case analyses in HSE. The reader should think of
these characteristics as independent variables in regressions, where the
dependent variable is the measure of a sanction’s economic effects.

Trade and Financial Dependence

The more dependent the target economy is on sender trade, the more
powerful the sanction should be ex ante. This seems obvious: if targets
are sufficiently dependent on senders for key goods and financing,
leakages should be few and costly. Sanction effects depend on trade
routes, trade partners, and international availability of credit. Trade
dependence is a function of a target’s ability to reallocate resources
not only from sources outside its borders, but within its domestic pro-
duction as well. Financial dependence may be more likely, especially in
countries where financial markets are not domestically or internation-
ally integrated or available. Given the potential profits in sanction
busting, trade dependence on a solitary country seems more unlikely
as globalization continues. The more universal sanctions can become,
the more likely economic effectiveness would result. “The more
28 ECONOMIC SANCTIONS

difficult it is for the target to find alternative foreign sources of supply,

alternative foreign markets or substitutes [domestic or foreign] for the
goods covered by the sanctions, the more effective the sanctions will
be” (Miyagawa 1992, 25).
Financial dependence is a parallel factor to the sender’s trade con-
trols. The integration of capital markets requires communications
infrastructure, a central clearinghouse for transactions, and interna-
tional recognition of the target’s domestic stock exchange. Even
China, whose star has risen hard and fast since 2000, struggles to have
integrative, domestic capital market transactions. China National
Petroleum Company’s initial public offering is an example of raising
financial capital in the United States without a thought of using a
Chinese source of equity; Hong Kong’s exchange was not discussed as
a place to raise the money (Steil and Litan 2006, 58). Many countries
depend on American stock markets for financial stability, fast transac-
tions, and lower costs. This characteristic facilitates the U.S. ability to
engage in financial sanctions. Countries that depend on foreign finan-
cial markets to attract and produce wealth open themselves to the use
of financial statecraft. The sender’s relative economic size may also
make the target vulnerable.

Relative Economic Size

Political and economic hegemony may be determining factors in
sanction effectiveness. However, does relative economic size mean
international political potency through embargoes? GNP ratios act as
a measure of relative size in HSE; while a good measure of an econ-
omy, GNP is somewhat archaic. In some historic cases, however, GNP
is the only measure of national income readily available in the data.18
Gross Domestic Product (GDP), as an alternative, focuses on each
country’s domestic ability to generate income. Percentage changes in
GDP or GNP measure how well the respective economies absorb
trade or financial shocks. Variables, such as total trade between the
nations, exports plus imports alone or as a percentage of GDP (open-
ness), may provide better information for both trade dependence and
size. The assumption is that the economically larger the target, the
more likely it has many trading partners and potential alternative
markets. However, if a small country has a key product used world-
wide (e.g., oil), its risk exposure may be small versus its relative size.
A country like Venezuela can use its oil resources as a buffer against a
hegemonic sender; in 2003, Venezuela had a GDP ratio of approxi-
mately 1.29 percent of U.S. GDP (IMF 2006).
 BASIC SANCTION ANALYSIS 29

Alternative Trade Routes and Partners

Every sender economy contends with an inability to perfectly restrict
goods and capital flows to the target. Sender nations must identify and
form compacts with target trading partners to minimize leakages.
Trade and financial dependence is likely to be a function of a nation’s
economic activity, driving international market availability and trade
integration. The number and type of alternative trading partners is a
classic, obvious aspect of economic sanction effectiveness.19 The
elasticity of target demand for the sender’s goods is a reflection of
how substitutable the sender’s goods are; the elasticity of the target’s
supply of goods purchased by the sender reflects how many markets
still exist if the sender embargoed the target’s exports. Perhaps multi-
lateral and universal sanctions are automatic winners for sanction
efficacy; however, the evidence is not conclusive to date and the
debate is likely to continue as more and more sanctions are used.20
HSE (1990) is the most cited work in this literature. Their study
provided comprehensive case details and made conclusions about
each case, well-grounded in political economy. Their data, while sub-
jective in many areas, make sense once the case details are synthesized.
Their conclusions and methods have created another strand in this
literature with scholars trying to either refine or argue against their
conclusions. HSE is this literature’s backbone to date, regardless of its
shortcomings. Cooper Drury’s work on the HSE data serves as a final
look at seminal studies. Drury’s work has become the new foundation
of political economy in this literature.

Cooper Drury
Drury (1998) confronted the literature’s results head-on with an
insightful critique, acting as a grand addendum to the HSE (1990)
study. Using their data set and adding political dummy variables and
data regarding trade flows and sender size, Drury estimated the cor-
relation between a sanction’s effectiveness and the basic data that
HSE provides. He developed a model, derived regression coefficients,
and made conclusions from them. Drury’s conclusions add to our
knowledge of sanctions.21 He suggests that the presence of a prior,
friendly relationship between the sender and target had no effect on
sanction effectiveness. This was in contrast to the conventional wis-
dom that when targets have large economic friends they do not want
to jeopardize that relationship and acquiesce under sanctions. Also,
alternative markets do not always negatively affect sanction efficacy.
30 ECONOMIC SANCTIONS

This result provides evidence that eliminating all markets may not be
necessary. Also, pre-sanction trade flow do not affect the sanction’s
outcome. “When explaining sanction effectiveness, the actual pre-
sanction trade level is relatively unimportant when compared with the
damage done to the target” (Drury 1998: 507).
Drury also suggests that of HSE’s nine policy recommendations,
only two hold up to further analysis.22 These include: the higher the
cost of sanctions, the more successful the episode; and the more dis-
tressed the target, the more successful the sanctions (ibid.). These are
important conclusions as the first is counterintuitive and the second
leads to an empirical challenge not easily solved. Logically, if costs
mount higher and higher, senders are more likely to ends the mea-
sures. Also, measuring economic distress prior to sanction imposition
has many possibilities, and debate could rage over any such as inflation,
unemployment, political instability, and so on.
Drury (2005) updates and expands on these issues and the U.S.
president’s role in sanction cases. Drury’s works provides three
insights moving forward. First, he identifies serious flaws in the HSE
conclusions and data. Second, his work incorporates many new inde-
pendent variables into the literature, as well as a continued use of
discrete choice modeling. Finally, his works show this literature has a
long way to go in explaining the economic effects of sanctions.
Chapter six in this text examines new ways to empirically investigate
economic statecraft, expanding directly on empirical methods in
Drury (2005) that provide these three insights.

Stylized Facts from Basic Sanction Analyses

There are four stylized facts drawn from the studies previously men-
tioned concerning political economy and sanctions. Chapter six exam-
ines these questions empirically, while Appendix 1’s histories of the
empirical cases provide a context per these stylized facts. This text’s
remaining chapters discuss these facts theoretically as mentioned in
each section that follows.

Marginal Improvements are

Successful Sanctions
Recent developments in North Korea remind us that the brinksmanship
antics of a dictator continue because the target’s people allow it. Can
sanctions easily change such polity? Case histories illustrate the difficulty
in doing so, as few sanctions have led to target government turnover. It is
 BASIC SANCTION ANALYSIS 31

important that scholars and policy makers begin to move away from
measuring success in absolute terms and begin to focus on relative
measurements. North Korea recently coming back to the negotiation
table shows the policies are effective. The simple model of sanction
effects described earlier suggests that forcing economic welfare losses
on a target is not difficult. The case studies tell us that lower welfare act-
ing as a conduit to political change is where sanction efficacy as a whole
breaks down. The analysis of public choice models in chapter four dis-
cusses how policy applies pressure through imposing costs on target
interest groups. If a certain political system exists such that the only
organized interest group is the regime in power, then focal sanctions
may reduce a sanction’s bluntness and force marginal change. Chapter
four also looks at these smart sanctions in both theory and reality.

Both the Sender’s and Target’s Diplomatic

Partners May Act as Sanction Busters
In many cases, interested third parties show their face on either side.
In some cases, there is international consensus and support for sanc-
tions. During the 1990s, as the United States expanded its embargo
policies widely, there was somewhat inconsistent international cooper-
ation. Martin (1992) points out that economic theory dictates that as
targets become more economically isolated, the sender’s policy poten-
tial strengthens. Alternative markets relieve economic pressure and
lead to spillover effects from policies onto otherwise uninvolved nations
and their populace; thus sanctions create both positive and negative
externalities. Measurement of these spillovers may be very difficult,
especially where data is parsimonious or non-credible, or where the
external market’s economic ties to either the sender or target logically
exist but are difficult to assess. Cooperative and competitive efforts
over the target’s polity are discussed in more detail in chapter three,
and also empirically tested in chapter six.

Sender Financial and Trade Hegemony is

Not a Sanction Necessity
Another stylized fact concerning sanction effectiveness is the role of
the sender’s economic strength and a perceived inability to use that
strength to force political change. The point of multilateral, coopera-
tive efforts is to act as a policy cartel to sway countries morally to
reverse their deviant policies through either positive or negative incen-
tives. Knorr (1975) and Baldwin (1985) are seminal works in how
32 ECONOMIC SANCTIONS

power relationships are used in policy such as economic coercion.

Their work on positive incentives versus punishment as different
structures of policy influences much of chapter six’s empirical work in
this text. Martin (1992) made a great case concerning the connection
between hegemony and multilateral sanctions, concluding that hege-
mony was immaterial. Drury (1998) corroborated this conclusion
empirically. If sanctions act as cartels, they are subject to cheating and
leakages as any collusive economic arrangements. Reducing profit
incentives to either cheat on or exist outside the policy cartel’s collu-
sive boundaries is the major challenge of diplomacy among potential
sender economies. Chapters three and four look at how game, price,
and public choice theories in economics may help policy makers and
scholars further understand the incentives of nations to act as sanction
busters. Chapter six also looks empirically at institutional sanctions, as
did Drury (2005).

Time is Not a Factor in Sanctions

and Statecraft
Bolks and Al-Sowayel (2000) suggest that the target’s characteristics,
specifically its institutions and current regime’s vulnerability to oppo-
sition, determine the length of sanction episodes. “Internal or exter-
nal pressures may cause [target] leaders to change the structure of
decision making in the regime. In turn, the development of counter-
measures may be altered as well. Consequently, the target’s reaction
to sanctions is a function of these confluent factors” (245). Sanction
analyses and debate would not exist if policy effects were straightfor-
ward to measure and track over time. Policy makers must imagine
resources wasted in using sanctions where the target’s political econ-
omy insulates its decision makers from sanction effects for decades.
Cases such as Cuba, North Korea, Iraq, Iran, and Libya all show how
long-term sanctions may devastate an economy and not change its
political regime. Empirically, the challenge is gaining access to time
series data, tracking economic coercion over time, and testing for
changes using econometrics. The model in chapter five provides a
theoretical framework to watch exchange rate volatility as a measure
of economic effectiveness, while chapter six employs time series analy-
sis to identify the direction and magnitude of exchange rate shocks
from sanctions.
Sanctions are aimed at affecting decision-making in another coun-
try. Senders may enjoy a time when philosophical agreements about
problematic target policy leads to multilateral sanctions, economic
 BASIC SANCTION ANALYSIS 33

incentives generally rule the strength of that policy cartel in the long
term. For example, concerns with communism and authoritative
governments in the Western world eroded over time due to cheap
sugar and coffee sources in Central America and the world’s affinity
for Cuban cigars. It is important that policy makers see sanctions as
aggressive diplomacy using economics as the weapon.

Conclusions
Sanctions are nontariff barriers to trade at their core. The sender’s
sanctions curtail economic connections between itself and the target
in an attempt to change the target’s political decisions. Sanctions,
regardless of their breadth, are either trade or financial sanctions.
There are three economic aims of sanctions, all to reduce the target’s
wealth. Sanctions aim to increase the target’s price of imported
goods and financing, decrease the target’s income from its sales of
exports or current assets, or reduce unilateral transfers and financing
availability. In sum, the sender’s policy actions are meant to unam-
biguously bring about target welfare losses, at some accepted, parallel
cost. Basic models come from seminal studies in sanction analysis,
beginning with Galtung (1967) and expanded upon by Knorr (1975)
and Baldwin (1985).
HSE (1990), the literature’s current foundation, is comprehensive
in breadth and applies basic international economics and empirical
techniques to estimate a sanction’s economic and political effects. It
does so by investigating case studies in depth and applying economic
logic. Baldwin (1985), derived from Knorr (1975), was the early
literature’s peak, expanding the nomenclature and political economy
theory of sanctions. Doxey (1987) provides different perspectives on
economic statecraft, specifically its international legality. Drury
(1998) added a much-need empirical expansion to HSE (1990), and
has sense expanded on those themes in his work in 2005. The critical
feature of all these studies is their use of basic economics and political
science to provide a framework in which to look at any sanction
policies. The four stylized facts provided earlier link the lessons of
earlier works to the remainder of this text. Chapter three discusses
why sanctions begin, continue, and end by reviewing game theory
models of economic statecraft.
This page intentionally left blank
 Chapter 3

Sanction Initiation and

Continuance: Enter Game Theory

The Games Nations Play

The War of Attrition game was originally posed as a model of conflict
between two animals over prey. Each animal chooses the time at which it
intends to give up. When an animal gives up, its opponent obtains all the
prey. If both animals give up at the same time, then each has an equal
chance of obtaining the prey. Fighting is costly: each animal prefers as
short a fight as possible.
—Osbourne 2004, 77

Sanctions act like wars of attrition. The sender attempts to manipulate

the target’s wealth and income such that the target acquiesces and
give up its prey. Instead of lions fighting over a gazelle in the sub-
Saharan plains, the prey is the ability to engage in human rights
abuses, aboveground nuclear testing, invasion of a neighbor, technol-
ogy piracy, and so on. As embargoes continue to be acts of diplomacy,
costs mount on both sides and reduce the benefits of respective
actions: the deviant behavior of the target and the sender’s economic
statecraft. The imposition of sanctions represents a deadweight loss of
utility for both nations, providing an incentive to reach an agreement
before imposition (Drezner 2003, 644).
Why do sanctions begin, continue, and end? Chapter two provided
a simple model of sanction effects, like those in Carbaugh and Wassink
(1988) or in Carbaugh and Olienyk (1998). In such models, the
world is worse off concerning overall welfare, but wealth is redistrib-
uted among the nations and entities involved. Game theory, models
describing strategic choices by parties involved in economic and polit-
ical transactions, provides multiple avenues of insight for sanction
36 ECONOMIC SANCTIONS

analysis. First, conflict resolution models easily apply to economic

statecraft, helping explain why sanctions begin, continue, and end
through their political economy. The expected net benefits or payoffs
are changing to both parties constantly during the sanction game.
Decision-making on both sides may also shift before and during the
embargo. Game theory models provide an exploration of cooperative
versus competitive outcomes under uncertainty. The cooperative
outcome where each side forms a strategy conditional on the other
side’s strategic choices, so-called Nash Equilibrium, may explain
strategic economic statecraft decisions. This theme of cooperation
transcends the sender’s decisions and involves other nations in either
support or circumvention of the sender’s coercive measures.
It is a mistake to think that sanctions do not imply strategic out-
comes. Policy may not be formulated in a strategic way initially. As
sanctions continue, strategic goals become apparent, and each party
adjusts accordingly. Game theory tells us that sanction initiation and
continuance lies in the sender believing and proselytizing that the
embargo’s expected net benefits exceed those derived from doing
nothing; as a corollary, the target’s costs exceed the benefits of
continuing the political action for which the sanction is imposed.
While this may sound like rational decision-making, as economics is
based on rational agents allocating resources to maximize individual
benefits in the midst of scarcity, these decisions are conditional on the
other party’s decisions. Because the sender and target lack informa-
tion about each other, the parties interact given each other’s best
strategies.
When a sender initiates an embargo, it must simultaneously moni-
tor all economic, human, and political costs to gauge continuing the
embargo, based on this new information. Since sanctions come in
many different forms and are case-specific, game theory models easily
adapt to explain why and how sanctions change or end. From basic
examples, where a mere threat is used, to the most complex measures
using comprehensive trade and financial restrictions with a military
option in the wings, game theory can explain the strategic choices of
economic diplomacy from beginning to end. Sanctions end when
continuance no longer provides a net positive benefit to the sender
due to the target’s acquiescence or policy impotency.
This chapter is split into six sections. The first section provides a
game theory primer and basic models of game structures, including
the rationale for Nash Equilibrium. The second section discussed
threats and blockades, the beginning of economic coercion, and the
brink of the military option. Do threats provide enough information
 SANCTION INITIATION AND CONTINUANCE 37

to force a quick settlement? If the threat does not work, then the
sanction begins. Next, a discussion of applied game theory models
discusses some seminal works on sanctions. Eaton and Engers (1992)
provide a highly theoretical model of sanctions, with some practical
applications when considering the toughness and resolve of each party
to continue an embargo. The fourth section looks at models of
cooperation and whether senders can act as a policy cartel. Economic
isolation is a key ingredient in sanction policy; cartelizing the target’s
sources and markets facilitates this isolation task. The following
section discusses why sanctions end, and ties together the idea of
strategic decision-making, which may either keep sanctions from hap-
pening, make sanctions rhetorical, or reduce the potency of future
sanctions due to the sender’s reputation eroding through policy
impotency. This chapter concludes by connecting these ideas to the
basic models in chapter two, the public choice models in chapter four,
and chapter five’s analysis. The mathematics of game theory will be
held to a minimum, as most ideas can be discussed in a nontechnical
way. This section also discusses measurements of efficacy and inte-
grates the movement along the Sanction Effectiveness Continuum
from chapter one with theory.
As an introduction to the chess match of international diplomacy
using economics, a primer on game theory sets the stage for the liter-
ature and analyses that follow. It is important that the reader does not
look at this primer as a substitute for a full lesson on game theory, or
to be as technical as studies such as Garoupa and Gata (2002). This
first section is meant to simply illustrate how game theory works in the
sanction context.

A Game Theory Primer

Game theory applied to sanctions may explain why diplomatic deci-
sions are made and why the actions of each country continue, espe-
cially when uncertainty reigns constantly over statecraft.1 The game’s
players are assumed to be rational and seek individual benefit maxi-
mization through their actions. Strategic games are defined by the
players, actions, and preferences over possible actions. In the sanc-
tion’s context, players are nations or international coalitions. The
rules are that targets engage in deviant actions, where sender reactions
by initiating sanctions are a constant threat. From there, the sender
reacts, then the target reacts, and so on, where other nations may join
the game as they see fit as either “white knights” in cooperation with
the sender or “black knights” to help the target, at each decision
38 ECONOMIC SANCTIONS

stage. Game theory predicts national behavior by following the net

benefits derived at every decision stage. Each country’s strategic path
is set initially, the choices to continue or cease these actions, assuming
full information.2
Each new stage provides a new set of potential decisions based on
uncertain net benefits. Players initiate and continue sanctions because
they stand to benefit from these decisions more than the decisions’
cost. For example, to initiate sanctions the sender must feel the bene-
fits exceed the costs, given how those benefits and costs are defined.
This holds true for sanction continuation, as each new decision stage
allows both the sender and target to consider acquiescing or not.
Explaining threat credibility based on the anticipated detriment to the
sender or target for initiating or continuing sanctions, the resolve of
the sender and target, and the formation of cooperative strategies
are major additions to this literature from game theory models.
Ultimately, conflict resolution models have been adapted to sanctions
and originate from both political scientists and economists. These
models also act as complementary models to public choice theory
discussed in chapter four.

Basic Concepts
Sanction games explain why targets concede or why senders terminate
sanctions. However, there are many different games and rules. Three
concepts that help to understand game theory applications are: exten-
sion versus normal form; sequential games; and Nash Equilibrium.
The game’s form is simply how the net payoffs from decisions are con-
ceptualized. The normal form is generally a table of net benefits for
each nation, outlining the possible outcomes. A game’s extension
form uses a diagrammatic approach, or a game tree. Table 3.1 illus-
trates a simple sanction game in normal form, where the numbers in
parentheses represent the action’s net payoffs: (sender, target).

Table 3.1 A sanction game in normal form

Target

Continue Acquiesce

Sender No sanctions (1,2) (2,0)

Sanctions (2,1) (3,3)
 SANCTION INITIATION AND CONTINUANCE 39

This normal form describes a game where the sanction ends if the
target capitulates. Most historic cases are sequential games. Sequential
games are those where Player 1 moves, then Player 2 moves, then
Player 1 moves again; in this case, there would be a new game in each
decision period, with a new payoff matrix depending on how the pre-
vious period’s game was resolved. Sanctions can be initially investi-
gated as examples of perfect information, sequential games because
both senders and targets react to each other with knowledge about
what the other player has done, reactions based on changing payoffs.
In table 3.1, the sender reacts to the target by sanctioning or not sanc-
tioning. If the sender imposes an embargo, the target knows the
embargo is on and reacts. The strategic behavior of each nation, if
based on the assumption of individual benefit maximization, leads to
a decision path that optimizes the net payoffs for each party. In con-
trast to the normal form, figure 3.1 shows the extensive form of a
sanction game, where each node on this game tree is a new decision
stage.
The extensive form shows the succession of decisions, which sum
to strategic decision-making. Some games are competitive, where the
payoffs reflect what each player can take from the other players. Some
games are cooperative, where each player shares in the decisions’
benefits. A cooperative outcome may take place because forcing

Sender

No Sanction Sanction

Target

(1, 2) (2, 0) (2, 1) (3, 3)

Continue Acquiesce Continue Acquiesce

Figure 3.1 A sanction game in extensive form.

40 ECONOMIC SANCTIONS

self-interested outcomes is not optimal if each player stands to lose

more by competing rather than cooperating. This illustrates the con-
cept of Nash Equilibrium, differentiating between cooperative versus
competitive games.

Economic Statecraft and Nash Equilibrium

Sanctions have both competitive and cooperative elements, and
national decision-making may depend on prior relationships and
alliances that exist ex ante the target’s malfeasance. Some models,
both theoretical and empirical, suggest that the level of belligerency
in the relationship between target and sender prior to sanctions
may be an explanatory variable in sanction effectiveness.3 There is
also competition and cooperation among the potential senders.
Alliance-building and forming consortia are acts to reduce the uncer-
tainty of rogue nations coming to the target’s assistance. Once an
alliance is struck, the sender consortium may have many possible
strategic paths. Individual members of a consortium can act in many
ways, which makes the game theory aspects of multilateral coercion
extremely complex.
The first game is competition and is simple; table 3.1 and figure
3.1 illustrate such a game. Each side is competing over the target’s
ability to make specific policy choices. The target engages in a specific
policy that the sender believes is imprudent. The sender tries to
change incentives for both the international community and the
target populace by directly reducing target decision makers’ resource
availability to continue their actions. The sender and target are com-
peting in these ways over control of resources.4 A competitive game
is a zero-sum game in its optimum: the gains of one party equal the
losses of the other. Because spillover effects are obvious in sanctions,
however, the pure competition game is unlikely. When looking at
the world split into three pieces, sender, target, and third parties,
sanctions are zero-sum games by definition. However, because the
sender pursues more benefits and nations to support sanctions, forc-
ing the target and its potential allies to bear increasing costs of the
target continuing deviant policies, the competition model is a strong
foundation.
In the competition game, each player’s strategic behavior pursues
individual benefit maximization. When viewing the payoffs, the
assignment of numbers makes for better illustration than reality. How
the payoffs relate to one another concerning the choices of each
nation is more important. For example, in table 3.1, it is not
extremely important that the sender’s sanctions lead to either two or
 SANCTION INITIATION AND CONTINUANCE 41

three payoff units, but that they exceed the payoffs of the sender not
sanctioning. Given its information set, the sender forms her strategy.
The next game is one of commitment. Credible threats collateral-
ize sanctions. The sender, once committed to a sanction strategy, is
assumed to continue until the target reverses its original policy stance.
A lack of commitment makes other sanctions somewhat incredible
and ineffectual. Commitment game choices must be irreversible and
provide future target economies with historic knowledge of the
sender’s intentions and resolve in economic coercion.5 This game is
similar to a cooperation game, where commitment acts as signal to
potential sanction partners that the sanction leader has resolve that is
unshakable. The target’s commitment may also be in question and act
as the crux of the sanction’s strength.6 If commitment is credible,
then the game has a cooperative solution.7
The final game is a cooperative game.8 One way to visualize differ-
ences between cooperative and competitive games is the dichotomy
between oligopoly and perfect competition from microeconomics.
Each nation reacts to the other dictating the game’s play, as each
nation uses the other’s reaction as new information. A classic game,
the prisoner’s dilemma, suggests that because players are unsure about
the other’s reaction, each player makes strategic decisions based on
what they believe the reaction of other players will be and how
net benefits are affected. For example, if two criminals are caught,
one may believe the other will confess to reduce her sentence. It is
assumed that each prisoner has an incentive to confess, given their
knowledge that the other prisoner also has a larger payoff from
confessing versus not confessing. However, if they both chose not to
confess, they would both be better off. Neither prisoner can trust the
other to follow the no confession strategy. Thus, there are two strate-
gies the prisoners could follow for individual maximization, given the
other’s action, but only one social optimum available.
The dilemma solves itself by the prisoners choosing the strategy
that leaves them at a second-best result, even though they are consid-
ered “rational”: they both confess. Moreover, the prisoners choose
the noncooperative strategy, rather than the cooperative one. In this
“prisoner’s dilemma,” there is no dominant strategy. In game theory,
a dominant strategy is a path of choices where any other strategy
leaves the player worse off. In table 3.1 and figure 3.1, the sender’s
dominant strategy is to impose sanctions. The sender has a credible
threat to do so.
This outcome illustrates Nash Equilibrium, a cooperative solution
to the sanction game that is not socially optimal because deadweight
42 ECONOMIC SANCTIONS

losses exist. There are many sources for a definition of Nash

Equilibrium; here is yet another specific to economic statecraft:

Nash Equilibrium: A nation chooses a policy strategy, such that no

other nation, as either an additional sender or target, can increase their
own nation’s welfare with a different action, given that every other
nation’s reaction strategy is formulated in the same way.9

There can be multiple Nash Equilibria in a game. This latter char-

acteristic typifies examples of the prisoner’s dilemma and many sanc-
tion cases. For example, suppose the United States sanctioned China’s
ability to use American equity markets because it was unsure where
the funding was going, a classic financial sanction against international
terrorism. There was a question of this money potentially funding
genocide in an African nation or terrorism in the Middle East because
the equity ultimately would be used for investments in these coun-
tries.10 Assume the sanctions would be positive in net for the United
States, and China would experience a net reduction in benefits. China
could retaliate and reduce American benefits if it so chooses. The
United States, knowing this may happen, chooses a strategic path that
does not maximize its own benefits, but avoids trade losses due to
Chinese retaliation, and a compromise is formed.
The existence of an information asymmetry between China and the
United States about each other’s intents is the essence of Nash
Equilibrium, a mix of uncertainty and cooperative outcomes because
information is not perfect. In the sanction context, Nash Equilibria
are a simple outcome of diplomacy, where each side gives up some-
thing to gain new policy directions. In the following sections, game
theoretical components are explained as extensions of the basics given
earlier, where two overriding principles hold:

1. Game theory suggests that nations may seek cooperative outcomes

because they lack perfect knowledge of other nations’ reactions;
and
2. The political economy of national decisions may not be well-
known at the sanction’s outset.

From a political economy perspective, policy shocks to wealth are

assumed to define payoffs in the sanction game. There are three basic
game structures for most case studies, all based on a sequential game
structure. Each decision stage in a sequential game is itself a subgame.
Each subgame is itself a type of new sanction episode. The sender and
 SANCTION INITIATION AND CONTINUANCE 43

target are choosing their “best-response” strategies, repeating the

decision process as a reaction to the other nation’s move.11

The Repeated Sanction Game

In this game, the target moves first, which is a slight deviation from
the way the sanction literature views game models. The target’s polit-
ical deviation causes potential sender economies to react, where they
may choose to sanction or do nothing. After the sender’s reaction, the
target may change their political actions or not acquiesce to the
sender’s political demands. If the target continues their policy, it leads
to a new game, based on the same rules and structure. The sender has
the choice of stiffening or relaxing sanctions, the target reacts by con-
tinuing or acquiescing, and so on. This game is repeated until either
side ends their respective actions. The sanctions and political deviations
can go on forever.
The repeated game is a game of attrition. The sender repeats the
sanction, year after year, in an attempt to slowly unravel the fabric that
allows the target to continue its actions, hoping the target will say
uncle. Unilateral sanctions of the United States against a dictator have
this characteristic.12 Every year, as new aid packages are debated in the
U.S. Congress, America approaches yet another game node on aid and
arms sanctions, for example. The target knows this, as it knows the
budget process of the United States, and waits to see if the aid restric-
tions will continue. The net payoffs to each nation are functions of
how long the sender and target intend to continue their policies.

The Two-Stage Game

The two-stage game is a repeated game with only two stages. Suppose
the target engages in deviant actions, the sender reacts and the target
may then have to make a choice to continue or not continue because
their decision will end the game. A real-world example is brinksman-
ship. If the sanctions are a precursor to war, and a blockade is enforced
by the sender’s military, the target must either yield or react violently.
In stage one, the target’s actions are sanctioned; in stage two, the
target moves to end the sender’s statecraft one way or the other.
All sanctions can be compressed into a two-stage game; however,
the dynamics of changing payoffs and choice and players are lost in
such a truncation. The credible threat of target losses dictates the pay-
offs, but this threat is not credible in every sanction case. This two-
stage game depends on how the target’s decisions end the game. The
uncertainty over the reactions of each player leads to mixed strategy
games. The military option need not be involved to have a two-stage
44 ECONOMIC SANCTIONS

game. Suppose that the Gambia experiences a governmental over-

throw, and the new government restricts antigovernment speeches,
elections, and freedom of its press. The United Nations may react in
such a way that does not change until the Gambian government
allows more freedoms. The Gambia reacts to this and faces sanctions
until their reaction is an acceptable compromise—stage one is the tar-
get’s action and sender’s reaction. Stage two is the target’s final action
and sender’s final reaction, which ends the game.

The Mixed Strategies Game

The most realistic and difficult game analytically is a mixed strategy
game, where an action creates an array of possible reactions with spe-
cific probabilities. Few target nations know how potential senders will
react, how many of them there will be, and the sanction’s compre-
hensiveness. These conditions may also change after a certain number
of stages depending on the information set at that stage. For target
countries, information should be close to perfect based on history and
current international relationships; for example, any terrorist act is
likely to bring the United States in as the leading sender nation.
Mixed strategy games involve defining the probabilities of actions and
reactions as a function of history, politics, and economics. Few coun-
tries are deterred by policies that are openly rhetorical. Each nation’s
pursuit of specific goals, in terms of payoffs, dictates strategic decision-
making. The sender, target, and third parties define themselves through
their actions at different stages of the sanction game.
Current sanctions against Iran act like a mixed strategy game. Both
Iran and the United States share uncertainty over the other’s actions
and reactions. Each country releases press statements berating the
other’s policy choices, engaging in a new type of cold war. Diplomacy
is a gamble, a speculation based on current knowledge of each player,
where sanctions act as a trump card. New events can shift these prob-
abilities, as Israeli action may trigger a new set of possible reactions
from both Iran and the United States. The more volatile and strong
Iran becomes, the more uncertain their policy decisions. While most
sanctions can be simplified into the games discussed earlier, mixed
strategy games are more realistic as decision models when the players’
moves are highly uncertain.
The underlying concept is that sender policy makers can predict the
target’s reactions, as targets can predict sender actions, because both
sides are pursuing the best response strategy. If both players move
such that they play their best response every move, the game ends in
Nash Equilibrium. The scholarly difficulty is not in defining the players
 SANCTION INITIATION AND CONTINUANCE 45

or rules, but in defining the payoffs. The reader should think about
the way game models are set up, and what small but interesting twists
scholars put on the rules, strategies, and payoffs of the two sides. The
next section describes possible initial reactions to target political
deviance, threats, and blockades.

Threats and Blockades

Initially, all sanctions are threats.13 Whether the threats are credible is
another matter. In an economic coercion game model, sanction effec-
tiveness is typically measured by reaction functions and the sender’s
ability to force the target into new decisions based on manipulating
the target economy. Can the sanction affect the target economy at all
and can the sanction effects transmit to a political outcome deemed
effective in the sender’s eyes?
If the sender cannot credibly threaten the target with sanctions or
military action, the target is unlikely to react “correctly” from the
sender’s viewpoint. One way to collateralize sanction threats or
actions is to engage in a physical, military blockade. If any military
action is used to enforce sanctions, the sender is de facto using a
blockade. However, because blood generally follows blood, physical
blockades spell the end of economic coercion and slip into military
action. Also, sanctions may be seen as pure rhetoric from initiation
forward by targets if economic relationships with the sender economy
are relatively small compared to third parties or the sender has no
credible use of force.
A blockade is the oldest sanction weapon. In a typical case, a large
sender used naval resources to cut sea routes to and from other
economies. To truly restrict trade, senders must also blockade land
and communication connections otherwise. Britain and France cer-
tainly fit this description in the nineteenth century, as the blockades of
the Napoleonic era are testaments to the tenuous nature of a general
blockade. The possibility of military force being used may be a good
thing, especially if threats are themselves sanction tools. “The founda-
tion of modern international law on blockades was laid during the
series of naval wars between 1776 and 1815, and the tendency
of international discussion after 1815 had been to further legal defini-
tion on the more innocuous practices of the earlier wars” (Medlicott
1952, 4). Theoretically, multilateral cooperation is meant to mock
such a physical embargo.
Economic coercion has as its extreme act a military blockade. Some
studies have used the phrase “military coercion” to illustrate the fine
46 ECONOMIC SANCTIONS

line walked by an aggressive sender (see Askari et al. 2003, 14).

Blockades represent sanctions as a parallel act of aggression, and thus
fall under the umbrella of economic warfare. Renwick suggests the use
of economic warfare began in earnest with World War I. “There was
nothing new about siege warfare, a naval blockade, or the attempt to
deprive an adversary of supplies of food and ammunition. The new
element in the industrial age was the effort to interrupt supplies of
essential raw materials” (1981, 70). In 2006, Israel resorted to a naval
blockade of Beirut as a reaction to its short skirmish with a Muslim
extremist group, Hezbollah. This episode is packed with the reasons
why blockades are difficult to assess, from brinksmanship to third-
party effects to the use of military force, as if the blockade was an
economic sanction.14
Blockades are dangerous choices. If military conflict is to be
avoided, one problem the sender contends with is the brinksmanship
issue. Doxey (1971) argued that using military blockades may not
only be dangerous if war is to be avoided, but the historic evidence
provides no insight as to how powerful a force a blockade may be in
changing outcomes. “Economic warfare, even in its most unrestrictive
and inhumane application, is not thought to have been a decisive
factor in either world war” (21).
However, this logic is somewhat counterintuitive. First, it is diffi-
cult to surmise embargo effects from military conflict, especially in
the context of the World Wars. Having naval or air superiority over
an enemy simply provides more hegemony to the superior forces. In
the Third Reich’s eyes, they were the sender; and the Allies thought
the same. Strategic bombing and unrestricted submarine warfare are
also possible ways to view blockades as extensions of economic war-
fare.15 The recent Israeli use of naval and air forces showcases this
decision-making, but this force was ultimately not decisive. Force
was used by the United Kingdom parallel to sanctions after the
Argentine invasion of the Malvinas in 1982, reinforcing the threats
and economic measures. Can threats of military force be made at
all times?
Drezner (2003) examined strategic interactions between senders
and targets and suggested that most sanctions should end before they
begin to eliminate the deadweight losses experienced by both coun-
tries. In fact, Drezner suggested that threats are the strongest sanction
tool when used correctly, and the perceived lack of success in historic
episodes is a result of not giving threats their due credit. “These
[threat] cases are far more likely to generate successful outcomes than
 SANCTION INITIATION AND CONTINUANCE 47

when sanctions are [actually] imposed. Underestimating the utility of

economic coercion calls into serious doubt the argument that eco-
nomic inducements are a more useful tool of statecraft than economic
coercion” (655).
Are threats effective sanction episodes in economic terms? The
answer is yes if the sender believes that the maximum benefit parallels
political capitulation at little or no economic cost. If physical blockades
and military complements to sanctions both bias and cloud sanction
effectiveness for the scholar, it is difficult to include such sanctions in
historic analysis.16 In game theory, the military option should linger
throughout a sanction’s duration as a game tree node because it is a
natural next option depending on the target’s reaction.
North Korean brinksmanship during 2006 shows that certain
countries are not afraid to bring themselves, as current or potential
targets, to the brink of war as an act of diplomacy. Once threats
become actions, the sender moves on the game tree and waits for the
target’s reaction. The target does the same if it, for example, engages
in human right abuses.17 The following two sections discuss how the
literature sees game theory applications and why sanctions begin and
continue in pursuit of elusive payoffs.

Initiating and Continuing Sanctions

Eaton and Engers (1992) studied why sanctions are initiated and con-
tinue by focusing on each country’s “toughness” concerning sanction
costs in an alternating move game, where sequential moves overlap each
other. “The more patient a party and the lower its cost of sanctions, the
tougher [the target] is and the better [the target] does. At the extremes,
if the target is much tougher, then it can attain its most preferred feasible
outcome” (902). Commitment is also a major factor in this model. The
model itself represents a sequence of overlapping commitments on each
side, where this sequence is the basis for assuming a sequential game.18
“Each time it chooses [a move], each party takes the entire sequence of
future choices into account in deciding its best current choice” (905).
While heavy on technique, Eaton and Engers concluded qualitatively that
the relatively tougher the target, the less effective sanctions are, where
toughness is a function of a sanction’s costs to each party.
This conclusion is obvious and logical—analyzing payoffs under dif-
ferent commitment levels provides policy makers with ex ante informa-
tion about whether to impose sanction or not. Eaton and Engers (1992)
suggest that the target acts in a reversible or nonreversible way that itself
48 ECONOMIC SANCTIONS

shows commitment. Further, their modeling of decision timing makes a

larger case for sanctions as sequential games, as each party learns the
other’s choice with delay and then reacts with delay, and so on (918).
The “preferred feasible outcome” for the target is assumed to be its
deviant action, and the sender wants to end that action. Sanctions
only continue if a subgame equilibrium can be reached by their con-
tinuance, where each subgame is defined once one nation learns of the
other’s actions.19 Subgames are decision nodes where the information
is completely known. For example, once Iraq had invaded Kuwait in
1990, it showed the world a level of commitment to occupying
Kuwait. As the United Nations debated its reaction, it did so in the
form of a proper subgame. A new subgame would be Iraq’s reaction.
The UN reaction likely tested the commitment level of Iraq to remain
in occupation; Iraq’s reaction tested the UN commitment to liberate
Kuwait. In the end, of course, military action was taken instead of
continued comprehensive sanctions by the United Nations, which did
continue once Iraq was forced from Kuwait.20
Can sanctions be perpetual and still have Nash Equilibria? Looking
at perpetual cases such as Cuba and North Korea, the lack of a Nash
Equilibrium may explain their lack of information about each other’s
toughness level. However, the guaranteed finite life of each country’s
current leader suggests that the game’s finiteness depends largely on
the sanctions’ stated goals: the demise of the target’s leader. Eaton
and Engers (1992) assume strategies are renegotiation-proof: once
the target knows their action’s continuance is dominated by the strat-
egy of ceasing the action, the sanction ends. The same holds for the
sender in continuing sanctions. The challenge is in knowing which
side will acquiesce, if either side does. Hence, commitment can lead to
perpetual sanctions.
Do countries wait for each other to react or are they proactive and
move simultaneously? Theoretically, sender economies should only
initiate policy once targets act and continue only if there are positive
net benefits given the target’s anticipated reaction. The target contin-
ues its deviant action under conditions that the action is part of a strat-
egy leading to Nash Equilibrium. The simultaneous move seems more
likely as a result of finding new information and acting immediately.
For example, Malawi experiences a bloodless coup, which then installs
a military government. The United States is likely to cut military and
maybe development aid until it can assess the overall effects of the new
Malawi government on human rights, southern African stability, and
so on. However, if new information comes to light, perhaps this new
government has ties to terrorist organizations, for example, the U.S.
 SANCTION INITIATION AND CONTINUANCE 49

sanctions may stiffen without any new actions by Malawi. If toughness

is a function of the net benefits derived from commitment, then a
sender nation being proactive showcases its resolve to the world.
Eaton and Engers (1999), a refinement of Eaton and Engers (1992),
focuses instead on the target’s tenacity as a function of the sender’s
resolve. The players are assumed to engage in an infinitely repeated
game. The sender controls sanction costs, where the target and sender
suffer costs as sanctions are imposed. The probability that the target
will be stubborn and not concede is equal to  . Since sanctions are
perpetual, the sender assumes its threshold to continue sanctions is
the discount factor’s weighted average, the payment that defines the
sender payoffs from pursuing sanctions in net. The target is assumed
to know the sender’s goal; the target’s punishment costs must exceed
its payoffs from being stubborn. The target also faces a probability
that the sender is engaging in rhetoric, equal to . This probability is
a function of the sender’s past actions, and decreases toward 1⁄2 as the
sanction continues.
The game’s outcome is decided by the interplay between the target’s
experience with the sender and the sender’s ability to keep up the pres-
sure year after year. Each player lacks specific knowledge of the other’s
payoffs and actions, but learns with each new period. The sender’s
resolve increases the probability that the target will experience costs.
The model also suggests that the sender can, under these conditions,
suppress the target by strong action. “Commitment eliminates the
complaisant target’s incentive to balk simply to raise the chances that
the sanctions will be lifted subsequently” (Eaton and Engers 1999:
413). The bond between the lead sender and other potential coalition
members may depend on the spillover effects caused by statecraft.
How do we model spillover effects, or treat sanctions as public
goods? Multilateral commitment is a major asset for organizational
senders such as the United Nations or European Union. The potency
of universal sanctions, if such coverage is attainable, has been debated
since Galtung (1967). One way sanctions may be modeled is a sender
cartel, acting like a trade barrier monopoly or monopsony. This coop-
erative strategy is parallel to Lisa Martin’s model (1992) on how
nations choose to cooperate with one another.

Cooperation among Nations:

Of Collusion and Cartels
Pape (1997), Drezner (2003), and Drury (2005) suggest that the
literature is still somewhat uncertain concerning multinational
50 ECONOMIC SANCTIONS

cooperative efforts and their efficacy. On the surface, the logic behind
multilateral sanction success seems obvious. If a monopolistic group
of trading partners suddenly eliminated their ties with the target econ-
omy, would this not naturally lead to major economic pressure on tar-
gets to cease deviant policies? Is this not what the United Nations
meant in Articles 41 and 42 in Chapter 7 of its charter? One way eco-
nomic theory may help understand why cooperative arrangements,
such as multilateral sanctions, do not work is cartel theory. A cartel
forms a monopoly where price is held at a specific, profit-optimizing
level, through restricting quantities in a market.21 What if a multilateral
sanction was modeled like a cartel in game theory?

Sanction Cartels
Cartels are organizations that explicitly collude price and quantity
among member firms, insulating the group from outside competition.
The agreement between the member firms is simple: do not compete
with other cartel members on price or quantity. This ultimate style of
cooperation leads to the Cournot–Nash Equilibrium in game theory.
The Cournot game focuses on two firms, a duopoly, controlling quan-
tity supplied in a market to reduce competition and increase each firm’s
returns by acting as a monopoly. The Cournot–Nash Equilibrium
results from the firms choosing to cooperate.22 The Stackelberg game
is very similar, assuming perfect information through a sequential
game, rather than the Cournot simultaneous game. The Stackelberg
game assumes perfect information and the existence of a price leader
that controls and dictates market conditions onto other firms. In the
sanctions example, the firms are simply nations, where the United
States has become the dominant “firm” in the sanction market.23
Countries involved in sanctions “compete” over the target’s
markets. A sanction cartel must deter countries from rescuing the
sanctioned nation and competing, showing resolve along with other
economies that certain acts will not be tolerated. If a strong economy,
say the United States, were to embargo Iran, its abandonment of
Iran’s markets would begin a natural competition over supply left
unfulfilled. The United Nations was built, in certain ways, to be such
a policy cartel, where international diplomacy and peacekeeping
flowed from a “market” controlled by UN member nations.24 Senders
colluding on sanction policy should realize the same basic benefits.
Few substitute markets exist when the cartel is made up of a market’s
largest players.
A cartel’s characteristics are simple and are similar to those for any
monopolist. First, the good must be homogenous, and not subject to
 SANCTION INITIATION AND CONTINUANCE 51

reduced demand based on product differentiation. If the target’s

imports, exports, and financial flows are bundles made up of similar
goods, regardless of trading partner, the homogeneity characteristic
holds. Second, the cartel’s demand must be inelastic. A price reduc-
tion by a cartel member, in an attempt to steal market share, provides
less total revenue when demand is inelastic. International cartels exist
primarily for this reason. In export sanctions, forming a sender cartel
works best if the target is dependent on trade and financing from poten-
tial senders and if the target’s demand for imports is inelastic.25 In 1974,
Turkey invaded Cyprus. The Unites States cut off military aid from
Turkey, increasing the “price” of continued action. The sanction cartel,
the United Nations, broke down very quickly as European countries
were quick to sell Turkey weapons, thus lowering the price of Turkey’s
policy. To Turkey, military hardware was interchangeable with elastic
demand. With willing sanction busters, Turkey circumvented the U.S.
arms embargo.
Third, there must be little competition elsewhere. New markets
erode a cartel’s power very quickly. This erosion is the source of much
frustration in sanction episodes. Profits soar for countries that sell lost
resources to targets due to sanctions. Arms sales are classic examples.
In 2006, Russia brokered a deal with Venezuela to sell them arms.
Venezuela recently experienced civil strife and purchased weapons
from Russia as an act of international defiance in response to other
nations using arms embargoes (The Economist, September 2, 2006).
Russia was simply taking advantage of a situation where arms embargo
left an oil-rich country with a resource need.
In international diplomacy, the benefits of cheating on sanction
policy can be reduced in net by the remaining sender nations penaliz-
ing the cheater. The United States put into law such a measure with
the 1996 Iran–Libya Sanctions Act (ILSA).26 The difficulty, however,
lies in knowing which coalition partners are more likely to cheat or if
new countries will appear as black knights. Suppose the United States
sanctions Iran and asks other countries to join the sanction because
Iran is a nuclear threat; the United States does so knowing the incen-
tives are high for countries to remain trading partners of Iran. Because
saying no to the United States may carry with it a cost under the
ILSA, certain countries join. On the margin, these additional senders
have reaction functions that suggest the cost of becoming a black
knight and trading with Iran, or not acting at all, are higher than the
costs of joining the coalition. Others do not.
Policy cartel leaders may provide incentives not to cheat, punishing
those that do cheat. “While the leading sender attempts to organize
52 ECONOMIC SANCTIONS

sanctions, other states often appear willing to free ride on its efforts
and need extensive persuasion before they will agree to cooperate”
(Martin 1993, 408). This extensive persuasion may be in the form of
most favored nation status, lower trade barriers otherwise, or financial
aid.27 Nations choose to cooperate or compete on sanctions much like
firms choose the same acts on price and quantity in cartelized markets.
The benefits of cooperation must exceed the costs of noncooperative
choices, but these choices are made in an uncertain environment. It is
important to understand the possible ways multiple senders may or
may not interact, and Martin (1992) began to use politics and game
theory to explain these agreements.

Multilateral Sanctions
The game theory models of Eaton and Engers (1992, 1999) and
Drezner (1999) are about sender and target interactions. Martin
(1992) began the latest discussion of how senders engage with other
senders in economic statecraft, following the tradition of Baldwin
(1985) and Doxey (1987). The gains from acting as a monopoly pol-
icy maker, especially when using instruments of economic coercion,
are too large to ignore; the profits to be made from cheating on this
policy cartel, especially if the punishment is weak and somewhat
incredible, may also be too large to bypass. Such costs are referred to
as “audience” costs. These costs face senders who have partners not
willing to commit to the sanction completely. Multilateral sanctions
must encompass a large percentage of the target’s trade and financing
to be successful (Galtung 1967). However, we know from history
that economic and political hegemony does not guarantee sanction
success.
Martin (1992) focused on the effects of multinational cooperation
in sanction episodes, concluding that cooperative efforts are not guar-
antees of sanction success and third-party leakages are not necessarily
the reason why sanctions fail.28 In sanction games, a strong leader
generally exists. This strong leader may act alone or lead a multilateral
action. A strong leader may only influence others to cooperate in cases
where there is either a dominant strategy for others to join in sanc-
tions or a credible threat from the strong leader to impose costs on
other countries in cases where they do not act parallel. Do the bene-
fits of joining the sanction outweigh the costs? Is joining a sanction
coalition a Nash equilibrium decision? These are difficult questions to
answer. The three examples of Martin (1992) tie the ideas of a cartel
and basic game theory well.
 SANCTION INITIATION AND CONTINUANCE 53

The first game is coincidence, where two or more countries act simul-
taneously to affect a target nation. In this game, two nations choose to
impose sanctions in equilibrium (Martin 1992, 25). Countries cooper-
ate toward a common goal (coincidental), the restoration or keeping of
peace, supporting and enforcing human rights, and reactions to other
violations. The United States and United Kingdom, outside of UN
sanctions, have acted in this way toward Afghanistan and Iraq in 2001
and 2003, respectively. The coincidence game takes place when multi-
ple senders move as a unit. Martin (1992) makes the case that whether
there is a strong leader or not in the coincidence game, the other coun-
tries must have sanctions as dominant strategies to join. In the presence
of a weak leader, countries may still join a sanction coalition due to
coincidental goals.
Can a strong leader impose its will on others in the international
community and is this a good strategy? Because of the uncertainty in
the coincidence game mentioned earlier, senders may try and impose
their will on other countries to join the coalition. This game is called
coercion, as one sender emerges as a leader trying to force other coun-
tries to join. Martin (1993) suggests that in the coercion case, a cred-
ible threat must exist to provide other countries with incentives to act
multilaterally with commitment. The leading sender’s strength is a
function of threat credibility and commitment. The strong leader uses
both positive and negative economic incentives toward other senders
dictating their audience and, ultimately, coercion costs.
The final game is a coadjustment game. In this situation, no sender
country has a dominant strategy involving sanctions; the dominant
strategy is then to not impose them. Suppose a country like the
United States wants to sanction a large trading partner, say China, and
wants Japan to join in that sanction. Japan may have no incentive to
sanction China, nor feel the United States has a credible threat against
them; thus sanctions are not a dominant strategy for Japan. As a
result, the United States may not initiate sanctions as a dominant
strategy, and sanctions are imposed as either rhetoric or as short-term
measures.

Unilateral sanctions may force the target to bear some transition costs
as it finds new trading partners; it may have to pay higher prices for
imports or accept lower prices for exports. Unless [sender] states
achieve a significant level of international cooperation, however, market
forces tend to make these effects transient and small in size. (Martin
1993, 408)
54 ECONOMIC SANCTIONS

Without a credible threat, sanction imposition may hurt the

sender’s commitment in future sanctions, as new target nations may
view the sender as inherently weak but reactive to international polit-
ical trauma. This lack of credibility may also reduce incentives for
other nations to join the leading sender in these actions, reducing the
possibility of forming a sanction cartel and an economic blockade
around the target. The three games mentioned earlier show there are
many possibilities in cooperative sanctions. Even if a dominant strat-
egy exists to use economic statecraft, the benefits may be difficult for
a coalition member to derive.
Logic dictates that cartelizing the sanction market, where the cartel
is made up of the target’s large trading and financial partners, should
be relatively more effective versus unilateral measures. These cooper-
ative efforts may not work, however, because most sanctions are either
coercive efforts or coadjustment situations (Martin 1992). Cuba is a
classic case of coadjustment sanctions, as the European Union had lit-
tle sanction activity against Cuba, while the United States continues
sanctions on Cuba virtually alone. The United States seemingly
intends to continue sanctions nominally until Fidel Castro dies. No
longer are demands made on other nations to join in, as the threat of
retaliation for not joining is now incredible. Current sanctions against
North Korea and Iran have similar characteristics, where the United
States is more focused on black knights.29 The logic of multilateral
and universal sanctions was envisioned in Galtung (1967) and Doxey
(1971) when UN sanctions were in their infancy. The future of eco-
nomic sanctions may be in jeopardy if the United Nations cannot
sanction effectively.
Cartel theory provides a basic economic framework to understand
what a cooperative sanction tries to achieve. The sanction cartel seeks
to provide its members with enough benefits to keep them from cheat-
ing (increasing trade with the target or leaving the coalition) or con-
structs negative incentives to punish a coalition member that strays
from current policy. The economy that initiated the sanctions must
possess the credible threat to impose these costs, or possess interna-
tional market power that provides disincentives to countries cheating
or acting as substitute markets. Of course, all this harkens back to basic
sanction theory discussed in chapter two. The more restrictive the
sanctions, the more possible economic damage.30
It is important to remember that damage and success are not
synonymous. Coincident, coercive, or coadjustment games provide
frameworks to realize why cooperative sanctions work or not; these
games also provide insight as to why sanctions continue or end.
 SANCTION INITIATION AND CONTINUANCE 55

To Continue or End Sanctions:

Strategy at Work
Game theory suggests that a sanction will end because to continue
sanctions is not a Nash Equilibrium strategy within each subgame.
Sanctions end because the target succumbs to the embargoes or the
sender terminates the measures due to her lack of willingness to con-
tinue. History teaches us that most sanctions end because the target’s
political conditions change enough for the sender to accept the new
state of affairs, or another policy option (diplomatic or military, e.g.)
is taken by the sender. Credible threats must also be addressed when
discussing sanction continuation and termination, as sender credibil-
ity will fall if sanctions end without some goal achievement. Perpetual
sanctions bring into question how long a sanction should last, regardless
of a sender saving face.
Do sanction effects erode over time? If they do, then the decision
to continue sanctions should face a finite timeline. Two examples of
long-term sanctions are North Korea and Cuba.31 Common ground
exists between these cases in that the sanctions are comprehensive,
have no end in sight, and have a common goal: the death or ousting
of the current dictator.32 How is this possible through economic coer-
cion alone? Policy makers must forecast what the target economy
views as payoffs and try to perceive the target’s reactions.33 This is
also true in monitoring third-party nations, which act as impartial
bystanders, sender coalition members, or as sanction busters.
For multilateral sanctions, some senders may experience this before
the dominant sender, thus eroding the sanction’s power while the
leading sender remains stoic. If the net benefits derived by the sender
are less than the maximum achievable, the sender has an economic
incentive to reduce the sanctions. Some senders continue to embargo
over long periods of time, where it is obvious sanctions are either not
achieving the statecraft’s overt objectives or have become unnecessar-
ily punitive to the target populace. The sender’s reaction function has
three general variables.

Political benefits derived by enforcing international law, rules, or

treaties. A sender economy may also play the role of world diplomatic
leader, where political deviance leads to automatic policy reactions.
Because international laws and agreements are difficult to enforce
without military intervention, benefits may be derived, both now and
in the future, by a sender that acts as in international law enforcer. For
example, the United States may initiate sanctions, reacting immediately
56 ECONOMIC SANCTIONS

to its own interpretation of international law contraventions. This

reaction shows commitment to using economic statecraft as a diplo-
matic tool rather than defaulting to the military option; such is
the vision of Article 41 of the UN Charter. The sender’s allies and
enemies know the sender is willing to act consistently and coalition
membership may rise and facilitate even more cooperation.
Economic benefits from increased trade and financial flows after
political change. The sender’s goal with sanctions may be political
turnover, as well as an implicit economic goal. A sender that influ-
ences political change may open up new markets or at least increase its
exposure to the target’s markets versus the ex ante situation. Many
sanction studies concentrate on trade flows as a variable driving and
describing sanction effects on both nations.34 If the sender perceives
that political deviance leads to reduced trade, the sender must forecast
that a policy reversal increases trade flows with the target.
Economic benefits from target’s political certainty. In general, politi-
cally stable nations have stable and developed economies over time.35
Larger, industrialized nations represent this characteristic, though we
must be careful in generalizing. Suggesting that an embargo aimed at
reducing political deviance necessarily stabilizes the target’s politics is
imprudent; the Haitian cases of 1987 and 1991 show that the best
intentions may lead to terrible results in economic statecraft.36 Does
the sender face costs of ending sanctions before goal achievement?
The decision to terminate sanctions, from a game theory standpoint,
is when the dominant strategy to do so exists. If the sender ends sanc-
tions, there are certainly costs of credibility in future sender statecraft.
Retaining credibility with other senders and future targets shows
international resolve in the face of target toughness. This brings our
analysis full circle, back to Eaton and Engers (1992, 1999). Such is
the repeated game.

Conclusions
Game theory provides insight as to why sanctions begin, continue,
and end. Models are based on the target country’s reaction to
the sender’s threat or actual imposition of sanctions. Each coun-
try involved seeks a Nash Equilibrium, an outcome that is a
specific nation’s best strategy given the other country’s moves. In
general, Nash Equilibria are not Pareto optima, which help further
define sanctions as second-best policy outcomes. There may be
multiple senders and targets, acting collectively or as independent,
parallel players in sequential games based on reaction functions
 SANCTION INITIATION AND CONTINUANCE 57

dictating how each party continues or ends policy once the other
nation has moved.
Eaton and Engers (1992) suggested that toughness was a key
factor in sanction success. Eaton and Engers (1999) suggested that
threats by themselves could and should be powerful enough to
change policy, and sanctions that last long periods of time are simply
wasting time. Bonetti (1994) argues the opposite, suggesting that
sanctions do not last long enough in many cases, or sanctions should
be imposed rather than just used as threats.37 Drezner (2003) consid-
ers threats the best sanction tool available, whereas embargoes are
simple acts of coercion, and feels that there is a serious problem in the
Hufbauer, Schott, and Elliott (1990) data because it does not include
many cases where threats alone were used and worked.
Using price theory as a foundation, sanctions may be most effective
when an entity like a policy cartel is formed. This cartel dictates the
trade and financing terms worldwide to the target economy, and does
not deviate from its policy stance. The leading sender may be domi-
nant and possess the ability to dictate policy to other senders and ulti-
mately to the target. Unfortunately, as quantities are restricted, the
profits derived from either acting as a cheater from within the cartel
rises rapidly. While theory suggests that monopoly power over the tar-
get should lead to large policy effectiveness, few countries have this
position internationally. Certain authors, especially Martin (1993)
and Drury (1998), argue that sanctions have little to no effect on tar-
gets because senders lack the trade linkages, treaties, and economies of
scale to control price and quantity; cooperative efforts do not guarantee
the elimination of these deficiencies.
While sanctions are played as sequential games, the type of sequen-
tial game depends on how senders interact. Martin (1992) proposed
three types of games to explain historical cases. Coincident games take
place when two or more sending nations have similar goals and are will-
ing to work together to change the target’s politics. Coercive games
involve a strong leader, where a policy cartel is enforced by the leader’s
credible ability to economically threaten nations that do not join the
cartel, cheat on the cartel’s policy focus, or openly act as black knights.
This threat credibility typifies this game. Finally, coadjustment games
take place when senders do not have sanction imposition or continua-
tion as a dominant strategy. The coadjustment game is triggered by the
lack of a credible threat to the target or potential coalition members
because of a weak leader. Drezner (2003) sees threats and pre-sanction
actions as some of the most powerful statecraft tools or economic
coercion, many historical cases of which have been lost in major studies.
58 ECONOMIC SANCTIONS

Game theory models of sanction effectiveness begin to move us

from economic effectiveness to humanitarian and political effective-
ness along the Sanction Effectiveness Continuum. The game theory
models of Eaton and Engers (1992) and Martin (1992) described
how senders force reactions from target countries; if sanctions begin
to damage the innocent populace, how do senders react to those data?
Humanitarian concerns in the reaction function of senders turn
embargoes into smart sanctions. In chapter four, public choice models
and smart sanction analyses provide connections between basic sanc-
tion effects, the strategic decision-making of both senders and targets,
and the reactions of all parties as sanctions endure.
 Chapter 4

Public Choice Theory and

Smart Sanctions

The interest-group state, unlike its contractual or public interest counterpart,

is an agent of a particular group or class . . . The state designs the system of
property rights in such a way as to achieve a redistribution of wealth away
from those groups with little or no influence and toward the ruling
groups . . . Dissatisfied subjects (even if disenfranchised) can signal their
unhappiness with the package of policies “produced” by the state through
engaging acts of political resistance that raise the costs of enforcing and
administering these policies.
—Kaempfer and Lowenberg 1986, 382

On the face of it, to “smarten” sanctions may be thought somewhat self-

defeating. After all, if the purpose of sanctions is to coerce the target state,
lessening the sharpness of that coercion may be thought to lessen the chances
of their success. It might also be argued that targeting sanctions may
actually increase the human cost by encouraging their deployment in cases
where [the sanction’s] utility is only marginal.
—Craven 2002, 46–47

The Interest-Group State

Sanctions are seen throughout the world as blunt diplomatic
instruments, potentially causing collateral damage to an innocent
populace as a penalty for its government’s actions. This belief is
exactly correct. Economic statecraft attempts to influence current or
potential disenfranchisement, providing incentives for the target’s
people to demand political change. Sanctions simultaneously attempt
to influence interest groups within the target economy to rise up,
either through a political process or more violently, against the tar-
get’s policies or rulers. How much are common citizens hurt by these
60 ECONOMIC SANCTIONS

policies? Is the sender’s populace harmed such that those groups that
influence sender decisions pressure for sanction’s end? Do the payoffs
to sender interest groups exceed the cost to humanity?
Beginning with Kaempfer and Lowenberg (1986), a literature
strand in the economic statecraft canon focuses on public choice the-
ory, arguing that sender interest groups influence decisions to initiate,
continue, and end sanctions, as if a “market” for sanctions existed.
This market for sanctions may also exist in the target economy, as
the target’s rulers may “demand” policies that result in the sender
“supplying” sanctions. Interest groups in each country act as the pri-
mary force behind economic statecraft, influencing policy makers to
act. These studies focus on the sender’s decisions, and are extensions
of game theory. This chapter holds a review of these analyses, as well
as their possibilities and pitfalls.
The Sanction Effectiveness Continuum is now expanded to include
humanitarian effectiveness. Sanction episodes have been seen as
generally unsuccessful in goal achievement, and continue to be ques-
tioned as policy choices because their success or failure remains greatly
unresolved.1 The human damage caused by sanctions has recently
been the focus of studies by O’Sullivan (2003) and edited volumes by
Cortright and Lopez (1995, 2002). These studies and others are in
part a reaction to a perceived lack of efficacy, examining certain cases
and suggesting that there must be a better way to engage in economic
statecraft. O’Sullivan (2003) concentrates on embargoes to constrain
terrorist activities, asking whether sanctions and targets are chosen
correctly to achieve this goal; her expansive case studies on Iraq, Iran,
and Libya are among the literature’s best. The key insight is that the
sanction, target, and goal choices must all be in alignment with reality
and the sender’s abilities. If the sanction’s goal is to minimize collat-
eral damage and damage the target’s ruling class at its budget source,
the sanction can be considered “smart” by the current definitions.
Cortright and Lopez (2002) provide an edited work that encom-
passes many different perspectives on smart sanctions, following
Cortright and Lopez (1995). A smart sanction is a focal sanction
method that ranges from arms embargoes to restricting financial aid
packages to travel sanctions, and so on, all in an attempt to minimize
collateral damage. Smart sanctions focus on specific entities, such as
the target’s ruling elite or its military machine. For example, freezing
the assets of specific individuals deemed to be either target decisions
makers, terrorists, or their funding sources, or other violators deemed
sanction-worthy. The smart sanction idea suggests that by the correct
 PUBLIC CHOICE THEORY 61

sanction mix, collateral damage is minimized because the target

commoners’ assets, income, and life are left mainly untouched by the
sender’s policies.
It is naïve to believe the target’s rulers will not, in the least,
impose a pass-through cost to make up losses if possible. These costs
could be higher taxes, lower wages in nationalized industries, or
fewer services provided by a more constrained government. It is also
possible that the target’s rulers will use such focal harm as propa-
ganda, blaming the sender for the pass-through costs. There is no
argument that if sanctions can focus on target rulers, the rulers
choose to impose a cost on their own people. The two arguments
are whether the sanctions can focus and to what extent will target
rulers use sanctions as fuel for anti-sender slogans. In this way, there
is an intrinsic connection between smart sanctions and public choice
theory.
However, embargoes of trade and financing are difficult to focus
on specific citizens or entities because that simply is not their intent. If
smart sanctions are synonymous with focused embargoes, analysts are
recommending an oxymoron in some ways. A sanction’s generality
flows from its fundamental property as macroeconomic policy.
The collateral damage can be seen as the opportunity cost the target
government is willing to impose on its citizens. In the 1985 UN sanc-
tions against South Africa, this cost was referred to as the apartheid
tax (Kaempfer and Lowenberg 1986). An embargo’s intent is to
undermine the economic engine that fuels the target’s political
actions. Smart sanctions are best examined case by case, as the case-
specific parameters are unique; the same target country, decades after
a previous sanction episode, may greatly change. As discussed later, a
sanction’s ability to truly focus is still debated, and is likely not to be
easily resolved.2
When constructed correctly, smart sanctions are likely to be pre-
ferred to generalized embargoes. If a market for sanctions exists, and
smart sanctions can center on interest groups, these groups or the
target’s citizens actually could change the government’s decisions
concerning statecraft through a political process. Public choice theory
is a way of explaining the sanction process in all countries involved,
and should help explain why smart sanctions would be sanctions of
choice. Sanctions ultimately must be initiated and continued because
their perceived benefits exceed their perceived costs. Certainly, there
can be a public outcry in the sender economy if sanctions are taking
too large a human toll on the target economy or at home, and policy
62 ECONOMIC SANCTIONS

makers are not reacting. The Iraqi cases of both the 1990s and early
this decade produced some criticism of UN policy.

Such [human] suffering [in Iraq] has become acceptable to the UN

Security Council as “unavoidable” in the broader interest of global
peace and security. Why is it that the sanctions against Iraq are not
imposed in such a way that civilian suffering is limited to tolerable
economic consequences? Instead they are extended to all aspects of life
in violation of existing international legal instruments of protection.
(Sponeck 2002, 83)

This chapter is broken into five sections. First is a discussion of

public choice theory and sanctions. This includes seminal works in
public choice, but mostly applications to economic statecraft. The
next section looks into the theory of interest groups driving policy,
from both the sender and target perspectives. The bridge between
public choice and smart sanctions discusses why interest groups would
demand economic statecraft tools that are more focused than other,
generalized sanctions. The third section is an overview of smart sanc-
tions, as compared to classic embargoes and their strengths and weak-
nesses. Next is a discussion as to the future of sanctions, given the
arguments of smart sanction proponents, international law, and the
United Nations’ increasing use of economic policy measures to
enforce its charter. A conclusion follows and summarizes these ideas,
reminding the reader that these policies are macroeconomic as a
harbinger of the model in chapter five.

Markets for Sanctions: Public Choice Theory

In economics, markets are generally described by their participants:
buyers and sellers. These participants, if allowed to do so by the
governing laws and regulations (if they are enforceable or exist),
negotiate freely over price and quantity until, as if by an invisible
hand, the two sides are drawn toward equilibrium. Policy markets also
exist, where the market “good” is policy. The policy’s magnitude is
the quantity “purchased” by citizens and “sold” by the government.
The price in such a policy market can be stated as the opportunity cost
of pursuing certain policies versus other choices. This holds true in
both the sender and target nations. Doxey recognized the importance
of sanction effects on both the target and sender home fronts. “The
imposition of economic sanctions will affect commerce and industry
at home at the same time it affects the economy of the target. In
democratic countries where governments must retain the support of
 PUBLIC CHOICE THEORY 63

the electorate, it will be necessary to explain and justify the imposition

of such burdens or risk being defeated at the polls” (1971, 107). Both
the sender and target interest groups are affected, though the sender
is likely to have a relatively smaller effect.
The policy market basics are as follows. Economic statecraft is sup-
plied by the sender government at a level greater than zero because
interest groups demand sanctions take place, bidding up what they are
willing to pay. If the government chooses an action where the cost,
price P, is too high, there will be little demand, thus reducing the
equilibrium sanction quantity toward zero. The intuition here is to see
the market “price” as the cost of engaging the government to act. The
market “quantity” is the level of action the respective governments are
willing to take. As a corollary, if the sender’s citizens demand that
sanctions take place, the sender government’s cost of not supplying
these measures rises, delivering sanction imposition. Figure 4.1 shows
the sender’s market for sanctions.
Target economies face similar policy markets with one small twist.
Targets indirectly demand sanctions when they engage in actions that
lead to sender reactions to supply embargoes. In many ways, one can
think of the sender and target policy markets as interlinked, much like
exports in one country are linked to imports in the other. As the target
nation supplies policies that it knows will draw economic statecraft in
reaction, it engages in that policy with an expected cost. That
imported policy cost is in addition to the domestically produced
policy cost of pursuing the deviant action. The target’s citizens must
demand the policy deviance, able and willing to pay the opportunity
cost (demand), in the face of international economic reactions.

S Sanctions

P*Sender

D Sanctions
Q* Q Sanctions

Figure 4.1 The sender’s policy market.

64 ECONOMIC SANCTIONS

Sanctions continue, once initiated, if both countries’ policy markets

clear at a nonzero quantity of actions. Figure 4.2 illustrates the
target’s policy market, analogous to the sender’s market in figure 4.1.
In chapter three, game theory models described why sanction
begin, endure, and end. To continue “consuming” actions that lead
to sanctions means the perceived benefit derived by target interest
groups must exceed the action’s perceived cost. This is a classic out-
come of basic economics. When producing and selling a good at the
market price, the firm is assumed to be profit (net benefits) maximiz-
ing by that choice. The target, by demanding the policy deviance,
maximizes utility (net benefits) paying for a quantity of international
reactions. The sender continues sanctions because the benefits from
doing so exceed the costs, just like the target. How large the sanctions
are is determined by the equilibrium outcome in the “policy” markets.
The payoff functions for each nation in sanction theory using public
choice must include measures for both the benefits and costs of
pursuing sanctions.
Public choice theory is also tied to the international trade litera-
ture. In textbook international economics, tariff and nontariff
barriers to trade help specific subgroups of the domestic economy,
but lead overall to a negative net change in societal benefits from
trade. Politically, these subgroups lobby for the trade barrier if they
gain by the barrier being imposed; producer groups may lobby
the domestic government for import taxes, knowing that their
actions reduce consumer’s welfare by raising price. The government
is, of course, willing to pursue the tariff as it stands to gain tax
revenue. Though sanctions are not tariffs, this rent-seeking activity
by domestic producers is analogous to special interest groups

SSanctions

P*Target

DPolicy
Q* QPolicy, QSanctions

Figure 4.2 The target’s policy market.

 PUBLIC CHOICE THEORY 65

in the sender economy lobbying for sanctions. Each side is discussed

here in the public choice context.

The Sender Economy and Interest Groups

Public choice theory begins with a simple idea of how policy decisions
are made. National policy decisions are driven by interest group pres-
sure on politicians. These interest groups act like citizens, pursuing
maximum net benefits from policy. In contrast to the indirect rela-
tionship experienced by the random citizen, interest groups may gain
or lose from policy directly. In sanction analysis, no one individual has
the ability to influence policy. Interest groups determine why sanc-
tions begin, continue, and end; these groups are where the wealth
gained from policy directly flows.3
Kaempfer and Lowenberg (1988) built a model where utility is
maximized with respect to sanctions by discussing sender agents with
specific interests in applying sanctions.4 Utility is assumed to be max-
imized by agents of both economies, where it is based on an agent’s
income. Income in the sender economy is a function of the sanction’s
magnitude imposed against the target economy, and also provides the
sender’s ability to fund the “purchase” of policy. Interest groups
perceive that income is generated (or income losses are avoided) by
pressuring the sender’s government to react to target actions, at a net
gain given the policy influence payments.
The sender’s interest groups pressure their government to send
sanctions to the target, and collect the net benefits from that action.
Much like other sanction models, how policy costs are measured is a
major source of uncertainty and makes empirical studies of these
models problematic.5 The citizenry of each country can influence the
respective governmental actions. Kaempfer and Lowenberg (1992)
built the current foundation of public choice theory as applied to eco-
nomic statecraft, where agents earning marginal income greater than
zero win the aggregate political battle for sanction imposition. These
agents push the sender to continue the embargo because of income
gains from these policies, and cease the pressure if net gains from the
embargo disappear.6 Other authors suggest that interest groups pay
for sanctions because these groups, in a similar way to trade barriers,
stand to make the most from the policy’s wealth gains.

The public choice interpretation of the sanctioning process implies that

the incentive to devote resources to lobbying against trade sanctions
will be greater the larger is the trade linkage.7 This in turn means the
66 ECONOMIC SANCTIONS

larger the trade linkage the greater are the domestic political costs of
imposing trade sanctions. Consequently, the public choice theory
implies an inverse relationship between the pre-sanction trade linkage
and the probability of imposing trade sanctions. (Bonetti 1997a, 729)

At too low a price, there is a shortage of sanction policies. Interest

groups would like to take full advantage of cheap policy initiatives and
fund the sender government, which is currently unwilling to initiate
policy due to its low value. Sanctions against China concerning nuclear
proliferation in the 1990s may constitute such sanctions, as the cost of
imposing sanctions on nuclear material was low for the government
but the American public outcry was high. In contrast, at too high a
price, there is a surplus of sanctions. The government supplies too
large a magnitude of sanctions for what domestic groups demand.
U.S. policy concerning Cuban sanctions is an example. It is likely
that Americans no longer see the threat Cuba once posed to the
United States, and American travelers would likely choose Cuba as a
destination if the direct travel ban was lifted. If a country is pressured
hard enough either endogenously or exogenously by interest groups,
policy continues. According to public choice theorists, this is exactly
why sanctions such as those on Cuba endure: political organizations,
corporations, and individuals provide financial resources to lobby for
their continuance.
The free rider problem is active in public choice models also, as these
policies are public goods. When interest groups pursue sanction impo-
sition in the sender nation, there are other interest groups that both
oppose and want sanctions. Groups that want sanctions but do not
want to pay may in fact reduce the special interest groups’ demand for
sanctions. The government becomes a group also, either for or against
sanctions, and either willing or not willing to pay. Opposition groups
are willing to pay to keep sanctions from happening; the more the pro-
sanction forces within the sender economy pursue sanctions, the more
expensive sanctions become. Thinking as an economist, the opposition
group forces the supply curve of sanctions to the left, reducing supply
at all prices, thus increasing price, ceteris paribus. The government
imposes sanctions at the level where the groups maximize their respec-
tive welfare functions. It is possible that the opposition will be so
strong, or the political will to impose sanctions will be so small, that the
costs of sanctions far outweigh their benefits such that equilibrium is at
zero quantity. It is also possible, as in sanctions against Iraq, North
Korea, and other dictators throughout history, there will be a call for
infinite sanctions, a complete close of all economic connections.8
 PUBLIC CHOICE THEORY 67

In the United States, the difference between Congressional policy

and presidential decisions is large when considering what groups
within the government may be on the pro-sanction or opposition
side. Kaempfer and Lowenberg (1992) argue that Congress is more
likely to be influenced by special interest groups, where the president
has historically responded to a foreign policy objective. The 1990
sanctions against Iraq, as a result of their invasion of Kuwait, illustrate
this point.

[The 1990 Iraq] episode suggests that although there is a preference

for export sanctions in certain departments of the executive branch,
there can be special interest pressures against export sanctions coming
from other executive departments. The establishment of a total
embargo shows the president’s tendency to act on national rather than
special interest pressures. (57)

Drury (2005) examines U.S. sanctions from presidential decisions

as a central focus. His empirical analysis suggests that U.S. president
chooses targets that have an immediate economic impact from trade
reductions, and that the president is not concerned with the domestic
political environment. This confirms in some ways the hypothesis of
Kaempfer and Lowenberg (1992) given earlier.9
In sum, interest groups seek to influence sanction policy, but the
evidence counters the theoretical models concerning the importance
of such groups. Targets face similar forces, where the fight is over
whether to pursue and continue policies that attract sanctions. The
following section discusses how public choice theory may be extended
to the target.

The Target Economy and Interest Groups

Public choice studies of sanctions have focused on the sender econ-
omy. Two reasons exist for this focus. First, the sender is seen as
commanding the sanction, imposing sanctions based on how sender
interest groups pressure for action. Second, these interest groups exist
in the major industrial nations of the UN Security Council. Thus,
wealth redistribution from sanctions may imply hundreds of millions
of dollars changing hands, and thus makes their analysis noteworthy.
However, the market for sanctions in the target’s economy may be
more important than the sender’s analogous market: the target both
supplies and consumes each political action. Measuring the level of
target political interest groups’ influence is at the heart of empirical
68 ECONOMIC SANCTIONS

public choice theory. The target’s choices are shaped by sanctions as

sanctions change the benefits and costs of political actions. A simple
theory of how the target “demands” sanctions is somewhat straight-
forward from the earlier analysis.
Suppose the target government wants to supply an action that may
lead to sanctions. These sanctions impose a cost onto the target citi-
zens and specific interest groups.10 As the target government engages
in its deviant policies, the opportunity cost of sanctions rise. In a
sense, this is a demand shift to the right in the target’s policy market,
where the target government is demanding sanctions as a result of
choosing certain policies. The supply of sanctions is exogenously
determined by the sender economy. Interest groups in the target
pressure for the policies, and must pay the cost imposed as the sender
supplies sanctions. This “tax” is ultimately borne by citizens in higher
prices and generalized scarcity. The “price” in the international policy
market is a combination of costs and income reductions from the
embargoes.
Sanctions seek to reduce target interest groups’ marginal utility.
Cuba again is a great example of this; the Soviet Union and Europe
directly or indirectly funded the Cuban regime for years as if they were
external interest groups. The Soviet Union wanted communism in the
Western Hemisphere; Europe wanted a vacation destination, sugar,
and cigars. In the United States, persistent fears of communism in the
Western Hemisphere became sanctions and military actions against
most nations in Central America since 1945. Concerning Cuba, the
United States was like an opposition group, paying for sanctions to
continue. Figure 4.3 shows how the sender and target policy markets
are linked in terms of sanctions. Notice that as the target’s supply
curve shifts to the right, the price to impose sanctions falls in the
sender economy. This makes it easier for sender interest groups to
pressure for economic statecraft.11
The sender’s decisions signal to the target a sanction’s magnitude,
at quantity Q*. The sender country resolves its policy market and
transmits this information to the target through policy. The target
country takes those sanctions as given and decides, through their
interest groups, how much to demand. If the sender’s interest groups
demand more sanctions, from DSanctions to D, it is in reaction to new
target policy: DPolicy to D. The target engages in policy, and the
sender reacts with sanctions, constrained by the past and current eco-
nomic and political relationships between the sender and target. If the
target demands more deviance, the sender reacts by lobbying the
 PUBLIC CHOICE THEORY 69

The sender’s policy market after sanctions

SSanctions

P⬘

P*

D⬘

DSanctions
Q* Q⬘ QSanctions

The target’s policy market after sanctions

SSanctions

P⬙

P*

D⬙

DPolicy
Q* Q⬙ QPolicy, QSanctions

Figure 4.3 The policy markets’ international connections.

government, shifting demand to a higher level, matching policy for

policy. This move is shown by the DSanctions curve shifting in the sender
market to match the DPolicy shift in the target market. The slope of the
supply curve reflects the sensitivity (elasticity) in sender interest
groups’ opposition level or to the target government, respectively. As
the policy supply curve becomes steeper, new policy becomes more
costly due to increased opposition.
This is intuitive because the target must pay a higher price to pur-
sue more deviant policies, assuming the sender will react. Kaempfer
and Lowenberg (1988) tell the tale a little differently. The target
country pays a cost to avoid sanctions and thus avoid the disutility that
70 ECONOMIC SANCTIONS

flows from economic coercion. Using two demand curves, where the
demand of those that oppose sanction policy has an upward slope, the
optimal sanction is found where the two demand curves cross. As
demand increases for the deviant policy, the policy’s price increases
and the groups opposing the policy are only willing to accept the
choice if they are compensated for living with sanctions.
This wealth transfer from one interest group to another is the point
of public choice. If interest groups are willing to pay to continue
policy, an equilibrium outcome occurs at a quantity where the gov-
ernment is compensated for administering the policy. Sanctions may
also signal to the target’s opposition group that there is help available,
and external pressure on the current regime is just the beginning.
Kaempfer and Lowenberg (1988) conclude that sanctions concentrat-
ing on reducing the wealth of groups demanding the deviant policy
are most effective, especially if they simultaneously support target
interest groups that oppose the policy being sanctioned.
The interest groups seek to redistribute income toward them
through lobbying for specific government action. How the govern-
ment comes into the picture, either opposed to or for the special inter-
est groups, dictates much of the political effects. It is possible, for
example, that regime opposition can be turned into regime support
under the right conditions, the so-called rally-around-the-flag effect.

The Rally-Around-The-Flag Effect

A rally-around-the-flag effect is a dangerous problem in economic
statecraft. This effect leads to target decision makers using economic
statecraft and the damage they do as propaganda. For example, target
countries that fund terror groups may use such a cry to rally citizens
along religious lines. Galtung (1967) was the first to apply this idea to
statecraft directly, and Baldwin (1985) expanded on the idea; both did
so in their respective analyses of UN sanctions against Rhodesia.
Haass (1997) suggests that this effect can upset the use of focal sanc-
tions that are meant to minimize collateral damage. “Sanctions some-
times trigger a ‘rally around the flag’ nationalist reaction; by creating
scarcity, they enable governments to better control the distribution of
goods; and they create a general sense of siege that governments can
exploit to maintain political control” (80). The sender wants a loss of
political control, and this effect may block the embargo’s effects.
As the rallying cry becomes larger, sanctions should lose effective-
ness. The price the target populace is willing to pay rises; it is as if the
demand curve for the target’s original action shifts to the right
 PUBLIC CHOICE THEORY 71

because of a positive change in tastes and preferences. Kaempfer and

Mertens (2004) examine sanctions against dictators, where rallying
around the flag is likely the autocrat’s reaction to economic coercion.
“Loyalty is viewed as a capital asset accrued to assist in the perfor-
mance of political exchange between ruler and subjects; it is pur-
chased by awarding rents, such as governmental contracts or other
favors, to recipient groups” (32). Of course, a dictator walks a fine line
concerning how to rally citizens, especially if the dictator uses draconian
methods to maintain order.
It is also possible that the sender experiences a domestic rallying
phenomenon in either direction. Where the human cost is large, the
target economy would both want to rally its own citizens around the
sender’s attack, and transmit some of that pressure onto anti-sanction
groups in the sender economy. This is one reason why repression and
loyalty may be seen as co-dependent acts (ibid.). Haass (1997, 1998)
suggested that in the 1990s the United States was engaged in seventy-
five sanctions at one time.12 While Haass (1998) states that the obvi-
ous benefit of using sanctions is the same as not using the military
option, sanctions can be damaging to sender opinion. “Mass hardship
can also weaken domestic and international support for sanctions, as
with Iraq, despite the fact that those sanctions have included from the
outset a provision allowing Iraq [in the 1990 case] to import human-
itarian goods and services” (Haass 1997, 79). Public choice theory
plays a vital role in understanding sanctions from debate to termina-
tion. Ultimately, market activity dictates sanction effects and duration.
Smart sanctions are the latest attempt to minimize the damage
sanctions put onto innocent target citizens and other nations.

How Much Are Interest Groups

“Demanding” Sanctions?
The evidence that public interest groups are forcing the U.S.
Congress or the UN Security Council into decisions of employing
economic statecraft suggests no one group is constantly seeking sanc-
tions. The evidence gathered has instead focused on how the sender
and target economies have dealt with the populace in order to influ-
ence the respective government’s ability to pursue policy. Kaempfer
and Mertens (2004) applied public choice theory to sanctions where
dictatorships looked to be displaced.13 They conclude that the level of
domestic opposition influences the sanction’s effectiveness, and that
the level of sanctions must be monitored as to not put that opposition
in jeopardy. A sanction’s attempt to impoverish the populace may
72 ECONOMIC SANCTIONS

weaken the domestic opposition’s ability to exert influence (ibid.). Using

a variant of Wintrobe’s (1990) dictatorship model, Kaempfer and Mertens
(2004) juxtaposes loyalty and repression as factors influencing a dictator’s
power. The relevance of investigating how sanctions influence specific
regime types is obvious, and the public choice approach is one way of
explaining how groups define and fight each other for the target’s polity.

Are Specific Groups Defined in Both Countries?

In an attempt to provide public choice modeling with some empirics,
Bonetti (1997a) applied a simple logit model using the HSE (1990)
data and public choice variables. Bonetti (1997b) suggests that the
main way sanctions have been viewed is as an instrument of coercion
(339), where coercion is seen as slightly different from “embargo” or
“sanction” because the intent is to force a change among specific
groups within the target nations, as well as convince the sender
government to act. Bonetti (1997b) concludes that sender interest
groups must engage in multiple coercion measures. Further, these
groups’ demands may not be to damage the target, but to profit from
the sanction directly and transfer wealth to the sender interest group.
Bonetti further suggests that because confusion exists as to whether
interest groups are individuals, political interest entities, or govern-
ments, finding evidence and empirically testing the public choice
hypotheses can be difficult.
There has been a stronger preference for export sanctions because
of interest group existence; this is not a prediction but a tautology
(ibid., 341). Groups form if the political conditions are right for
individuals to act collectively against government policy. Gathering
data that reveals the propensity of the target populace to oppose its
own government to cease current policies may be a link to how
powerful public choice theory is in explaining how sanctions work.
However, this data has not been found to date.

Does a Transmission Process Exist?

The other major factor in public choice theory is the assumption that
a transmission mechanism exists between the sender’s and target’s
interest groups that is economic. In international economics, a major
link between two countries is trade, and the transmission mechanism,
for acts such as trade barriers, is very specific. Exports in one country
are imports to another, and vice versa. Public choice theory assumes
that interest groups are self-motivated, seeking to augment current
wealth from policy. The political economy aspects of such groups are
accepted as fact, as the sender expects policy to redistribute wealth in
 PUBLIC CHOICE THEORY 73

a specific way; the target groups engage their government for the
same reasons. However, to build an empirical model to test for the
existence of these conduits entails discovering data that either describe
the transmission mechanism or make sense of the interest group link-
ages. Since countries differ in how wealth is distributed and the infra-
structure for an interest group to engage its own government, it is
difficult to generalize this characteristic and empirically test these
public choice models.

Summary
The public choice approach to sanctions blends political economy
with international economics. This approach suggests that interest
groups within each nation involved seek out sanctions. In the target
economy, sanctions are an accepted by-product of pursuing deviant
actions: the pursuit of political actions that may lead to sanctions is
considered rational because interest groups seek increased income as a
result. In the sender economy, interest groups are assumed to lobby or
pay directly for sanctions as a reaction to target malfeasance. Sender
interest groups want their government to impose sanctions because
their income rises with policy, dictating sanction demand. The combi-
nation of these two policy markets leads to an international policy
connection based on the interest groups in action.
This combination also leads to a global market for sanctions. The
target demands sanctions by its actions, and the sender supplies the
reactions, where equilibrium is struck at a price and quantity such that
each country maximizes individual welfare by pursuing their respec-
tive actions. These rent-seeking actions, according to public choice
theorists, drive sanction outcomes. Bonetti (1994, 1997b) argues that
while public choice theory is logical explanation of sanction effects,
the ability to test the hypotheses and conclusions of the theory is non-
existent. Chapter six expands somewhat on the role of interest groups
and empirical studies. Another way to view public choice theory and
sanctions is to test for the influence of groups calling for smart or focal
sanctions. While this text suggests that sanctions are ultimately macro-
economic, focal sanctions that try and circumvent widespread damage
are gaining momentum as recommended policy.

Mitigating a Cruel Paradox: Smart Sanctions

The issue of collateral damage in sanction episodes has recently
become a focus of policy makers and scholars alike. The United
74 ECONOMIC SANCTIONS

Nations has debated the “legality” of sanctions in terms of how the

policy restricts market access. Doxey (1971) was the first to discuss
the international law complications for economic statecraft and in a
sense began the discussion of how senders must consider both the
target’s economic and human costs. Humanitarian concerns have
followed because there are cases where sanctions are perceived to
cause or exacerbate problems of disease, starvation, reduced water
resources, and so on. However, damage to the target’s people outside
of its rulers must be expected by all parties involved or observed
during a sanction.
Smart sanctions claim to focus on decision makers while mitigating
the cruel paradox of parallel punishment to the target’s citizens. Can
statecraft focus on those who are most likely to change the deviant
behavior, and cause a behavior change quickly and efficiently? Smart
sanctions are an idea that at their core are as old as sanctions them-
selves: economic statecraft can indeed focus more on the ruling elite
in sanctioned countries, rather than being broad measures, and by
doing so be more effective.
Certain studies argue, however, that the efficiency of sanctions is a
priori enhanced if the greatest costs fall on innocent bystander groups
within the target, regardless of how antagonistic or friendly they are to
the sender (Major and McGann 2005, 338). The public choice frame-
work is important here, as the definition of which groups are “inno-
cent” and which are not depends on how the groups influence target
political behavior. One way to view collateral damage is to see them as
punitive against a country allowing such actions to take place. One
example is the 1985 UN sanction case against South Africa. These
sanctions were designed to eliminate apartheid explicitly, reform the
political structure in South Africa, and allow the black vote. Since
those in power were mixed among the elite residents of South Africa,
especially land and business owners, how could the effects not
penetrate to the masses? By punishing these wealth holders, there
must be some effects on workers, consumers, and non-black citizens.
“Although these [factors of production curtailed by sanctions] could
be had, the premium that South Africa had to pay to ‘bust’ the
embargo on these items made them costly commodities to their
domestic consumers” (349). However, to claim that South African
business people and consumers had larger ties to groups in the United
Nations beyond the philosophical or that there was a sense of
philanthropy on the part of the Security Council is an overstatement
of the international importance of South Africa during the lengthy
sanctions.14
 PUBLIC CHOICE THEORY 75

Can policies that originate in another economy, whose ultimate

goal may be the foreign government’s complete political turnover,
not affect the foreign populace that keeps its current ruling party in
power? Cortright and Lopez (2002) edited a volume that discussed
smart sanctions in depth. In many of the included articles, smart sanc-
tions were considered difficult to focus in practice and the success of
such sanctions remains highly subjective. Another highly debated case
is the 1990 UN sanctions against Iraq, which concluded by force
when the Anglo-American invasion took place in 2003. These sanc-
tions were comprehensive in scope from the United States and United
Kingdom, but the United Nations had looked to remove some of its
sanctions because of a carrot–stick approach taken concerning
weapons of mass destruction.15
The sanction market can be bifurcated into a market where the
interest groups and the government interact; the other market is
where the remaining citizens interact in the face of a sanction’s costs.
What smart sanctions should do theoretically is impose costs directly
on target interest groups lobbying or funding the government’s poli-
cies deemed problematic by the sender, which then reduces the equi-
librium quantity of the policy. The sanction should shift the target
government supply curve to the left, reducing the supply and thus
reducing the quantity demanded in this policy market. If the benefits
derived from engaging in deviant actions are reduced, the supply of
such actions falls.
Smart sanctions attempt to make economic statecraft more micro-
economic. Forcing specific target agents to change their behavior
implies the sanctions must affect individual policy makers directly, as if
making policy was within the policy maker’s utility function much like
any other good, service, or policy. Smart sanctions act like policy
barriers, trying to reduce the policy market’s activity by forcing a tax
levy because of domestic decision-making. Smart sanction measures as
policy cannot be ignored, as the ideological ground upon which they
stand is solid. However, smart sanctions suffer from three main prob-
lems that affect all such policies and put their use as more than rhetoric
in danger.

Sanctions Are Meant To Be Blunt Instruments

While all sanctions should aim to minimize human costs, how can
sanctions by any design not spillover onto an otherwise innocent
populace? Interest groups in the target economy that pressure for the
deviant action should always be the focus of sanctions. The ultimate
76 ECONOMIC SANCTIONS

focus of sanctions is political, without an economic bull’s eye. The

unfortunate economic outcome is that sanctions flow to target deci-
sion makers by damaging the populace, which in part makes up the
interest groups. In the best scenario, sanctions provide a credible
threat of such collateral damage such that a reversal of target policy
takes place. So much of a sanction’s perceived power is through its
larger effects.
Sanctions are macroeconomic phenomena that must credibly have
breadth and depth in their effects. Threatening the populace must be
to threaten those in power, and vice-versa. The United States currently
sanctions Iran comprehensively. There are also financial and travel
sanctions on Syria, where both packages are meant to reduce the threat
of terrorism. Iran’s ruling elite are likely hurt by sanctions because of
their breadth, but less than the populace. The ruling elite may have an
incentive to slowly pass a “sanction tax” onto its people, using higher
prices and generalized scarcity as propaganda. Sanctions against Iran
have become, over time, a classic sanction case involving black knights.
The existence of external markets acts as an enforcement problem; the
United States has reacted by passing legislation to expand sanction
coverage to parties that act as sanction busters, both domestically and
internationally.16 This leads to another layer of credibility on the
sender’s part that must display commitment, leading possibly to
engaging some of the larger economies in separate embargoes, which
generally have little credible threat of effectiveness.

Target Governments May Avoid Focus

through Alliances and Diversification
As with any other economic measures, a sanction must act like a cartel
and insulate itself against substitute policies or internal cheating.
Financial sanctions involve a worldwide network of transactions and
potential market participants to act in concert with an embargo to be
effective. Reid et al. (2002) provided three differences between coun-
tries that generally lead to problems in forming long-lasting coali-
tions. First, there is a question of statutory authority. Some nations
simply disagree with certain sanction policies, especially in the face of
potential human damage. Second, the implementation mechanism is
different from country to country. Drury (2005) focuses on how U.S.
presidents have used unilateral authority in imposing sanctions. Some
countries and institutions do not have such a trigger mechanism by
which sanctions can be immediately unleashed in the event of an
international crisis. In July 2006, the slow and heavily debated UN
 PUBLIC CHOICE THEORY 77

reaction to North Korea’s test launch of long-range missiles provided

a lot of real-world evidence for the mechanics of multilateral sanction
imposition. Diplomatic efforts take time and even then a resolution
may be weak due to statecraft within statecraft.
Finally, authorized agencies that impose and monitor sanctions dif-
fer among nations. In the United States, the Office of Foreign Asset
Control (OFAC) is meant to regulate sanctions such as freezing of
assets and the seizing of foreign property.17 Reid et al. explain that
asset seizures are difficult sanctions in practice because jurisdiction
dictates that the country imposing the sanction describe the financial
claims as criminal (2002, 81). Steil and Litan suggests that not only is
it difficult for such agencies as OFAC to track sanction busters, but an
international money laundering system is in place currently that can-
not be fought by U.S. personnel alone (2006, 35). There are many
profit incentives here as well, all of which undermine perceived and
explicit alliance structures.
Initiating such financial measures also has its challenges due to the
speed of financial transactions. The sender nation must act swiftly and
almost completely under the international radar initially to keep the
target country from making a few phone calls or sending a few e-mails
to move assets from one country to the other. If alliances are in place,
this secrecy is even more tenuous, and the major financial markets
must all ban new trades and income flows simultaneously to work.
Arbitrage is just a phone call away or those seeking to profit from new
assessments of country risk initiating market signals.
This lack of control over foreign asset movements is just the begin-
ning. Trade sanctions and rationing the target’s available credit easily
provide incentives for third-party nations, black knights rescuing the
target from sanctions.18 The Iran–Libya Sanctions Act (ILSA) of 1996
was meant to punish countries, firms, and individuals that helped
U.S.-targeted nations during sanctions. If the United States sanctions
arms shipments to a belligerent, oil-producing nation, these seem like
smart sanctions. They look like very dumb sanctions once another
country sells weapons at a higher price and profits, because America
cannot credibly sanction countries such as France and the former
Soviet Union historically, or countries of the Russian Federation now.
How can sanctions focus on the target’s ruling class if other entities,
at the public or private level, act on the target’s behalf and provide
new markets? With the profits available to third parties, smart
sanctions as envisioned need some adjustments.
The proliferation of foreign currency from the larger, industrial
nations all over the world has made financial sanctions less powerful,
78 ECONOMIC SANCTIONS

and made targeting these sanctions even more difficult in practice; it

also further sweetens the deal for economically weak third countries to
help the target (Newcomb 2002). Alliance structures should avoid
many of these problems if well-devised and enforced strongly, where
certain studies suggest positive economic incentives are more efficient
than negative incentives in enforcing alliances.19

Arms Embargoes are Also Subject

to Black Knights
One sanction that seems very specific to reducing a target’s deviant
behavior is arms sales curtailments, the archetype smart sanctions
(Brozska 2002). If political deviance is enforced by domestic military
or paramilitary units, reducing arms trade with such a country hits at
belligerency’s heart. However, the world arms market is wide in
breadth, and offers bellicose nations a large array of choices. These
choices are somewhat restricted to small arms, especially for sub-
Sahara African nations (Elliott 2002). The experience of the late 1980s
and 1990s showed that if the price is high enough, markets will exist.
The United States and United Kingdom have both brokered deals for
domestic defense firms in cases where arms sanctions were in place
through the last quarter century; Bondi (2002) recommends that
tighter monitoring and an increase in penalties is the best way to
reduce arms sanction busting. Brozska (2002) suggests that the black
market structure of arms transactions makes enforcement almost
impossible, and that the current monitoring is weak. “It is sometimes
argued that the U.N. is not the best body for monitoring and verifi-
cation [of arms embargo violations]. Because of its nature as a diplo-
matic institution, [the United Nations] is inclined to uphold very high
standards of proof for allegations that imply misconduct by member
states” (137). However, even if the United Nations found a violator,
what court would hear the case that has jurisdiction?
Since 1990, it is reported that over ten arms embargoes have taken
place where the United Nations acted as the sender.20 However, trian-
gular trade is used to circumvent direct evidence of an arms deal, typified
by the Iran-Contra affair involving the United States trading arms for oil
in the 1980s. Since then, arms deals to bust sanctions have been more
overt, as in cases involving Angola and Nigeria, because their resources
allowed certain countries to profit from sanction busting. “Realizing
that [Angola’s] UNITA rebels use diamond profits to finance their
weapons purchases, the U.N. Security Council imposed an embargo
on uncertified diamond exports from Angola. This episode suggests
 PUBLIC CHOICE THEORY 79

that, as a stand-alone policy, arms embargoes are unlikely to curtail

local conflicts” (Hufbauer and Oegg 2000, 2).
The profits available to a private arms manufacturer rise quickly
during embargoes, and thus stiffer penalties must be assessed if such
sanctions are to be effective. This problem may be worse in Africa
than any other place on earth (Brozska 2002). The problem is in any
country undermining these sanctions’ purpose.21 The target’s ruling
elite is not hurt if second or third markets for these weapons exist.
However, a higher arms price means a larger economic burden on the
populace through inflation, higher interest rates, or more taxes to pay
for the price change. Analogous to arms embargoes, there are some
problems with financial statecraft as well.

Financial Embargoes and Currency Crises

Financial sanctions have been the sanction of choice of the United
States and are quintessential smart sanctions: they are generally easy to
enforce, difficult to evade, and market reactions are likely to reinforce
the sanctions (Elliott 2002). Like arms embargoes, market reactions
are the key problem as they are uncertain and must be monitored
closely. Monitoring means more cost to the sender, which reduces
sanction potency in the context of being smart; the cost increase likely
leads to the sanction’s burden being shifted onto the populace.
In a financial sanction, the market reaction increases the cost of
credit, reduces income flows to both the target’s government and citi-
zens. The governmental cost may be simply passed on to the target’s
people. Analysts should view these sanctions as restricting these mar-
kets that lack substitutes and are necessary to the target’s economic
and political functions. The freezing of assets is highly focused on
individual level, but individuals generally do not make policy choices.
If individual investors happen to also be political decision makers, they
are unlikely to expose themselves to an asset seizure if a fund transfer is
done quickly through some communication medium. This debate over
sanction bluntness rages throughout the smart sanction literature
(Cortright and Lopez 2002). The creation of a currency crisis, where
the target’s currency value is put into jeopardy, may come from strong,
focused financial sanctions, delivering a macroeconomic problem.
A currency crisis could come from very harsh financial sanctions,
especially if the target economy is already on the brink of economic
collapse. Almost every major currency crisis over the last ten years has
a “currency mismatch” at its root (Stein and Litan 2006, 99). A cur-
rency mismatch is when a country has a significant amount of its debt
80 ECONOMIC SANCTIONS

in terms of another currency, which opens a country up to economic

fluctuations based on another country’s activities versus their own.
There are many studies that claim a currency mismatch is a function of
a flexible currency.22 A weak banking sector, as well as poor and mis-
managed monetary policy may also be to blame. However, sanctions
that impose currency-crisis conditions on a target will have widespread
effects and no longer be smart per se.
Market reactions must be taken into account. As a currency falls
in value, the balance of payments immediately shifts to adjust the tar-
get nation’s wealth. “Financial markets also become disturbed and
balance-of-payments problems, currency convertibility and a host of
related difficulties appear or worsen. Transaction costs become dis-
proportionate. [Pre-war] Iraq is a classic example of all these prob-
lems” (Losman 1998, 39). There may also be a large amount of target
politicking as the currency spirals downward. “The political element
of currency production and supply management is so powerful and
pervasive that we must be extremely wary of policy conclusions that
ascribe the obvious, and frequently calamitous, failures of the world of
national currencies to those who must decide, in their own interests,
whether to hold or jettison them” (Steil and Litan 2006, 113).
Does this mean, potentially, that public choice theory can be
applied to currency crises and sanctions? The case could be made that
countries with a history of currency problems, perhaps sovereign debt
renegotiations or outright default, perceive credit sanctions as precur-
sors to currency crises. The mere threat of such sanctions may be
enough to force a policy change, especially in light of how generalized
the effects of a currency debacle would be.
The market reactions to financial sanctions can also work against
the sender economy, as both goods and financial market transactions
slow down. As the target currency falls in value, the target builds
international incentives to export more goods. When other countries
buy the target’s goods, they provide the target with desperately
needed foreign currency. These purchases support the reduced value
of the target currency and mitigate the original effects. The way
sanctions should be mixed is discussed further in chapters five and six.
Regardless, financial sanctions are likely to be the smart sanction of
choice into the future. They are relatively straightforward to both ini-
tiate and manage for the sender. They send worldwide signals con-
cerning the target’s creditworthiness and its potential currency
conditions. The target populace, in the threat of both goods and
credit market inflations, may immediately call for an end to the
deviant policy rather than suffer the consequences. All these effects are
 PUBLIC CHOICE THEORY 81

part and parcel of the smart sanction idea. Smart financial sanctions
will be discussed further in chapter seven when humanitarian effec-
tiveness is compared to economic and political efficacy. To conclude
on smart sanctions and close the loop connecting focal embargoes and
public choice theory, debates over sanction legality provide another
dimension as to why smart sanctions may be the future of economic
statecraft.

Smart Sanctions and Legal Issues

Smart sanctions seem the most “legal” of any embargo, but so much
of that perceived legality depends on two definitions: that of a legal
sanction and that of a smart sanction. First, a legal sanction describes
an embargo that originates from an international agreement on using
economic statecraft after or parallel to over-the-table diplomacy and
before military action. Because these policies transcend borders, it is
important to recognize that no one country’s jurisdiction controls
this definition. However, the UN Security Council is the most likely
place to define such measures for the international community, as
should other international alliances and economic unions through
their charters.
The legal debate has recently shifted to discuss humanitarian issues.
A new phase is developing around proportionality, a principle devel-
oped to monitor sanctions for how much damage the populace
endures versus its culpability (O’Connell 2002). Craven (2002) sug-
gested that while it is powerful to define when sanctions are legal in
terms of human cost, ultimately there is little to be done concerning
enforcement. “The evident discretion available to the Security
Council in this regard is further reinforced by the presumption that it
is largely competent to determine its own jurisdiction, and by the
absence of available mechanisms for judicial review” (50–51).
There simply is no legal forum to debate these cases. Even if such
an international court existed with any authority, the data on cause
and effect would likely be a central debate item. This debate is far
from resolved. This is seemingly the current argument of the North
Korean government that the lack of proportionality in U.S. sanctions
against this regime has forced a humanitarian crisis. As a result,
nuclear arms are produced to retaliate against an economic aggressor.
One way in which sanctions can be monitored or made “smarter”
is to watch how businesses are reacting, both in the sender and target
nations, during sanctions. The effects on target businesses are the
foundation of economic statecraft, as business reactions are likely to
82 ECONOMIC SANCTIONS

transmit themselves to a sanctioned populace and government.

Losman (1998) discusses the effects and costs on U.S. businesses as a
result of sending sanctions, which is a sanction literature topic discussed
indirectly with a lot of work to come.
If costs are truly escalating for American businesses as a result of
sanctions, there could be legal repercussions in the U.S. legal system.
“[As a result of U.S. sanctions on Panama in 1989], a number of
American-owned properties in Panama had to be sold at distress
prices. If import restrictions are part of the sanctions package, it is
likely that higher cost substitute suppliers, either domestic or foreign,
will have to be utilized” (41). Can and should American businesses
pursue financial recourse when able to prove the sanction has injured
profits? Is this not a classic case of the rally-around-the-flag effect in
reverse?
Thinking as a public choice theorist may help again. If both sender
and target businesses have influences on political decision-making,
then their realization of new costs drives new policy decisions. As a
result, public choice theory suggests that as the sender government
pursues sanctions, the effects transmit themselves to the target coun-
try but also influence the costs of pushing for such action domesti-
cally. Special interest groups may be influenced by business costs as a
result of sanctions, as common sense suggests that businesses stand to
gain from foreign policy more than individuals. While sanctions have
their effects on targets, leading to public reactions, they are working
against themselves in the sender economy by imposing a domestic tax
on businesses. Of course, sending sanctions comes with costs. Losman
(1998) called for U.S. sanctions to be completely dropped from the
policy toolbox, suggesting they be made illegal because of domestic
costs.
Ultimately, a sanction’s legality is not as important as defining
smart sanctions. This chapter shows how difficult it is to see sanctions
as having an ability to affect a specific subgroup of the target popula-
tion, making focal sanctions less focused in a hurry. Sanctions such as
travel, arms, and financial embargoes are defined immediately by
scholars and policy makers as smart, but can certainly hurt the popu-
lace of both economies. The future of smart sanctions lies in goals of
avoiding human costs and major redistributions of an embargo’s
damage from the ruling elite to the poor. If these policies achieve
these goals, then smart sanctions are truly economic statecraft’s
future. However, as sanctions are imposed more and more to enforce
what could be considered international law violations, a gray area may
appear that makes smart sanction effects even trickier to analyze. “The
 PUBLIC CHOICE THEORY 83

humanitarian law of armed conflict fits the law enforcement context

more closely than the law of human rights. Humanitarian law regu-
lates otherwise unlawful behavior resorted to in response to a prior
wrong” (O’Connell 2002, 73). Are not all sanctions a reaction to a
perceived violation of international law or standards in the first place?
The interest group state ensures that international political deviance
that reduces wealth domestically leads to a reaction, which is the
essence of applying public choice theory to economic statecraft.

Conclusions
Public choice theory provides a foundation for the political economy
of economic statecraft. Kaempfer and Lowenberg (1986) is the begin-
ning of this literature strand and they have authored many subsequent
studies. Bonetti (1997b) has been critical of this approach on empiri-
cal and theoretical grounds, also providing a great literature review. If
special interest groups within each economy, the sender and target,
influence political decision-making, then a market for economic sanc-
tions may exist. These markets are much like the markets for goods
and services traded between two countries, as a policy shock in one
market is transmitted to the counterpart markets in other nations.
However, public choice theory suggests that groups within the sender
economy pay for sanctions to derive wealth-enhancing effects from
them. Embargoes may protect the group’s interest and potentially
redistribute statecraft costs to other industries or the populace as a
whole. The target economy faces the same problem, as its interest
groups are willing to pay to subsidize governmental actions in the first
place. How each group receives income from the sanctions drives
their decision-making and whether or not the respective governments
continue to supply the demanded policies.
The more blunt the sanction instrument, the more widespread its
effects. Many studies contend that collateral damage renders the
sanction archaic and imposes more costs than benefits, especially as
sanctions continue for years and decades. The imposition of focal
sanctions, such as restrictions on arms, travel, financial flows, and
other specific goods and services, is said to embargo the decision
makers’ (or interest groups’) wealth directly. Cortright and Lopez
(1995, 2000, and 2002) have contributed, through direct science and
edited volumes, much to our knowledge of these measures.
Such measures are known as smart sanctions, because they focus on
specific subtargets within the sanctioned economy. There are still
debates raging over how focused these embargoes really are. Haass
84 ECONOMIC SANCTIONS

points out that the opportunities to employ sanctions effectively yet

with great precision are rare (1998, 202). It is likely, however, with
the constant call to monitor the humanitarian damage caused by
economic statecraft that smart sanctions are here to stay, providing ex
ante protection to the masses. Policy makers and scholars alike must
obviously devise strategies to limit damage and focus as much as pos-
sible, perhaps providing positive economic incentives to countries that
help in a parallel way to relieve suffering, a major international policy
challenge. One small item scholars cannot forget is that smart sanc-
tions are just that: sanctions. It is hard to imagine any strategy that
makes these instruments more microeconomic.
The first four chapters of this text provided background literature
and theory for why sanctions take place and continue. In chapter five,
a macroeconomic model is introduced to further this text’s thesis that
sanctions are macroeconomic phenomena and should be treated and
analyzed as such.
 Chapter 5

Open Economy Macroeconomics

and Sanctions

One way to combine both the qualitative analyses that have been a
constant throughout the sanction canon and quantitative tests of sanc-
tion effectiveness is to view sanctions as macroeconomic policy choices,
analogous to monetary or fiscal policies. Recent advances in macro-
economics and international finance allow economists to potentially
settle issues concerning sanction measurement and tracking sanction
effects as for both the sender and target economies. This chapter
extends the Obstfeld–Rogoff “Redux” (1995) model to provide a
macroeconomic theory of sanctions.1 Their model and the extensions
that followed became the “New Open Economy Macroeconomics,”
called NOEM models from here.
Using the dynamic behavior of optimizing agents to study policy
transmission from one country to another allows analysts to model
how contending nations interact through their trade and financial
relationships. This interaction tracks how a larger country’s policies
may become negative welfare shocks for a smaller country through
policy transmission. Seminal research in international policy transmis-
sion began with Fleming (1962) and Mundell (1963), continuing
with Dornbusch (1976). The original Mundell–Fleming model is one
of static exchange rates, which led to fixed price, IS-LM model exten-
sions.2 The Dornbusch (1976) extension of Mundell–Fleming’s
model used intertemporal ideas of household welfare maximization, a
precursor of NOEM models, to describing exchange rate overshoot-
ing as a reaction to policy under a rational expectations framework
and flexible exchange rates. The exchange rates overshooting result
was based on sticky prices in goods markets and rational expectations
in asset markets. The mix of goods market imperfections and asset
86 ECONOMIC SANCTIONS

market perfection imply two possible exchange rate effects from

policy: (1) post-policy exchange rate movements such that policy is
domestically neutral; or (2) exchange rates movements that cause
expenditure-switching behavior between domestic and foreign goods
in such a way that initial welfare shocks were either augmented or
reversed in each country.
The political economy of sanctions fits this idea well. Sanctions are
meant to transmit, from the sender economy, welfare redistribution
incentives to a target. In macroeconomic theory, transmissions that
reduce welfare are reasons not to initiate policy. In the case of eco-
nomic coercion, sender economies want to transmit policy to affect
targets in deleterious ways, to “beggar thy neighbor” as a matter of
statecraft while limiting “beggar thyself” effects (policy-driven,
domestic reductions in welfare). These transmitted policy effects are
the foci of the sanction extension given later, where the policy shock
is an economic sanctions mix. Sanction-caused exchange rate fluctua-
tions may not nullify welfare shocks on sanctioned countries, and thus
sanctions may not be neutral in their effects on either country. The
economic effectiveness of a sanction can now be based on differential
welfare movements between sender and target, helping to bridge the
literature’s gap in estimating a sanction’s relative costs between sender
and target.
How prices are set in each country is of primary importance. The
NOEM models allow for both asset and goods price stickiness as
assumptions, where market frictions exist between the target and
sender that lead to sanction policy transmitting non-neutral results.
The pricing-to-market (PTM) models focus on goods price stickiness
rather than asset markets working incorrectly (Betts and Deveraux
2000). The sanctions NOEM model here uses a PTM model as a
foundation.
This chapter is split into four sections. The first further describes
the literature and basic assumptions of NOEM models, especially the
intuition and implications specific to sanctions. The second section
describes a NOEM model, where sanctions cause welfare shocks and
exchange rate fluctuations based on their mix and connections to both
sender and target welfare. This set of models builds to a statement of
the “steady state” or long-run equilibrium where households maxi-
mize their individual net benefits or utility over their lifetime. This
steady state is shocked by economic coercion, and those effects are
explained in the third section. The chapter concludes by suggesting
future research directions in modeling sanction policies as macroeco-
nomic phenomena. While this model is mathematically complex, the
 OPEN ECONOMY MACROECONOMICS 87

intuition behind an embargo’s political economy from basic models

prevails here. We see that sanctions have predictable effects on exchange
rates, some ambiguity on welfare overall, and that using financial sanc-
tions may be the strongest coercion tool available. For the reader’s
convenience, a symbol list appears at the beginning of chapter five A,
the mathematical appendix to this chapter.3

The NOEM Sanction Model: The Basics

The interesting idea behind the NOEM models is that a welfare analy-
sis of both the sender and target describes how their interplay may
affect economic statecraft’s potency. The NOEM model’s initial
insight was that the overshooting and resultant welfare effects pre-
dicted by Dornbusch (1976) may not take place under a sticky price
assumption, where prices are set in advance of policy being made. The
baseline NOEM model suggests that macroeconomic policy will not
force a deviation from purchasing power parity (PPP), driven by the
law of one price (Mark 2001). The law of one price is crucial, as it
leads to a non-neutral result concerning welfare: the country originat-
ing policy “beggars its neighbor” by reducing another country’s
welfare to augment its own. Since most domestic macroeconomic
policies are meant to enhance domestic welfare by definition, the
NOEM model provides a general equilibrium approach to policy’s
differential effects.
The price-setting assumption is important for the sanctions model.
In the PTM extension, prices are set one period in advance in each
market faced by each country (the domestic price differs from the for-
eign price of a good), disintegrating commodity markets between
countries. This is due to the assumed existence of monopolistically
competitive firms, which can price-discriminate, and may have more
information about policy than households. The policy neutrality
predicted by Dornbusch (1976) returns in this framework, and the
differential effects fall away. In most NOEM models, households are
assumed to optimize a utility function composed of consumption,
money demand, and work effort; the households also own the firms
that maximize profits from the domestic good’s production and sales.
In the sanction extension later, prices are set when policy is made.
Sanctions are macroeconomic policies, like their monetary and
fiscal counterparts. Unlike other policies, the sanction’s direct intent
is to affect sovereign decision-making in another country by transmit-
ting welfare-reducing, beggar-thy-neighbor effects. The NOEM con-
clusions are that long-run welfare losses depend on how responsive
88 ECONOMIC SANCTIONS

domestic and foreign goods are to substitution incentives created by

policy and transmitted through subsequent exchange rate fluctua-
tions. The reason policy may be neutral in these models is that the
resultant exchange rate fluctuations eliminate policy’s initial effects.
Most target countries are smaller, more volatile economies than the
sender nation or coalition. NOEM studies assume that policy in a larger
economy transmits its effects to smaller countries generally. Because of
the larger country’s perceived economic hegemony, the smaller country
faces price-taking behavior except in its domestic markets for its own
good. As we will see later, the PTM model and these pricing assump-
tions may lead to policy neutrality in certain cases, which make the case
for sanction ineffectiveness pervasive in the literature.
A related, seminal study is by Fender and Yip (2000), who investi-
gate the welfare effects of tariffs in the NOEM model. They found
that tariff effects are shared by both countries. The sender effects are
ambiguous, analogous to those on the tariff-setting country in the
Fender/Yip model. This ambiguity depends on the elasticity of sub-
stitution between the imported and import-substituting, domestic
good. Fender and Yip’s approach allows sticky prices in the short run
and flexible prices in the long run, splitting the agent’s life into two
discrete periods. As in other NOEM models, the policy shock lasts
one period (short run) and resolves itself in the next (long run) where
the overall effects are the sum of each period’s effects. Policy neutral-
ity results if the long-run effects completely mitigate the short-run
changes. If the tariff is anticipated, there are welfare costs in the target
economy; if not, neutrality results for the target (Fender and Yip 2000).
A sanction, however, should not be neutral in its effects. Economic
statecraft relies on sanctions to be detrimental to target welfare.
Sanctions should be modeled as quantity restrictions, like quotas, and
analyzed as such as in chapter two.
In Betts and Devereaux (2000), prices are set by monopolistically
competitive firms that charge one price domestically and another in
the foreign sector rather than one world price. Prices are assumed to
reflect a mark-up from the wage, such that the firm’s profits are
maximized. As firms set their prices in advance, the exchange rate
changes from monetary policy are magnified versus the typical
NOEM model. As a result, the exchange rate change moves to
eliminate any consumption-switching behavior between the goods
in each country, leaving domestic monetary policies neutral in their
effects. The importance of the PTM extension to sanctions is that
each economy involved has some pricing power over its own
good. This seems more realistic, especially in terms of oil-exporting
 OPEN ECONOMY MACROECONOMICS 89

nations that become sanction targets. The PTM conventions concern-

ing pricing are employed here.
The pricing decision’s timing is important. This model assumes
that prices are set when sanctions are imposed. As such, prices and
interest rates are known when policy is made, and thus trade embar-
goes are simply reductions in firm revenue in the short run, and may
affect terms of trade in the long run. Financial sanctions affect the
household budget and the consumption-saving decision by reducing
available funds and asset income. This restriction affects the long run
because the target has less flexibility to substitute consumption for
savings through an embargoed credit market.
In the model given here, the policy variables become simple linear
terms in the pre-sanction equilibrium, the steady state, similar to
Fender and Yip (2000) but without the tariff. As the sanction is
imposed, there are specific changes to the optimal choices of con-
sumption, money demand, and work effort because the sanction
reduces the target’s income (import and debit sanctions), its ability to
borrow or lend (credit sanctions), and its available goods markets
(export sanctions). These changes affect target utility aiming to
reduce it.
The effects on both the sender and target can be derived simulta-
neously in this general equilibrium framework. By imposing sanctions,
the sender is choosing to reduce its own welfare explicitly, where the
costs avoid larger political and subsequent economic costs in the
future by not reacting to target malfeasance. The focus here is on
the welfare reductions.4 The extent to which exchange rates fluctuate
once policy takes place, and the specific channels leading to these
changes, are still major questions in this literature (Vanhoose 2004).
The model given here extends NOEM models to include sanction
policy effects on both countries involved and provides more founda-
tion for sanctions to be viewed as macroeconomic phenomena. The
reader is asked here to refer to chapter five A for the model’s mathe-
matics, walking parallel to the mostly qualitative descriptions here.

The Model’s Structure

The basic set-up of the NOEM sanction model is that there are three
countries, two goods, and one financial asset traded between these
countries to link their balance of payments. The third country acts as
either a coalition member with the sender or a black knight for the tar-
get. The target is distinct from the third country, called ROW for
“Rest of World” from here on. If there are no sanctions, this model
90 ECONOMIC SANCTIONS

Sender Target ROW

0 n m 1

Figure 5.1 The world trichotomy.

collapses to the basic model of two countries, the sender and all
foreign economies amalgamated.5 Figure 5.1 is a standard diagram of
the world economic composition in these models, as in Mark (2001).
The concept here is that macroeconomic policy transmission origi-
nates in the sender’s economy and moves through its international
connections to the other economies. A unique basket of goods is pro-
duced and purchased in each country, where the target and ROW are
seen as two markets that sum to an aggregate but face separate policy
outcomes. It is in the goods markets where trade sanctions appear.

Goods Markets in the NOEM Model

Sanctions redistribute revenue to “pay” for the policy; the sender
chooses to lose revenue or forces the target to do so through export
and import sanctions respectively. The key difference between this
idea and the Fender/Yip (2000) model is that a tariff provides rev-
enue that must be accounted for in the household’s budget. Sanctions
do not generate revenue, such as the monopoly rents from selling
quota licenses; the sanction is instead modeled as lost revenue, where
this is the opportunity cost for the sender to force change in the tar-
get economy.6 However, there are many arguments in the interna-
tional trade literature that quotas have “tariff-equivalent” effects.7 An
export sanction reduces revenue specific to the sender’s domestic
good (z), and an import sanction reduces the supply of the foreign
good (z*) originating from the target only, based on the proportion
of imports from the target. This import sanction reduces target
revenue as a mirror to export sanctions.
 OPEN ECONOMY MACROECONOMICS 91

Before we discuss the sanctions effects, we must build the basic

model. Focusing on the sender as the domestic country,8 total pro-
duction of good z, h(z), is equal to the sum of the consumption and
government spending on good z worldwide in equilibrium, equal to
y(z); aggregate supply (hours of work  h) equals aggregate demand
(y(z)). Government spending is assumed to be exogenous of house-
hold consumption decisions, and funded by period-specific money
supply (seigniorage profits) and taxes. In the steady state, government
spending will be equivalent to taxes alone under the assumption all
money held is spent in the long run.
Prices and consumption are based on a constant elasticity of substi-
tution (CES) between the domestic and foreign good.9 In their natu-
ral logarithms, the prices take an intuitive form in the aggregate from
their more complex forms in equations 5A.1 and 5A.2 in chapter five A.
When sanctions are placed on another country, the target’s propor-
tion of the world economy, mn, is greater than zero; when there are
no sanctions, the sender’s world collapses back to itself and the ROW,
with the ROW equal to (1n). The aggregate price faced by the
sender for the bundle of goods z and z* is this weighted sum, with E
as the price of sender currency in terms of the target and ROW
($/baht, e.g.), the exchange rate:

Pt  [npt(z)1  (m  n)(Et q*(z*))

t
1

 (1m)(Et q*(z*))
t
1 1/(1)
] (5.1)

qt(z) 1
t  [n(  (mn)(p*(z*)) 1
P* ) t
Et

 (1m)(qt (z*))1]1/(1) (5.2)

There are three prices for each good from equations 5.1 and 5.2, a
la PTM models. Price p is the domestic price of good z, the good
produced by the sender. Price p* is the domestic price of the target
good, z*. The sender charges a foreign price of q*, where sanctions
force a difference in the relative price of q* once policy is initiated
between the target and ROW. The elasticity of substitution between
the sender and foreign good within each country is . Notice that
prices in each country are simply the weighted sum of individual
goods prices in terms of the local currency, where variables with an
asterisk (*) are target variables. The ROW walks parallel to the target,
where two asterisks (**) are used for any initial differences. The
92 ECONOMIC SANCTIONS

consumption of each good is a function of the specific good’s price

relative to the country’s aggregate price.
Intuition: To this point we have simply described the pricing struc-
ture. We must now add both the asset markets that the household uses
to borrow or save and the firms that employ the households and pro-
duce the goods they purchase. The sum of those parts provides the
budget constraint for the household and shows where sanctions
effects enter and flow through the economies. This model is building
toward a general equilibrium where households, firms, and foreigners
interact such that sanctions can impose welfare costs on citizens of
both countries. Since the household can bridge between different
periods in consumption by borrowing or saving money, the asset mar-
kets also need an explanation.

Asset Markets in the NOEM Model

Financial sanctions have two forms. Credit sanctions reduce the
amount of borrowing or lending originating from the sender. Debit
sanctions reduce income flows from the sender to the target for past
borrowing. Financial sanctions affect household utility by changing
its planned budget and the intertemporal choices of both households
and firms. Households use asset markets to bridge between con-
sumption today and consumption tomorrow by borrowing (more
consumption today) or lending (saving today, consuming tomor-
row). We assume a bond exists to be bought and sold, and is a nom-
inal bond, denominated in the sender’s currency; there are distinct
markets for debt and goods such that borrowing is not repaid in
goods directly.10
It is important to recognize the intertemporal aspects and require-
ments for these sanctions to be effective and more than rhetorical. To
provide a credible threat of financial sanctions, a capital market must
exist between the two countries. Further, if debit sanctions are used,
the sender must have outstanding net borrowing owned by the target.
If the sender owns more assets in the target economy than the target
does in the sender economy, the threat of debit sanctions is not cred-
ible: the sender must expect retaliation if debit sanctions are used and
the target can credibly retaliate.11 However, for our purposes, debit
sanctions affect the income flows from past borrowing and credit
sanctions affect contemporaneous decisions concerning consumption
because they restrict the current ability to borrow or lend. Following
Mark (2001), the nominal interest rate (it) is simply the sum of the
 OPEN ECONOMY MACROECONOMICS 93

Pt1
real interest rate (rt) and the inflation rate , related in the following,
Pt
Fisher equation:

Pt1
(1  it)  (1  rt) (5.3)
Pt

Interest rate parity connects the nominal interest rates between the
two countries:

Et1
(1  it)  (1  i*)
t (5.4)
Et
There is a zero-net supply condition for bonds held worldwide. We
will assume for tractability that the ROW has net zero borrowing with
the sender and target, such that their balance of payments (BOP)
adjusts depending on the sender and target interactions. Equation 5.5
reflects this zero net supply, where B represents the net borrowing or
current account deficit of each country:

Bt  m n n B*  1 n m B** (5.5)

Expected goods consumption drives the borrowing decision; the

household first decides how much it wants to consume and borrows
or lends as needed. Nominal interest rates are determined in each
economy by assumption, where real interest rates adjust based on
changing prices and exchange rates. Credit sanctions are envisioned
where the sender is a net creditor to the target, placing the sender in
a relatively hegemonic position. By reducing the amount of credit
provided, the sender forces higher interest rates onto target borrowers.
This should reduce current consumption and utility.
Intuition: Financial sanctions affect both economies, but there are
preconditions that must be in place to credibly threaten and carry out
these embargoes. First, certain sanctions are ineffectual unless both
current and past decisions allow the sanctions chosen to take place. As
described in chapter two, the ability to sanction an economy depends
on the pre-sanction relationship between the sender and target. In the
earlier-mentioned asset markets, the target and sender households buy
bonds to bridge between the consumption of goods today and tomor-
row, the sanction period, and the new steady state. If the correct prior
financial relationships do not exist, financial sanctions are pure rhetoric.
94 ECONOMIC SANCTIONS

The financial markets may realize asset price inflation through

scarcity, coupled with goods market inflation. As credit is more dif-
ficult to find, interest rates rise. The goods and asset markets are
tied to each other through the BOP. Theory suggests that negative
overall welfare effects are potentially mitigated because of the
exchange rate effects that flow from this connection reversing initial
BOP imbalances. The following section describes the firms and
introduces how trade sanctions enter this model due to lost firm
revenue.

The Firm and Household Budgets

Firms sell their respective domestic product to both domestic and for-
eign citizens. The PTM idea begins with the firm, which prices differ-
ently to each market (Betts and Devereux 2000). In fact, the sender
government is in effect asking its firms to reduce sales to the target
under export sanctions. Wages are paid to households that supply
hours of work, and the wage is assumed to be sticky or nominally rigid
in the short run and flexible in the long run. The following equations
describe the firm’s profits assumed to transfer directly to households,
which own the firms.

t  pt(zt) . xt(zt) + Etq*(z

t
.
t) vt(zt) + Etq*(z
t
.
t) vt**(zt)  Wtht (5.6)

qt (z*)
*
t 
t
. x*(z
t *)
t + p*(z
t *)
t
. vt*(z*)
t
Et
+ qt (z t*) . x**(z
t t  W*h
*) t *t (5.7)

The first term in each equation is the revenue derived from selling
the domestic product in its home market. The second and third terms
are revenues from selling the domestic good in foreign markets.
Foreign demand is represented by the symbol v. The final term is the
cost of production, assuming the labor cost is the only expense.12
Sanctions affect these profit functions directly by reducing revenue
to firms: export sanctions affect the sender’s profits, while import
sanctions affect the target economy’s profits. By affecting these prof-
its, sanctions affect household budgets that rely on the firm’s dividend
payment for income. Wages are set to maximize profit based on the
hours hired by the firm.
The household budget constraint is a key equation in any economic
model; in this model, the budget transmits the sanction shock to the
 OPEN ECONOMY MACROECONOMICS 95

household. Equations 5.8 and 5.9 represent the sender’s and target’s
consolidated budget constraints with the sanction parameters.

Ct  Ωt  Bt1  Bt  Gt  EX  FS (5.8)

n  Bt  Bt1
t  Ω*
C* t   G*
t  IM FS (5.9)
1n Et

Ω represents firm revenue. The sanction parameters show the

reduction in firm revenue (EX, IM) and the credit market shock
(FS), where EX  export sanctions, IM  import sanctions, and
FS  financial sanctions. Consumption and taxes are uses of funds,
while firm profits and household wages plus government spending
round out the funding sources. Notice borrowing exists in both
equations, specific to the economy; the sender chooses the credit
market shock that directly affects both economies. Economic coer-
cion affects the optimal decision-making of sender and targets as
follows:

EX  mEX[Etq*(z
t t)
. vt(zt)] (5.10)

IM  nIM p*(zE *)  x*(z*)

t

t
t
t t (5.11)

FS   (mCR . Bt + mDB . (1+r)Bt1) (5.12)

where FS  CR  DB.

These shocks force deviations from the initial steady state equilib-
rium, where mEX, mCR, mDB (0, mn) and nIM (0, n). The sanc-
tion’s effects are discussed in more detail later as the steady state is
derived explicitly. From there, this model concludes with solving for
welfare deviations for each country in the long run, the post-embargo
steady state.
Intuition: These budget constraints show that the target house-
hold’s ability to consume is based on specific sources of income that
sanctions aim to reduce. Sanctions are meant to affect these household
budgets. The intertemporal structure of these models is necessary to
understand how using economic statecraft can have effects on choices
today and tomorrow through sanction effects on both goods and asset
96 ECONOMIC SANCTIONS

markets. Financial sanctions are potentially the sender’s most potent

weapon, and may be one of the reasons why the historic trend in sanc-
tions has shifted toward financial embargoes and fewer stand-alone
trade sanctions. How the target can shift consumption between the two
time periods is a function of how available worldwide credit is to the tar-
get’s households. The mere threat of financial sanctions may persuade
certain targets to acquiesce simply because they will lose relative stabil-
ity in consumption and welfare. The consolidated budget constraint is
where the labor, goods, and asset markets all merge.
The overall economic effects of sanctions are dependent on three
major factors: (1) the connectivity between the sender and target (the
size of parameter mn); (2) the elasticity of substitution between the
sender and target goods (); and (3) the way the sanctions either
accommodate or mitigate each other through subsequent exchange
rate fluctuations. The next section describes the sanction transmission
process as the household optimizes its utility.

Steady State, Economic Statecraft, and Welfare

Redistribution
The Household’s Optimization Problem
Households consume, demand money, and supply work effort. Work
effort is measured in work hours, where one hour is equal to one unit
of output. The utility function of each household is the sum of pres-
ent discounted value of utility derived from choosing consumption (C),
nominal money demand (M), and work effort (h):

 lnC   
 Mtj 1
2
max Ut  j
  h tj (z) (5.13)
Ct,Mt,yt j0
tj
1 Ptj 2

As consumption grows, utility grows; we assume logarithmic pref-

erences to simplify the analysis later.13 The more the money that is
demanded, the more the utility derived by the household. This
assumes the household gains satisfaction from holding money.14
Finally, the negative sign on work effort signifies that as the household
increase its hours of work supplied to firms, its utility falls. Thus,
for sanctions to be economically effective, these policies must affect
the target’s utility negatively in sum.15 NOEM models investigate
macroeconomic shocks and their effects on the steady state, the
long-run equilibrium of each nation. The adjustment from the
 OPEN ECONOMY MACROECONOMICS 97

sanction period (the “short run”) to the long run depends on many
factors.
For the sanction model, the sender does not want policy neutrality
or a situation where the short-run effects are negated by long-run
adjustments. How the original versus the post-policy steady states
compare is the welfare shock from policy. In macroeconomic policy-
making, policy neutrality is a good thing for a foreign country as a
result of domestic monetary or fiscal shocks; neutrality is, in fact, the
sought-after result to minimize differential effects on other countries.
If policy neutrality from economic statecraft takes place, a lack of
long-run potency builds negative reputation effects for the sender, as
discussed in chapter three.
The steady state equations define the pre-sanction setting, the
benchmark for comparison to the post-sanction economics of each
country. The budget constraints, the labor supply conditions from the
optimal solutions to the household’s problem, along with the demand
for each good in terms of consumption and government spending,
define the steady state. Once steady state solutions are found, we can
then discuss sanctions and their predicted effects. The exchange rate
fluctuations that result from sanctions may force welfare costs where
the relative magnitude is based on country size, sanction mix, and
sender economic hegemony over the target.
The next two issues are related and mathematically intensive, where
the technical portion is relegated to chapter five A. First is solving for
the sender’s first-order, optimal conditions to maximize household
utility; for the sender, this is optimizing equation 5.13 subject to the
budget constraint equation 5.8. The household chooses consumption
(through optimal borrowing choices), money demand, and work
effort to maximize utility over all time periods. Intertemporal con-
sumption links the short run to the long run, and the choice of bond
holdings by the sender economically links the belligerent nations.
Equations 5.14–5.16 show the sender’s optimal, Euler conditions;
the target’s equations are analogous and in chapter five A.

Consumption: Pt1Ct1  PtCt (1  it) (5.14)

Money demand:
Mt
Pt


C
1 t  (5.15)

Wt
Work effort: ht(z)  (5.16)
PtCt
98 ECONOMIC SANCTIONS

Simply put, these conditions equate the marginal benefit and

marginal cost of choosing levels of consumption, money demand, and
hours of work effort for the household. Sanctions aim to shock
these choice variables through shocking the household budget.
Equation 5.14 tells us that the present discounted value of consumption
tomorrow is equal to consumption today, as we assume that the
representative individual wants to smooth consumption over their
lifetime. Equation 5.15 tells us that money demand today is equiva-
lent to the nominal cost of goods given up by holding money rather
than spending it. Equation 5.16 equates the amount of hours worked
by the individuals to the value of goods lost if those hours were not
worked.16

The Steady State with Sanctions

The steady state is when percentages changes in the endogenous vari-
ables are equal to zero simultaneously. The percentage changes are the
time derivatives of the original variable’s natural logarithms that lin-
earized the original equations. This process and subsequent equilib-
rium provides a starting place for the embargo’s policy shock.
Equations 5A.29 through 5A.35 make up the initial steady state equa-
tions, including the potential sanction effects, defining the endoge-
nous variables. Looking at the budget constraints in the steady state
provides intuition as to how both trade and financial sanctions enter
each economy in Equations 5A.33 and 5A.34.
Budget constraints: Sanctions reduce household budgets by reducing
income derived from ownership of the firm during trade sanctions and
reduced funds availability during financial sanctions.
In other versions of this model, two sources of ambiguity over the
final model’s effects are expectations and policy duration. One version
of shocks is where policy is expected, both in timing and magnitude.
There is another where the policy’s timing or magnitude or both are
not expected. Sanction policy is likely to be anticipated in the real
world. Even U.S. decisions by its president to initiate sanctions take
time to begin, and it is conceivable that most targets know the sanc-
tion will be the international reaction or have time to react to an
embargo’s announcement through the media. We focus here on just
anticipated economic statecraft, where the target’s uncertainty lies in
magnitude and duration. In the general PTM version of Betts and
Deveraux (2000), if policy is anticipated and the effects are temporary,
then policy neutrality results from statecraft; neutrality comes from
 OPEN ECONOMY MACROECONOMICS 99

adjustments to the anticipated policy, specifically through exchange

rate adjustments in NOEM models.
In the sanctions model, policy is generally non-neutral as shown
later. The effects on the steady state may be permanent regardless of
the sanction’s duration, as exchange rates may not react enough to
nullify short-run shocks because of market frictions, thus welfare
reductions result for both countries in the long run. The change in
exchange rates and bond holdings help determine the welfare effects.
Intuition: Most papers and studies involving NOEM models claim
there is little to no intuition in the steady state results’ derivation.
There is, in fact, some intuition here. This section, especially its mir-
ror in chapter five A, outlines household decision-making, its links to
the economy’s general equilibrium, and the sender–target relationship
when no shocks take place in the economy. The model assumes
households decide on a level of consumption, which implies net sav-
ings through bond holdings or borrowing if net savings in negative,
based on the household assessment of lifetime income. Money
demand is driven by consumption in subsequent time periods, as is
work effort. The major piece of intuition is when deriving the steady
state is recognizing that the changes to consumption choice drive the
resultant welfare effects.
There are three macroeconomic markets: goods, labor, and asset.
Each of these markets is affected by sanctions, and their changes in
combination affect welfare. This is the other piece of intuition: the
steady state conditions define how each type of sanction enters each
economy. If we assume the only shock in the current period is an eco-
nomic sanction, then defining and solving for the steady state shows
how an economist can watch economic coercion evolve from post-
policy, macroeconomic effects on household utility. From the data
presented earlier, we see that the marginal benefit of each choice vari-
able equals its marginal cost when utility is optimized. The terms of
trade, the ratio of the export to import prices, must be quickly men-
tioned here. The effect on the exchange rate helps determine the
terms of trade shock.

 
1
n Etq*(z)
t
TOTSender  (5.17)
(m  n) qt(z*)

When thinking about the damage caused by sanctions to the sender,

it is convenient to think or relative export prices, as economic statecraft
100 ECONOMIC SANCTIONS

may cause the sender to lose comparative advantage, reducing exports

otherwise, and may lead to lower unemployment. Those are large
intellectual jumps to make, but theory does suggest that a deterioration
of the terms of trade may lead to more unemployment.17

The Welfare Effects of Economic Sanctions

In the NOEM literature, there is a split between models that have
welfare effects and those that do not. Under certain conditions, wel-
fare in each country remains intact after a policy shock, or is neutral,
due exchange rate fluctuations that result naturally from market forces
reacting to policy. In the PTM model, generally, welfare remains
intact when new policy takes place because the exchange rate fluctu-
ates to equate policy-driven changes in prices; in the short run, pur-
chasing power parity (PPP) is violated and restored with a resultant
exchange rate fluctuation.
The derivation of sanction effects on welfare in this model is shown
in chapter five A. These derivations split the welfare effects into both
long- and short-run effects. In the sanction version of this model,
sanction effects may be neutral due to ambiguous effects on exchange
rates. To understand the sanction effects, we must first look at the
welfare components of this model and see how sanctions may change
them. Both countries’ utility functions are affected by sanctions, and
the sanction effects are then affected by the parameter values.
Tables 5.1, 5.2, and 5.3 show these effects from the model’s derivation
in chapter five A.
Table 5.1 provides a brief overview of changes in welfare under dif-
ferent sanction mixes, given the NOEM model of the target economy.

Table 5.1 Sanctions effects on utility functions

EX IM CR DB

Uc    
Uh    
U  ? ? ?
Ê t   0 0
TÔTSender   0 0
Uc*    
Uh*    
U* ? ?  

Note: Assumes the following parameter values: n  0.3,  4,

  2,  0.95.
 OPEN ECONOMY MACROECONOMICS 101

Notice that full trade sanctions have ambiguous effects on target

utility. This is expected, and is likely why trade sanctions have mixed
results in practice. Basic trade theory suggests world welfare is unam-
biguously reduced by barriers such as quotas. While that still hold true
here, the wealth and welfare redistributions may be ambiguous.
Notice financial sanctions alone have unambiguous effects on the tar-
get and those effects are welfare-reducing changes; in practice, these
measures have become the economic statecraft tool of choice.18
Welfare increases as a result of export sanctions through con-
sumption, while falling under import sanctions. This makes intuitive
sense, as lower domestic prices under export sanctions become
larger consumer surplus in the long run, where welfare is redistrib-
uted toward the household; in the import sanction, which acts like a
classic quota, the sender household loses because of relatively higher
import prices.
The terms of trade and exchange rate effects follow this logic. The
exchange rate rises from export sanctions, falls from import sanctions,
and the terms of trade follow. The intuition here is simple. Export
sanctions reduce the worldwide demand for the sender’s goods by the
sender’s own hand, and thus reduce the sender’s currency demand.
This causes the target’s currency to appreciate, the sender’s to depreci-
ate, and the exchange rate as defined overall to rise (sender per target
currency). The terms of trade follow behind: as the sender restricts
exports, the export price rises (not necessarily by the same magnitude as
the domestic market’s price because of the PTM assumptions) versus
the import price. In this way, the sender pays for export sanctions
through deteriorating terms of trade and currency values, while experi-
encing increased consumer surplus. The reduction in utility from
reduced leisure is due to reduced budgets to spend on leisure activities,
and may or may not be offset by the gains in consumption from the
import sanctions, which have the opposite effects.
Financial sanctions are ambiguous in their effects on senders
because policy restricts their own ability to borrow or lend as it
restrains the target. This restriction affects consumption in a positive
way, leisure in a negative way, leading in sum to ambiguity. For the
target, the financial sanction is unambiguous, reducing welfare from
both consumption and leisure. Fewer options and reduced income
from savings is like a shift in the budget constraint toward the origin
that reduces the target’s ability to consume both goods and leisure.
This lack of ambiguity makes financial sanctions theoretically attrac-
tive; we see in chapter six that the empirical results on financial
sanction effects are less convincing.
102 ECONOMIC SANCTIONS

One result that remains from the original PTM assumptions is that
the financial sanctions have no effect on exchange rates and the terms
of trade. Because financial sanctions are reductions in the current sup-
ply of financing, and must still follow the bond’s zero-net supply
condition in the long run, the exchange rate change simply restores
interest rate parity. The terms of trade effects from financial sanctions
follow suit. The magnitudes of any sanction mix are affected by the
model’s parameters changing, as table 5.1 reflects just one vector of
parameter choices. Table 5.2 and 5.3 reflect sensitivities to changing
parameters in both economies.
Tables 5.2 and 5.3 summarize changes in sanction magnitude as
the parameters change. Notice in table 5.2 that as the sender’s size
increases (n ↑), the effects on all sanctions change but the exchange
rate does not change. This may seem like a confounding result, as
hegemony is logically a major factor in a sanction’s effectiveness. One
way to interpret this result is that the relatively larger the sender, the
more exposure it has to the risks of voluntarily restricting economic
flows with another economy. This is reflected in reverse for the target
in table 5.3, outside of import sanctions.
An important result is the shock’s magnitude being influenced by
alternative market availability. At high values of  (many substitutes),
export sanctions strengthen while import sanctions weaken. The exis-
tence of willing and low-cost alternative markets is truly the bane of
sanction policy for the sender, shown by the elasticity of substitution, .
Alternative markets change the basic assumption upon which sanc-
tions rely to provide credible threats to intransigent target nations: the
more the substitutes available, the more the target can circumvent a
sanction’s effects.
The final parameter changes reflect that as the rate of discount and
the elasticity of money demand rise, trade sanctions follow the elasticity

Table 5.2 Sanction effects sensitivity to parameter

changes: sender

EX IM CR DB Ê t

n     0
     
    
1/     

Note: These results reflect how the welfare shock is exacerbated

(), mitigated (), or not affected by a change in the parameter,
ceteris paribus.
 OPEN ECONOMY MACROECONOMICS 103

Table 5.3 Sanction effects sensitivity to

parameter changes: target

EX IM CR DB

n    
    
   
1/    

Note: These results reflect how the welfare shock is

exacerbated (), mitigated (), or not affected by a
change in the parameter, ceteris paribus.

of substitution between the two bundles of goods. At high rates of

discount and low elasticities of money demand, there will be an incen-
tive for each economy to save and hold more money. If the sender
restricts exports or financing to the target, there is a lower demand for
the more valuable money and financing, increasing sender welfare to
the target’s detriment. Import sanctions have the opposite effects.
The NOEM model provides a literature-based framework for sanc-
tions, and also delivers insight for the perceived economic ineffective-
ness of sanctions. Theoretically, sanctions must force currency
depreciation, inflation pressure, and real resource losses for the target.
If it can achieve unambiguous currency depreciation, theory suggests
that target entities suffer economically. The export and financial
sanctions are similar in their effects on each economy.

Conclusions
The NOEM sanction model provides three basic conclusions about
economic statecraft as macroeconomic policies. First, sanctions con-
flict with each other in certain cases. Using an export and import sanc-
tion simultaneously conflict concerning welfare changes, while
financial sanctions used alone are unambiguous in their effects.
Second, certain parameters are vital to sanction potency. The elasticity
of substitution dictates how easily the target can shift purchases and
sales under sanctions to other countries and away from the sender.
However, the sender’s hegemony over the target may play only a small
role, which provides more theoretical foundation for pundits of mul-
tilateral or universal sanction efforts.
Third, sanctions have similar effects as other macroeconomic poli-
cies, and policy makers should expect the exchange rate to experience
post-policy movements in response to new market forces. Table 5.1
104 ECONOMIC SANCTIONS

summarizes the sanction effects on welfare, while tables 5.2 and 5.3
map sanction potency at different levels of the model’s major param-
eters. Economic theory tells us that changing trade or financial flows
causes exchange rate shocks that may balance these flows between two
countries. However, policy makers continue to use sanction mixes in
which ambiguity may exist. While the number of sanction tools is
small, the sanction’s magnitude may be varied. Moreover, if the
sender economy lacks the hegemony to affect the target’s trade and
financial flows to credibly pursue the stated sanction goal, sanctions
may be less potent as tools of statecraft.
Bridging between these economic effects and political goals of sanc-
tions is the policy’s ultimate task. If the stated goal is to economically
punish the target sovereign and populace for deviant political actions,
sanctions can easily achieve such a goal. Ultimately, nations must craft
their diplomacy in such a way as to provide a credible threat of eco-
nomic damage or sanctions are useless. For both policy makers and
economists alike, consensus around what sanctions are meant to do is a
large step forward. The challenge all sanction studies face is to describe
how much shock is enough to reverse a target’s deviant action.
Further research should be done on how targets can use domestic
monetary and fiscal policies as reactions to reduce sanction problems.
The fact that these policies can actually help reduce sanction effects
could be a major step forward in sanction research. Since financial
sanctions have unambiguous results on target welfare while using
every tool in the sender’s arsenal has ambiguous effects, the sanction
mix may attract third markets because of profit opportunities that
senders hand out to noncooperative nations like candy to children. In
chapter six, we examine econometric models of these relationships
and those implied by humanitarian and political effects of these
economic shocks.
 Chapter 5A

Mathematical Derivations
of NOEM Sanctions Model

T his chapter focuses on the equations and derivations discussed in

chapter five. New Open Economy Macro (NOEM) models are math-
ematically intensive and tedious to derive. However, in taking the
steps to solve this model, the way sanctions enter and affect the econ-
omy is illuminated. Empirical implications of this model are discussed.
Any errors or omissions that lead to confusion are my own doing and
do not reflect any other author’s work. Table 5A.1 provides a symbol
list for the equations in chapters five and five A.

The Basic Set-Up

Most of this chapter is applied mathematics, delivering insight as to
how macroeconomists can inform policy makers about economic
statecraft’s effects. This model, as stated in chapter five, provides
economists with a way of tracking economic coercion’s welfare effects
on both the target and sender economies.1 The target country is seen
as the foreign country, the sender is the domestic country, and the rest
of the world (ROW) is the foreign sector beyond the target when
economic statecraft is used. World production is of two goods,
domestic and foreign in terms of the sender, delineated by z and z*.
We can think of z and z* as baskets of goods consumed in the sender
and target economies; thus each country produces a unique basket of
goods. This dichotomy allows for exports and imports to be different
goods in each country. In the pricing-to-market (PTM) version of this
model, each country sets its own price because monopolistic compe-
tition exists for each country over their respective good. The ROW
acts as the Walrasian clearinghouse for consumption and production,
taking prices as given by the sender and target.
106 ECONOMIC SANCTIONS

Table 5A.1 Symbol list for sanctions NOEM model

p Price of sender good in sender economy

q Price of target good in sender economy
q* Foreign price of target good
p* Foreign price of sender good
E The exchange rate
z/z* Sender/target good
P/P* The sender’s/target’s price index
U/U* The sender’s/target’s utility
C/C* Sender/target consumption
M/M* Sender/target money supply
 Multiplier of money demand in utility
1/ Consumption elasticity of money demand
 Elasticity of substitution between the sender and target goods
mn Proportion of target of world economy
n Proportion of ROW in world economy
 Work effort multiplier
 Present value discount factor
y/y* Aggregate demand
CW World consumption
GW World government spending
G/G* Sender/target government spending
W/W* Sender/target wages
/* Sender/target profits
R/R* Nominal interest rates in sender/target economy
B/B*/B** Bonds held by the sender/target/ROW
x/x* Sender/target demand for sender good, z
v/v* Sender/target demand for target good, z*
h/h* Hours worked by sender/target household y/y*
EX Export sanction parameter
IM Import sanction parameter
FS Financial sanction parameter
mEX The percentage of sender trade involved in export sanctions
nIM The percentage of sender trade involved in import sanctions
mCR The percentage of sender financial flows involved in credit sanctions
mDB The percentage of target income involved in debit sanctions

The Goods Market

In equilibrium, consumption and government spending in each coun-
try are the aggregate expenditures. We take government spending in
each country to be exogenously determined. Consumption is a func-
tion of prices, where each good’s price is preset for each country-
specific good in terms of their currency. For example, good z is
purchased in the sender economy at price p(z). Good z is sold to the
target economy and ROW at target currency price q*(z). The same
 NOEM SANCTIONS MODEL 107

Sender Target ROW

Imports

v(z*) v*(z) v**(z)

x(z) x*(z*) x**(z*)

Domestic

0 n m 1

Figure 5A.1 Demands for z and z* in each nation.

holds true for the prices of good z*: good z* is sold to the sender
economy and ROW at q(z*) in terms of the sender’s currency, and
purchased domestically at p*(z*) in the target. Figure 5A.1 illustrates
the world demand for each good, each country’s domestic and import
consumption.
The sender restricts only a portion of its foreign households and
suppliers. The world demand for sender output is thus in three parts,
based on domestic and foreign consumption, where n represent the
sender’s size in terms of the world economy; m  n represents the tar-
get’s relative size. The ROW is the remainder (1  m).
Goods prices are aggregated in a constant elasticity of substitution
(CES) form, where the elasticity of substitution between z and z* is
equal to  1. The price index illustrates the CES functional form that
follows:
1

p (z) (q (z*)) (q (z*))

n m 1


1
1 1 1
Pt t dz t dz t dz (5A.1)
0 n m

(q*(z))  (p*(z*))
n m
1 1
P*t t dz t dz
0 n

 (q*(z))
1


1
1
t dz (5A.2)
m

This equation in its natural logarithm appears as equations 5.1 and

5.2 in chapter five. Since price is a CES function, consumption is also
108 ECONOMIC SANCTIONS

a CES form. The sender and target consumption levels as functions of

price are defined as follows:

p(z)
P  q(zP ) qP(z)* 
 *  * 
c(z) C; c(z*) C; c*(z) C*;

 
p*(z*) 
c*(z*) C*
P*
We assume aggregate demand equals aggregate supply for each
good. Total production of each good is equal to y(z) and y*(z*),
respectively. The demands for each good are symbolized in figure 5A.1.
They are as follows for the sender and target, respectively:

x(z) n · y(z); v*(z) (mn) · y(z); and v**(z) (1 m) · y(z).

v(z*) n · y*(z*); x*(z*) (mn) · y*(z*); and x**(z*)
(1m) · y*(z*).

The PTM model assumes that prices are set in advance of policy.
The sanction’s imposition, as we see later, initially affects firm rev-
enues and household budgets. The next section describes the asset
market that links the countries today and tomorrow through interna-
tional borrowing and lending.

The Asset Market

The asset market defines how bonds are allocated among the countries,
where Bt is the value of assets purchased by the economy in time t, a
measure of net savings.

nBt (mn)B*
t (1m)B**
t 0 (5A.3)

This sum is equal to zero because it represents a global market for

assets where there is a zero-net supply: if Bt 0, the sender is lending
to foreigners in net. The sender currency price of these one-period
bonds is normalized to 1  , where i is the sender’s nominal
1 i
interest rate.
An interest parity condition links the domestic and foreign interest
rates. Assuming uncovered interest parity, the sender’s nominal interest
rate is equal to the foreign nominal rate and the expected appreciation
in nominal exchange rates, where the nominal interest rate is determined
by the Fisher relationship. Both are described in equations 5.3 and 5.4
 NOEM SANCTIONS MODEL 109

in chapter five. The exchange rate in this model is defined as the num-
ber of sender currency units per target currency to match other NOEM
extensions.2 The bond’s availability allows households to borrow or
lend to allocate consumption between the short and long runs in order
to maximize their lifetime utility. The next section details the firm and
sets up the budget for the household’s optimization problem.

The Firms
The firms earn profits from the sale of goods they produced to all
countries, less wages paid to domestic households. Since households
are the firms’ owners, profits are assumed to be distributed to the
households. The sender and target profit functions are:

t pt(zt) . xt(zt) Etq*(z

t
.
t) vt(zt)

Etq*(z
t
. t *(zt)  Wtht
t) v* (5A.4)

qt(z*)
*
t
t
. x*(z
t *)
t p*(z
t *)
t
. v*(z
t *)
t
Et
qt (z*)
t
. x*t *(z*)
t  W*h
t *t (5A.5)
The firm uses one hour of labor to produce one good. Firms employ
only labor in each country3: y(z) h(z) and y*(z*) h*(z*). Firm
also sets prices such that price is equal to marginal cost to maximize
profit:


pt (z) W (5A.6)
1 t

This is true for the target’s prices also:


p*t (z* ) W* (5A.7)
1 t
qt(z*)
t
where price pt(z) Etq*(z
t t) and p*(z*)
t in the sender econ-
Et
omy in terms of target currency.
Aggregate demand is a combination of household and government
spending of goods z and z* from total world production. Aggregate
demand equals aggregate supply (aggregate work effort) in equilib-
rium, where aggregate demand is simply a country-specific proportion
of world consumption and government spending.
110 ECONOMIC SANCTIONS

y(z) x(z) v(z) v**(z) (5A.8)

y*(z*) x*(z*) v*(z*) x**(z*) (5A.9)

The sender’s good, z, is purchased by domestic consumption, x,

and exports, v and v**.

 
pt(z) 
xt nCt (5A.10)
Pt

E q*(z*)

t t
vt (m  n)Ct (5A.11)
P t

 
Etq*(z*) 
t
v**
t (1  m)Ct (5A.12)
Pt

The target economy has similar demand functions for good z*,
where domestic consumption is v*, and target exports are x* and x**:

qE*(z*)
P* 

t
x*t nC*t (5A.13)
t t

 
p*(z*) 
v*t (m  n)C*t (5A.14)
P*t

qE*(z*)
P* 

t
x** (1  m)C*t (5A.15)
t t

The ROW takes form if m  n 0; when m 0, there are no

sanctions. While one-period bonds are available, the government
finances itself through taxes and seigniorage, such that the fiscal
budget constraint is as follows:

Mt  Mt1
Gt Tt (5A.16)
Pt

Trade sanctions affect the firm’s revenue in each country: export

sanctions reduce sender revenue and import sanctions reduce target
revenue. Further, financial sanctions affect wealth redistribution from
previous financial flows and credit availability in these budgets. The
representative consumer faces the following, initial budget constraint
 NOEM SANCTIONS MODEL 111

in terms of real consumption, repeated here from chapter five for

convenience:

Ct Ωt Bt1  Bt  Gt EX FS (5A.17)

n  Bt  Bt1  G*
C* Ω*t IM (5A.18)
t
1n Et t FS

where Ω pt(zt) . xt(zt) Etq*(z

t
.
t) vt(zt) Etq**(z
t
.
t) vt(zt), and

qt (z*t)
Ω* . x*t (z*)
t p*(z
t *)
t
. v*t (z*)
t qt (z*)
t
. x**(z
t *).
t
Et

The sender’s net sources of income to use in consumption include

the revenue derived by the firm, Ω current wealth augmentation,
Mt1 Bt1; less wealth taken to period t 1 Mt Bt; and
current taxes paid, Tt. The same sources exist for the target.
Government spending funds itself from the sum of taxes and the net
money supply change through seigniorage, and that bonds are in net
zero supply.

The Household’s Optimization Problem

The utility function is a linearly separable, discrete time model, where
the choice variables are consumption (bonds), nominal money
demand, and work effort. While most interpretations of the NOEM
model use an unconstrained version of this optimization problem, it is
actually more convenient here to use a constrained version because
sanctions act on the budget constraints first and foremost. The indi-
vidual’s lifetime utility is, following both Fender and Yip (2000) and
Mark (2001),

  lnC   
  Mt j
1  2
Ut j
 h (z) (5A.19)
j 0
t j
1   Pt j 2 t j

  lnC*   
  M*t j 1 
U*t j
 (h*t j(z))2 (5A.20)
j 0
t 1
1   P*t j 2

where 1
 is the consumption elasticity of money demand, and , 
are parameters of the marginal utility derived from money holdings
and work effort, respectively.
112 ECONOMIC SANCTIONS

Each household maximizes their lifetime utility, subject to

constraints 5A.17 and 5A.18, respectively. The next section explains
the model’s optimal conditions and their derivation.

The Model’s Euler Equations

The following Euler conditions, the equality between marginal
benefit and marginal cost from the earlier-mentioned constrained
optimization problem, show the tradeoffs households face in choos-
ing the optimal amount of bonds, money demand, and work effort in
hours.4 These are repeated from chapter five here because of their
importance.

Consumption: Pt 1Ct 1 PtCt (5A.21)

 

Mt 
Money demand: C (5A.22)
Pt 1 t

Wt
Work effort: ht(z) (5A.23)
PtCt

The target’s equations are analogous, where is multiplied by the

ratio of the exchange rate tomorrow, Et 1, and the exchange rate
today, Et, and put asterisks on all the variables, but not the parameters:

Et 1
Consumption: P* C* P*t C*t (5A.24)
Et t 1 t 1

 
1
M*t  
Money demand: C*t (5A.25)
 
P*t Et 1
1
Et
W*t
Work effort: h*(z) (5A.26)
t
P*
t C*t

These conditions are derived by taking the partial derivative with

respect to each choice variable in the utility function, and equating the
result with the respective partial derivative from the consolidated
budget constraint multiplied by the Lagrange shadow price, t, for
each time period. In the case of consumption, the time periods are
 NOEM SANCTIONS MODEL 113

linked by bonds (net savings), thus the household is implicitly

choosing bonds to distribute consumption between the two periods.
Equation 5A.24 is the present value of consumption tomorrow
equal to the present value of consumption today. Using  t
PtCt
and t 1 , replace the Lagrange multiplier in the first-order
Pt 1Ct 1
conditions for money demand and work effort to get the results given
earlier. The target’s conditions are found in similar ways. Interest rate
parity allows us to eliminate the target’s interest rate from the analysis
and focus on the sender’s nominal interest rate and the expected
change in the exchange rate. Any fluctuation in the exchange rate
adjusts the local price of imported goods.5 The interactions between
the sender and target, again assuming the ROW follows Walras’ Law,
determine general equilibrium in this model. The starting place of
measuring changing welfare, and its resting place after the shock has
played itself out, is called the steady state.

The Steady State Derivations with Sanctions

Trade sanctions enter the model here, as a shock to the availability of
good z in the target economy (export sanctions) and good z* in the
sender economy (import sanctions). We assume the sender makes pol-
icy choices exogenous of price. Credit and debit sanctions are shocks
to the wealth derived from credit availability through bonds (credit
sanctions) and the income or costs from current bond holdings (debit
sanctions); in sum these are the financial sanctions.
The steady state provides two key insights concerning the conclu-
sions of this model. First, it defines the “0-steady state” equilibrium.
This starting place helps define a long-run benchmark for responses to
shocks, the economic adjustment toward a new steady state after a
one-period shock. To solve for the steady state, the convention is to
take the natural logarithm of the variables in key equations, then take
their derivative, essentially log-linearizing around the optimal condi-
tions. This time derivative is now the percentage change in the vari-
able, as a function of other percentage changes. NOEM models
generally begin these derivations with the aggregate price level in the
steady state, from equations 5A.2 and 5A.3 given earlier:

P̂t np̂t (m  n) q̂ t (1  m) q̂*t *(z*) (1  n)Ê (5A.27)

P̂ *
t n(q̂t  Êt ) (m  n) p̂*t (z*) (1  m) p̂**(z*)
t (5A.28)
114 ECONOMIC SANCTIONS

Variables with a “^” on them represent the time derivative of the

original variable’s natural logarithm, the percentage change in the
variable. The percentage change in the aggregate price level is the lin-
ear combination of percentage changes in the prices of individual
goods z and z*. Notice that the exchange rate change is also a part of
these equations.
The following seven equations are made up the endogenous variables
in the steady state, our variables of interest. Assuming h(z) y(z) in
equilibrium:

Labor Supply:

ŷt (z) ˆt
p̂ t (z)  P̂t  C (5A.29)

ŷ*t (z) ˆ C*

p̂*t (z)  P* ˆ (5A.30)
t t

Aggregate Demand:

ŷt (z) ˆw
(p̂t (z)  P̂t )  C (5A.31)
t

ŷ*t (z) ˆ wt
 ( p̂ *t (z) P̂*t ) C (5A.32)

Budget Constraints:

Ĉ t p̂t (z)  P̂t ŷt (z)  b̂ ˆ

 FS ˆ
 EX (5A.33)

b̂ FS  ˆIM

P̂*t n ˆ
Ĉ *t p̂ *(z) ŷ *t (z) (5A.34)
t
1n
ˆw
C n [ p̂ t (z)  P̂t ŷ t (z) ] (1 n)
t

[ p̂*t (z*)  P̂*t ŷ*(z*)

t ] (5A.35)

These seven equations define the sender’s and target’s steady state,
where Walras’ Law determines the ROW’s steady state. The labor
supply equations flow from the Euler labor supply rules given earlier,
equations 5A.23 and 5A.26. The aggregate demand equations flow
from the world demand for the sender and target good, respectively.
The budget constraints are derived from equations 5A.17 and 5A.18;
the percentage change in world consumption, Ĉwt , is derived from the
sum of the proportional changes in consumption for the world
economies. Sanctions enter these steady state equations through the
linearized budget constraints, equations 5A.33 and 5A.34.
 NOEM SANCTIONS MODEL 115

The solutions of this seven-equation system provide the steady-state

equilibrium of the seven endogenous variables: ŷ(z), ŷ*(z), p̂(z)  P̂,
p̂*(z)  P̂, Ĉ, Ĉ*, and Ĉ w, with the potential sanction shocks.

Steady State Solutions

The following equations provide solutions to the endogenous vari-
ables given earlier, and begin our ability to analyze the welfare effects
of sanctions on both countries. The time subscripts are dropped as
these describe a snapshot of each macroeconomy after a policy shock.6
Note that the percentage change in bonds held is a key variable
remaining to be solved. Its solution in the steady state determines the
equilibrium values of the other endogenous variables. This under-
scores the consumption-savings decision’s importance in maximizing
utility, given the parameter values of the utility functions and how
these variables relate to one another.

1[ b̂  ˆ
ŷ(z) FS  ˆEX] (5A.36)
2

ŷ*(z) 1
2 1n 
n ( b̂ ˆ
FS) ˆ
IM (5A.37)

p̂(z)  P̂  1 b̂
2  (1  n)(ˆ IM  ˆEX)

ˆ (2n 1)
FS  (5A.38)

p̂*(z)  P̂* n
2(1  n)b̂ ( ˆIM  ˆEX)(1  n)

ˆ (2n 1)
FS  (5A.39)

Ĉ  1 (1
2  )b̂  (1  n ) ˆEX

(1  n) ˆ IM ˆ (2n  1  )
FS  (5A.40)
116 ECONOMIC SANCTIONS

Ĉ *
(1 )
2(1  n)
b̂  ( ˆIM  ˆEX)(1  n)

ˆ (2n 1)
FS  (5A.41)

Ĉ w
2 
 1 n( ˆEX ˆ )
FS 
(1 n) ˆIM (5A.42)

To begin the welfare shocks analysis, subtract 5A.41 from 5A.40 to

solve for the steady state change in relative consumption as a function
of the percentage change in sender bond holdings and sanctions:

Ĉ  Ĉ * 
(1 )
2  (1  n) 
b̂ ˆ (2n  1)
FS

(ˆEX  ˆIM) (1  n)  (5A.43)

We call
(1 )
2  (1  n) FS 
ˆ (2n  1) ( ˆEX  ˆIM)(1  n)
 S1
from here. The steady state change to the exchange rate is an implication
of purchasing power parity (PPP):

P̂ P̂* Ê M̂  M̂*  1
 (Ĉ  Ĉ*) (5A.44)

From equations 5A.43 and 5A.44, the steady state change in the
exchange rate is a function of sanctions. Much like other macroeco-
nomic policies, sanctions force a deviation from PPP in the short run
(t), to be restored somewhat or completely in the long run (t 1).
Assume prices are set in time t due to the PTM assumptions. This
assumption reflects the pricing decision’s timing, as prices are initially
known when policy is made. The money demand functions are:

ˆ 1 Ĉ  ˆ
M  t (5A.45)
(1  ) t

ˆ* 1 Ĉ*  ˆ E
ˆ ]
M  t (1  )
[ˆt E t (5A.46)

Subtracting 5A.46 from 5A.45 isolates the differential between the

steady state changes to the exchange rate and the short-run change.
 NOEM SANCTIONS MODEL 117

This is where the potential overshooting result of Dornbusch (1976)

takes place in similar NOEM models (Mark 2001, 234).

1 ( Ĉ  C
ˆ *)   ˆ )
M̂ t  M̂ *t  (Ê  E (5A.47)
t t
(1  ) t

Setting equation 5A.47 equal to equation 5A.43, and assuming the

money supply remains constant (M̂ t and M̂ *t and both equal zero),
provides

 1 (Ĉ  Ĉ*) 1 (C
ˆ ˆ
(Ê  Ê t)  t ⇒  t  C *)
(1  ) t t

   (1 ) ( Ĉ  Ĉ *)
2

Ê
(1  ) t  (5A.48)

which uses the assumption that Ĉ  C* ˆ Ĉ t  Cˆ*t Ê t in the

steady state due to the PPP restoration from the short-run deviation.
Plug 5A.48 into 5A.44 to get:

ˆ (1  )  ˆ ˆ *)
E (C t  C (5A.49)
t
(1 ) t

The PPP restoration is important here, and we see soon that equa-
tion 5A.49 is also important for solving this model: equating con-
sumption changes in each economy over the two periods to the
resultant change in the exchange rate. In short, equation 5A.49 tells
us that as consumption is affected by sanctions in the short-run, so
goes the exchange rate shock.
Using the consolidated budgets in percentage change form, equa-
tions 5A.40 and 5A.41, and assuming that the percentage change in
the demands for specific goods are equal to the percentage change in
overall consumption for that economy (where x̂t v̂*t C ˆ ), with
t
government spending in each country constant ( ĝ, ĝ* 0):

Ĉ t nĈ t (1  n)(Ĉ*t Êt ) ˆ  b̂ ˆ (5A.50)

EX FS

ˆ n (b̂ ˆ )
Ĉ *t n(Ĉ t  Ê t) (1 n)(Ĉ*t ) (5A.51)
IM
1n FS

Subtracting these equations from each other gives the difference in

the relative percentage changes in short-run consumption as a func-
tion of sanctions and the short-run change in the exchange rate.
118 ECONOMIC SANCTIONS

Solving for this change less the exchange rate is a function of the
percentage change in the current account (net savings) between the
sender and target (b̂ ) and sanctions:

 
b̂ ˆFS 1  2n ˆ
Ĉt  Ĉ *t  Êt   ˆ IM (5A.52)
(1  n) 1 n EX

Equate this result to 5A.43 and solve for the percentage change in
bonds held by the sender as a function of sanctions:

ˆ 2(1  n) ˆ
b 
(1 )
[Ct  Cˆ *t  Eˆ t S1] (5A.53)

Plugging equation 5A.53 into 5A.52, we produce

2 1 
Ĉ t Ĉ*t Ê t  S1
1 1

ˆ 112nn 
FS
ˆ
EX  ˆIM  (5A.54)

Plugging this result into 5A.49 and solving for Ê t leads to:

Êt
(1 )
(2  1)

2
1
 S1  1 
1  ˆ (112nn )
FS

ˆ
EX  ˆIM  (5A.55)

ˆ as a function of sanctions:
Using 5A.55 allows us to solve for b

b̂ 
2(1  n)
(1  )
 S1
1 
1  ˆ (112nn )
FS

ˆ
EX  ˆIM .  (5A.56)

(1 )
where again [ ˆ (2n  1)
2  (1  n) FS
( ˆEX  ˆIM)(1  n)] S1.

In the Betts and Devereux (2000) pricing-to-market model, policy

is considered neutral. Notice if there are no sanctions, the percentage
 NOEM SANCTIONS MODEL 119

change in bond holdings in the steady state is zero; neutrality results

as the households have no incentive to change consumption from its
current steady state level. However, with sanctions, the percentage
change in bond holdings is only zero under specific conditions, not in
general, and thus there are welfare effects. This result suggests that eco-
nomic coercion, from the slightest restriction to a complete embargo,
has non-neutral effects. The welfare costs to both the sender and target
are summarized here.

Sanction Effects on the Open

Macroeconomy
In the NOEM literature, the welfare effects of changes in real
balances are seen as negligible versus changes in consumption and
work effort. Because we assume monetary policy is not in play here,
we assume the same concerning the welfare effects of sanctions on
real balances.7 We assume there are three periods: t1 is the steady
state, t is the shock period, and t 1 is the long-run. The only role of
t1 is to define when previously held bonds were purchased or
issued to augment wealth (create interest and principal payments) in
time t for initial net savers.
Changes in welfare are symbolized by U: U UC Uh. For
consumption, the percentage changes have already been defined as
Ĉ t and Cˆ* . For the sender, the change over time of consumption is
t

Ĉ t ˆ is the change in consumption after the shock

Ĉ , where C
1
and into the long run Ĉ t 1. The ratio ofreflects the present
1
value of the perpetual change in consumption in the long-run.

UC Ĉ t Ĉ (5A.57)
1

UC* Ĉ*t Ĉ* (5A.58)

1

For work effort, the calculation is a bit more complex. First, because
we want to focus on the change in work effort squared, which is what
resides in the utility function. To do so, we need to do a linear approx-
imation of work effort around the steady state h0. Following Mark
(2001), this approximation around t 0 is h2t h20 2h0(hth0),
which implies that h2t  h20 2h0(hth0). Multiplying the right-hand
120 ECONOMIC SANCTIONS

side of this equation by h0/h0, we get the following for the sender:

Uh  12 [2 h20 ĥ t h20 ĥ ] (5A.59)

1

Because work effort is assumed to be one-for-one with production,

we can use the steady state equilibrium to replace h02 in 5A.59:

  1
1/2
h0 C0 Cw0 . Substituting this expression into 5A.59
and its analog from the target economy, we get the following welfare
shocks from changes in work effort.

Uh [   1 ĥ t
1   1 ĥ ] (5A.60)

U*h [   1 ĥ* t
1   1 ĥ*] (5A.61)

Focusing on the sender, and using the model’s symmetry to ulti-

mately provide similar solutions for the target, we can now replace the
unknowns given earlier with their solutions in terms of the exogenous
variables and the sanction shocks. Starting with the long-run values,
Ĉ and ĥ from Equations (5A.43) and (5A.39), where ĥ ŷ in the
aggregate. Substitute the value offrom equation (5A.56) in each to
get the long-run change in consumption and work effort:

Long-Run Changes

Ĉ  1 [11 FS 12 ˆEX  13 ˆIM)] (5A.62)

2

where 11 (1 )u11 (2n1); 12 (1 )u12(1n);

13 (1 )u12(1n)

1[(u  1)
ˆ
h 11 FS (u12  1) ˆEX  u12 ˆIM)] (5A.63)
2
(1  n) 
where u11 (2n  1)( 1) and
(1  )(1  n)

(1  n  )
u12  (1 )
(1  )
 NOEM SANCTIONS MODEL 121

Short-Run Changes

Using the relationships in 5A.54 and 5A.55, and that

ˆ
C wt  n Ĉt
Ĉ*t
(1  n)
⇒

Ĉt 3(2 
 1) 
u13  1[n ˆEX
2
( ˆ
FS ) (1  n) ˆIM] (5A.64)

Using the following equality of ĥ t (1  n) Ê t Ĉ wt :

(1 )
ĥt (1  n)  u  1[n ˆEX
(2  1) 13 2
( ˆ
FS )
(1  n) ˆIM ] (5A.65)

where u13 1 2   S

1  (
1  ˆ 1  2n
1   FS 1  n ) ˆ
EX .
 ˆIM

At this stage, the long-run and short-run shocks to utility are in terms
of exogenous variables. The target analog equations are found by
replacing (1n) with n, which holds throughout these final deriva-
tions (Mark 2001). For the sender economy, the final changes in
utility that result from the sanction shock are as follows:

U UC Uh

UC 3(2 
 1) 
u13  1 [n ˆEX
2
( ˆ
FS ) (1  n) ˆIM]

1

1
[
2 11 FS 12 ˆEX  13 ˆIM] (5A.66)

(1 )
Uh  ( 1)[(1 n)  u  1[n ˆ EX
(2  1) 13 2
( ˆ
FS )
(1  n) ˆIM]]
1
[(  1) 21 [(u 11  1) FS

(u12  1) ˆ EX  u12 ˆIM ]] (5A.67)

122 ECONOMIC SANCTIONS

For the target, the symmetry of the model’s symmetry leads to

similar solutions. The final change in utility is as follows:
U* U*
C U*h

UC*  1 n n 3(2 
 1) 2 
u13  1 [n ˆEX ( ˆ
FS )
(1 )
(1  n) ˆIM] [( u11 (2n  1))
1 2(1  n) FS

( u12  (1  n))(ˆEX  ˆIM) ] (5A.68)

(1 )
U*h  (  1) [n u ]
(2  1) 13

1
[  1 n 1 [( u  1)
1n 2 11  FS

(u12  1) ˆIM  u12 ˆEX ] ] (5A.69)

The welfare effects from sanctions are summarized in tables 5.1, 5.2,
and 5.3 in chapter five at specific parameter values initially, then through
changing the parameters: n, , , and .
The final stop is to solve for the terms of trade, as this variable is
also a function of sanctions and another potential monitor of sanction
effects. Because the exchange rate is changing, the relative export
price in the sender economy is likely to change as well. The terms of
trade are solved for as follows, where the respective export prices are
simply portions of each country’s aggregate price index from
equations 5A.2 and 5A.3:

  (E q*(z)) dz n(E q*(z)) 

n 1 1
1 1 1
1 1
PEX
t, Sender t t t t 0 n1 Etq*t
0

 
m 1 1

qt(z*)
    qt(z*)
 
1 1 1 1

PEX
t,Target dz* m
Et Et
0

  
1
qt(z*) 1 1

 n (m  n)1 Etq*t
Et
 NOEM SANCTIONS MODEL 123

1
PEX n Etq*(z) n 1
Etq*(z)
( )
1
t, Sender t 1
t
TOTSender (m  n)
EtPEX qt(z*)
1 qt(z*)
t, Targer
Et(m  n) 1

Et
(5A.70)

In the steady state, the terms of trade simply become a function of

exchange rates, and thus of sanctions.

TÔTSender Ê t
(1 )
(2  1)

2
1 
 S1
1 
1

ˆ 112nn 
FS
ˆ
EX  ˆ IM  (5A.71)

Chapter six begins with an empirical assessment of these results, as

well as some econometric examples of how the sanctions shock
exchange rates, consumption, and production in certain sanction
cases.

Conclusions
This chapter shows the tedium and length of solving this general
equilibrium. It is logical to assume that sanctions affect the budget
constraints of each country involved. The NOEM mathematics and
symmetry provide intuition and ease in adding sanctions as simple
policy variables to extend this class of models. As NOEM models
evolve both mathematically and in their detail, it is likely this baseline
model may change. Again, any errors in judgment, logic, or mathe-
matics are my own. I am much indebted to Mark (2001), as that text
taught me a great deal about these models.
This page intentionally left blank
 Chapter 6

Empirical Analyses of Sanction

Effectiveness

Since the goal of the [empirical] analysis [of sanction effectiveness] is to

determine what makes economic coercion succeed, including the contribu-
tion of sanctions as part of that measure creates a potential endogenity
problem. That is, determining the factors that make a sanction successful
is similar to determining what contribution the sanctions made toward
success.
(Drury 2005, 45)

Introduction
This quote alludes to the challenges in using current data sets on
sanctions without some adjustments. Other challenges include an
inability to pinpoint what represents economic coercion’s “contribu-
tion” to sanction effectiveness. This chapter continues this text’s thesis
that economic statecraft is ultimately macroeconomic, because sanc-
tions are initiated as if they were macroeconomic policy. Further,
decoupling the economic contribution from the political and human-
itarian effects of statecraft reduces the endogenity problem. However,
the endogenity problem is difficult to conquer; endogenity in this
model reflects reality. Sanction effects are based on feedback loops
surging through the target economy to create a political critical mass
for change.
These estimations combine the intuition from chapters two, three,
and four, where the effects play out in the exchange rate as discussed in
chapter five. This model may also predict the economic effects in new
embargo cases. The hypotheses tested later are all made to understand
how policy makers move along the Sanction Effectiveness Continuum.
Testing a sanction’s economic effects using macroeconometrics focuses
126 ECONOMIC SANCTIONS

on how changes to exchange rates may take place under sanctions and
act as economic shock data. Next, discrete choice modeling is used to
investigate the humanitarian effectiveness of sanctions, a la smart
sanctions. Smart sanctions are assumed to be focused on decision
makers to minimize collateral damage. Thus, comparing social and
demographic factors in the target economy before and after sanctions
may be used to track humanitarian changes. Finally, political effective-
ness is estimated based on Hufbauer, Schott, and Elliott (HSE 1990)
data, Drury (1998, 2005) and the author’s calculations of what HSE
would likely say the political success was in cases after 1990 as well as
current ones.
Empirical studies provide researchers with ways to explain the past,
but also help forecast how policy may affect the future. There have
been many studies of how sanctions in the past have been effective or
not vis-à-vis qualitative goals. There have also been many review arti-
cles analyzing past studies, especially the HSE data, methods, and con-
clusions. Recent studies suggest that the sanction literature strayed too
far away from its major empirical issues. Drezner (2003) and Drury
(2005) argue that investigating sanction success, whether sanctions
achieve stated goals, is not worth a researcher’s time at this point.
Multiple statistical issues, coupled with data problems, make current
empirical studies on statecraft immediately questionable at best.
The sanction effectiveness question is still a viable one, however,
and must be studied. Policy makers need a body of research to use as
a foundation for their decision-making. The ordinal success measures
of HSE (1990), Lam (1990), and Van Bergeijk (1989) provide ad
hoc, but logical measures. Because a success statistic can be stated as
an ordered dependent variable, using discrete choice modeling has
now become the standard. Lam (1990) was the first to investigate the
HSE (1990) data in more detail using discrete choice modeling.
Martin (1993) provides an initial, empirical test of the HSE data,
whereas Drezner (2003) investigates threats versus sanctions as an
extension of the HSE data. Drury (2005) suggests that a restructure
of the HSE success measures is a good beginning, given the data’s
breadth. Drury’s (2005) results are compared to this chapter’s output
later. These studies are strong examples of using discrete choice mod-
eling to look at different aspects of sanction effectiveness; however, it
is both difficult and impractical to walk away from the HSE data set
completely, and it is again employed in part here.
There are three empirical exercises that act as the path through this
chapter, breaking the chapter into four sections. The first is estimating
the economic shock that sanctions cause. This is done through vector
 ANALYSES OF SANCTION EFFECTIVENESS 127

autoregressions (VARs) and impulse-response functions. VARs provide

a way to look at the time dynamics of economic coercion, estimating
how damaging the effects may be and how long the effects may last.
This is done through impulse response functions, where the sanction
shock is the impulse, a one standard-deviation shock to target trade or
financing with the sender. Because of data limitations, I use only a
subset of cases with the United States as primary sender, sixty-five
cases in all. This first exercise is to show how viewing sanctions as
macroeconomic phenomena can tie itself to exchange rate fluctuations
as a measure of economic effectiveness, analogous to other macroeco-
nomic policies. The permanence and explanatory power of the shock
on U.S. exchange rates with the target is a new measure of economic
effectiveness.
Next the humanitarian effectiveness of sanctions is estimated using
discrete choice modeling. If the dependent variable is an ordinal
measure of effectiveness, then the independent variables are simply
driving the probabilities of effectiveness higher or lower. This text
looks at both humanitarian and political effectiveness, where each
regression uses similar explanatory variables with the idea that human-
itarian success is a station passed by the sanction train on its way to the
political goals; the factors assumed to exogenously affect political out-
comes also affect human costs.
The next section estimates political effectiveness, which is the
empirical focus of this literature. We must look beyond sanction suc-
cess and make that view operational, looking beyond binary measures
of sanction efficacy also. Using ordinal measures and discrete choice
analysis of political effects, as in Drury (2005), captures the essence of
policy effectiveness. Policy is not meant to pass or fail the test of goal
achievement, as it may or may not achieve its stated goal. Marginal
movements toward the sanction’s goals dictate how economic coercion
is effective.

A Brief Set of Caveats

There are two large caveats to any empirics involving sanctions that
must be addressed, as in Drezner (2003). Each section here uses data
from respected international sources, such as the International
Monetary Fund (IMF) and United Nations. However, data collection
may be tainted because the data’s original sources are flawed. Further,
the continued use of a subset of the HSE (1990) data brings in the
caveats discussed earlier. In all, these estimates are toeholds on new
ways to view these policies, using readily available data as well as
128 ECONOMIC SANCTIONS

subjective and objective control variables that are generally accepted

by sanction scholars. The reader should take these caveats into con-
sideration when viewing the results; the conclusions here discuss further
empirical research opportunities.

The Economic Effects

Time-series econometric techniques allow a prediction of sanction
effects as if sanctions were macroeconomic policies. Such estimates
show a consistent way to track sanction effects over time, comparable
regardless of targets across cases. The sender is hurt if, with respect to
the target, the exchange rate falls; the target is hurt if the exchange
rate rises, where exchange rates are defined as target currency per
sender currency units. This inverse view of exchange rates is used here
to make the estimates more intuitive for their humanitarian and polit-
ical effects. The reader should always think of the exchange rate
reflecting the relative value of a currency.
Eyler (1998) suggested an ordinal measure of a sanction’s eco-
nomic effectiveness, after calibrating a neoclassical, open economy
model.1 This measure was based on an estimated loss of welfare over
the policy’s duration.2 However, such a measure should also be con-
structed for the sender. Because chapter five’s model focused on the
exchange rate changes that result from consumption and work effort
effects in both countries, the model here does the same. Other models
of macroeconomic effects exist, however, that have provided some
control variables used here. These are mainly in the form of gravity
models.

Gravity Models
Gravity models come from the international trade literature and are
applied in two general ways to economic coercion.3 First, sanctions
are modeled as shocks to bilateral trade. Gravity models measure the
expected trade between two countries based on country size, distance
between the countries, and other factors. As a result, the economic
effects of embargoes are either assisted or mitigated in the target’s
eyes by gravity factors; HSE (1990) includes the relative size of the
economies for this reason. By design, gravity models easily become
linear regressions, where the level of bilateral trade acts as the depend-
ent variable. Independent variables may include the product of GDP
per capita in the two countries, GDP per capita in each country indi-
vidually, the distance between the trading partners, and other control
variables.4 The key hypotheses of gravity models are that trade flows
 ANALYSES OF SANCTION EFFECTIVENESS 129

are proportional to the product of the “size” of the trading partners

and inversely proportional to the distance between them, an idea
drawn from physics where two objects gravitate toward each other
based on size and distance between them.5
Recent sanction gravity models include Hufbauer et al. (1997), Yang
et al. (2004), and Askari et al. (2004). Yang et al. (2004) expanded the
time series used in Hufbauer et al. (1997) and focused on a smaller
number of cases. They claim that these models are extremely sensitive
to sanctions labeling, where the range of labels are either selective (set
in specific ways to combat specific problems) or comprehensive.
Sanction effectiveness was seen as relatively smaller in magnitude than
the results of Hufbauer, Schott, and Elliott (1990). “This [lack of a uni-
form method of defining sanctions] may provide an explanation for the
weak overall effect of selective sanctions—they simply are in too narrow
a range of products to have much impact on overall imports, exports, or
bilateral trade flows” (Yang et al. 2004, 58). Comprehensive sanctions
are assumed to have larger effects; the data are somewhat suspect in
certain important cases, including Iraq and Cuba. In short, gravity
models provide an easy-to-use method of estimating sanction effects on
trade, and may also be applied to financial sanctions easily. Can we fur-
ther apply macroeconomic techniques to economic coercion, isolating
the economic effects? Sobel (1998) took this idea of wealth transfer to
another level, employing VARs, the basis of this study’s estimations.

Vector Autoregressions
Sobel (1998) used exchange-rate dynamics as a way to infer sanction
effects. If a country’s exchange rate can appreciate as a result of stabi-
lization, then destabilization should cause currency depreciation.
Sobel’s question was whether sanctions are temporary in their effects,
regardless of their stated duration. He examined the cases of Lebanon
and South Africa, comparing two cases with different contexts of UN
sanction packages.6 The hypothesis was that tracking the change in the
real exchange rate as a result of sanctions was a way to at least partially
reflect the internal political and economic consequences of statecraft
measures (Sobel 1998).
The model here is a VAR analysis, where the sanction shock is
modeled like any other macroeconomic shock to the target economy.
Impulse response functions provide insight to economic effects from
embargoes but also show how policy makers may track future uses of
economic coercion before initiating policy. The impulse response
helps test the hypotheses of duration, magnitude, and direction of a
shock’s effects on the economies involved.
130 ECONOMIC SANCTIONS

Sobel (1998) used VARs and impulse response functions as a

means of measuring the forecasted effect of sanctions on exchange
rates; his use of a transfer function assumes that the exchange rate
shock that results from UN economic coercion fades as time moves
forward. In the macroeconomics literature, there are large differences
and implications for permanent versus temporary shocks.7 The VAR
analysis here investigates sixty-five cases, using data on changes to
exports, imports, and the financial flows between the United States
and selected target economies as the policy shock variables.
In VAR models, the dependent and independent variables are both
lagged on the right-hand side of the equation. A vector of lagged
variables is employed to account for autoregressive properties in the
dependent variable (in this case the exchange rate) and how this
variable may be affected by the exogenous variables’ histories. The
number of lags is determined by the Akaike Information Criterion
(AIC) in order to compare and contrast specifications.8 Further, the
lag structure and converting each variable into its natural log when
appropriate helps ensure the model is cointegrated, such that the time
series properties of the included variables are the same.9
The impulse response functions are based on the final specifica-
tions. For example, if export sanctions are used by the United States
on Guatemala, the sanction is modeled as a one-time shock to the
Guatemala–U.S. exchange rate in percentage terms through exoge-
nous shocks (sanction policy) to exports, imports, or financial flows
also changing in percentage terms. The dynamic structure of the VAR
transmits the effects over time, and can forecast the effects over the
embargo’s potential life. These functions represent new forecasts
based on the previous period’s realization, innovations of the error
term between the endogenous variable (the exchange rate) and the
shock variables, the sanctions. These regressions are meant to simply
gain access to new measures of a sanction’s economic effects.

Dependent Variable Choice

Using exchange rate fluctuations as the dependent variable ties this
initial empirical analysis to both the NOEM model in chapter five and
to Sobel (1998). This is in lieu of using binary statements, such as the
HSE “contribution” score, as a dependent variable.10 Choosing a par-
ticular effectiveness measure has been debated and analyzed through-
out this literature’s history. In many ways, the perceived ineffectiveness
of economic coercion may be a function of using an incorrect set of
economic variables. Assessing the damage of selected past cases and
 ANALYSES OF SANCTION EFFECTIVENESS 131

how policy makers can predict the economic damage from future
sanctions is the driving force here.
The exchange rate is defined as target currency units per U.S. dollar.11
Sobel (1998) uses transfer functions as theory concerning the sanction
effects, and uses the monthly change in the nominal exchange rate as the
dependent variable. This assumes that a real exchange rate exists above
the street level that is both not fixed and has enough independent
variability such that sanction shocks can have a measurable effect. The
exogenous variables represent the sanction shocks.

Exogenous Variable Choice and the Data

The independent variables are the exports, imports, and net financial
flows between the nations involved in each sanction case. The sanc-
tion mix in each case determines which shocks are included in the esti-
mations. The lagged dependent variable is implied by the VAR model
itself as well. There are two hypotheses to be tested for each shock
variable included, after converting the target’s exports and imports
with the United States to their natural logarithms, as well as using the
U.S. net financial flows with the target. These hypotheses are more
global, and not coefficient-specific.

Hypothesis 1: A shock to these variables should have predictable effects

of exchange rates during the sanction’s duration. For example, export
sanctions should increase the U.S.––target exchange rate; import sanc-
tions should lower the exchange rate; and financial sanctions should
increase the exchange rate. The rationale for these movements is
presented in chapter five, albeit with an inverted definition.
Hypothesis 2: Over the policy’s life, the shock’s effect should not
erode. Whether the exchange rate effects are permanent or temporary
is of major importance to economic coercion’s credibility. If the
impulse does not create a permanent change in the exchange rate,
then the policies can be considered neutral in their effects. If the
target knows this, because of the structural relationship between the
exchange rate and the international flows, all it needs to do is wait.
The sender’s expectation is that the shock is permanent in their effects
on the target.

The VAR results provide data that can be used as economic shock
variables for the humanitarian and political effects of sanctions tested
in this chapter.12 The equations are the same in each case study, where
the number of lags () is specific to each case and is chosen using a
132 ECONOMIC SANCTIONS

case-specific AIC. Equation 6.1 describes the general specification in

each case.

FXTarget-Sender,t  Xt-1  Z1,t-1 Z2,t-2 Z3,t-2 Target-Sender,t (6.1)

where FX is the exchange rate, target currency units per sender, in

month t; Xt is the 1 x  vector of lagged dependent variables at time t. 1
is the   1 vector of coefficients associated with the variables in Xt. Zk,t
is the 1   vector of the kth policy variable, lagged  times, where k rep-
resents the number of included variables, and k is the associated   1
coefficient vector for the respective policy variable. The error term,
Target-Sender,t is assumed to uncorrelated white noise between time peri-
ods. This basic VAR structure sets us the impulse response functions,
where the embargo’s imposition is that impulse.
Once estimated, the error term is converted to a Wold moving
average representation before employing the impulse response func-
tions.13 Once this conversion is complete, the impulse response func-
tions are built. By standardizing the error matrix such that the error
variance is equal to one (and thus the standard deviation is equal to
one), the moving average representation of the error can be rewritten
and standardized. Once this is completed, the original VAR equations
can be rewritten as variance decompositions, a two-term sum. The
first term is the shock term, with coefficients representing the weight
of previous lag periods, and the second term represents the dependent
variable’s time series found to be random and not changed by the
shock.14 The impulse response functions ultimately estimate what the
deviations from trend would be if the error term in the original VAR
was shocked by one standard deviation.
The real exchange rate is used, deflated in year 2000 terms. The data
sources are either the IMF International Financial Statistics Yearbook,
Pick’s Currency Yearbook (for years up to 1984 when needed), or the
World Currency Yearbook (after 1984 when needed). The export data is
the real value of exports between the United States and the specific tar-
get. The import data is also the real value of imports between the two
countries with 2000 as the base year. Net financial flows are represented
by the real U.S. net liabilities between itself and the country in question
from the Survey of Current Business; in some cases, regional financial
flows or the target’s overall net liabilities with the world are used because
the data were unavailable or unreliable otherwise. This is an unfortunate
problem in analyzing target countries that are developing historically.
Two issues make using VAR analysis for the entirety of the HSE data set
problematic. First, the economic data for some sender countries
 ANALYSES OF SANCTION EFFECTIVENESS 133

involved is parsimonious as is the target’s data; the data’s credibility

may also be an issue even if data is available. China and the former
Soviet Union are good examples of senders with troubling data. The
cases examined here may help minimize data problems because the
United States is either the lone or a partial sender each time. This
choice may also introduce bias because the United States is involved
every time. However, the following section, taking these caveats into
account, provides the VAR and impulse response results.

Results
The results are broken into multiple tables, starting with table 6.1.
Table 6.2 provides information concerning the tests of hypothesis 1
given earlier, the direction of the impulse response. Table 6.3 provides
the variance decompositions for each case, representing the exchange
rate innovation in percentage terms explained by specific shock
variables. Table 6.3’s data, in combination with Table 6.2’s, tests
hypothesis 2.
Table 6.1 summarizes the basic information from Appendix 1’s
cases to be empirically tested. The “lags” represent the VAR specifica-
tion for each case that minimized the AIC. The approximate duration
in months is simply from HSE (1990), Drury (1998, 2005), or the
author’s assessment of duration in years multiplied by twelve. The
sanction mix data tells the reader which variables were shocked. This
information helped build each model’s specification, as described
earlier. Table 6.2 shows the exchange rate shock’s direction initially
versus the policy’s end.
Some shocks are estimated to have permanent effects on target
exchange rates, while others have temporary effects. The first month’s
measure (“1 mo”) represents the initial response of exchange rates to
policy, while the last month’s (“Last mo”) sign represents what hap-
pened over time. If the last month’s sign is different than the first
month’s, and is also counterintuitive or equal to zero, then sanction
effects were not permanent.
To better interpret the results, an example helps. Every case in
table 6.2 has either a 1 or 1 in its first-month position if a sanction of
that type took place. For example, in the 1992–95 case of Nicaragua,
there are 1, 1, and 1, respectively, in each of the first-month
positions. Those numbers tell the reader that the sanctions against
Nicaragua were full sanctions, employing export, import, and financial
sanctions simultaneously. The data also suggest that the sanction’s
impulses were initially ineffective ( 1 represents policy effects moving
against the sender’s intent) for both export and import sanctions, while
Table 6.1 Cases and Basic Data for VAR Analysis

Target Dates Sanction Contribution Approx. Duration Lags

Tools Score (HSE) (months)

Argentina 1977–84 EX, FS 2 72 2

Argentina 1978–82 EX 2 48 2
Brazil 1962–64 FS 3 24 2
Brazil 1977–84 FS 3 84 8
Cameroon 1992–98 EX, IM, FS ? 72 9
Chile 1965–66 IM, FS 4 12 2
Chile 1970–90 FS 3 36 2
Chile 1973–90 FS 2 204 2
China 1989–98 EX, FS 1 12 2
China 1991–98 EX, FS ? 96 2
Colombia 1996–98 IM, FS ? 24 7
Cuba 1960-Present EX, IM, FS 1 Ongoing 2
Egypt 1963–65 FS 4 24 2
El Salvador 1977–81 FS 3 48 2
El Salvador 1987–88 FS 4 12 2
El Salvador 1993–93 EX, IM, FS ? 36 2
Ethiopia 1976–92 IM, FS 3 168 2
Guatemala 1977–86 FS 2 108 2
Guatemala 1990–93 EX, IM, FS ? 36 2
Haiti 1987–90 FS 3 36 2
Haiti 1991–94 EX, IM, FS ? 36 2
India 1965–67 FS 4 24 2
India 1971–72 EX, FS 1 12 2
India 1978–82 EX 2 48 2
India 1998–2005 EX, FS ? 84 2
Indonesia 1963–67 FS 2 36 2
Indonesia 1993–95 FS ? 168 2
Iran 1979–81 EX, IM, FS 3 24 2
Iran 1984–98 EX, IM, FS 3 72 3
Iraq 1980–87 EX 2 120 2
Iraq 1990–2003 EX, IM, FS 3 12 4
Liberia 1992–98 EX, IM, FS ? 72 2
Libya 1978–98 EX, IM, FS 2 144 3
Malawi 1992–94 FS ? 12 5
Myanmar 1988–90 FS 3 24 2
Nicaragua 1977–79 EX, FS 3 24 2
Nicaragua 1981–88 EX, IM, FS 2 108 2
Nicaragua 1992–95 EX, IM, FS ? 36 3
Nigeria 1993–98 IM, FS ? 60 3
Pakistan 1971–72 EX 1 12 4
Pakistan 1979–86 FS 1 132 2
Panama 1987–93 IM, FS 1 36 2
Paraguay 1977–81 FS 3 48 2
Paraguay 1996–96 EX, IM, FS ? 12 2
Peru 1968–74 FS 1 12 3

Continued
 ANALYSES OF SANCTION EFFECTIVENESS 135

Table 6.1 Continued

Target Dates Sanction Contribution Approx. Duration Lags

Tools Score (HSE) (months)

Peru 1991–95 EX, IM, FS ? 48 2

South Africa 1975–82 EX 2 84 2
South Africa 1985–92 EX,IM,FS ? 84 3
South Korea 1973–77 FS 2 48 3
Sri Lanka 1961–65 FS 4 48 2
Sudan 1989–94 FS 1 12 2
Sudan 1993-Present IM, FS ? Ongoing 6
Syria 1986–94 EX, FS 3 48 2
Taiwan 1976–77 EX 4 12 2
Thailand 1990–93 EX, IM, FS ? 36 2
The Gambia 1996–98 FS ? 48 2
Togo 1992–94 EX, FS ? 24 3
Turkey 1974–78 FS 1 48 5
Uruguay 1976–81 EX, FS 2 60 2
USSR 1975–90 IM, FS 2 180 2
USSR 1978–80 EX 1 24 2
USSR 1980–82 EX 1 24 3
USSR 1981–81 EX 1 12 2
USSR 1983–83 IM 1 12 2
Zimbabwe 1983–88 FS 2 60 2

Sources: HSE (1990), Drury (1998, 2005) and author’s calculations.

financial sanctions initially were effective (1 represents policy moving

with the sender’s intent).
These data allow a comparison to what happened over the sanc-
tion’s duration, as told in the adjacent cells. For Nicaragua, the last
month’s values are 1, 0, and 1, respectively. By the sanctions’ end,
export and financial sanctions succeeded in decreasing the target
currency’s value, while import sanctions had no permanent effects.
Coupled with the initial results, it is estimated that the intent of
economic statecraft’s impulse was reversed in the export sanction to
the correct direction, eroded in the case of the import sanction, and
became stronger over time in the financial sanction. The initial con-
clusion from this example is that U.S. economic coercion against
Nicaragua was mixed initially, and became stronger over time through
the combination of export and financial sanctions. The variance
decomposition data in table 6.3 enhances this output.
This output adds the element of how much the policy shock
explained the percentage change in exchange rates. For example, the
results of table 6.2 show that sanctions against Brazil in 1962–1964
were financial only and at first were causing economic damage (1 in
Table 6.2 Initial and cumulative effects of sanction shock—first month (1 or 1 or
0) versus last month (1 or 1 or 0)

Target Dates Export sanction Import sanction Financial sanction

1 mo. Last mo. 1 mo. Last mo. 1 mo. Last mo.

Argentina 1977–84 1 0 0 0 1 1
Argentina 1978–82 1 1 0 0 0 0
Brazil 1962–64 0 0 0 0 1 1
Brazil 1977–84 0 0 0 0 1 1
Cameroon 1992–98 1 1 0 1 1 1
Chile 1965–66 0 0 1 1 1 1
Chile 1970–90 0 0 0 0 1 0
Chile 1973–90 0 0 0 0 1 1
China 1989–98 1 0 0 0 1 0
China 1991–98 1 0 0 0 1 0
Colombia 1996–98 0 0 1 1 1 1
Cuba 1960–Present 1 1 1 1 1 1
Egypt 1963–65 0 0 0 0 1 0
El Salvador 1977–81 0 0 0 0 1 0
El Salvador 1987–88 0 0 0 0 1 1
El Salvador 1993–93 1 1 1 1 1 1
Ethiopia 1976–92 0 0 1 1 1 1
Guatemala 1977–86 0 0 0 0 1 0
Guatemala 1990–93 1 1 1 1 1 1
Haiti 1987–90 0 0 0 0 1 1
Haiti 1991–94 1 1 1 1 1 1
India 1965–67 0 0 0 0 1 0
India 1971–72 1 1 0 0 1 1
India 1978–82 1 1 0 0 0 0
India 1998–2005 1 0 0 0 1 1
Indonesia 1963–67 0 0 0 0 1 1
Indonesia 1993–95 0 0 0 0 1 1
Iran 1979–81 1 1 1 1 1 1
Iran 1984–98 1 0 1 0 1 1
Iraq 1980–87 1 1 0 0 0 0
Iraq 1990–2003 1 0 1 0 1 0
Liberia 1992–98 1 1 1 1 1 1
Libya 1978–98 1 1 1 1 1 1
Malawi 1992–94 0 0 0 0 1 1
Myanmar 1988–90 0 0 0 0 1 1
Nicaragua 1977–79 1 0 0 0 1 0
Nicaragua 1981–88 1 1 1 1 1 1
Nicaragua 1992–95 1 1 1 0 1 1
Nigeria 1993–98 0 0 1 1 1 1
Pakistan 1971–72 1 0 0 0 1 0
Pakistan 1979–86 0 0 0 0 1 1
Panama 1987–93 0 0 1 1 1 1

Continued
 ANALYSES OF SANCTION EFFECTIVENESS 137

Table 6.2 Continued

Target Dates Export sanction Import sanction Financial sanction

1 mo. Last mo. 1 mo. Last mo. 1 mo. Last mo.

Paraguay 1977–81 0 0 0 0 1 0
Paraguay 1996–96 0 0 0 0 1 1
Peru 1968–74 0 0 0 0 1 0
Peru 1991–95 1 1 1 1 1 0
South Africa 1975–82 1 0 0 0 0 0
South Africa 1985–92 1 1 1 1 1 1
South Korea 1973–77 0 0 0 0 1 1
Sri Lanka 1961–65 0 0 0 0 1 0
Sudan 1989–94 0 0 0 0 1 1
Sudan 1993–Present 1 1 0 0 0 0
Syria 1986–94 1 1 0 0 1 1
Taiwan 1976–77 1 1 0 0 0 0
Thailand 1990–93 1 0 1 1 1 1
The Gambia 1996–98 0 0 0 0 1 1
Togo 1992–94 1 1 0 0 1 1
Turkey 1974–78 0 0 0 0 1 0
Uruguay 1976–81 1 1 0 0 1 1
USSR 1975–90 0 0 1 0 1 0
USSR 1978–80 1 1 0 0 0 0
USSR 1980–82 1 1 0 0 0 0
USSR 1981–81 1 1 0 0 0 0
USSR 1983–83 0 0 1 1 0 0
Zimbabwe 1983–88 0 0 0 0 1 0

the first-month position), but then eroded completely and reversed

themselves over the two years ( 1 in the last month position). On the
surface, these results would be interesting, but once table 6.3’s results
are integrated, we see that the shock to financial flows explains little of
the innovations both initially and at sanction’s end: 0.16 percent and
0.79 percent, respectively. The explanatory power of economic coer-
cion measures is case-specific. In combination, like an interaction
term, tables 6.2 and 6.3 provide the economic damage data to be used
in the humanitarian and political effectiveness regressions given later
by multiplying each case’s directional term with its percentage.

Summary
The cases with the largest economic effects should be those with impulse
responses causing exchange rates to move in the intended direction
(hypothesis 1) and be permanent in those effects (hypothesis 2). The
impulse response direction and the error term’s variance decomposition
Table 6.3 Variance decompositions (%) for VAR results

Target Dates Export Import Financial

sanction sanction sanction

1 mo. Last mo. 1 mo. Last mo. 1 mo. Last mo.

Argentina 1977–84 0 0 0.34 3.01 0.32 15.76

Argentina 1978–82 0.02 0.49 0 0 0 0
Brazil 1962–64 0 0 0 0 0.16 0.79
Brazil 1977–84 0 0 0 0 0 99.96
Cameroon 1992–98 0.01 17.88 0.36 2.74 0.03 75.45
Chile 1965–66 0 0 0.41 1.82 3.14 44.26
Chile 1970–90 0 0 0 0 0.31 8.55
Chile 1973–90 0 0 0.64 11.85 0.11 10.37
China 1989–98 0.02 1.56 0 0 4.07 5.99
China 1991–98 0.05 4.89 0 0 0.79 0.2
Colombia 1996–98 0 0 0.08 8.42 0.9 22.12
Cuba 1960–Present 0.23 7.46 2.08 16.17 7.01 35.88
Egypt 1963–65 0 0 0 0 0.24 0.66
El 1977–81
Salvador 0 0 0 0 0.04 0.11
El 1987–88
Salvador 0 0 0 0 4.13 9.08
El 1993–93
Salvador 0.02 1.72 0.45 0.39 0.12 0.95
Ethiopia 1976–92 0 0 2.43 3.25 0.01 95.17
Guatemala 1977–86 0 0 0 0 0.56 7.57
Guatemala 1990–93 0.01 0.93 0.0005 0.88 0.02 0.09
Haiti 1987–90 0 0 0 0 0.13 4.98
Haiti 1991–94 0.03 0.48 0.3 3.18 0.06 1.66
India 1965–67 0 0 0 0 0.36 0.4
India 1971–72 0.16 1.72 0 0 0.04 0.05
India 1978–82 0.08 0.75 0 0 0 0
India 1998–2005 0.04 0.52 0 0 0.43 63.94
Indonesia 1963–67 0 0 0 0 0.64 18.59
Indonesia 1993–95 0 0 0 0 0.76 15.32
Iran 1979–81 0.001 1.22 0.14 3.94 0.48 5.68
Iran 1984–98 0.08 77.83 0.08 6.84 0.33 4.41
Iraq 1980–87 0.02 16.88 0 0 0 0
Iraq 1990–2003 0.06 4.04 0.35 11.33 1.28 22.44
Liberia 1992–98 0.49 2.75 0.81 47.45 1.24 28.64
Libya 1978–98 0.08 1.64 0.17 1.49 0.1 74.57
Malawi 1992–94 0 0 0 0 1.04 1.82
Myanmar 1988–90 0 0 0 0 0.003 0.008
Nicaragua 1977–79 0.95 1.05 0 0 3.8 27.48
Nicaragua 1981–88 0.48 9.61 0.16 36.49 4.13 35.92
Nicaragua 1992–95 0.007 8.23 0.15 0.39 0.22 29.58
Nigeria 1993–98 0 0 0.05 23.13 0.0004 20.55
Pakistan 1971–72 0.08 3.92 0 0 0.31 1.61
Pakistan 1979–86 0 0 0 0 0.03 71.49

Continued
 ANALYSES OF SANCTION EFFECTIVENESS 139

Table 6.3 Continued

Target Dates Export Import Financial

sanction sanction sanction

1 mo. Last mo. 1 mo. Last mo. 1 mo. Last mo.

Panama 1987–93 0 0 0.66 4.96 0.37 2.06

Paraguay 1977–81 0 0 0 0 0.2 0.28
Paraguay 1996–96 0.04 0.03 0.63 1.35 0.93 0.08
Peru 1968–74 0 0 0 0 0 39.88
Peru 1991–95 0.33 0.41 0.18 28.89 0.38 1.12
South Africa 1975–82 1.66 0.09 0 0 0 0
South Africa 1985–92 0.001 3.45 1.33 17.45 0.48 18.61
South Korea 1973–77 0 0 0 0 0.01 11.47
Sri Lanka 1961–65 0 0 0 0 3.7 25.87
Sudan 1989–94 0 0 0 0 1.57 31.77
Sudan 1993–Present 0.39 9.93 0 0 0.0004 0.02
Syria 1986–94 0.01 0.96 0 0 0.002 5.49
Taiwan 1976–77 0.62 25.96 0 0 0 0
Thailand 1990–93 0.03 4.91 0.54 7.15 0.002 3.43
The Gambia 1996–98 0 0 0 0 0.07 3.78
Togo 1992–94 0.03 0.057 0 0 0.7 12.11
Turkey 1974–78 0 0 0 0 1.3 12.68
Uruguay 1976–81 1.78 0.19 0 0 0.47 34.08
USSR 1975–90 0 0 0.58 0.69 0 0
USSR 1978–80 0 0 0.69 42.29 0.003 4.68
USSR 1980–82 0.03 0.19 0 0 0 0
USSR 1981–81 0.04 1.48 0 0 0 0
USSR 1983–83 0 0 0.55 0.44 0 0
Zimbabwe 1983–88 0 0 0 0 4.29 10.82

Note: Remainder of variance decomposition is from exchange rate itself.

combined provide an interaction term that represents the estimated

economic damage caused by sanctions. Economic effectiveness, as
defined in this text, is unlikely to be present unless the target perceives
permanent effects from economic coercion. A database for all possible
targets should be assembled for such a purpose. Data on humanitarian
issues is more parsimonious. Some data have been identified by the UN
Inter-Agency Standing Committee (IASC) study and handbook (2004)
on sanctions assessment.

The Humanitarian Effects

If gaining international support is important to sanction success, then
monitoring the collateral damage caused by sanctions is also impor-
tant. The smart sanction literature, discussed in chapter four, has ini-
tiated a new set of factors for this literature to monitor and test
140 ECONOMIC SANCTIONS

empirically. If sanctions are blunt instruments that act like macroeco-

nomic policies transmitting their effects to a target nation, collateral
damage is an unfortunate consequence of economic statecraft. The
Sanction Effectiveness Continuum suggests that the sanction’s eco-
nomic success must be separated from humanitarian success because
of this social element.
Two major challenges exist in estimating humanitarian damage to
the target during sanctions. First, the omnipresent data problem con-
tinues. While the United Nations and World Bank make great attempts
to track the evolution of developing economies concerning their level
of freedoms, available goods and technologies, and education, these
agencies also struggle due to low resources, poor participation by the
populace, and unreliable data sources. Linking measures from a census
taken under one government to another comes with the hope that
logic not politics dictated the data’s collection. Across countries, data
matching and splicing may work if a global organization gathers, syn-
thesizes, and disseminates the information, but still caveats remain.
There is one recent study that has focused on humanitarian effects
from sanctions taking these two problems into consideration.
In October 2004, Bessler et al. (2004) released the UN Inter-
Agency Standing Committee published a report called the Sanctions
Assessment Handbook, called IASC 2004 from here. This guide is the
foundation of the human effects estimations given later.15 IASC 2004
outlines the potential problem that may flow from sanctions for an
assumed innocent populace. Using data from the World Bank’s World
Development Indicators (2006), this section attempts to utilize the
exchange rate effects given earlier as exogenous shocks to test for
humanitarian effectiveness in the included cases. The largest initial
challenge once a data source is found is deciding on what specific data
to include as indicators of humanitarian conditions.
As chapter four discussed, the debate continues to swirl concerning
the legality of sanctions and how much collateral damage is too
much.16 Much like the economic results given earlier, it is more precise
to talk in terms of effectiveness and not humanitarian success or failure:
the less negative the human impact, the more effective the sanction.
Instead of using binary measures that force this analysis into the box of
success or failure, measuring these effects seems better done by looking
at ordinal measures and comparing them across episodes.

Dependent Variable Choice

The dependent variable is a composite indicator built on the sum of
the increases and decreases of subindicators. The difficulty, as with all
 ANALYSES OF SANCTION EFFECTIVENESS 141

indicators, is to choose the components well. Each individual indica-

tor of the target’s human conditions is given equal weight in this
analysis, as debating over which indicator should receive more weight,
given the data parsimony, is like two bald men fighting over a comb.17
This is where IASC 2004 plays a helpful role.
IASC 2004 gives eight categories to measure human conditions in
a country: access to health care, water, education, and food; gover-
nance and economic conditions; environmental and demographic indi-
cators. The dependent variable in this study looks at six of these
categories; due to data parsimony in the World Development Indicators
(2006) database, all but the economic conditions and demographic
indicator are used.18 Because the other data categories have some indi-
cator that touched all the case studies and their targets, they were the
final subindicators of choice. Putting together an index avoids prob-
lems of missing data somewhat, as there may be periods in which no
data exists for certain indicators. Table 6.4 shows this data, where a
qualitative description of each indicator precedes the data.

Health Access: Prevalence and death rates associated with malaria and
tuberculosis; deaths of children under five years old; infant and
maternal mortality rates.
Food Access: Prevalence of underweight children under five years old.
Water Access: Proportion of population with access to improved water
and sanitation sources.
Environment: Proportion of land covered in forest; carbon dioxide
(CO2) emissions; consumption of ozone-depleting chlorofluoro-
carbons (CFCs).
Education: Proportion of pupils that start grade one and reach grade five;
literacy rates; ratio of girls to boys in secondary and higher education.
Governance: Proportion of seats held by women in national parliament.

The categories have an interpretation, as shown in table 6.4,

when the data are available.19 If the post-sanction condition is worse
than the pre-sanction condition, the subindicator is given a value
of 1. If conditions were better after the sanction, the value is 1.
If there is no change or no data, the condition is zero.20 Like other
ad hoc measures in the sanction literature, these data are merely a
toehold and a beginning toward empirical models of economic
statecraft where the human outcome can be compared to past cases,
given the pre-sanction data. Figure 6.1 provides a histogram of the
final indicator data.
Table 6.4 Humanitarian data estimates for sanction cases

Target Dates Humanitarian Indicator

Health Food Water Environment Education Governance Total

Argentina 1977–84 1 1 0 1 0 0 1
Argentina 1978–82 1 1 0 1 0 0 3
Brazil 1962–64 1 1 0 1 0 0 3
Brazil 1977–84 1 1 0 1 0 0 3
Cameroon 1992–98 0 1 0 0 0 0 1
Chile 1965–66 1 1 0 1 0 0 1
Chile 1970–90 1 1 0 1 0 0 1
Chile 1973–90 1 1 0 1 0 0 3
China 1989–98 0 1 0 1 0 0 0
China 1991–98 0 1 0 1 0 0 2
Colombia 1996–98 1 1 0 1 0 0 1
Cuba 1960–Present 1 0 0 0 0 0 1
Egypt 1963–65 1 1 0 1 0 0 1
El Salvador 1977–81 1 1 0 1 0 0 3
El Salvador 1987–88 1 1 0 1 0 0 1
El Salvador 1993–93 0 1 0 1 0 0 0
Ethiopia 1976–92 1 1 0 1 0 0 1
Guatemala 1977–86 1 1 0 1 0 0 3
Guatemala 1990–93 0 1 0 1 0 0 0
Haiti 1987–90 0 0 0 1 0 0 1
Haiti 1991–94 0 1 0 1 0 0 2
India 1965–67 1 1 0 0 0 0 2
India 1971–72 1 1 0 1 0 0 1
India 1978–82 1 1 0 1 0 0 1
India 1998–2005 0 1 0 1 1 1 2
Indonesia 1963–67 1 0 0 0 0 0 1
Indonesia 1993–95 0 1 0 1 0 0 0
Iran 1979–81 1 1 0 1 0 0 1
Iran 1984–98 0 1 0 1 0 0 2
Iraq 1980–87 0 0 0 1 0 0 1
Iraq 1990–2003 1 0 0 1 0 0 2
Liberia 1992–98 1 1 0 1 0 0 3
Libya 1978–98 1 1 0 1 0 0 3
Malawi 1992–94 1 1 0 0 0 0 0
Myanmar 1988–90 1 1 0 0 0 0 0
Nicaragua 1977–79 0 1 0 1 0 0 0
Nicaragua 1981–88 1 0 0 1 0 0 2
Nicaragua 1992–95 0 1 0 1 0 0 0
Nigeria 1993–98 0 0 0 1 0 0 1
Pakistan 1979–86 1 0 0 1 0 0 0
Pakistan 1971–71 1 1 0 1 0 0 1
Panama 1987–93 0 1 0 0 0 0 1
Paraguay 1977–81 1 1 0 1 0 0 3
Paraguay 1996–96 0 1 0 1 0 0 0
Peru 1968–74 1 1 0 1 0 0 1
Peru 1991–95 1 1 0 1 0 0 1
South Africa 1975–82 1 1 0 1 0 0 3
South Africa 1985–92 0 1 0 0 0 0 1
South Korea 1973–77 1 1 0 1 0 0 1

Continued
Table 6.4 Continued

Target Dates Humanitarian Indicator

Health Food Water Environment Education Governance Total

Sri Lanka 1961–65 1 1 0 1 0 0 3

Sudan 1989–94 1 1 0 1 0 0 1
Sudan 1993–Present 1 1 0 1 0 0 1
Syria 1986–94 1 1 0 1 0 0 1
Taiwan 1976–77 0 0 0 0 0 0 0
Thailand 1990–93 0 0 1 1 0 0 0
The Gambia 1996–98 0 1 0 1 0 0 2
Togo 1992–94 1 1 0 1 0 0 1
Turkey 1974–78 1 1 0 1 0 0 3
Uruguay 1976–81 0 1 0 1 0 0 0
USSR 1975–90 1 1 0 1 0 0 1
USSR 1978–80 0 1 0 1 0 0 0
USSR 1980–82 0 1 0 1 0 0 2
USSR 1981–81 1 0 0 1 0 0 0
USSR 1983–83 1 0 0 1 0 0 0
Zimbabwe 1983–88 1 0 0 1 0 0 0

Source: World Development Indicators (2006) and author’s calculations.

 ANALYSES OF SANCTION EFFECTIVENESS 145

An example such as U.S. sanctions against India between 1978 and

1982 illustrates the data interpretation. The zeros represent no data
change; health care and water availability did not change over the
sanction period. However, environmental conditions worsened, while
food and education availability and governance conditions were aug-
mented. Are sanctions meant to worsen these conditions? Absolutely
not, especially if the sanctions are “smart” and focused. Since the
human indicator (HUMAN from here) is a discrete dependent
random variable, the model chosen here is an ordered logit model,
with similar exogenous variables to Drury (2005).

Exogenous Variable Choice and Hypotheses

This model’s hypotheses revolve around the exogenous factors that
may affect human conditions during sanctions. The idea of smart sanc-
tions is that unless sanctions are made more focal, their human costs
are larger than they need to be. Further, factors that may affect the
political outcome of sanctions may also affect the human conditions.
Other factors follow Drury (2005) and their links to HSE (1990).
In short, factors such as the economic hegemony of the sender over
the target, the presence of third-party sanction busters, or whether
other policies are in place or not should also influence the human
costs of embargoes. The hypotheses are as follow, where worsening

35.0%

30.0%
% of observations

25.0%

20.0%

15.0%

10.0%

5.0%

0.0%
–6 –5 –4 –3 –2 –1 0 1 2 3 4 5 6
Possible values of humanitarian indicator

Figure 6.1 Humanitarian data distribution (% of total observations [65 obs]).

Source: World Development Indicators (2006) and author’s calculations.
146 ECONOMIC SANCTIONS

human conditions or higher human costs are HUMAN becoming

relatively more negative as these exogenous factors change in value.

Hypothesis 3: Sanctions that cause target exchange rates to depreciate

resulting in worsening human conditions. On the surface, if higher
exchange rates (target currency falling in value) indicates economic
damage from coercion, human costs should rise. The cross-sectional
effects of export, import, and financial sanctions are used as variables
to test this hypothesis.
Hypothesis 4: If the target is under economic distress before sanc-
tions, the human cost of sanctions is higher. If the target economy is
already experiencing economic problems, human costs should rise
even without sanctions. This is a dummy variable (distress  1, no
distress  0).
Hypothesis 5: If a black knight is present, the human cost should be
less. Analogous to political damage, if a sanction buster exists, their
substituting for lost target markets supports better human conditions
(black knight present  1, 0 otherwise).
Hypothesis 6: If there is more than one sender, human costs rise.
This is the opposite of the black knight problem. If other countries are
involved as senders, this condition should force more scarcity and
more human costs (multiple senders  1, unilateral sanctions  0).
Hypothesis 7: If the total trade between the sender and target is
relatively high, the human costs will also be relatively larger. If sender
economic hegemony is measured by the size of total trade between
sender and target, the sum of the pre-sanction percentages of target
exports and imports involving the sender, then the larger the size of
total trade involving the sender, the larger the human costs from
sanctions.
Hypothesis 8: If the sender’s costs are relatively large, the target’s
human costs should fall. If the sender’s costs rise due to sanctions
creating a target reaction that affects the sender economy adversely
(i.e., a retaliatory oil embargo), the shorter the sanction duration
should be and the lower the target’s human costs (sender costs are
integers between one and four, higher sender costs closer to four).
Hypothesis 9: If other statecraft policies are in place originating
from the sender, the higher the human costs of sanctions. Generally, other
policies are military or diplomatic restrictions. The more restrictive
the overall policy package, the higher the relative human costs should
be (other policies  1, 0 otherwise).
Hypothesis 10: The more recent the sanction, the lower the human
costs. For more recent cases based on start year, humanitarian
 ANALYSES OF SANCTION EFFECTIVENESS 147

Table 6.5 Humanitarian logistic regression results: initial effects of

sanction shock

Variable Results

Coefficient Std Error

Export sanction 1.220* 0.660

Import sanction 1.506* 0.896
Financial sanction 0.104 0.183
Target under economic distress? 0.506 0.412
Presence of a “black knight”? 0.329 0.961
Multilateral cooperation? 0.638 0.367
Total trade  export %  import % 0.011 0.007
Sender cost 0.564 0.366
Other policies in place? 0.249 0.616
Start year of policy 0.103*** 0.033
National security sanction? 0.695 0.669
Institutional involvement? 0.687 0.669
Prior relationship good? 0.756 0.576

Global Results
Psuedo-R2 0.194
Observations 65

Data Sources: HSE(1990), Drury (1998, 2005), and author’s calculations.

*** significant at the 1 percent level; ** significant at the 5 percent level;
* significant at the 10 percent level.

monitoring and aid is more likely to flow in the face of sanctions, even
from the sender’s themselves.
Hypothesis 11: If the sanction is for reasons of U.S. national security,
the target’s human costs are higher. Since sanctions involving national
security are likely to be more comprehensive and blunt, the human
costs should rise under these sanctions (if for national security  1,
0 otherwise).
Hypothesis 12: If institutions, such as the United Nations or
European Union, are involved in sanctions, the lower the human costs.
On the surface, the involvement of an institution should bring
more countries and more monitoring of human costs if any are
overtly involved. This hypothesis could go either way (institutional
involvement  1, 0 otherwise).
Hypothesis 13: If the prior relationship between the sender and target
is amicable, the human costs should be lower. If the nations are historically
friendly with each other, human costs are likely to be reacted to more
quickly, especially if prior relationships mean cultural connections.
148 ECONOMIC SANCTIONS

Table 6.6 Humanitarian logistic regression results: cumulative

effects of sanction shock

Variable Results

Coefficient Std Error

Export sanction permanent 0.007 0.018

Import sanction permanent 0.046** 0.020
Financial sanction permanent 0.003 0.006
Target under economic distress? 0.200 0.203
Presence of a “black knight”? 0.530 0.462
Multilateral cooperation? 0.553*** 0.175
Total trade  export %  import % 0.006 0.004
Sender cost 0.253 0.205
Other policies in place? 0.328 0.379
Start year of policy 0.022 0.016
National security sanction? 0.178 0.372
Institutional involvement? 0.715* 0.405
Prior relationship good? 0.700** 0.311

Global Results
Psuedo-R2 0.203
Observations 65

Data Sources: HSE(1990), Drury (1998, 2005), and author’s calculations.

*** significant at the 1 percent level; ** significant at the 5 percent level;
* significant at the 10 percent level.

Given these hypotheses, building the model is our next task, where
the exogenous variables are specific to the hypotheses given earlier.
Our major focus is on hypothesis 3, to test this hypothesis when
including the economic results given earlier.

The Model
This model’s specification is a discrete choice model, an ordered logit
model.21 Because the dependent variable, HUMAN, was built to have
ascendancy, the exogenous variables help determine the probability
that an exogenous shock causes more human costs (a coefficient that
is negative) or less (a positive coefficient on the respective independ-
ent variable), if the effect is significant. The results in tables 6.5
and 6.6 show the effects of the initial and permanent economic effects
from the VAR analysis given earlier on the human costs, holding the
pre-sanction control variables otherwise constant.
This model is not meant to be a causal model. No claims to causal-
ity are made here because without time series data for HUMAN or
an analog measure, there is difficulty testing for causality between
 ANALYSES OF SANCTION EFFECTIVENESS 149

economic coercion and human costs.22 In a similar way to the earlier

analysis, economic coercion flowing through exchange rate changes
acts as the shock; the evolution of the humanitarian conditions result
from these shocks.
The ordered logit regression is shown by the model in equation 6.2,
and the results follow. This cross-sectional analysis takes the selected
cases as observations. The following model is the regression formula
used in each case:

HUMANi  kXi,k i (6.2)

where HUMAN is the indicator calculated earlier, the data of which is

in the “Total” column of table 6.4. k is the 1 x k vector of regression
coefficients on the included exogenous variables, and Xi,k is the k  1
vector of explanatory, exogenous variables that test the hypotheses
given earlier.  is the error term in the regression, which has a logistic
distribution in this regression. Tables 6.5 and 6.6 test for the deter-
minants of human suffering in sanctions, where the economic shock is
first measured (table 6.5) by the one-month interaction term for each
sanction involved for sanction case i, and then for the cumulative
effects of the sanctions (table 6.6). If the exchange rate was
unchanged, the economic effect is considered zero, as if no sanction
took place. The remaining exogenous variables are control variables.

Results
From table 6.5, we see that export sanctions and import sanctions,
through their exchange rate effects, initially have ambiguous results on
human costs. Financial sanctions are not significant in their effects.
Export sanctions having a strong, negative effect on human costs is log-
ical. As a sender restricts its exports to an economy, those restrictions may
include specific goods that inhibit economic development: technologies,
agricultural equipment, other capital, and so on. Such sanctions were at
the heart of South African coercive measures, for example.
Import sanctions augmenting social conditions is somewhat con-
founding, but not when considering the exchange rate effects. If the
sender restricts travel to the target, for example, this may lower export
prices and attract other countries to buy the target’s exports. The
costs of such sanctions on the populace may be mitigated by natural
market effects; the scarcity issue is not the same as under export sanc-
tions. However, in both cases, the significance level is not extremely
high, which acts as a caveat on making too large an inference about
these effects.
150 ECONOMIC SANCTIONS

Among the control variables otherwise, the start year variable from
Drury (2005) is negative and significant in its effects. More recent
sanctions are correlated with more humanitarian problems, providing
evidence to support increased awareness concerning collateral damage,
a la smart sanctions. Since many U.S. sanctions against less-developed
nations are with formerly friendly nations (Iraq’s 1980 case is an exam-
ple), this problem may also be controlled by a less-developed nation
dummy variable. That variable is not included here because the level of
ad-hoc variable usage is already high, and determining which countries
are less developed versus others may also be definitional and problem-
atic. However, further research may find such a variable useful. It is
also perplexing that variables concerning multilateral sanctions, institu-
tional involvement, and the prior relationship between the sender and
target were not significant in their effects.
In table 6.6, which changes the model by dropping the initial eco-
nomic effects and including their permanent counterparts, the results
change. Import sanctions continue to be significant, positive in their
effects on human conditions, and have become more significant;
export and financial sanctions are both insignificant in the long run.
The import sanction’s significance and direction again is somewhat
intuitive, as targets likely adjust easier to these sanctions over time
concerning human costs. These results help the smart sanction case, as
sanctions have either positive or no effects on human conditions in the
long term, using exchange rate fluctuations as the economic shock
variable.
Smart sanction arguments to monitor sanction effects closely are
supported by the other significant coefficient results. Multilateral
sanctions are now significant, as are institutional involvement and a
positive prior relationship between the sender and target. For each of
these variables, the effects are detrimental to the populace. As multi-
lateral sanctions continue, the fact that more than one sender exists
harms the population through general scarcity. The logic here is the
same for institutional involvement, where some consortium is reduc-
ing the alternative options for the target economy. The positive prior
relationship likely hurts the populace because common people gained
from that prior relationship. The ease in goods flow suddenly reversed
due to sanctions may be this detriment. The multilateral sanctions
effects are the most significant of all the results, and the most
intuitive.
Import sanctions minimize humanitarian damage to the target
concerning human costs and are thus effective sanctions. Export
 ANALYSES OF SANCTION EFFECTIVENESS 151

sanctions are initially ineffective from a humanitarian perspective;

both financial and export sanctions are neutral on social costs in the
long run; in this way, these sanctions are also effective. However, if
there is multilateral cooperation or institutional involvement in policy,
especially against a prior friend in the target, the human costs could be
significant. The global performance of the model, as stated by the
pseudo-R2 in each case, is not extremely high.23 This is a general out-
come of cross-sectional studies, as there are many other explanatory
variables that exist to explain the true human costs of a sanction.
However, the variables are somewhat ad hoc here, and that caveat pre-
vails, unfortunately, throughout most sanction studies because of data
parsimony. Using a similar methodology, this empirical analysis con-
cludes with the political effectiveness tests.

The Political Effects

These effects are simultaneously the most important and most elusive.
There are three issues with defining the political effects of sanctions:
(1) the stated goal and its dynamics during the sanction episode;
(2) the target’s political conditions in place before, during, and after
sanctions; and (3) finding a variable that measures the sanction’s
relative political success. The HSE (1990) data, supplemented by
Drury (1998, 2005), provide an ad hoc statistic, but one that is more
robust than its contribution score counterpart or the original HSE
(1990) success score. Using the data generated from the economic
effectiveness results given earlier, the political effectiveness is a cross-
sectional examination, where each sanction case is tested for how the
economic shock affects the political outcome comparing the initial
effects to the longer term. The same exogenous variables are used as in
the humanitarian effects model, which parallels HSE (1990) and
Drury (2005). In fact, their results are directly compared to the fol-
lowing results when applicable.
Certain attempts stand out as strong examples of testing a sanc-
tion’s political efficacy using first the HSE (1985) data. Lam (1990)
used a probit model to test the HSE hypotheses and conclusions,
employing the four-value, political success score of HSE (1985) as
the dependent variable by converting it into a binary variable of suc-
cess or failure.24 Dehejia and Wood (1992) basically repeated Lam’s
(1990) analysis, and suggest that econometric analyses of sanctions
are inherently biased to the point where any assessment has so many
data and methodological flaws that any estimation is problematic.
152 ECONOMIC SANCTIONS

25

20
% of observations

15

10

0
1 2 3 4
Possible values of political effectiveness

Figure 6.2 Political effectiveness indicator for sixty-five observations (distribution of

values by percentage of total).
Sources: HSE (1990), Drury (1998, 2005), and author’s calculations.

Dashti-Gibson et al. (1997) underscore the key consideration

underlying these analyses. “The purpose of statistical estimation is
precisely to find generalizable relationships between variables. The
contribution of sanctions—or the components thereof—is precisely
what is to be estimated” (611–12). Their analysis uses the same scal-
ing as Lam (1990) to show how sensitive the results are to small
changes in specification and data choice. Consensus is followed here,
where the dependent variable is ordinal rather than binary, a la
Drury (2005).

Dependent Variable Choice

The HSE (1990) data, as in the other studies, provides a beginning
to these answers. The focus here is on combining several political
indicators and their links to both economic and humanitarian
variables above by modeling the target’s macroeconomy and its
links to political change. The approach here mirrors Drury (2005)
using a discrete dependent variable that is ordinal in tracking the
relative political change in each case. This variable is called “RESULT”
 ANALYSES OF SANCTION EFFECTIVENESS 153

from here, which ranges from one to four in value. Figure 6.2 pro-
vides a histogram of this variable’s values for the sixty-five cases
examined.

Exogenous Variable Choice and Hypotheses

Using the economic shocks and control variables from the hypotheses 3
through 13, the following hypotheses focus on the political outcome of
each case. The RESULT variable increases as the policy is more effec-
tive. These descriptions mirror Drury (2005) for comparison when in
common.

Hypothesis 14: Sanctions that cause target currency to fall in value

are more politically effective. If higher exchange rates (target currency
falling in value) indicate economic damage from coercion, the current
target regime will have pressure, from interest groups or a lack of
funding otherwise, to change policy.
Hypothesis 15: If the target is under economic distress before sanc-
tions, sanctions are likely to be more effective. If the target economy is
already experiencing economic problems, sanctions are likely to exac-
erbate political pressure on a regime already in trouble.
Hypothesis 16: If a black knight is present, the sanction will strug-
gle to be politically effective. This is the classic way in which the target
avoids the deleterious effects of sanctions. If a sanction buster can
replace the markets abandoned by the sender, the political pressure for
change is likely to fall.
Hypothesis 17: If there is more than one sender, political effective-
ness increases. This tests the ideas of the sanction cartel on the target’s
politics. The cooperation issue continues to be debated, as tested for
in Drury (1998, 2005).
Hypothesis 18: If total trade between the sender and target is
relatively high, the political effectiveness will also be relatively larger.
Sender hegemony is a classic variable claimed to help determine
economic coercion’s efficacy.
Hypothesis 19: If the sender’s costs of sanctioning are relatively large,
the target’s rulers are likely to be less pressured for change. If the sender’s
sanctions costs are large, the sanction’s potency is in jeopardy, and
should reduce the political pressure on the target’s decision makers.
Hypothesis 20: If other policies are in place originating from the
sender, the political effectiveness of economic statecraft should rise. Drury
(2005) states that certain sanctions are less likely to work without the
assistance of covert operations; thus this variable controls for that in
testing political effectiveness (42).
154 ECONOMIC SANCTIONS

Table 6.7 Political logistic regression results: initial effects of economic coercion

Variable Results and comparisons

Coefficient Std error HSE Drury

Export sanction 0.077 0.283

Import sanction 0.048 0.362
Financial sanction 0.146 0.120
Target under economic distress? 0.401 0.276 ** *
Presence of a “black knight”? 0.228 0.488
Multilateral cooperation? 0.549** 0.261 * ***
Total trade  export %  import % 0.011** 0.005 ** *
Sender cost 0.087 0.279
Other policies in place? 0.436 0.454
Start year of policy 0.029 0.027 ** **
National security sanction? 0.633 0.379 *
Institutional involvement? 0.195 0.464 **
Prior relationship good? 0.152 0.428

Global results
Psuedo-R2 0.138 0.078 0.119
Observations 65 114 114

Note: Drury (1998, 2005) is the source of both the HSE and Drury data (49).
*** Significant at the 1 percent level; ** significant at the 5 percent level; * significant at the 10 percent
level.

Hypothesis 21: The more recent the sanction, the more politically
effective it is. This is an ad hoc measure of international interdepend-
ence from Drury (2005); however, for less-developed nations, which
have been the focus of recent sanctions, their lack of interdependence
may lead to more damage (ibid.). This is a difficult hypothesis to
assess ex ante.
Hypothesis 22: If the sanction is for U.S. national security purposes,
the political effectiveness is likely to rise. Sanctions involving national
security are likely to be more painful for targets than other sanctions.
The political effectiveness, however, may be in question for such a case
due to a rally-around-the-flag effect.
Hypothesis 23: If institutions, such as the United Nations or
European Union, are involved in sanctions, the more politically effective
the sanction. Institutional involvement is meant to be an extension of
diplomatic prowess and hegemony as senders.
Hypothesis 24: If the prior relationship between the sender and
target is amicable, political effectiveness should rise. If the nations are
 ANALYSES OF SANCTION EFFECTIVENESS 155

Table 6.8 Political logistic regression results: cumulative effects of economic

coercion

Variable Results and comparisons

Coefficient Std error HSE Drury

Export sanction permanent 0.046 0.034

Import sanction permanent 0.039* 0.021
Financial sanction permanent 0.002 0.004
Target under economic distress? 0.384 0.268 ** *
Presence of a “black knight”? 0.313 0.452
Multilateral cooperation? 0.652*** 0.222 * ***
Total trade  export %  import % 0.009* 0.005 ** *
Sender cost 0.115 0.283
Other policies in place? 0.365 0.434
Start year of policy 0.259 0.427 ** **
National security sanction? 0.547 0.406 *
Institutional involvement? 0.026 0.447 **
Prior relationship good? 0.005 0.026

Global results
Psuedo-R2 0.157
Observations 65

Note: Drury (1998, 2005) is the source of both the HSE and Drury data (49).
*** Significant at the 1 percent level; ** significant at the 5 percent level; * significant at the
10 percent level.

historically friendly with each other, the cultural links may work with
the sanction’s efficacy on the target’s politics.

Using the ordered probit model for the initial effects and ordered
logit for the cumulative changes, the results of the regressions
follow.25 The regression results appear in tables 6.7 and 6.8. The
following equation outlines the model of political effects:

RESULTi  kXi,k  i (6.3)

where RESULTi ranges from one (not effective) to four (most effec-
tive) for each sanction case i. k is the 1  k vector of regression
coefficients on the included exogenous variables, and Xi,k is the k  1
vector of explanatory, exogenous variables that test the hypotheses
given earlier.  is the error term in the regression, distributed normally
in the probit and logistically in the logit.
156 ECONOMIC SANCTIONS

Results
From both tables 6.7 and 6.8, the results on the common variables
with Drury (2005) and his interpretation of the HSE (1990) results
match in terms of significance for total trade (a measure of the sender
economic hegemony over the target) and multilateral cooperation.
The cooperation variable is negative and significant in all the studies,
and this provides additional evidence that cooperation is not necessary
for sanction success, and in fact may be detrimental. Martin (1992),
Pape (1997), and Drezner (2003) all consider the issue of coopera-
tion deeply and the question remains somewhat unanswered; these
results corroborate Drury (2005) and HSE (1990) that cooperation
may be detrimental with fewer cases, fewer explanatory variables, and
a higher level of goodness of fit. The sender’s size driving political
change is extremely intuitive, and connects to the target’s elasticity of
substitution falling in cases where the sender dominates the target’s
international market options. One way this variable should be
enhanced is to include the sender’s financial dominance (or lack
thereof) over the target as well. The shock variables pick up some of
this explanation as well.
The significant common variables from Drury (2005) and HSE
(1990) that are not significant in this study are the economic distress
variable, and the control variables of institutional involvement, the
start year of policy, and whether sanctions are for national security
purposes or not. This may be because the economic shock variables’
inclusion picks up the explanatory effects of these variables in past
regression, or the sample size adjustment takes away cases from the
other studies that would otherwise replicate their results. Since the
goodness-of-fit measure is marginally greater in this study than in
Drury’s results, the inclusion of the exchange rate shock variables
positively influences the regression’s explanatory power, as other
included variables are essentially the same in both studies. To repeat as
a caveat of comparing these studies too much, using just the U.S.
cases here may bias these results upward in fit.
In the initial regression results in table 6.7, the coefficient on dis-
tress is negative as in other studies, with a p-value of 0.14, marginally
insignificant. Also notice that the start year variable is not significant
initially and then becomes significant and negative in the cumulative
effects shown in table 6.8. This could be explained somewhat by
globalization, as targets have been able to take advantage of global-
ization in more recent sanctions than in older cases, also a highly
debatable result.
 ANALYSES OF SANCTION EFFECTIVENESS 157

None of the economic shocks are significant initially. Once sanc-

tions begin, the target’s political reaction is not immediate. The power
in these results lies in the comparison to the long-term effects of eco-
nomic sanctions, shown in table 6.8. The cumulative results show that
the effects are stronger in the long term, becoming more significant
and with anticipated results per the hypotheses given earlier.
Import sanctions, in their cumulative effects, are positive and sig-
nificant. As import sanctions continue, the correlation with political
change is significant. The p-value on this coefficient is 0.06. The
other sanctions are not significant, though export sanctions have the
correct sign and a p-value of 0.17. Financial sanctions do not have
significant effects on political outcomes in these cases. Additionally,
this could be an anomaly of the sample size and concentration on
cases where the United States was the sender only. However, assum-
ing American hegemony over financial markets worldwide through-
out the last sixty years, and that most financial sanctions in the data
originate from or flow through the United States and its currency as
transactions, these results feel counterintuitive. Easily circumvented
and difficult to precisely define, financial sanctions may be victims of
rhetoric and lack a credible threat to targets. Import sanctions in con-
trast, and export sanctions to a lesser extent, may be more detrimen-
tal to the target and its rulers, translating to marginal changes in the
target’s politics.
In short, these results both differ from and mirror the results in
Drury (2005). The economic shock as characterized by a change in
exchange rates shows how sanction mix affects political outcomes.
Import sanctions are now seen as effective in both human concerns
and political changes. The other sanctions either have little effect or
are close to significant. Adding other cases may enhance these results,
again because the United-States-only cases may bias these results
significantly.

Conclusions
These estimations combine the intuition from chapters two, three,
and four, where the effects play out in the exchange rate as discussed
in chapter five. This model may also predict the economic effects in
new embargo cases. This chapter engaged in three empirical studies,
expanding the sanctions literature. Using exchange rate shocks as
measures of economic sanction effects in the first empirical exercise,
these outcomes are used as exogenous variables data for regressions
158 ECONOMIC SANCTIONS

and hypotheses tests concerning humanitarian and political effectives

of economic coercion. The results show that import sanctions are
more effective on human costs and simultaneously on changing the
target’s politics than any other sanction. Financial sanctions show lit-
tle efficacy, while export sanctions show mixed results. The exchange
rate shocks tie to chapter five’s model as a measure of economic effi-
cacy in sanctions. Future research may use a similar time series analy-
sis but different dependent variables such as the terms of trade,
domestic target prices, target unemployment, sender unemployment
in specific industries, and so on. From a humanitarian and political
perspective, it is likely the cross-sectional work will continue, where
the expansion will be better data and new control variables.
Following Drury (2005), some control variables show intuitive
results on both humanitarian and political outcomes. Cooperation
between nations, where the sender may be alone but other nations
implicitly assist in statecraft, was seen as having little effect.
Multilateral sanctions were detrimental to both humanitarian and
political outcomes from sanctions. The total trade between the coun-
tries, a connection to gravity models, is significant in the political
effects initially; as sanctions continue, total trade is less important.
These results show that sanction scholars can go beyond the models
and data currently available in conducting empirical studies, especially
where sanctions are viewed as macroeconomic policies.
 Chapter 7

Conclusions and Policy

Recommendations

Economic Coercion in a Spartan Fashion

Economic coercion, as an option in international diplomacy, has been
in existence since the Megarian Decrees of 435 BC. Many historians
debate whether these decrees, which acted to prohibit Megarians
from using Greek ports or markets, initiated the Peloponnesian War.
Current economic powers continuously debate over how to impose
sanctions on upstart, rogue nations, remaining somewhat divided on
actions and perceived consequences. North Korea’s missile and
nuclear tests in 2006 showed the world three key aspects of economic
sanctions and statecraft, as the United Nations debates new measures.
The first is how sanctions walk a fine line between economic coer-
cion and provoking war; the current case of North Korea and the
Megarian case parallel each other in this way. “It does seem clear that
in imposing restrictions of Megarian trade the Athenians were seeking
a middle course. To have done nothing may have encouraged hostil-
ity to Athens, whereas resorting to the military option would have
violated treaty obligations and provoked a military response from
Sparta” (Simons 1999, 14). War is a constant specter hanging over
many sanctions, though some targets cast a much longer shadow than
others concerning geopolitics.
Second, certain target countries feel the weight of sanctions over
time and retaliate. North Korea, under full U.S. financial sanctions
since 1993, repeatedly states she is willing to negotiate over missile
and nuclear programs if the United States began to lift its financial
embargoes. This shows that current sanctions are having some politi-
cal effects, pushing North Korea to develop a bargaining chip for the
negotiation table. However, are they hurting the regime, the people,
160 ECONOMIC SANCTIONS

or both? If Kim Jung Il is willing to provoke a military episode, it

seems that sanctions are hurting the regime; war is likely to be very
detrimental to the North Korean people ultimately. The UN reaction
to the July 2006 missile tests, which the international community
agreed was itself a move of North Korean statecraft to awaken the
world, was initially split. China and Russia, two of North Korea’s
major trading partners and historic sources of military hardware,
refused to initially endorse Security Council Resolution 1695 to
impose sanctions on Kim Jung Il’s nation in 2006, while Western
Europe, South Korea, Japan, and America pleaded for this resolution.
Third, economic statecraft struggles to achieve political goals when
the bar is set too high. Analogous to Cuba, Libya, Iran, and Iraq, it is
unlikely that North Korea’s international stance will change before
the death of its current leader, a death that does not guarantee the
regime’s demise. If the sanctions had worked correctly, vis-à-vis their
stated goal, the brinksmanship of 2006 would not have occurred.
However, this is the bane of economic statecraft: focus on a political
goal is unlikely to be consistent or precise. As in Ancient Greece,
economic coercion can provoke exactly what it is trying to avoid. That
leads to a classic and constant question in this literature.

Are Sanctions Effective?

This text outlines the political economy of economic sanctions, where
the perspectives and studies are wide in academic genesis, focused
throughout on what sanctions do. In a time where the world powers
have generally moved away from the military option unless provoked,
and where the United States has become the world’s leading sanction
sender, three basic economic ideas drive the literature’s generalization
that sanctions are ineffective. First, sanctions credibly reinforced by
military options make economic coercion meaningless. Sanctions are
no longer a precursor to military conflict; they have become a substi-
tute or complement for military conflict. The empirical analysis of
chapter six suggests that U.S. import sanctions have shown humani-
tarian and political effectiveness, while export and financial sanctions
are mixed in their effects. Since most sanction packages are a mix of
tools, this ambiguity may lead to these perceptions of ineffectiveness.
Second, the inability to gain enforceable international coalitions
implies there are many third-party options for targets to circumvent
sanctions further, reducing the credibility of sanctions. Ownership of
a highly demanded, natural resource, for better or worse in our world,
provides some insurance against effective sanctions. Nigerian sanctions
 CONCLUSIONS 161

of the 1990s show evidence that certain countries decided not to

sanction a military, human rights violations machine because of poten-
tially losing Nigerian oil resources (O’Sullivan 2003). As Asia’s
demand for oil continues to increase parallel to its economic growth,
the U.S. ability to sanction oil-exporting nations is likely to fall
further. However, larger nations provide avenues for countries to
avoid deleterious sanction effects by providing alternative trade
routes. Cultural and ideological connections, as between China and
North Korea, may be as powerful a force as profit.
Third, and maybe most important, the sanction goal may be
defined such that sanction effectiveness is doomed to failure ex ante.
Cuba has become a foundational case as to why sanctions are not
effective, especially in penalizing the populace while targeting the rul-
ing class. How could sanctions defined by complete political turnover
ever work outside of a confluence of circumstances or military force?
North Korean sanctions have become another important example of
continued sanction ineffectiveness based on stated goals. Kim Jung Il
remains as the North Korean leader while sanctions continue to
mount.
The Sanction Effectiveness Continuum has four points along
which a sanction travels: ineffectiveness, economic success, humani-
tarian success, and political success. This is a continuum because sanc-
tions are not meant to discretely jump from one success level to
another, but to move marginally toward stated political goals.
Sanction effectiveness must be seen as steps toward a goal, not based
completely on goal achievement. Policy makers must change their
view on sanctions mechanics. This chapter concludes the text and
makes policy recommendations under the belief that there are various
levels of sanction effectiveness. The Continuum begins with sanction
ineffectiveness.

Sanction Ineffectiveness
If the payoffs from sanction initiation provide a dominant strategy to
engage in economic coercion measures, the policy maker must moni-
tor transition points. Theoretically, the reason why sanctions end is
because continuing sanctions are no longer the sender’s dominant
strategy. If the stated goal is to apply economic and political pressure
on the current target regime to curb their behavior, many sanctions
apply such pressure forcing the target economy to seek new markets,
imposing higher costs on the target economy. Sanctions are effective
generally in one of the following three ways.
162 ECONOMIC SANCTIONS

Economic Effectiveness
Cuba and South Africa are strong, contrasting examples of economic
versus political effectiveness. In the Cuban case, the stated goal was
and remains Fidel Castro being removed from power. This goal has
not changed, nor will it until Castro’s death.1 Economic success is a
function of how the target economy is injured by policy. Chapter five
of this book explained a new model of sanction effects using open
economy macroeconomics, where the target’s real exchange rate path
should follow the sanction effects. If the sender truly has the ability to
cause exchange rate fluctuations, reflecting macroeconomic reactions
of one country to another’s policies, then sanctions are simply another
type of macroeconomic policy. What may force a sanction episode’s
end is the imposition of large human costs.

Humanitarian Effectiveness
Smart sanction analysis has developed due to the damage economic
coercion is perceived to impose on an “innocent” populace. Less-
developed nations have been, and likely will be in the future, the
quintessential sanction target due to inherent political instability. Can
sanctions be constructed in such a way so as to minimize collateral
damage, or are they meant to harm the populace as a means to an end?
From the previous chapters, it seems unlikely that sanctions can avoid
damaging the public at large, and as a result struggle to focus on any
one group.
For sanctions to be effective from a humanitarian perspective,
policy must constrain its effects to the target rulers. Socioeconomic
and human development measures provide data sources to potentially
track a policy’s collateral damage. In many sanction cases, there are
other countries to provide food, clothing, medical supplies, and other
essentials while sanctions take place. Of course, depending to which
theory the policy makers subscribe, the more the harm done to the
populace, the more likely it is that the sought-after change takes place.
Others argue that the more the populace suffers, the more likely they
will rally around their current leader, especially if effective propaganda
campaigns takes place or a specific regime exists that imposes its will
on its people.
Economists struggle with measuring international human condi-
tions due mainly to poor data rather than poor methods, defaulting to
income inequality measures such as the Gini Coefficient, net mortal-
ity rates, misery indices, education statistics, and others, as broad stabs
 CONCLUSIONS 163

in tracking socioeconomic conditions. These data, unfortunately,

simply do not exist for many countries. Humanitarian effectiveness in
economic sanctions may be seen as an afterthought due to a lack of
consensus over what data to track for humanitarian conditions and
data availability, which is a mistake. Human suffering is likely to
mitigate economic effects that would otherwise lead to political
efficacy.

Political Effectiveness
For policy makers, defining political success should not be organic, it
should be realistic. Of course, the luxury that analysts have is clear
vision concerning the past. Political effectiveness in economic coercion
comes in two general ways.

Intended Change in Target Behavior,

Including Governmental Overthrow
This is true political success. The target actions are reversed, and the
sender’s intention comes to fruition. The sender is now challenged to
release the target economy from the embargo and monitor how the
target continues as a member of the international community. Also,
how the populace reacts to a new government may be a key factor in
how these policies truly become effective sanctions. Haitian sanctions
are an example of sanction efficacy on high one day and looking terri-
ble the next: problem government falls, democratic government
enters, humanitarian problem ensue with disastrous consequences
(Simons 1999, 107). The Sanction Effectiveness Continuum is meant
to keep policy makers and scholars focused on the steps sanctions
ascend to take policy from initiation to success or failure. Most rulers
do not leave office willingly, and thus sanctions rely on positively
affecting opposition forces in the target economy, and negatively
affecting the current ruler’s ability to govern. Chapter four suggested
that public choice drives the sanction’s political effectiveness and is
tied to the level of opposition in the target economy.

Marginal Change in Target Behavior

The sender’s political goal may be on the road to achievement
through coercive measures, but doomed to never be achieved. This is
a major problem with many sanctions, as the sender’s original intent is
not realized, even though the target’s politics actually turn toward the
stated goal. Rhetoric, for better or worse, is its own worst enemy,
especially in supporting the credibility of further action. Chapter three
164 ECONOMIC SANCTIONS

suggested that game theory models exist where the commitment level
of senders is based on issuing credible threats. The more the target’s
behavior turns toward the sender’s demands, the more successful the
sanction episode on the margin. Chapter six showed that political suc-
cess is most likely through import sanctions, as is with humanitarian
effectiveness from the same policies.
This text has discussed many models and ways to track sanctions,
where effectiveness is viewed in stages rather than as a single political
triumph at any cost. General conclusions and policy recommendations
follow.

General Conclusions
Economic Statecraft Must Act Like a Cartel
Sanction regimes that act like cartels split the world in three pieces.
Certain countries either do not react or pledge their support for or
against the embargoes. The policy cartel forces markets to accept a
cost of political deviance that the countries within the policy cartel
agree is the benefit-maximizing level of sanction imposition. In this
way, the target’s political market consumes the cartel’s policy at a
monopoly price. Once multilateral sanctions have been initiated, the
sender must worry about nations cheating on the agreements, substi-
tutes markets coming to the surface, and the fact that the sanctions are
not homogenous products. If the sender can act as a dominant firm
(the United States), driving the coalition (the remaining United
Nations) because it is a dominant leader, then the game and public
choice theory lessons fall quickly into place toward explaining sanction
efficacy.

Sanctions Are Macroeconomic Phenomena

While there is much discussion and hope that sanctions can be
analyzed and devised as microeconomic phenomena, this view is
naïve. Economic coercion aims to affect national political decisions,
whether that is to isolate a portion of or the entire target economy,
they must be viewed analogous to monetary and fiscal policies. It is
important to also recognize economic statecraft is not a trade policy
or a capital control or exchange rate policy in classic ways. Trade and
financial connections between two countries are manipulated for
political reasons in a sanction episode. Export and capital controls,
 CONCLUSIONS 165

nontariff barriers to imports, and sanctions are just different terms for
similar policies. The key is why the policy is applied.
Sanctions should be viewed as macroeconomic policies because
the policy’s intent is to transmit costs upon the target as a whole. The
beggar-thy-neighbor effects displayed in the New Open Economy
Macroeconomics (NOEM) models is exactly what sanctions intend to
do. In these models, the analysis centers on how macroeconomic policy
originating in a larger country may lead to lower welfare in other coun-
tries due to exchange rate effects. In chapter five, the model showed how
sanctions can be modeled as such policies; chapter six provided estimates
as to the exchange rate effects in sixty-five episodes. While the results
were ambiguous under some policy combinations, examining macro-
economic variables investigates what sanctions are at their core.

Smart Sanctions Are Great In Theory,

Difficult in Practice
From their definition to implementation and beyond, smart sanctions
as devised in the literature are rife with problems. The theoretical idea
of smart sanctions is very appealing: focused sanctions that minimize
collateral damage and reduce the target government’s resources to be
pernicious internationally or oppressive domestically. Smart sanctions
are a light form of trade and financial sanctions. They are simply non-
comprehensive sanctions on specific goods and financing deemed
focal to the stated political goal; these measures are simply specific
statecraft and one should expect generalized effects from a blunt
instrument (Craven 2002).
Sanction policy should be focused rather than blunt instruments if
possible. Policy makers and scholars all should agree that minimizing
damage to the target populace while maximizing the political effects is
exactly what sanction efficacy and efficiency is all about. Sanctions
seen as blunt, broad instruments initially, where the sanction mix
determines the effects, may be a compromise. However, in any sanc-
tion, collateral damage must be expected. Differentiating smart sanc-
tions from others is currently a semantic exercise rather than a research
question. Chapter six showed that sanctions generally have little effect
on human conditions to date through a cross-sectional analysis of his-
toric cases.
These conclusions summarize the main points concerning sanc-
tions made in this text, leading to simple policy recommendations for
any economy deliberating sanctions as statecraft.
166 ECONOMIC SANCTIONS

Policy Recommendations
Nations Must Seek Sanction Alliances
with Nations Regionally
For countries such as the United States, slipping economic hegemony
signals a need to reassess alliances and agreements such that a united
approach to economic statecraft can take place. Drury (2005) elo-
quently described how the U.S. president has initiated sanctions on
many occasions. These unilateral measures may have been more effec-
tive if a coalition of nations, especially those countries possessing large
percentage shares of alternative trade and financing with a specific
target, was in place regionally to act as sanction allies.
Alliances are only as strong as their weakest members. Such coali-
tions would diminish the American president’s power in engaging
sanctions. Such alliances cannot hurt economic statecraft’s cause.
Any sanction alliance must be built with the strongest world
economies, not just the larger democracies, and retain vision enough
to allow emergent economies to join rapidly. Russia, which recently
failed to join the WTO because of antipiracy problems and a lack of
commitment to international legal issues involving trade, could
make a better case to join in American eyes if it were willing to act as
a parallel sender on sanctions. Such an alliance would also take the
focus off of alleged American neocolonialism and on the task at
hand: maintain a world where economics is used rather than the mil-
itary to solve regional diplomatic problems with renegade govern-
ments. Ideological issues must also be secondary to what ultimately
is a change in the way the United Nations uses tools of economic
diplomacy.

Build and Agree On Empirical Models

to Predict Sanction Effectiveness
This text has steered away from discussing sanction “success” and
instead focuses on marginal movements as measures of “effective-
ness.” Sanctions have effects that can be viewed by various parties as
successful or not: relative effectiveness is the key measure to watch.
Economic effectiveness is primary. Most sanctions pass this test.
Chapter six’s economic model provides this evidence through simple
econometrics to predict how a sanction shock will change the target
economy’s exchange rates. The exchange rate shock should be a reac-
tion to the economies changing, moving in a predictable fashion, and
may also affect the sender.
 CONCLUSIONS 167

Humanitarian effectiveness is not secondary per se, but is only

observable once the economic effects take place. The estimation of
humanitarian efficacy in chapter six provides a beginning to predict
how exposed the target is to social costs. Social problems can become
political problems, for either the sender or target, depending on how
the target government spins the social costs that flow from sanctions.
Humanitarian effectiveness must be autonomous of political effective-
ness and keep from initiating rally-around-the-flag effects anywhere in
the world.
The sanction’s political effects are based on a specific, stated goal
but really on moving the target nation marginally toward that goal. If
the goal is stated as political turnover, the sender is overstating its eco-
nomic power. Even at the height of American economic hegemony in
the twentieth century, using economics alone could not force such
political turnover. Political success, if based on goals of substance
rather than rhetoric, is much easier to achieve marginally.

A Sanction Must End

Sanctions, like any other economic decision, are subject to the law of
diminishing marginal returns. As time wears on in perpetual sanctions,
such as U.S. sanctions on Cuba, the measures’ power fades for two
reasons. The price of continuing sanctions rises over time for all par-
ties. The sender’s maintenance and oversight costs alone rise in cost
due to higher wages, attempts to circumvent the measures, and pres-
sure groups lobbying for sanctions to end. There are also diplomatic
costs: if sanctions continue, their goals have not been achieved. If
social costs result, the longer the policy remains, the more the target
populace suffers. The international community may call for sanctions
to end simply to mitigate the target populace’s suffering. The sender
nation also suffers in both explicit and implicit ways by supplying
sanctions for long periods of time, such that the demand for sanctions
within the sender nation is likely to fade with time. The original polit-
ical rationale is likely to erode through social change in the least, and
the price the public is willing to pay is likely to fade. The key problem
with ending sanctions is what price is paid to credibility in ending
these measures prematurely, another diplomacy tradeoff for the ages.

* * *

At the time this text is being written, the United Nations is deciding
what to do about North Korea. Three months after Security Council
168 ECONOMIC SANCTIONS

Resolution 1695 was passed, North Korea conducted a nuclear test

underground, in violation of the 1974 Non-Proliferation Treaty. It is
likely that if Kim Jung Il’s regime remains belligerent, and continues
to upset its largest ally, China, due to the possible trade problems they
would face by not following new UN proposals, these sanctions are
likely to increase in intensity and their effects. Only time will tell as we
enter year fifty-seven of North Korean sanctions. Unfortunately,
brinksmanship and other acts of provocation are driving for just that,
the antithesis of economic statecraft. And so it goes.
 Appendix 1

Brief Cases Histories of Selected

Sanction Episodes

These brief histories focus on U.S. economic statecraft initiation and

continuation since 1960.1 Providing background information, this chap-
ter pays homage to and derives much from the extensive chronology
provided by Hufbauer, Schott, and Elliott (HSE 1990) plus new studies
and data. These brief histories provide other information, not empha-
sized by HSE, concerning certain macroeconomic variables. Other
sources are quoted as needed case by case. This appendix provides the
reader a taste of these cases’ diversity while not substituting for texts with
more focus on specific, multiple cases: HSE (1990) remains the ency-
clopedic work. O’Sullivan (2003) looks at recent cases in outstanding
depth and detail; Askari et al. (2004) looks at multilateral cases.
Table A1.1 provides the cases and dates for the examples in this
chapter, all with the United States as unilateral sender or a partner in a
multilateral policy. Sanction studies have now focused on effectiveness
in many different ways, not just a binary success or failure statistic.
Appendix 1 presents historic cases and shows differences and similari-
ties in each sanction’s political economy, providing background for the
empirical work in chapter six. Some of these cases are not mentioned
elsewhere in the text qualitatively, but all cases are part of chapter six’s
empirical analysis. Stated data made in each case, though available from
multiple sources, originate with the International Monetary Fund’s
website, www.imfstatistics.org, unless otherwise referenced.

An Isolated Case: North Korea, 1950–Present

Due to extreme data parsimony, North Korea is for now only a quali-
tative study. Some of the classic sanction issues exist. The sanctions are
170 ECONOMIC SANCTIONS

Table A1.1 Selected cases

Target country Sanction dates Target country Sanction dates

Argentina 1977–84 Libya 1978–98

Argentina 1978–82 Malawi 1992–94
Brazil 1962–64 Myanmar 1988–90
Brazil 1977–84 Nicaragua 1977–79
Cameroon 1992–98 Nicaragua 1981–88
Chile 1965–66 Nicaragua 1992–95
Chile 1970–90 Nigeria 1993–98
Chile 1973–90 Pakistan 1971–72
China 1989–98 Pakistan 1979–86
China 1991–98 Panama 1987–93
Colombia 1996–98 Paraguay 1977–81
Cuba 1960–Present Paraguay 1996–96
Egypt 1963–65 Peru 1968–74
El Salvador 1977–81 Peru 1991–95
El Salvador 1987–88 South Africa 1975–82
El Salvador 1993–93 South Africa 1985–92
Ethiopia 1976–92 South Korea 1973–77
Guatemala 1977–86 Sri Lanka 1961–65
Guatemala 1990–93 Sudan 1989–94
Haiti 1987–90 Sudan 1993–Present
Haiti 1991–94 Syria 1986–94
India 1965–67 Taiwan 1976–77
India 1971–72 Thailand 1990–93
India 1978–82 The Gambia 1996–98
India 1998–2005 Togo 1992–94
Indonesia 1963–67 Turkey 1974–78
Indonesia 1993–95 Uruguay 1976–81
Iran 1979–81 USSR 1975–90
Iran 1984–98 USSR 1978–80
Iraq 1980–87 USSR 1980–82
Iraq 1990–2003 USSR 1981–81
Liberia 1992–98 USSR 1983–83
Zimbabwe 1983–88

Sources: HSE (1990), Drury (1998, 2005), and author’s calculations.

long term; in HSE (1990) it was the longest running sanction in their
data set and remains so to date.2 The sanctions changed in 1994,
amidst the loss of allies in the former Soviet Bloc and reduced conces-
sion trade with China in lieu of hard currency trades. Further, the
philosophy of self-reliance, juche, drove North Korea toward a nuclear
program and profiting from selling arms and weapons systems to
belligerent nations in the Middle East. Sanctions that began in 1950
 BRIEF CASES HISTORIES 171

as parallel acts to the Korean War continued as a way of stemming

communism’s tide during the Cold War. It is likely that the lack of
dynamics and data in this episode has led to few studies before 1994.
Recently, authors such as Dreszner (1999), Simons (1999), and
Martin (2000) have discussed North Korea at length. HSE (1990)
provide extensive background, likely to be repeated in their third edi-
tion. Martin suggests that U.S. sanctions against North Korea became
driven by incentives rather than aimed at punishment (2000, 107).
The carrot–stick idea is also featured in Dreszner (1999), where positive
incentives and punishment are discussed as policy options.

The threat of economic sanctions [rather than positive incentives] was

unsuccessful in generating all the desired concessions [concerning
nuclear fuel reprocessing] for two reasons. First, North Korea’s expecta-
tions of future conflict were so extreme that it preferred stalemate to total
acquiescence . . . Second, the United States could not compromise its
demands due to the normative importance it placed on nonproliferation
issues. (283)

This case is unique because of its duration, as coercive policies con-

tinue with worsening geopolitics that have forced the United Nations
to reassess its sanctions package against Kim Jung Il’s regime. North
Korea claims economic isolation of its people, save relationships with
Japan, China, and South Korea. U.S. sanctions are full sanctions, and
deemed a function of national security. Elliott (2003) suggested a
lifting of all nonproliferation and non-security related coercion
policies (6). Rennack (2003) provides an outstanding overview of the
case, stating that sanctions continue because North Korea remains a
potential terrorist nation through funding and committing direct acts
of terror, a nonmarket state under the Trade Agreement Extension
Act of 1951 and the Trade Act of 1954, and engaged in nuclear pro-
liferation (5–11). However, tracking sanction effects outside of the
anecdotal is virtually impossible because no acceptable data exist.
In October 2006, as this study was being completed, North Korea
conducted an underground nuclear test. This is likely another act of
brinksmanship, as were the missile tests of July 2006. North Korea may
see the end of sanctions very soon, with full-scale conflict. The following
are brief histories of the cases empirically examined in chapter six.

Argentina: 1977–84
Argentina experienced two different, related sanctions between 1977
and 1984. The first was a reaction to human rights abuses, according
172 ECONOMIC SANCTIONS

to UN stipulations. Political murders took place by the score, as

Argentina’s military government attempted to eliminate its enemies.
In February 1977, President Jimmy Carter suspended aid to
Argentina; Argentina immediately refused aid from the United States
(HSE 1990, v. 2, 445). It decided to finance its military and economy
through its domestic capital market. Debt rose steadily, and the
human rights problems subsided from 1977 to 1982.
Argentine president Galtieri in 1982, amidst a large military
buildup and economic implosion, sent 4, four thousand troops to the
Malvinas (Falkland Islands), a British protectorate. Immediately, the
United Kingdom and United States began full economic sanctions
against Argentina, and Argentina retaliated by beginning a military
conflict. The war ended quickly, though the sanctions lasted another
full year. By the end of 1983, the Argentine government was over-
thrown. The economic situation in Argentina from 1975 to 1984 was
a roller coaster of mismanaged central banking, large debt excesses for
military control purposes, and international policies that restricted
many markets (Di Tella and Dornbusch 1988, 311). Economic and
military aid from the United States was cut off by 1977, not to return
until 1984. In 1982, the United Kingdom cut all its Argentine aid and
trade. In 1976, 7 percent of Argentine exports went to America or
Britain, while 19 percent of their imports originated in these sender
economies. The 1977 sanctions from America were still on during the
Malvinas conflict. The decision to invade the Malvinas put additional
economic pressure on the Argentine capital markets already suffering
from monetary policies that had caused three devaluations in five
years (Corradi 1985, 132). The sanctions against Argentina were full
sanctions, consisting of both trade and credit. Once the overthrow
took place, sanctions were lifted.

Brazil: 1962–64, 1977–84

During the early 1960s, the threat of U.S. financial sanctions cast a
cloud over every decision Brazilian president Goulart made. In 1962,
Brazil wrote into law a bill limiting the amount of income international
firms could derive from their Brazilian subsidiaries and investments.
Goulart ordered capital expropriation, petroleum-producing capital in
particular, from U.S. firms in 1962 (Wesson 1981, 22). This decision
was blamed on pressure from leftist forces within Brazil, forces taking
a stand against aggressive capitalists (U.S. investors in specific) invad-
ing Brazil. Once the capital was expropriated, the United States
formally cut 75 percent of its economic and military aid to Brazil, on
 BRIEF CASES HISTORIES 173

the advice of a Brazilian government official that sanctions would have

large effects on the Goulart government’s ability to retain power (HSE
1990, v. 2, 222). America influenced the International Monetary Fund
and UN agencies to delay loans to Brazil. From 1940 to 1965, the
United States was the largest supplier of loans to Brazil internationally,
a pattern developed between many Latin American nations and
America in the postwar era (Salazar-Carrillo and Fendt 1985, 179).
Robert Wesson reported that six hundred million dollars in loans
were approved by the United States between 1960 and 1962, but
only 20 percent of these loans landed in Brazil (1981, 107). Goulart
was ousted in 1964, and the U.S. aid packages returned. The military
government that followed Goulart was a human rights disaster, limit-
ing political freedoms and holding no free elections. A purge of com-
munists took place in Brazil, including book-burnings and labor
union disintegration. Inflation increased 86.6 percent in 1965, due to
monetary expansion and economic stagnation.
In 1977, Brazil’s military government refused military aid from the
United States after perceived human rights abuses by Brazil’s govern-
ment. America nominally cut aid for the next seven years. In 1984,
Reagan established a new military aid relationship with Brazil; these
sanctions suppressed Brazil’s international capital market by direct
intervention. The sanctions restricted only aid and credit directly. The
Brazilian economy grew between the first and second sanction
episodes; in the 1962 episode, U.S. real GDP was thirty times her
Brazilian counterpart, shrinking to twelve times the size by 1977.
Brazil’s trade dependence fell from 1962 to 1977 also, as exports to
the United States fell significantly from 62 to 17 percent of total
exports, and Brazil’s imports from America fell from 35 to 20 percent.
In the first episode, Goulart was forced out of office in 1964 in a
military coup. Burns (1993) saw the U.S. economic sanctions as tanta-
mount to the military overthrow of Goulart, as Goulart became unable
to control his leftist government’s position after losses of aid and loans.
In contrast, Wesson (1981) saw the sanctions as doing little in the face
of U.S. military presence and an unprecedented call for military rule in
Brazil.3 The second episode’s end came with less excitement, as America
simply reversed its position on Brazil’s human rights situation, electing
to grant aid that the Brazilian government had earlier renounced.

Cameroon: 1992–98
In early 1992, soldiers engaged in summary executions of political
demonstrators. Political violence continued to sweep through
174 ECONOMIC SANCTIONS

Cameroon through 1993, as the Cameroon Democratic Union’s vice-

president was abducted and killed; this assassination was suggested to
be politically motivated (U.S. Dept. of State, 1994a). Constitutional
changes took place such that there was less democratic and personal
freedom in Cameroon, worse its citizens human rights situation. As a
reaction, the United States cut aid and oil imports from Cameroon.
The Cameroon economy remains based on oil as its main export,
but it also exports coffee and other agricultural products. Cameroon
sold 7.5 percent of its exports in 1990 to the United States, down to
0.4 percent in 1991, while it purchased 6.6 percent of its imports
from America in 1991. Cameroon’s economy suffered in the early
1990s. Real GDP fell at a rate of 2.9 percent per year from 1991 to
1994 (U.N. Statistical Yearbook 1997). From 1995 to 2000,
Cameroon’s economy rebounded well, fueled by better politics and
economic growth worldwide.

Chile: 1965–66; 1970–90

Chile was the world leader in the copper markets in the early 1960s.
In 1965, due to rising international demand for copper, Chile
increased the raw copper price from $0.36 to $0.38 per pound. The
copper price increase was followed by American copper and aluminum
companies augmenting their prices, as aluminum and copper were
close substitutes on the world market. Copper was Chile’s principal
export market, making its copper earnings almost entirely dependent
on export prices (Corbo and Fischer 1994, 31). The United States
restricted imports of Chilean copper, and aid was nominally cut. By
1966, America reinstated aid to Chile, and the raw copper price
settled to the pre-sanction price of $0.36 per pound. Chile’s depend-
ence on American sales fell dramatically, possibly a function of the
1965 sanctions; by 1970, the United States made up only 9 percent of
Chile’s export market, down from 43 percent in 1965. America sold
Chile 30 percent of her imports in both 1965 and 1970, however.
HSE (1990) suggested that the import restrictions on copper, cou-
pled with aid and loan sanctions, forced Chile to rethink its pricing
policies (276–79).
The 1970 sanctions against Chile were much longer in duration.
These sanctions focused on ending the presidency of Salvador
Allende. Before Allende’s government, there was the Chilean Christian
Democracy. Unfortunately, both the political right and left (Allende)
opposed tax and social reforms. In 1970, by a slim margin, Allende
won a parliamentary election. The United States consequently cut
 BRIEF CASES HISTORIES 175

loans and aid to Chile, as asset expropriation invoked sanctions via the
Hickenlooper Amendment.4 Allende was killed during a coup d’etat
in September 1973. Of the many groups attempting to seize power,
the military took control. The sanctions against Chile continued
through the 1970s and 1980s, shifting focus to new human rights
problems during the new government’s regime.
Between 1973 and 1990, a military government ruled Chile,
suspending parliament, restricting freedom of press, and participating
in other human rights violations. The United States reacted by cutting
financial and military aid. Credit sanctions continued into the 1980s,
with many battles in the U.S. Senate over aid reinstatement. U.S.
sanctions became more nominal over time, and by 1990 were off
completely. In certain years, 1976 and 1985 particularly, U.S. restric-
tions eased, with loans and aid given for “humanitarian” purposes
(HSE 1990, v. 2, 360–61). The Chilean dependence on international
trade began to slow the economy, leading to currency devaluations
and revaluations of debt (Dornbusch and Edwards 1994, 85).
Financial sanctions were the only restrictions used in the 1970
sanctions. Politically, Allende’s assassination in 1973 ended his regime,
bringing a new government into power. This new government, how-
ever, would compile just as bad a human rights record as Allende’s.
Political goals were achieved, one quickly and the other over a long
stretch of time. Chile’s imports were 27 percent from the United
States in 1973. Chile experienced high inflation and volatility in real
GDP growth, though private investment from foreign sources grew
between 1977 and 1981 by three billion dollars.

China: 1989–98
China was first sanctioned in 1949, measures that continued through
1970. These sanctions were concerned with the spread of communism
along the Pacific Rim, especially after the Korean War began in 1950.
Once the Korean War was over in 1953, international participation in
the Chinese embargo shrunk to only the United States and United
Kingdom; by 1958, Britain retreated as well. The U.S. goal was
a blockade of communist expansion (Leyton-Brown 1987, 68).
Sanctions ended in 1970 because of changes in U.S. trade strategies
(HSE 1990, v. 2, 101). In the 1980s, these fear resurfaced somewhat,
culminating in 1989.
Beijing’s Tiananmen Square was the site of student-driven, pro-
democracy protests in May 1989. The Chinese government declared
martial law in Beijing and moved in troops. Hundreds were killed,
176 ECONOMIC SANCTIONS

hurt, and wounded as troops opened fire on demonstrating crowds.

America acted immediately by curtailing all government contract
activity, arms flow, and loans to China. Sanctions stiffened to influ-
encing loan activity between China and international agencies and
governments. In 1988, the Chinese purchased 12 percent of their
imports from the United States and sold 7 percent of their exports to
America. Many countries did engage in some form of embargo in
1989, with the United States pressuring some loan deferments, but
the Chinese economy was little affected. Over the first five years of
sanctions, Chinese real GDP grew at 10.2 percent per year, with slow
inflation until 1992. These sanctions were more political reaction than
action, a stand the United States felt morally obliged to take.
Chinese sanctions were imposed again in 1994 due to alleged sales
of missile systems to Iran and Pakistan, and over goods piracy issues.
To try and stop yet another arms race between India and Pakistan,
these U.S. sanctions reduced sales of missile technology to China,
which the American government saw as going straight through to
Pakistan for profit. By the end of 1994, the sanctions were partially
lifted due to an accord between the countries on third-party transac-
tions (Wall Street Journal 1994). China oscillated politically between
selling and not selling the weapons systems to Iran and Pakistan; issues
over Chinese piracy of U.S. technology and allowing a huge black mar-
ket for such things as videos and music exacerbated pressure for larger
sanctions (Simons 1999, 160). By 1999, sanction pressure ebbed.
While other, tighter sanctions were threatened for alleged human
rights abuses, U.S. trade with China has increased in trend since 1994.

Colombia: 1996–98
Colombia has been associated with narcotics trade, specifically
cocaine, for decades. In the 1990s, Colombia’s president Sampler was
associated with drug traffickers and accused by the United States of
undermining drug enforcement progress (Dow Jones News Services
1996b). Early in 1996, Colombia’s refusal to sign a trade accord with
the EU was pressured by U.S. sanctions (Dow Jones News Services
1996a). U.S. aid came to Colombia in many forms in the early 1990s,
from financial to military, specifically to reduce the amount of drug
trade taking place in the United States. The sanctions were on imports
of Colombian coffee, flowers, among Colombia’s largest export sec-
tors, as well as aid. Sanctions ended when Andres Pastrana was elected
as Colombian president in 1998. The problems of political corruption
and drug trafficking continue today, even under Colombia’s free-trade
 BRIEF CASES HISTORIES 177

agreement with the United States. “The armed insurrections and

narcotics trafficking that accompany [the insurrections], are among
the most significant threats to the Colombian economy” (Grieco and
Schott 2006, 63).
The Colombian economy is a primary products economy. Coffee,
roses, and apparel are among their chief exports (CIA Factbook
2000). In 1996, Colombia imported 36.1 percent of its total from the
United States, selling America 40.2 percent of its exports. Colombia’s
economy grew during the Sampler regime at a quick pace. Real GDP
grew at an average of 4.28 percent per year from 1991 to 1995 (U.N.
Statistical Yearbook 1997). After that regime ended, Colombia fell
into a recession for the remainder of the 1990s (ibid.).

Cuba: 1960–Present
Cuban sanctions are among the most heavily studied and debated.
Fidel Castro led a communist revolt in Cuba that finally overtook the
democratic Cuban government in 1960. The Soviet Union’s involve-
ment in both funding and supporting Cuba was perceived as a threat
to the United States and the stability of other Latin American nations.
In 1962, Soviet missiles deployment to Cuba pushed the Cold War
superpowers to the brink of a nuclear exchange. The Soviets acqui-
esced to U.S. demands, the missiles were removed, and the threat
subsided. Beginning in 1960, the U.S reduced its trade with Cuba,
and by 1965 the sanctions were as complete as any American measures
to date. Throughout the 1970s, the U.S. intelligence network
attempted to undermine Castro.

The regime withstood a continuous campaign of CIA-orchestrated

terrorism, military invasion, industrial sabotage, agricultural arson,
assassination, threat of assassination, the manipulation of regional
bodies, the pressures of the “missile crisis,” intimidation of domestic
and foreign companies, threats to trading nations, and all the many
burdens of the broad economic war being waged by a superpower
against a relatively small Caribbean island. (Simons 1999, 127)

These coercion measures became the template for much of

American, anticommunist activity in Latin America throughout the
1970s and 1980s. Cuban troops were deployed to other communist
revolts in Africa, including Ethiopia and Angola, increasing pressure
to increase sanctions. In the late 1980s, pressure built on Cuba, as the
Soviet Union began to break down economically and politically.
178 ECONOMIC SANCTIONS

Through the 1990s, Cuba began to suffer as a result of losing its main
trading and financing partner, which increased the potency of U.S.
coercion. “U.S. sanctions against the Castro regime have complicated
the economic agenda sought by Havana, namely reliance on foreign
investment as a substitute for lost Soviet subsidies. U.S. sanctions have
cost the regime hard currency” (Fisk 2000, 82).
Cuban sanctions remain in effect based on an anti-Castro stance,
and are simply part of everyday events in Washington. The policy’s
simple purpose is to punish the Cuban government and people for
allowing Fidel to remain in power. The Cuban case is hotly debated
but lightly examined quantitatively due to a lack of reliable data.
These sanctions are unlikely to end before Castro’s death, and then
may depend on his successor.5 Cuban economic statistics are
extremely suspect, though exchange rate and trade data are used in
chapter six.

El Salvador: 1977–81; 1987–88; 1990–93

In 1977, the United States was supplying El Salvador with 90 percent
of its military aid. Political problems within El Salvador made America
leery of continuing a high level of financial support. El Salvador
withdrew new requests for military aid to avoid being denied the
funding (Russell 1984, 120). U.S. arms sales to El Salvador continued
after 1977, though at reduced levels, and there were few economic
problems for El Salvador from this first sanction. Coffee is the historic
leader in El Salvador’s export market. After 1978, as her real GDP
moved up rapidly, the El Salvador economy began to slump as coffee’s
value internationally decreased. El Salvador’s real GDP fell almost
33 percent from 1977 to 1983. Her dependence on America
increased from the first to the second episode; in 1977, America
constituted 33 percent of El Salvador’s exports and 30 percent of her
imports. By 1986, these numbers increased to 44 and 39 percent,
respectively. American aid to El Salvador never ceased completely in
either sanction; U.S. economic aid to El Salvador doubled between
1980 and 1983 (Russell 1984, 122).
In 1987, four U.S. Marines died in a street attack in El Salvador’s
capital, San Salvador, and amnesty was granted shortly thereafter to
the perpetrators by El Salvador’s highest court (HSE 1990, v. 2, 606).
Amnesty was granted because the killings were “purely political” in
their nature. The United States threatened to withhold 10 percent of
an aid package to be delivered in 1987; President Duarte overturned
the amnesty ruling by executive order and the aid arrived. Human
 BRIEF CASES HISTORIES 179

rights issues were front and center in the 1990 sanctions on

El Salvador. The killing of six Jesuit priests in late 1989 led to another
aid cut in 1990. These sanctions lasted until 1993 when human rights
were considered to be augmented by the U.S. Congress.
El Salvador experienced 3 percent growth per year of real GDP
from 1989 to 1992, while inflation averaged 16.8 percent over the
same time period. El Salvador’s substantial trade ties to America, its
previous experience with sanctions, and its growing external debt
stock made this an effective sanction. HSE (1990) claimed the first
sanction a failure, as the military government was neither threatened
from within nor strongly pressured by America and continued to rule.
The second sanction was claimed successful, as the amnesty decision
was overturned. In 1990, El Salvador purchased 37.9 percent of its
imports from the United States and sold America 34.1 percent of its
exports. The late 1980s were a recessionary period for El Salvador,
during which time there was much political upheaval. However, the
early 1990s were strong economic years for El Salvador, as real GDP
grew at an average rate of 6.2 percent per year from 1991 to 1995
(U.N. Statistical Yearbook 1997).

Ethiopia: 1976–92
In 1976, Haile Selassie abdicated his rule under military pressure, and
a military junta replaced him. President Mengistu Haile-Mariam’s
new government immediately nationalized all land and most indus-
trial properties, including U.S. firms and assets. By 1978, with the
threat of U.S. aid losses, Ethiopia signed an agreement with the Soviet
Union. U.S. military and humanitarian aid was curtailed. Loans were
defaulted and, under new legislation and the Hickenlooper
Amendment, further sanctions were imposed.
In 1984, bad harvests and a lack of timely aid damaged the
Ethiopian economy. The United States did help, but only in humani-
tarian efforts. By 1986, an agreement was reached for expropriated
asset compensation, but human rights abuses continued. Conflicts
with Somalia and continuous civil war drove Ethiopia into further
domestic problems. By 1990, rebels made large advances; by 1991,
Mengistu was overthrown and the U.S. sanctions ended in 1992 as
free elections took place in Ethiopia after sixteen years of civil conflict.
Ethiopia’s real GDP growth before sanctions in 1976 was slow.
Between 1973 and 1976, real GDP increased at 1.89 percent per year;
prices climbed 14.5 percent per year. Her government spending
deficit was 5 percent of GDP in 1976 and the current account was in
180 ECONOMIC SANCTIONS

surplus. External debt was 18 percent of GDP in 1976. During the

sanctions, the current account was in deficit throughout, as expected
with droughts and subsequent famine; prices doubled between 1977
and 1987, while real GDP never grew more than 5 percent annually
but twice in ten years between 1977 and 1987. In the late 1980s,
however, there was a recovery, followed by problems until 1992. Real
GDP grew an average 6 percent between 1986 and 1988, then fell
3.3 percent per year between 1990 and 1992.

Guatemala: 1977–86; 1990–93

Sanctions were imposed against Guatemala in 1977 for suspected
human rights abuses and alleged mass political killings by both leftists
in power and conservatives inside Guatemala attempting to gain
power. In 1977, U.S. military aid and loans were reduced; in 1981, the
U.S. severed diplomatic ties and canceled all military sales, aid, and
loans due to continued poor reports on the Guatemalan civil rights sit-
uation. After democratic elections in 1984, the Reagan administration
renewed aid and loans; in 1986, Guatemala received aid and military
equipment, albeit at levels below pre-1977 agreements.
America purchased 37 percent of Guatemala’s exports and provided
36 percent of her imports in 1976. Guatemala also found assistance
from the USSR. As its economic situation fluctuated heavily between
1976 and 1986, Guatemala’s real GDP increased at 12.2 percent per
year from 1976 to 1978, and then fell at an average of 0.63 percent per
year from 1979 to 1986. Prices increased 10.7 percent per year from
1976 to 1980; from 1981 until 1986, prices grew at 12.5. In the
period between 1982 and 1984, there was 1.5 percent average infla-
tion. Guatemala’s debt to GDP ratio doubled from 8.8 percent in
1976 to 17.6 percent in 1986.
In March 1990, the New York Times reported that Thomas Strock,
the American ambassador to Guatemala, was recalled to protest polit-
ically motivated killings of U.S. citizens; the human rights situation in
Guatemala was reported as “deplorable” (New York Times 1990). In
response to political pressure concerning drug trafficking and the
aforementioned murders, President Bush threatened to again curtail
economic aid unless Guatemala’s government resolved these issues
overtly. After more killings in the fall, the American Congress sus-
pended $2.8 million in aid. In 1993, aid was resumed after Guatemala
had sufficiently improved the human rights situation.6
In 1990, the United States accounted for 25 percent of
Guatemalan exports and 20 percent of imports, down from the first
 BRIEF CASES HISTORIES 181

sanction episode where the United States made up 36 percent of

Guatemala’s total international trade. Guatemala’s inflation rate for
1985 through 1990 averaged 21.9 percent; however, real GDP grew
over the same period at a modest 2.33 percent average. Guatemala’s
external debt was stable over this period, but relatively large; at the
end of 1989, their debt was 34 percent of GDP, down from the pre-
vious two years. Guatemala’s government spending deficit grew
between 1985 and 1990, as did their current account deficit, in real
terms. Guatemala’s chief export was coffee, accounting for one-third
of its exports in 1990.
Guatemala was a Latin American success story throughout the 1990s.
From 1990 to 1999, real GDP grew at an annual average of 4 percent
(U.N. Statistical Yearbook 2001). This was during sanctions, which
shows the sanctions were low in their economic effectiveness.

Haiti: 1987–90; 1991–94

The United States suspended aid to Haiti after blatant human rights
abuses in November 1987; thirty people were killed attempting to
vote (HSE 1990, v. 2, 598). One month later, President Duvalier,
blamed for ordering these political murders, went into exile. A mili-
tary government stepped in and ironically the violations subsided.
In 1990, new elections were held, and aid restored somewhat by
the countries participating in the sanction.7 A priest, Jean-Bertrand
Aristide, was sworn in as president of Haiti on February 8, 1991.
The external debt of Haiti tripled between 1979 and 1987. Prices
in Haiti deflated by 11.4 percent in 1987, and then grew slowly until
1990, at an average of 5.5 percent per year. In 1990, Haitian prices
began to soar, and continue to increase throughout the decade. From
1987 to 1990, Haiti’s real GDP fell by 1.35 percent per year. This first
sanction embargoed loans and aid only, though America claimed
84 percent of Haitian exports and supplied 64 percent of her imports
in 1986.
In September 1991, violence erupted in Port-au-Prince with a
coup attempt, and President Aristide lost power. America threatened
renewed sanctions if the coup leaders did not step down and allow
Aristide back into the presidency. The coup leaders did not and Haiti
faced both trade and credit sanctions. U.S. oil companies were forbid-
den to deal with Haiti; Haitian assets and loans were also frozen. The
sanctions became an embargo from the United Nations with Security
Council Resolution 875; this increased the pressure on the military
regime and had an immediate impact (Cortright and Lopez 2000,
182 ECONOMIC SANCTIONS

90–92). Between 1991 and 1993, sanctions stiffened and eliminated

all aid, as the talks between coup leaders and the deposed Aristide
stumbled. The United States accounted for over 80 percent of Haitian
exports and 65 percent of imports by 1991. On October 17, 1994,
Aristide was back in power, and sanctions were lifted soon thereafter.
The 1991 policies against Haiti were punitive. Using both trade
and credit sanctions, the United States forced Haiti’s small, depend-
ent economy into economic ruin. The Haitian economy of the late
1980s mocked other military economies in the Caribbean and Latin
America. From 1991 to 1995, Haiti’s real GDP fell 4.1 percent per
year. This incredible slowdown was accompanied by 25.1 percent
inflation.

India: 1965–67; 1978–82; 1998–99

The Indian sanction episodes are spread over time, imposed for many
reasons. Balasubramanyam (1984) suggested sanctions imposed on
India in 1965 were to combat agricultural reforms considered prob-
lematic by the United States. U.S. food packages, initially meant to
help the poor, were perceived to be detrimental to agricultural
production in India by 1965; President Johnson food aid packages
should stop and force India to prioritize its agricultural planning bet-
ter (Balasubramanyam 1984). Pakistan then invaded Kashmir, and the
U.S. military aid to India was cut. With food and military supplies cut
off, the threat of widespread famine increased with a bad harvest. By
1967, India has its food and economic relief packages restored under
a new agreement that demanded India’s reformation of an agricul-
tural, five-year plan; food supplies from America came in smaller pack-
ages over time. America supplied India with 30 percent of its imports,
while buying 18 percent of India’s exports in 1964.
India’s economy in 1965 was neither large nor industrial. Real
output per capita grew over the 1960s, except for 1965 and 1966.
The aid that flowed into the Indian economy as foodstuffs was
30 percent of the total value of Indian aid by 1967 (179). In reality,
much of the U.S. aid came to promote Indian industry. Consumer
prices in India increased 11.4 percent per annum from 1965 to 1967,
while real GDP fell in 1965 by 6 percent, growing in 1966 by less
than 1 percent. Government expenditure, as a percentage of GDP,
rose from 1967 to 1977 steadily. The agricultural sector’s dominance
in the Indian economy fell over this time, and the manufacturing
sector began to grow. Real GDP grew at 4.6 percent per year in those
eleven years.
 BRIEF CASES HISTORIES 183

In 1978, the United States passed a bill restricting the flow of

high-grade uranium exports to countries not willing to meet new
standards and requirements for nuclear testing and use. India’s
aboveground, “peaceful” nuclear test in 1974 worried many nations
about Indian nuclear program safety. India had become much less
dependent on America by this time; only 12 percent of imports were
American, while India sold 13 percent of its exports to the United
States in 1978. In 1980, India refused to meet new UN safeguards
and standards, though uranium shipments never completely stopped.
When India requested new shipments in 1981, no action was taken
on the request. By 1982, however, India was receiving low-grade
uranium shipments from France. The Indian economy in 1978 was
growing; from 1975 to 1978, India experienced 6.5 percent average
growth of real GDP, with 2.2 percent average inflation.8 India’s
economy continued to grow, after a downturn in 1979 attributed in
HSE to energy prices, and was little affected by sanctions. The key
sanctioned good, uranium, was a minuscule part of the Indian econ-
omy, with an estimated value in trade of one ten-thousandth of
Indian GDP.
This issue continues to be of interest to the United States,
especially as border skirmishes are standard practice between Pakistan
and India over Kashmir and other issues. India’s economy, however,
has strengthened dramatically in the last thirty years, especially in the
last five. In May 1998, India conducted five, aboveground nuclear
tests in stark violation of the 1995 version of the UN Non-Proliferation
Treaty. Pakistan began to worry and an arms race ensued. The United
States acted immediately to sanction development aid. India felt
the wrath of these sanctions quickly. “Faced with slowing foreign
investment and widening trade and budget deficits, Indian Finance
Ministry officials are eager to see foreign development aid return”
(Karp 1999).
India’s economy continued its stagnant ways until the end of the
1990s. Through the 1990s, the Indian economy made a decisive turn
toward a better export mix of manufacturing and agriculture. In
1993, agriculture was 64.9 percent of exports and manufacturing
34.5 percent. In 1998, those proportions had become 53.8 percent
and 45.6 percent, respectively (UN International Trade Yearbook
2000). Real GDP growth in India increased over the 1990s at a
historic pace, an annual average of 6.275 percent (UN Statistical
Yearbook 2001). Since sanctions were basically focused on
weapons technology, they did little to change the Indian economy’s
direction.
184 ECONOMIC SANCTIONS

Indonesia: 1963–66; 1993–95

Indonesian rebels pushed for Malaysia’s formation in the early 1960s,
seeking asylum in this newly formed country. The United States
immediately came to the aid of Malaysia, a new state created in a dem-
ocratic image. Indonesia’s refusal to support and recognize Malaysia
led to U.S. sanction threats. Malaysia was formed and immediately
severed diplomatic and economic ties with Indonesia. Indonesia
ceased all trade and openly supported various pro-Indonesia forces in
Malaysia. The United States asked and received support for sanctions
from the United Nations. Many UN members began to pull invest-
ments out of Indonesia in response to Indonesia’s “Crush Malaysia”
policies; in reaction to U.S. sanctions and capital flight, Indonesia
nationalized U.S. oil interests in 1965, bringing more pressure for
sanctions (HSE 1990, v. 2, 254–55). Indonesia’s consumer prices
doubled in 1963, doubled again in 1964, and quadrupled in 1965. In
1966, Indonesia’s CPI increased tenfold. Skyrocketing inflation was
accompanied by stagnant real GDP growth, averaging 1.7 percent
between 1963 and 1966. The United States sold Indonesia 35 percent
of its imports.
The 1990s were great times for the Indonesian economy through
the Southeast Asian currency crises of 1997. Though the average
inflation rate has been 8.5 percent, real GDP growth was 8 percent
annually from 1990 to 1997. Government spending deficits were in
surplus during the early 1990s. The current account balance was in
deficit throughout the 1990s, a by-product of growth, macroeconomic
policy, and capital inflows; national debt stock equaled 99 percent of
GDP at the end of 1992.
In September 1993, alleged human rights abuses by the
Indonesian military caused sanction threats from the United States.
Worker rights in East Timor were eroding, and the Clinton adminis-
tration stopped the sale of Jordanian fighters to Indonesia (Cronin
1994). This news came on the heels of Indonesia’s unprecedented
economic growth. The United States nominally sanctioned Indonesia
over these abuses, and by 1995 the sanctions were lifted. In both cases
here, the United States sanctioned credit only, though the United
States accounted for approximately 20 percent of both Indonesian
exports and imports in 1993.

Iran: 1979–81; 1984–Present

On January 16, 1979, the Shah Pahlavi of Iran abdicated, seeking
asylum in the United States. The Ayatollah Khomeini announced his
 BRIEF CASES HISTORIES 185

seizing of power and appointed a provisional government. In

November 1979, the U.S. embassy was overrun; sixty Americans were
among one hundred persons held hostage in protest of the United
States granting the Shah asylum. The United States sent in a negotia-
tion team, which was refused. The next day, President Carter ordered
the suspension of arms sales and oil purchases with Iran. Iranian assets
in the United States were also frozen. Iran nominally announced a
suspension of oil shipments to the United States. In January 1981, the
hostages were released, but sanctions went away very slowly.
The United States played a modest role in Iran’s economy in
1979, supplying Iran with various goods, including rice, wheat,
manufactured goods, and military hardware in return for oil. In
1977, 12.8 percent of Iran’s exports were sold to the United States
(UN Statistical Yearbook 1980). Iran’s government spending was in
heavy deficit territory; by 1978, Iran’s external debt balance was
more than 100 percent of GDP. After 1977, real GDP fell 15 percent
per year over two years, and prices began to steadily climb. This sanc-
tion’s effectiveness was a function of the United States closing all its
markets to Iran officially; the World Bank also applied pressure for
countries to not readily give credit to Iran, which also helped the
American cause.
Alleged Iranian involvement in the Beirut car bomb death of two
hundred Marines in 1983 led to export sanctions. When it became
apparent that the Iran–Iraq conflict of the mid-1980s also involved
chemical weapons use, the United States immediately banned all
chemical exports to both countries. As Iran became more belligerent,
so did the sanctions. There were trade and financial sanctions levied
against Iran, attempting to punish their economy for supporting
terrorism and war with Iraq. Iran’s economy suffered from U.S. pres-
sure on international agencies to cut loans and other UN countries’
trade involving Iran. In 1987, the American import embargo was
augmented, but the effects were short-lived because of a trade diver-
sification strategy employed by Iran (O’Sullivan 2003, 65).9 In 1989,
Iran and the United States negotiated for the release of over five hun-
dred million dollars in frozen Iranian assets in America; almost one
billion dollars worth of assets remained frozen, however. Iran’s sup-
port of terrorist organizations led to additional sanctions in the mid-
1990s, though not as complete as previous sanctions. In 1996, the
United States passed the Iran–Libya Sanctions Act (ILSA). This act
was meant to not only provide breadth to direct sanctions against Iran
but to expand economic and legal punishment to domestic companies
and foreign entities that acted as sanction busters.
186 ECONOMIC SANCTIONS

In 1978, the United States accounted for 28 percent of Iranian

imports, and 12 percent of exports; the figures fell to 1 percent and
5 percent, respectively, by 1984. Oil cartel riches eroded as the petro-
leum price plummeted in the 1980s. The average growth rate of
Iranian real GDP between 1983 and 1988 was 0.7 percent. Because
of constant tension between Iran and Iraq, government spending was
in a deficit from 1976 on, while the current account fluctuated from
deficit to surplus as the oil price fell then rose, respectively. One sta-
tistic of note: the external debt of Iran fell over the second sanction
period, until 1990, when it doubled. Prices in Iran quadrupled
between 1983 and 1990. The hostage case can be viewed as a success,
as the hostages were released. The 1984 sanctions, as they did not rid
the world of Iran’s dictatorship or Iranian-funded terrorist activities,
are a political failure as Iran has continued to be an open supporter of
terrorism and anti-American activity.
Through the 1990s and into this decade, Iran has drawn broader
UN attention. With continued terrorist ties and a recent expansion of
its nuclear program, Iran has become a rogue nation. With a new
authoritarian government in power, it is likely that the current
sanctions will remain for some time and possibly expand among UN
members. Iran’s economy continued to grow through the 1990s and
into this decade. On average, real GDP grew in Iran at a 5.8 percent
annual average from 1990 to 2001, which is similar to the growth rate
of the U.S. economy over the same time period (UN Statistical
Yearbook 2003). This is where we need to be careful in generalizing
that sanctions have not worked, as the macroeconomic statistics do
not necessarily reflect the plight of the common person in Iran
(Amuzegar 1997).

Iraq: 1980–87; 1990–2003

Sanctions began in 1980, based on increased terrorism believed to
be originating from within Iraq. Also, Iraq expanded its weapons
technology, using chemical weapons on Kurdish rebels and Iranian
soldiers, a problem that continued through 2003. American oppo-
sition to Iran and the continuing conflict between Iraq and Iran
helped Iraq’s position with the United States during the 1980s.
The United States ceased exports of jet engines and nuclear fuel to
Iraq from 1980 to 1983. Export sanctions were the only sanctions
noted by HSE (1990). In 1983, Iraq was removed from the list of
terrorist nations, only to return in 1989. HSE (1990) see this first
case as a moderate failure, as terrorist activity originating in Iraq did
 BRIEF CASES HISTORIES 187

not cease. In 1990, Iraq began an explicit military campaign in

Kuwait, possibly indicating how little Saddam Hussein feared
sanctions.
Economic statistics for Iraq are extremely hard to both gather
and believe. Some possible sources include the IMF, World Bank,
UN, and CIA documents, and some splicing of data series for Iraq
through 1991. Between 1981 and 1989, the average real GDP
growth in Iraq fell by 2.9 percent; the average inflation rate over the
same time was 11.2 percent. Iraq was not dependent on American
goods, with only 7 percent of imports coming from the United States,
which purchased 3 percent of Iraqi exports in 1980. War debts, from
the war with Iran, were large and culminated with the Gulf War
buildup in 1990.
The second set of sanctions against Iraq was much different from
the first. Iraq invaded Kuwait on August 2, 1990. A long queue of
countries formed against Iraq, which made finding substitute markets
for exporting oil and importing food difficult. Whereas the 1980
sanctions were export only, the 1990 sanctions were comprehensive,
originating with the United Nations. During a time of economic
growth in Iraq, trade was closed from the outside and international
credit availability was virtually eliminated. During the 1990s, the Iraqi
economy was subject to weapons restrictions, surprise inspections,
and careful scrutiny by the international community. Iraq’s economy
became isolated from the major world powers. “In this endeavor, the
universal, multilateral structure of sanctions also was important. Had
a regional body, rather than the United Nations, mandated multilat-
eral sanctions on Iraq, its oil wealth would have enticed other
countries not bound by the regional sanctions to trade with Iraq”
(O’Sullivan 2003, 155).
Following the terrorist attacks of September 11, 2001, the Bush
administration looked at Iraq as a country who funded and harbored
the organizations responsible. Sanctions were then remade in complete,
originating this time more directly from the United States. Threats of
military action began in earnest, made credible by the invasion and
takeover of Afghanistan in 2002. The U.S. invasion of Iraq in March
2003 effectively ended economic sanctions. Before the 2003 invasion,
according to the United Nations, Iraq’s economy was volatile. Real
GDP grew at an annual average of 14.8 percent from 1997 to 2001
after falling at an annual average of 13.7 percent from 1991 to 1996
(UN Statistical Yearbook 1997 and 2003). Sanctions were a political
failure, but Iraqi economic welfare certainly plummeted as a result of
sanctions (Yousef 2004).
188 ECONOMIC SANCTIONS

Libya: 1978–Present
Moammar Gadhafi, as the leader of Libya, isolated Libya from much
of the world. Libya’s support of terrorist groups forced the United
States to sanction arms sales in 1978. From 1978 to 2004, Libyan
belligerency was a constant factor in American–Libyan relations. In
1982, the United States embargoed oil imports from Libya in an
attempt to curtail oil profits from becoming terrorist funding. In
January 1986, as a reaction to terrorist attacks in Rome and Vienna
believed to be funded by Gadhafi, comprehensive U.S. sanctions were
imposed. In March 1986, the U.S. military had two events take place
in the Mediterranean Sea involving Libya. First, U.S. warships sta-
tioned themselves in waters Gadhafi warned would provoke action.
This tension led to two Libyan gunboat sinkings and a missile site
being destroyed. International reaction to this U.S. retaliatory move,
Operation El Dorado Canyon, was poor. “Probably not in twenty
years had there been a period of such intense international criticism of
the United States as the last two weeks of April 1986” (Davis 1990,
145). As a result, few countries joined in on the U.S. sanctions. By the
end of 1986, the only other country involved was France.
However, a Libyan link to the bombing of Pan Am Flight 103 in
1988 strengthened world resolve against Libya, but the United States
continued to be the key sender nation. Libya provided asylum for two
individuals accused of the bombing, not acquiescing to international
appeal for extradition. As a result of that event 270 people died. In
1992, the United Nations voted to uniformly adopt full sanctions on
air travel and arms sales to Libya, making the existing sanctions that
much stronger (Lewis 1992). These were UN Security Council
Resolutions 748 and 883.
In 1996, Libya was named along with Iran in a new U.S. law to
penalize both domestic and foreign companies for dealings with these
countries (see the section on Iran earlier). The ILSA further isolated
Libya’s economy. The United States has recently dropped most of its
sanctions against Libya, and trade has begun again. The economic dam-
age of these sanctions was significant, especially on inflation, but human-
itarian conditions were supported by the regime. “Sanctions on Libya
did not precipitate a humanitarian crisis by any standard, although they
did significantly diminish the standard of living of most Libyans . . . The
main influence of U.N. sanctions on the humanitarian situation in Libya
was through the inflation that sanctions encouraged” (O’Sullivan 2003,
211–12). In 1999, Libya finally handed over the terrorists, and there was
momentum toward lifting the long-standing sanctions.
 BRIEF CASES HISTORIES 189

In 2004, UN sanctions were lifted as Libya came out and spoke

against international terrorism.10 Whether Libya continues its stance
against international terrorism remains to be seen. The overall effec-
tiveness of over twenty-five years of sanctions is still in question. First,
Gadhafi is still in power. Second, terrorism originating out of Libya
or funded by Gadhafi continued for decades, and may still exist.
However, Libya economically limped through the 1990s, a time
where worldwide economic growth should have naturally affected
Libya’s economy positively. Real GDP fell at an annual average of
0.74 percent from 1991 to 1999 (UN Statistical Yearbook 2001).

Malawi: 1992–94
Military assistance was cut by the United States to Malawi in 1992.
This resulted from human rights abuses on the part of the govern-
ment. Malawi’s self-proclaimed, life-president Kamuzu Banda began
to arrest opposition leaders and a demonstration over wages turned
violent, where twenty-two people were killed by pro-Banda forces.
Western Aid Donors had imposed economic aid sanctions on the
Malawi government of Dr. Banda beginning in 1991 (Kachala 2002).
Reports of prisoner mistreatment, especially those in political opposi-
tion to Banda, further turned friendly Western nations and the
World Bank against the regime (Toronto Star, May 13, 1992). Aid was
summarily cut until human rights were augmented. Aid was reinstated
in 1994 after elections voted against Banda’s regime. Human rights
are still in question for Malawi.
In 1991, Malawi exported 15 percent of its foreign sales to the
United States, while buying only 3.3 percent of its imports from
America. Malawi’s economy during the 1990s became one of intense
inflation. By 1999, food prices increased thirteen times from their
1990 levels; real GDP grew at a good pace of 4.45 percent, however
(UN Statistical Yearbook 2001). Malawi received over sixty-five million
dollars in development aid through UN channels in 1994, down to
forty-eight million in 1995; industrial production grew from 1993 to
2000 at an average of 4.4 percent per year (ibid.).

Myanmar (Burma): 1988–90

A military coup overthrew the government in 1988, and worries
increased that the new government would become a Chinese ally,
begin to violate human rights through martial law and suppression
of opposition and the press, and be nondemocratic by definition.
190 ECONOMIC SANCTIONS

These worries ceased in 1990 when elections were held and the
financial sanctions imposed were lifted. Japan was likely the most
threatened of the three nations involved; sanctions continue from
other nations, but the United States and United Kingdom have gen-
erally been opposed to major sanctions against Myanmar since this
episode (Simons 1999, 3). President Clinton, under the Burma
Freedom and Democracy Act, barred new investment originating
from the United States in 1997 (Hufbauer and Oegg 2003, 127).
However, in 2003, a motorcade ambush forced a switch in the U.S.
stance, leading to import and financial sanctions imposition on
Myanmar. This latter case is not empirically discussed in chapter six.
Myanmar’s economy is mainly agricultural. In 1987, Burma sold
8 percent of its exports to Japan, West Germany, and the United
States and purchased 30 percent of its imports from the same group
(Drury 1998).

Nicaragua: 1977–79; 1981–88; 1992–95

Civil war ravaged Nicaragua in the early 1970s as rebels took aim
at President Anastasio Somoza Debayle, who in December 1974
imposed a state of martial law to control the spread of Sandinista
rebels. U.S. sales of police equipment and economic aid were sus-
pended to Nicaragua.11 Pushing to end martial law and nationwide
labor stoppages, the United States persuaded the IMF to hold back a
debt service loan in 1978 (HSE 1990, v. 2, 452). In July 1979, the
Sandinistas overthrew Somoza, and full economic aid was restored.
Nicaragua engaged in relatively large government spending in the
1970s to fund its internal conflicts continuing into the 1980s. In
1986, 55 percent of the Nicaraguan government spending was mili-
tary. Nicaraguan trade dependence on the United States dictated this
sanction’s focus. In 1977, the United States accounted for 26 percent
of Nicaraguan trade; by 1981, it accounted for 35 percent.
The Sandinistas then began to fund leftist groups in El Salvador. In
1981, President Reagan suspended all aid to Nicaragua to combat the
Sandinista movement. The problem was exacerbated by the Soviet
Union’s willingness to make up Nicaraguan losses from U.S. sanc-
tions. Reagan pushed to fund the growing anti-Sandinista (contra)
groups in Nicaragua, but found stiff opposition in the U.S. Congress.
In 1984, general elections were held in Nicaragua, and a military
leader named Daniel Ortega came to power. The United States saw
these elections as bunk and completely controlled by the military;
Reagan opted to fund a CIA attempt to stabilize the Nicaraguan
 BRIEF CASES HISTORIES 191

situation, and the election process in particular (HSE 1990, v. 1,

179). Reagan, in 1985, cut all aid and trade with Nicaragua, seized
assets, and deported ambassadors. In 1990, elections were held again
and Ortega was voted out of office. The fighting stopped and trade
began again, including economic aid packages.
Close (1988) stated that Nicaragua is one of the richest resource
countries in Central America, but ranked among the poorest in terms
of GDP per capita by 1986. Nicaragua, due to U.S. sanctions, lost
approximately $341.8 million dollars in loans from 1981 to 1986
(103). Throughout the embargo period, Nicaragua attempted land-
use reforms, completely turning their agricultural policies upside
down. The United States, one of the major importers of Nicaraguan
goods historically, likely hurt those reforms through sanctions.
By the mid-1980s, the Nicaraguan economy faced stagnation.
Inflation in 1985 was 219 percent, followed by 681 percent and
911 percent the next two years. From 1984 to 1990, Nicaraguan real
GDP fell by an average 2.8 percent per year. The inability to gain
access to cheap development loans, via the IMF or the United States,
led to problems. Smith (1992) posited these credit problems led to
the end of the Sandinista regime. When Nicaragua was released from
sanctions in 1990, the new government attempted to use the coun-
try’s resources to reform the economy, rather than relying on U.S.
technology or foreign loans.
HSE (1990) saw the 1977 sanctions as a success with the end of
the Somoza regime. The 1981 sanction is seen by HSE as a failure,
with political success but little economic contribution. The lack of
assistance Nicaragua found internationally put pressure on the Somaza
government to fund its military internally. “[President Ronald]
Reagan froze Food for Peace aid to Nicaragua in 1981 and blocked a
$9.6 million wheat sale to that country” (Simons 1999, 122).
Nicaragua struggled at the macroeconomic level during this sanction,
with real GDP falling by an average of 1.5 percent per year during the
seven sanction years.
Sanctions in 1992 were again imposed on Nicaragua for human
rights violations. After over fifty killings by police officers in
Nicaragua, the State Department recommended that the president
impose sanctions on specific interest groups to stem the violence
(U.S. Dept. of State, 1994b). The Sandinista Popular Army was
singled out specifically as one of these groups. In 1995, the Sandinista
government was replaced, and sanctions went away; aid still remained
far below pre-1990 levels (Haugaard 1997). Nicaragua’s economy in
1992 was purchasing 25 percent of its imports from America and
192 ECONOMIC SANCTIONS

selling 26 percent of its exports to American markets (UN Statistical

Yearbook 1996). Real GDP growth was negative from 1991 to 1993,
but rose to 3.2 percent growth in 1994 (Haugaard 1997).

Nigeria: 1993–98
Nigeria’s economy was sanctioned by the United States in 1993, as
were other African nations, for human rights violations. A coup also
stopped Nigeria’s path toward democracy, and imposed a military
regime. The coup came after democratic elections, which were
Nigeria’s first in over a decade. “By the time the current ruler, General
Sani Abacha, seized power in November 1993, Washington had can-
celed the visas of important military personnel, restricted arms sales,
halted all U.S. economic and military aid, and cut off Nigeria’s access
to trade credits and guarantees” (Fadopé 1997). Most international
observers felt the best measure against the new regime was to focus on
reducing oil importation, as that was said to be the source of funding
for suppression.
Human rights activists were also murdered, which led to military
aid being cut and an asset freeze by the United States. The sanctions
were not universal, which likely hurt their potency. “Many human
rights groups and opposition organizations within Nigeria urged an
oil embargo and financial sanctions following the cancellation of elec-
tions in 1993 and the execution of nine Ogoni activists in 1995, but
the United Kingdom and other major powers opposed such action”
(Cortright and Lopez 2002, 30). Sanctions eased after elections were
held in 1998.
Nigeria’s economy is based on oil and mining, as over 95 percent
of her exports are in extractive industries (UN International Trade
Yearbook 2001). Nigeria experienced strong growth during the
1990s, amidst political strife and being a nation on the brink of
civil war. Real GDP grew at an annual average of 6.03 percent from
1991 to 1999; however, food prices increased nine times over the
same period (UN Statistical Yearbook 2001). Developmental aid to
Nigeria fell precipitously in the 1990s, partially due to sanctions. In
1993, Nigeria received $478 million from international sources, and
by 1995 that number was down to $40 million; aid picked back up in
1997 to $200 million and was back down to $149 million by 1999
(ibid.). In 1991, Nigeria purchased 7.9 percent of imports from the
United States, and sold America 46.9 percent of its exports,
mainly oil.
 BRIEF CASES HISTORIES 193

Pakistan: 1971–72; 1979–90

In 1971, Pakistan was sanctioned for pursuing nuclear arms sales as
well (see the section on India earlier). Pakistan, in 1976, agreed to
buy reprocessed nuclear fuel from France to begin producing atomic
power and nuclear armaments after the 1975 Indian invasion of
Kashmir.12 In 1977, the military overthrow of Premier Bhutto led to
more international worries, as the new government under General
Mohammed Zia ul-Haq pushed for nuclear energy and weaponry
(HSE 1990, v. 2, 505). In 1979, after much delay and worry over the
French deal, Pakistan constructed a nuclear power plant and began
operations. They refused to follow U.S. safeguards. The United States
promptly cut military and economic aid. In one year, Pakistan eased
its stand, and the United States became more congenial in terms of
financial sanctions; aid was sent again and divided evenly between the
economy and military. The obvious fear was a possible arms race
between India and Pakistan, which did take place in limited form.
Sanctions on nuclear technology and fuel continued throughout
the 1980s.
The 1977 military takeover by General Zia began a new phase in
the economic history of Pakistan. GDP growth was significant over
the sanction period. On average, GDP grew 6.1 percent from 1978 to
1983. In 1984, military expenditure was 28 percent of overall gov-
ernment spending. Between 1978 and 1983, all sectors of the econ-
omy, except services and construction, grew above the GDP growth
rate (Noman 1990, 181). Through the rest of the 1980s, Pakistan’s
economy grew at an annual average of 6.03 percent.

Panama: 1987–90
In 1987, Manuel Noriega headed the Panamanian Defense Forces.
Acting as dictator, Noriega suspended much of his people’s demo-
cratic rights. The United States vowed to restore these rights, by force
if necessary. Sanction threats led to violent demonstrations against
America, causing damage to the U.S. embassy. As an attempt to ruin
domestic sentiment toward Noriega, the United States suspended
trade preferences, held up international bank transfers, restricted U.S.
companies in Panama from paying local taxes, and stopped shipments
in Panamanian dollars (Conniff 1992, 159).
The United States accounted for two-thirds of Panamanian exports
and 35 percent of imports in 1986. The trade and credit sanctions
were swift and comprehensive. In December 1989, Noriega and other
194 ECONOMIC SANCTIONS

major military officers stepped down, giving in to the American

military presence in Panama and grassroots dissent toward Noriega.
The Noriega-governed economy of Panama was one of increasing
debt and military expenditure. Between 1987 and 1990, real GDP fell
by 0.65 percent per year while inflation moved at a 0.5 percent aver-
age. HSE (1990) cited the economic measures’ inability to force a
change, supporting evidence that this case was a marginal failure.13
Panama was somewhat dependent on the United States for its inter-
national trade, selling 66 percent of its exports to the United States,
and buying 34 percent of its imports from American firms. Panama’s
external debt before sanctions was over 100 percent of GDP in 1987.
The sanctions did not cause new problems, they just exacerbated
existing crises.

Paraguay: 1977–81; 1996–98

Human rights abuses in Paraguay stimulated the United States to
suspended aid in 1977. Paraguay’s problems centered on a tribe of
indigenous peoples know as the Ache Indians. Alfredo Stroessner’s
regime, in power since a 1954 coup d’etat, was believed to be killing
Ache by the dozen and taking political prisoners, trying to eradicate
the Ache from Paraguay. After conflicting evidence on government
involvement in these crimes, the United States suspended military aid
(HSE 1990, v. 2, 434). The United States also attempted to influence
credit flows with limited success. In 1981, the United States reversed
its policy after evidence of human rights improvements, and aid was
restored (435). Paraguay relied on the United States for about 15 per-
cent of both exports and imports in 1977; sanctions, however, were in
financial form only. HSE (1990) felt this case was not a success, as
Ache Indians continued to be incarcerated and Stroessner remained in
power until 1989. From 1981 to 1990, the Paraguay economy expe-
rienced an average of 22 percent inflation per year, and 3.1 percent
average growth of real GDP. The United States was able to pressure
for curtailed international loans flow and stopped a development loan
of $141 million from coming into Paraguay (HSE 1990, v. 2, 435).
The 1996 episode was to prevent a coup attempt by Lino Oviedo,
which came on the eve of new elections. The governing body, which
had been in place since the time of Stroessner’s departure, was led by
President Juan Carlos Wamosy. The struggle for power between
Oviedo and Wamosy became a failed coup attempt, and Oviedo
became defense minister as a compromise (Sonntag 2001, 136). The
United States embargoed military and financial aid in an attempt to
 BRIEF CASES HISTORIES 195

stop the struggle. Once elections were held in 1998, the sanctions
were eased. The Paraguayan economy was stagnant in the 1990s, not
taking advantage of global growth. Real GDP grew at an annual rate
of 2.24 percent from 1991 to 1999; consumer prices tripled over the
same time period (UN Statistical Yearbook 2001). In 1995, develop-
mental assistance to Paraguay was $180.4 million, which fell to
$102 million in 1997, and then again to $77.6 million in 1999 (ibid.).

Peru: 1968–74; 1995–98

Military buildup and U.S. asset expropriation led to restrictions of eco-
nomic and military aid to Peru in 1968. Peru bought French Mirage
fighters in 1967, indicating to U.S. president Johnson’s administration
that Peru was beginning a military buildup. Simultaneously, a Standard
Oil subsidiary’s assets became disputed ground in Peru. President
Fernando Belaunde Terry had negotiated with the United States over
the ownership of Peru’s northern oil resources in 1963. The subsidiary,
International Petroleum Company (IPC), was accused of tax evasion
and political meddling (HSE 1990, v. 2, 309–310). In late 1968, an
agreement was reached between IPC and Peru; shortly thereafter,
Belaunde was overthrown by the military. This military government
immediately took control of the oil resources by force. The United
States, in response to the expropriation, suspended some economic and
military aid. “Predictably enough, given the hostility of the World Bank
and the United States government to the brand of radical nationalism
displayed by Peru since 1968, there was a price to be paid for ready
access to the international capital market” (Thorp 1991, 76).
Financial sanctions continued until 1974, as the United States and
Peru agreed on compensation payments for expropriated property,
and aid was restored. The United States also pressured the diversion
of international loans. The United States sold Peru 8 percent of
exports and purchased 11 percent of its imports. Peru experienced
moderate growth from 1966 through 1974, averaging 4.5 percent
per year; inflation was also moderate at 8.8 percent per year over the
nine years.
Peru and Ecuador engaged in a brief border skirmish and political
tension thereafter from 1995 to 1998. The bombardment of a
Peruvian military outpost, on the heels of gun battles, put the
two countries on the brink of war. To curb hostilities, the United
States cut military aid to both countries. The Peruvian economy was
little hurt by this outside of higher arms prices. Peru’s economy was
volatile from the late 1980s through the 1990s. Real GDP fell
196 ECONOMIC SANCTIONS

between 1989 and 1993 by 3.52 percent per annum, growing at

7.7 percent from 1994 to 1997, to then grow from 1998 to 2001 at
a paltry average of 0.9 percent per year (UN Statistical Yearbook
2003). Prices in Peru were historically calm during the 1990s.

South Africa: 1975–82; 1985–92

In 1960, South African police massacred sixty-four civilians. Many
African nations lobbied for the United Nations to use sanctions in
response. Curiously, the United States lobbied against specific sanc-
tions initially, and was not originally a participant in initial UN embar-
goes against South Africa. In 1964, however, the United States
nominally began to sanction South Africa, by voting to restrict EX-IM
Bank loans availability. Military hardware sales were also restricted.
These sanctions continued until 1975, and then expanded.
In the early 1960s, South Africa lost foreign investment and import
markets due to sanctions. As South African domestic policies became
more restrictive, so did international economic sanctions. “Inflation
was stable, however, up until the late 1970s; South Africa grew at a
modest rate, fueled by countries willing to break the sanction lines
and a monopoly position in diamonds and gold mines. This market
power also gave the United States and United Nations easy-to-detect
goods in policing sanction breakers” (Moorsom 1986, 31). By the
1970s, sanction potency was fading, but South Africa was hurt by
increasing oil prices.
South African real GDP was growing at 2.9 percent per year
between 1958 and 1961; inflation grew at 2.1 percent per annum
over the same time. From 1964 to 1976, the South African economy
became more dependent on U.S. goods and markets. By 1976, South
Africa was importing 21.6 percent of its goods from America, and
selling the United States 10.2 percent of its goods. In 1964, the per-
centages were 18.9 percent of imports and 8.6 percent of exports.
South African real GDP growth was an annual average of 3.68 percent
from 1971 to 1975 (UN Statistical Yearbook 1978).
As a reaction to India’s 1974 nuclear test, the United States set
major restrictions on countries looking to buy raw uranium from
American sources. South Africa did not meet these new requirements
and uranium shipments were curtailed. The nuclear possibility, cou-
pled with South African sanction-busting for Rhodesia, expanded
sanctions already against apartheid. In 1986, sanctions were further
strengthened. An antiapartheid movement grew within South Africa,
as a reaction to a 1984 law stating “mixed” race and indigenous South
 BRIEF CASES HISTORIES 197

Africans would receive parliamentary representation, but the majority,

black African population would not be represented (Lipton 1988, 5).
The international protests, led by such activists as Desmond Tutu, fur-
thered UN sanctions. Augmented sanctions called for banning both
nuclear collaboration and new U.S. investment, prohibiting bank
loans and computer sales to the South African government, and
stopping Krugerrands purchases (Orkin 1989, 135–36).
South African support and participation in Angola’s civil war led to
further cuts in military hardware sales to South Africa. After some
worsening domestic violence, South Africa began to turn around
politically. By 1992, free elections were held. Nelson Mandela (long
imprisoned for his opposition to apartheid) was freed, and the world
saw South Africa slowly reform. Nelson Mandela was later elected
president of South Africa in 1994.
The sanctions restricted South Africa from finding inexpensive sub-
stitute markets for oil. By the 1980s, fewer economies were coming to
South Africa’s rescue, as the world moved uniformly against human
rights abuses. Helping Rhodesia during U.K. sanctions in the 1960s
and 1970s showed South Africa’s initial lack of fear over international
embargoes. The United States took many years to join the original
sanctions, as U.S. business interests seemed to outweigh the political
rationale for sanctions.14
From 1985 to 1992, South Africa’s real GDP grew at 0.5 percent
per year, while inflation averaged 15.3 percent over the same time.
These numbers are worse than any eight-year span in postwar South
Africa. Some authors, like Lewis (1990), Lundahl (1992), and
Hazlett (1992), tried predicting effects on apartheid directly. They all
claimed the end was near and were correct. These sanctions initially
struggled to achieve the goal of dissolving apartheid. In hindsight, it
is likely that the 1985 strengthening of sanctions put pressure on
apartheid to end.

South Korea: 1973–77

Nuclear safeguards, proliferation, and international security, similar to
recent measures against their northern neighbor, were reasons for
these sanctions against South Korea (ROK). Another reason was
human rights violations. Political challenges and fear of another con-
flict with North Korea (PRK) sparked restrictions of political freedom
from fear that PRK was trying to undermine the current ROK gov-
ernment from within; these internal measures were enforced by ROK
troops. Aid reduction, specifically military assistance, was initiated in
198 ECONOMIC SANCTIONS

August 1973 by the United States; there will still American troops in
South Korea, however, as part of a long-term peacekeeping effort.
Related to these reductions was the U.S. reaction to ROK seeking
to purchase weapons-grade nuclear materials or technology to
reprocess nuclear fuel for weapons from France. ROK’s transactions
began to stir international interest, especially after India’s above-
ground weapons tests in 1974. Dreszner (1999) suggests that the
U.S. reaction was a threat to reduce aid and trade if ROK went
through with the reprocessing plant purchase (257). In 1975, ROK
ratified the 1968 Non-Proliferation Treaty and America slowly backed
away from sanction threats. The human rights violations also eased as
did tensions by 1977.
The South Korean economy in the 1970s was a developing econ-
omy, moving quickly to become an industrialized one. In 1972, the
United States purchased 46.7 percent of ROK exports and sold ROK
25.7 percent of its imports. South Korea’s real GDP grew at a rate of
9.78 percent per year between 1969 and 1973, while inflation grew at
8.7 percent over the same time (UN Statistical Yearbook 1975).

Sri Lanka: 1961–65

Oil asset expropriation led to U.S. financial sanctions against Sri Lanka
(Ceylon at the time). In 1961, the Sri Lanka government alleged that
the level of foreign investment did not reflect profits generated by
foreign oil companies there (Karunatilake 1971, 270). America pur-
chased 9 percent of Sri Lankan exports; 3 percent of Sri Lanka’s
imports were purchased from the United States. In 1962, the United
States passed the Hickenlooper Amendment, legally restricting aid of
any country that expropriated U.S. assets. Sri Lanka lost more eco-
nomic and military aid due to this legislation. Sri Lanka also found no
friends at the International Bank of Reconstruction and Development
(IBRD). The IBRD declared that the World Bank would grant no
loans to Sri Lanka because of the inadequacy of compensation pro-
posed for nationalized foreign assets (Kodikara 1982, 123). In 1965,
the government of Prime Minister Bandaranaike was voted out of
office, and the new government quickly agreed to compensate oil
companies for their losses. Economic aid, the only sanctioned good,
resumed shortly thereafter.
The Sri Lankan economy was based on rubber, coconut, and rice
at the time. Sri Lanka experienced 3.9 percent average growth,
accompanied by 1.1 percent average inflation during the 1950s.
HSE (1990) felt these sanctions were incredibly successful, perceiving
 BRIEF CASES HISTORIES 199

the aid lost stimulated governmental change and subsequent

compensation for seized assets. With little external debt before
1961, and relative little trade dependence on the United States, the
financial sanctions working alone did little to affect the Sri Lankan
economy.

The Sudan: 1989–Present

Human rights problems, including mass starvation caused by civil
unrest and war, led the United States to look at the Sudan as a place
in need of assistance. However, when the Sudan was unable to repay a
U.S. loan in 1989, America cut all economic and military aid, except
for food. This sanction continued for six months, and the loan was
summarily forgiven. The Sudan was not largely dependent on
American markets, buying 11 percent of her imports from and selling
3 percent of her exports to the United States. After a military over-
throw of a democratically elected government, the civil war ended
poorly from the U.S. view and all aid was cut in mid-1989. In 1991,
amidst famine and dissolving political situation, the Sudan received
humanitarian relief. The allocation of those goods, however, was
under a multinational UN force to make sure they were used to feed
the hungry.
The 1980s were particularly harsh on the Sudan as a result of civil
unrest. Prices rose from 1986 to 1990 by 500 percent, and rose
2500 percent between 1990 and 1993. Fixed investment plummeted
after 1989; real GDP was highly volatile between 1984 and 1992.
Debt moved up during sanctions to 90 percent of GDP by 1992. The
current account moved deep into deficit territory over the sanction
years. The Sudan forged agreements with Cuba, Libya, and other
international undesirables for military hardware (HSE 1990, v. 2, 634).
The United States still provided humanitarian relief, in large quanti-
ties, as did the United Kingdom. O’Sullivan (2003), in an expansive
case study, suggests that a more positive approach with the Sudan be
taken; in 2002, the Sudanese Peace Act in the United States began
such a process. “U.S. policy also needs to underscore how positive
changes in Khartoum’s behavior will result in improvements in its
relationship with the United States” (O’Sullivan 2003, 279). The
Sudanese economy is agriculturally based. Real GDP grew at an
average of 5.75 percent from 1992 to 2000, which was as strong as
any Western nation (UN Statistical Yearbook 1997 and 2003).
However, prices soared over the 1990s, increasing tenfold between
1990 and 1995.
200 ECONOMIC SANCTIONS

Syria: 1986–94
In 1980, Syria found herself on a list of countries that the United
States believed to directly support and fund international terrorist
activities. In 1985, Syria helped the release of hostages from an air-
plane hijacking. The United States subsequently loosened export
controls on Syria and purchased more oil. In 1986, however, Syria was
accused of developing chemical weapons. America accounted for only
6 percent of Syrian imports in 1986. The U.S. ban on chemical sales
to Syria for weapons production also included limited military hard-
ware sales. After Syria was linked to a group taking responsibility for
an attempted airline bombing in 1986, many European countries cut
aid; the U.K. cuts were the most severe, also cutting diplomatic ties
with Syria. From 1984 to 1989, Syria’s real GDP grew an average of
0.4 percent per year.
The 1986 sanctions were eased when Syria spoke out against Iraq’s
invasion of Kuwait. Syria’s diplomatic role in the Gulf War, denounc-
ing Iraq’s actions, renewed a positive status with the West and sanc-
tions became less restrictive after 1990. The stance against Syria
lightened further when she agreed to the Arab–Israeli peace confer-
ence in 1991. However, due to Syria being named as a country fund-
ing terrorist groups and a recent reversal to speak out against the
West, the United States has imposed full sanctions again. Syria’s inter-
national relations worsened during the 2006 conflict between
Hezbollah and Israel, where Israel blamed Syria and Iran of using
Lebanon as a puppet state to fund anti-Israeli activities.
The Syrian economy is based on oil, though cotton and olives are
also large agricultural industries. Real GDP grew at a pace of 7.57 per-
cent from 1990 to 1998; between 1999 and 2001, this growth rate
was under 0.2 percent per year (UN Statistical Yearbook 1997 and
2003). Prices in Syria, one of the major concerns with comprehensive
sanctions, were held stable through the late 1990s after rising about
80 percent from 1990 to 1995 (ibid.).

Taiwan: 1976
U.S. sanctions against Taiwan in 1976 were similar to those against
India, Pakistan, and South Korea discussed earlier. The omnipresent
issue of nuclear proliferation, especially with the threat of its immedi-
ate use against an assumed aggressor by new nuclear powers, has made
even a modicum of evidence enough to initiate economic coercion.
Based on intelligence reports that Taiwan wanted to develop nuclear
 BRIEF CASES HISTORIES 201

weapons, licenses for Taiwan to buy U.S. nuclear reactors were

stalled, effectively acting as an export sanction. “Nuclear exports
to Taiwan were not cut off, but export licenses were deliberately
delayed” (HSE 1990, v. 2, 424). The Taiwanese economy, much
like the South Korean economy, was a market economy breaking
out of its development shell in the 1970s. Taiwan’s trade with the
United States increased quickly throughout the decade. Taiwan
purchased 22 percent of its imports from the United States and sold
America 41 percent of its exports in 1975. These sanctions ended
when Taiwan agreed to shut down its reactor program in 1977.

The Gambia: 1994–98

A bloodless coup took place in the Gambia in 1994, putting in a non-
democratic government and ousting then president Dawda Jawara.
Sanctions were imposed immediately on the Gambia to coerce the
new government, under Colonel Yahya Jammeh, to restore democ-
racy. A combination of international economic sanctions and domes-
tic protests impelled Jammeh to announce a two-year schedule for
returning the country to “civilian” rule (Saine 2002). Military and
financial aid was embargoed until the United States was assured that
free elections were restored and human rights protected. After free
elections and a clean bill of human rights health, sanctions ended
in 1998.
The Gambia has a small, agriculture-based economy. Real GDP
grew at an annual pace of 5.68 percent from 1997 to 2001, with
prices remaining stable over the same time period (UN Statistical
Yearbook 2003). The United States purchased approximately
$100,000 in Gambian exports in 1998, exporting $12.9 million
in goods and services to the small nation (UN International Trade
Yearbook 2001). The credible threat of damage to the Gambia
came in the form of aid restrictions. Developmental assistance drop
from $40 million in 1997 to $38 million in 1998 to $32 million in
1999; by 2001, aid was back up to $48 million (UN Statistical
Yearbook 2003).

Thailand: 1990–93
A military government ruled in Thailand from 1976 to 1988. In
August 1988, Chatichai Choonhavan was voted prime minister. In
1990, political killings and some civil unrest raised eyebrows interna-
tionally. U.S. aid was restricted, but in trivial amounts. Ultimately,
202 ECONOMIC SANCTIONS

measures against Thailand between 1990 and 1994 were a mix

of the 1974 Trade Act’s Section 301 sanctions and anti-proliferation
measures. Many U.S. firms lobbied for financial sanctions against
Thailand, complaining about the Thai government’s subsidies to
pirating firms. In late 1990, the U.S. Pharmaceutical Manufacturers
Association filed a formal trade complaint against pharmaceutical
imports from Thailand, and importation was curtailed under
Section 301 due to unfair trade practices and patent violations. After
both patent and human rights problems seem to settle down during
1993, America reinstated aid. Recently, Thailand has experienced a
military overthrow, though no UN resolutions have flowed to initiate
sanctions.
Thailand, with real GDP growing from 1988 to 1994 at an average
10 percent per year, with inflation averaging 5 percent, quickly
became an Asian financial center, competing directly with Hong Kong
until the Asian currency crises of 1997. Thailand was an industrial
economy in 1990, with 79.8 percent of her exports in manufacturing,
specifically food and textiles. Industrial supplies and machinery domi-
nated Thailand’s imports. Thailand was relatively dependent on
America for 30 percent of its trade in 1989.

Turkey: 1974–78
A political problem in Cyprus led to economic sanctions against
Turkey in 1974. After a Greek takeover, defying a U.S. warning,
Turkish troops invaded Cyprus in 1974. Problems began when Turks
living on Cyprus lost many of their rights in 1964, as the government
revoked any rights Turkish national and descendants had on the
island. In 1974, a power struggle took place, and the Greek govern-
ment attempted to seize control. In the name of protecting their
citizens, Turkey moved against Cyprus’ government.
Reacting to this invasion, the United States cut military aid. Turkey
renounced this sanction, claiming that the military aid and sales were
essential elements of the NATO Treaty (Denktash 1982, 78–79). The
largest failure of U.S. sanctions against Turkey was not preventing the
United Kingdom, Italy, and West Germany from acting as substitutes
arms suppliers. Also, the sanction’s concentration was on the military,
not the economy. The United States told Turkey not to use U.S.
weaponry on Cyprus, and to negotiate a settlement. The strain in
Turkish–American relations also led the USSR to Turkey’s rescue,
though military aid never completely ceased between the United
States and Turkey. The Turks also found friends in Libya, Saudi
 BRIEF CASES HISTORIES 203

Arabia, and Iraq. Fighting a bloodless war for the next three years,
Cyprus’ Greek government, for reasons other than this crisis, fell
apart. A new, Turk-friendly government took over Cyprus by
September 1978. The United States subsequently lifted the arms and
aid embargoes.
During the 1960s and early 1970s, Turkey had impressive eco-
nomic growth. Real GDP grew between 1963 and 1978 at a 6.5 percent
average; the numbers are even larger for the time period between
1970 and 1973. Foreign aid, in conjunction with agricultural and
industrial policy reforms, fueled the Turkish economy prior to 1973
(Harris 1985, 76). Turkey had rapid government spending and
money creation, increasing inflation after 1973. Turkey’s balance of
payments situation deteriorated so bad as a result of the 1973 oil
crisis, as their dependence on foreign oil mounted, the international
capital market stopped lending to Turkey (80). America claimed
12 percent of Turkish trade in 1973. Over the sanction period, real
growth fell sharply, government spending and debt and prices and
unemployment all rose.

Uruguay: 1976–81
Uruguay’s military regime was accused of holding political prisoners
in deplorable conditions in 1976. The United States reacted by
banning arms shipments to Uruguay, followed two years later by eco-
nomic aid restrictions. During this sanction, domestic price instability
led to poor economic growth, furthering a need for economic aid
(Rottenberg 1993, 305). In 1980, America sent an aid package to
Uruguay, though most international agencies claimed human rights
abuses continued afterward. In 1981, U.S. military sales resumed, to
the chagrin of many human rights organizations and Uruguay’s political
moderates.
Uruguay’s economy is an agriculture-based economy. America
bought 11 percent of Uruguay’s exports and provided 8.3 percent of
her imports in 1976. Cattle and sheep herding and processing were
Uruguay’s largest industries. Government spending was rapid over
the sanction period. Increased debt service pressured Uruguay
exports, which fell after 1980 in real terms (Harris 1985, 80). The
United States also provided Uruguay with military and economic aid
in small amounts.15 HSE (1990) felt that Uruguay’s human rights
situation was not resolved by sanctions, and deemed these embargoes
a failure, as evidence provided by Uruguay of better social conditions
was unconvincing.
204 ECONOMIC SANCTIONS

Union of Soviet Socialist Republics

(USSR): 1975–83
These U.S. sanctions were Cold War sanctions, reactions to the threat
of communism spreading through Soviet funding and trade support
of new governments, or rhetorical moves to posture on the world
stage. There were numerous episodes identified by HSE (1990), the
most of important of which was the grain embargo of 1981.16 Many
of these measures were small in their magnitudes. In 1975, the United
States finalized the Trade Act of 1974. In 1972, the USSR imposed a
tax on the emigration of educated citizens, which threatened the Most
Favored Nation (MFN) status of the USSR being discussed; the MFN
label for the USSR was a key issue for the 1974 Trade Act (HSE 1990,
v. 2, 386–87). In 1975, U.S. imports to and financial flows with the
USSR were restricted due to the delay in MFN status. This continued
nominally for years. In 1978, alleged U.S. intelligence activity in the
USSR was thwarted, and the spies were put on trial. As a result, the
United States restricted technology exports to the Soviet Union. This
sanction was also somewhat nominal.
In 1980, U.S. sanctions began to have more substance. The Soviet
invasion of Afghanistan initiated worldwide reactions, and U.S. sanc-
tions were an immediate policy reaction. American exports of grain to
the USSR were sanctioned in 1980, as diplomacy to remove Soviet
troops from Afghanistan fell apart. The decision by President Carter
for the United States to not participate in the 1980 summer Olympiad
in Moscow can be seen as an import sanction. The grain embargoes
are an important case for two reasons. First, the measures are seen as
a grand example of why sanctions do not work between major
economic and political powers, and may have cost U.S. workers and
farmers dearly due to sales reductions. Second, this episode continued
American statecraft’s use of an economic tool as diplomatic reputa-
tion; if a country acts as an aggressor against an otherwise innocent
nation, U.S. economic sanctions flow immediately thereafter. “The
U.S. decision to implement the grain embargo was framed in terms of
the concern for relative gains [the economic and political benefits
would exceed their costs] and reputation” (Dreszner 1999, 76).
Sanctions were imposed again in 1981 due to the Soviet support of
martial law in Poland. This is another important episode illustrating
the immediacy of U.S. economic statecraft due to international
human rights violations. These sanctions included travel, grain, and
technology exports that lasted only until 1982, as a new trade agree-
ment began new negotiations (HSE 1990, v. 1, 209). A South Korean
 BRIEF CASES HISTORIES 205

airliner was shot down in 1983 by the USSR, Korean Air Lines Flight
007, for crossing into Soviet air space and allegedly not reacting to
warnings to turn back across the border. It took seven days for the
Soviet Union to admit to the downing of Flight 007; U.S. travel
sanctions were augmented as a result.
Trade between the Cold War superpowers fluctuated in the early
1970s, but was relatively low as a percentage of the nations’ overall
trade. The United States purchased a negligible amount of Soviet
exports in 1974 and sold the USSR 3 percent of Soviet imports.
Soviet statistics on their domestic economy are parsimonious and
suspect during this time. Real GDP growth in the USSR from 1970
to 1975 is reported by the United Nations at an average of 5.33 per-
cent per year, where inflation was negligible over the same timeframe
(UN Statistical Yearbook 1977).

Zimbabwe: 1983–88
Zimbabwe was no stranger to economic sanctions when the United
States cut aid and credit in 1983. Zimbabwe was the white-ruled part
of Rhodesia (Southern Rhodesia) between 1965 and 1980; Zimbabwe
is adjacent to South Africa. The Rhodesian economy was heavily sanc-
tioned throughout the late 1960s and 1970s by the United Kingdom
and United Nations for its repression of blacks and its foreign policies.
The right-wing, white government of Ian Smith declared independ-
ence unilaterally from the United Kingdom in 1965, and sanctions
were swift, comprehensive, and punitive. These sanctions continued
through the 1970s; South Africa became Rhodesia’s trade station,
hoping the lack of success in Rhodesian sanctions would ease sanctions
against apartheid (Renwick 1981, 51–52). Petroleum purchases and
sales for Rhodesia would have the opposite effect on both countries. In
1976, South Africa ironically imposed its own sanctions on Rhodesia,
where sanction continuation was based on a turnover to majority
(black) rule of Rhodesia in two years. In 1978, the Rhodesian govern-
ment planned to hold elections, elections that in 1979 voted in a black
prime minister. By April 1979, the old government was gone, and
Rhodesia became majority-ruled. A violent power struggle ensued,
and the United Kingdom negotiated a cease-fire in December 1979.
In 1980, Zimbabwe became independent.
America and Zimbabwe disagreed on some key foreign policy
issues in the early 1980s. Zimbabwe disagreed with the U.S. accusa-
tion that the Soviet Union deliberately downed a Korean Air jetliner
and further believed that the Grenada invasion was unnecessary (HSE
206 ECONOMIC SANCTIONS

1990, v. 2, 568). Also, there was some civil unrest in Matabeleland,

Zimbabwe’s western frontier, as President Robert Mugabe alleged
South Africa was supplying rebel forces there with weapons and fuel.
In 1983, the United States cut aid and credit to Zimbabwe as a reac-
tion to anti-American statements made to the United Nations and
world press.
More aid was embargoed in 1984 because America felt that dis-
tributing aid would not happen efficiently in Matabeleland. In 1986,
after another verbal attack from Zimbabwe, U.S. aid was further cut.
In 1988, aid was restored in the hopes that the failing Zimbabwe
economy would be stimulated. The United Kingdom disagreed with
the U.S. reaction to Zimbabwe’s complaints, continuing to provide
assistance and trading goods.
Agriculture industries account for over 70 percent of Zimbabwe’s
labor force in 1985 (UN Statistical Yearbook 1990). From 1981 and
1986, real GDP per capita fell from $484 in 1980 U.S. dollars to
$454. Inflation averaged 15.2 percent per year between 1981 and
1986. HSE felt this sanction was a failure as sanctions hurt those who
needed food, and did not deter the Zimbabwe government from
speaking out on international matters.

Summary
This chapter provides brief case histories to springboard further case-
specific analyses. Chapter six investigates these cases empirically in an
attempt to see how economic effectiveness compares not only to the
measures by HSE (1990) and others, but how it may drive sanctions
along the Sanction Effectiveness Continuum. The case studies pro-
vided us with a historical picture of the political and economic situation
in each of the selected cases.
 Notes

Chapter 1 To Sanction or Not to Sanction?

1. See Sharp (2006) and his collaborators for an in-depth investigation of
the 2006 events in Palestine.
2. The situation has both worsened and eased since a North Korean
underground nuclear test in October 2006.
3. Hufbauer, Schott, and Elliott (1990) put together the most compre-
hensive summary of sanction cases to date. This study is mentioned
in almost every article and book in the literature, and will be often
mentioned in this text. The third edition of that study is imminent.
4. Because this text focuses on tracking the economic effects rather than the
legality of sanctions or the diplomatic aspects of using embargoes, the
differences between tools of statecraft and coercion are seen as negligi-
ble. Sanction nomenclature leads to confusion over terms and tedium in
reading and writing about these policies. However, this confusion
among sanctions, economic coercion, economic statecraft, embargos,
and so on is discussed in Drezner (1999), Bonetti (1997a, b), and Drury
(2005), e.g., and is a substantive issue in some disciplines.
5. The public choice literature suggests that the target populace is never
completely innocent in a target’s deviance, as interest groups are made
up of target citizens. However, it is a matter of perspective. The target,
acting rationally, engages in policy to better itself; the sender reacts for
the same reasons.

Chapter 2 Basic Sanction Analysis

1. See Bolks and Al-Sowayel (2000) concerning an empirical assessment
of sanction duration.
2. This is case 65–3 in Hufbauer, Schott, and Elliott (1990). There one
will find more details than in Galtung (1967).
3. See Gaisford and Sood (1996) for more on trade sanction optimality.
4. See Pugel (2003), chapter nine, for a textbook exposition of barriers to
trade versus free trade outcomes. Also see Carbaugh and Wassink
(1988).
208 NOTES

5. The South African case is the most-often quoted and referenced case
in this literature. It is also one of the longest in duration and had three
different incarnations seeking multiple goals over time. See Crawford
and Klots (1999) and Drury and Chan (2000) for edited volumes on
South African sanctions.
6. Hufbauer, Schott, and Elliott’s (1990) two-volume treatise on
sanctions is said to be updated in 2007.
7. The classic result of a quota is a “deadweight” loss of welfare. If the
sender government imposes an import sanction, sender citizens experi-
ence higher prices and lower quantities for target goods, and thus lose
some of their consumer surplus gained under unrestricted free trade. This
loss is more than sender producers’ gain from higher sales of domestic
substitutes for the target’s goods. The result is a net national loss, a
reduction in national welfare. In the target’s export market, there is a loss
of producer surplus greater than the target consumer’s gain as a result of
lower prices and higher quantities available in their domestic markets.
8. In chapter four, public choice theory provides some more similarities
between import sanctions and import quotas.
9. The idea of smart sanctions is expanded upon in chapter four of this
text, and is gaining steam in this literature. Arms sanctions, however,
like any other sanction, fall under the umbrella of trade sanctions in
theory. See Cortright and Lopez (2002) for a broad, edited volume
on smart sanctions, specifically arms embargoes.
10. The dichotomy of real versus financial sanctions is a large one in the
capital markets. Financial sanctions are capital market sanctions, but
affect entitlements directly rather than physical capital directly. This is
important because instead of the target losing a rate of return on a
physical asset, they lose the ability to financially invest in the sender
economy. Also, through the balance of payments, financial sanctions
have trade balance effects and vice-versa.
11. The financial sanction is much like credit rationing. If the sender acts
as a major target creditor, the sender merely rations credit, much like
a bank would to certain markets domestically.
12. The United Nations Charter is on the web: Chapter 7 is located at
http://www.un.org/aboutun/charter/chapter7.htm.
13. This case is briefly discussed in Appendix 1 under Zimbabwe’s case,
the current name for Southern Rhodesia. A reviewer suggested that
my treatment of Baldwin’s text here understates the broad scope
of Baldwin’s work. It is important to note the timing of Baldwin’s
(1985) book. The first edition of Hufbauer, Schott, and Elliott (1990)
came out the same year, and was subsequently updated. While many
authors have cited Baldwin, as I do here, Hufbauer Schott, and Elliott
(1985 and 1990) has dominated the literature because of the empiri-
cal dimension in each edition. Baldwin’s text is seminal beyond doubt,
and actually acts as a comprehensive theoretical and literature review
to 1985 on sanction topics.
 NOTES 209

14. This rally-around-the-flag effect of sanctions is discussed in more

detail in chapter four, as public choice models have assumed that cer-
tain target political systems are impenetrable due to the ruling class’
ability to make the sender the bad guy in citizens’ eyes. See Baldwin
(1985, 183) for more.
15. A. Cooper Drury, one of the foremost political scientists in this field,
has written extensively about the HSE study, its power and pitfalls. See
Drury (1998, 2005) and Chan and Drury (2000) for multiple angles
on the HSE (1990) data, hypotheses, and conclusions.
16. The product of the economic and political scores is also highly debated
because it lacks continuity. See Drezner (2003) and Drury (2005) for
more. This is an ad hoc measure in any case. If a sanction receives a
“nine,” this comes from one possible combination of political and eco-
nomic scores: three and three. If the score is an eight, there are two pos-
sible combinations: four for political and two for economic, or vice-versa.
17. Studies complain about the subjectivity of HSE’s conclusions, but
these data are used to not only refute the HSE claims, but to also
enhance other episodes’ analyses. Lam (1990) and Drury (2002) are
good example of specification changes in an attempt to enhance the
robustness of the HSE conclusions.
18. The GNP–GDP debate in this literature has not really begun, and
I believe that it is due to HSE’s continued use of GNP as the main
measure of both the cost of sanctions and the nation’s relative size.
Some of this use comes from gravity equation theory, where GNP is
used to measure relative size and includes all trade. Gravity equations,
and their role in this literature, are discussed briefly in chapter six.
The literature should update to GDP for future studies and case
analysis and break the foreign flows away from the other economic
measures.
19. See Galtung (1967) for the beginning of this idea.
20. See Drury (2005), chapter nine, for conclusions on this, and Martin
(1992) for counterpoint to Galtung (1967).
21. This statistical procedure is utilized in chapter six of this study. All
these conclusions flow from Drury (1998) and are basically repeated
in Drury (2005), which is in many ways a follow-up and expansion of
the 1998 study.
22. In HSE (1990), the nine policy recommendations are stated in
volume 1 on page 94.

Chapter 3 Sanction Initiation and

Continuance: Enter Game Theory
1. McCain (2004) is an amazing introductory text on game theory and
serves as general background for this primer. Osbourne (2004) is a
more technical but still approachable text, as is Watson (2003).
210 NOTES

2. See Drezner (1999), chapter two, for a simple model and background
on political game theory. Drezner (2003, 646) also provides a simple
example in the sanction context.
3. See Drury (2005), chapter three, for a recent use of such a variable
empirically as an update to HSE (1990) and its use of the prior
relationship.
4. The public choice framework in chapter four extends this basic idea.
5. The Sanction Effectiveness Continuum provides the insight that sanc-
tion decisions have multiple stages and payoffs, which stretch across
economic, humanitarian, and political efficacy.
6. See Eaton and Engers (1992) concerning the target’s toughness, and
Eaton and Engers (1999) concerning the sender’s resolve. These are
expanded upon later.
7. See McCain (2004), chapter twelve, concerning commitment. The
commitment game lies between the competitive and cooperative
games because its outcome hinges on the commitment’s credibility.
8. See Lisa Martin (1992, 1993). Her work is the game theory pillar in
this literature. Her studies are aimed at identifying cooperative games
played between the sender and other countries deciding to join the
sender or assist the target. Much of her work is on strong versus weak
leadership, discussed in more detail later.
9. See Drezner (1999, 38), chapter two, note 22, for another definition
of Nash Equilibrium.
10. Steil and Litan (2006) discuss a real-world analog of this example in
chapter six of their text. The real-world example concerns PetroChina
and the Sudanese government seeking funding for oil exploration and
production in 1999 through an initial public offering in the U.S.
equity markets.
11. See Bonetti (1997b) for a discussion of game theory and other models
in the context of sanctions, leading up to public choice models, in an
outstanding survey article.
12. See Kaempfer and Mertens (2004) for more on dictator sanctions and
theoretical issues specific to these cases.
13. Pape (1997) and Drezner (2003) are both excellent investigations as
to the credibility and power of threats versus their omission in the
HSE (1990) data and from the literature otherwise.
14. The Israeli blockade of Lebanon in July and August 2006 may be clair-
voyant of future sanctions. The naval blockade took place parallel to a
military invasion of southern Lebanon. As a result, Lebanon pleaded
with the United Nations to order a cessation of military activity and an
immediate lift of the blockade. The effects on the Lebanese economy,
coupled with the military action, had a devastating trajectory.
15. See Jack (1940) and Pigou (1941) for more on economic warfare,
Pape (1996) for the use of strategic bombing as a means of coercion,
and Pape (2005) for a role reversal concerning terrorist suicides acting
as coercive measures.
 NOTES 211

16. HSE (1990) include some sanctions with military conflict alongside of
sanctions. The bias in these episodes toward the military victor makes
statistical analyses imprudent. These sanctions are eliminated from the
set of those analyzed in chapter six for this reason. See Pape (1997) for
a breakdown of these cases from the HSE (1990) data.
17. Thailand recently experienced a bloodless coup of its democratic
system in favor of military rule with allegiance to a monarchy. It
is likely that the September 2006 coup will become yet another
sanction case.
18. This is typical of the bargaining literature; Rubinstein (1982) is a
seminal work.
19. The existence of subgame equilibria (given perfect knowledge, the
player will choose a Nash Equilibrium when making a decision) is
guaranteed under a finite and perfect information game. Information
is assumed both symmetric and perfect; future models are likely to
expand on the theme of information asymmetry.
20. The Iraqi sanctions of 1990–2003 when the United States invaded
Iraq are now a famous case study. See Askari et al. (2003, 48) for a
brief history, as well as Appendix 1 in this text. Also see Cortright and
Lopez (2000), chapter three.
21. Cartels are illegal in the United States, when prosecuted and proven
to be so in a court of law, on either side. However, implicit cartels
exist within the United States and explicit cartels exist internation-
ally. Regardless, the lessons to be learned from price theory are
enlightening.
22. Also from microeconomics, the Bertrand model is the Cournot model
application where price instead of quantity is controlled. However,
they have the same outcome if competition is allowed. The Cournot
model suggests that the quantity at which profit maximization takes
place, and indirectly price, is where the firms in the oligopoly choose
a cooperative strategy rather than take market share via expansion of
supply; the Bertrand model focuses on price, indirectly affecting
quantity. See Varian (2002) for an accessible version of each of these
models in the context of microeconomic theory.
23. See Haass (1998) for a strong argument against the proliferation of
sanction use in terms of the United States using sanctions as default
diplomacy.
24. Individual countries have brought international cartels to trial and
punished them for their price-fixing activities in domestic courts, but
international enforcement is a major issue. See Connor (2004) for
recent legal actions and successes against international cartels.
25. Looking at the target’s openness, exports and imports summed as
a percentage of Gross Domestic Product (GDP) may provide
insight here.
26. The following website has the basic details and background on this
Act, where some updating is needed, given recent relaxation of
212 NOTES

sanctions against Libya. See http://www.fas.org/irp/congress/1996_cr/

h960618b.htm for the Congressional record.
27. The “carrot–stick” approach of economic statecraft is discussed
extensively in Drezner (1999).
28. Martin (1993) extends her 1992 study, but the conclusions are very
similar. There is a larger discussion of audience costs and sender
credibility in the 1993 study.
29. This makes the question of sanction duration large, as time not only
naturally erodes sanction effects due to diminishing returns (if sanction
effectiveness is a function of time) but also provides more opportuni-
ties for cheating and new markets to emerge, reducing the present
value of economic statecraft’s benefits in both the numerator (net
benefits) and the denominator (the present value discount factor).
More studies on sanction duration are needed in this literature.
30. Pape (1997) argues strongly that sanctions may have no future
beyond this economic damage.
31. These countries are simply the longest episodes currently in place.
Iraq, Iran, Libya, and South Africa are all veterans of long-term
sanctions.
32. Recently, the United States softened its stance against Libya in
response to her supposed disassociation with al-Qaeda and her leader
expounding the tenets of “liberal internationalism.” See Hurd (2005)
for more.
33. See Pape (1997, 99–100), tables 1 and 2, for a breakdown of the
HSE (1990) cases, where “Political Destabilization” is analogous to
such a goal.
34. See Lam (1990), Drury in multiple studies (1998, 2005), Drezner
(2003), and Sobel (1998). There is some debate that the previous
trade relations between sender and target have little to do with sanc-
tion effectiveness, where Lam and Drury have explicitly included the
trade flows as independent variables in a regression.
35. See Dunning (2005) for a recent paper on this connection.
36. See the brief history of Haitian sanctions in Appendix 1 and in Askari
et al. (2004) and HSE (1990), case 87–1.
37. Hazard functions have been used to test for a sanction’s optimal dura-
tion. See Bonetti (1994) for a technical analysis of sanction duration.

Chapter 4 Public Choice Theory

and Smart Sanctions
1. See Pape (1997) for one of the most critical works on whether sanc-
tions are a good policy choice at all, and Craven (2002) for the use of
smart sanctions.
2. Craven (2002) discusses the flaws of smart sanctions at length. This
study is informative on the aspects of international law, but is not
 NOTES 213

preachy about the need for all sanction regimes to turn and focus
strictly on humanitarian factors and effects.
3. See Buchanan and Tullock (1962) for what is considered the seminal
work in public choice theory, as well Tullock (1987) for a brief his-
tory and summary. Becker (1983) is seen as the father of modern
public choice, and is the springboard for most of Kaempfer and
Lowenberg’s works.
4. William Kaempfer and Anton Lowenberg are leaders in the sanction
literature and have applied public choice theory in many studies. Their
1992 text on economic sanctions is not only a treatise on public
choice theory generally, but on political economy in statecraft as well.
5. See Bonetti (1997b) for a review of Kaempfer and Lowenberg (1992)
and others. Bonetti is critical in this survey about public choice theory
and its inability to easily test its hypotheses empirically (340).
6. Kaempfer and Mertens (2004) examine sanction imposition against a
dictator, which has many real-world applications.
7. The trade linkage signifies the percentage of the target’s total trade
with the sender before the sanction is imposed. This is more or less a
measure of openness with the sender. Bonetti (1997a) empirically
investigates the HSE (1990) data using openness as the focal inde-
pendent variable. Chapter six’s empirical investigations of sanction
effectiveness will discuss the use of openness and trade linkages as
explanatory variables in more detail.
8. See Kaempfer and Mertens (2004).
9. Drury (2005) is one of the best texts in this literature in its breadth on
political economy and empirical analyses. His models are the founda-
tion of the humanitarian and political models in chapter six. See
Drury (2005), especially chapters four, five, and six, for his examina-
tion of U.S. presidential sanctions.
10. An example of this is the current sanction package on Iran as a result
of its funding terrorist organizations such as Hezbollah. However, this
policy package is one of many the United States currently has in play
concerning international terrorists. The point is that governments
such as those in Iran and Syria choose to pursue policies that are guar-
anteed an economic statecraft reaction from the United States in the
least, and possibly the United Nations as well. A target may continue
a deviant policy simply because the economic and political pressure
specific to certain groups in these countries is not strong or focused
enough to act as a credible threat. Thus, the target government
continues and the sender’s policies become more like rhetoric, but
damage the populace.
11. This analysis is slightly different than the classic public choice model
of economic statecraft. Refer to Kaempfer and Lowenberg (1992) for
a full exposition.
12. USA Engage has more than that number listed on their website,
www.usaengage.org, and remains fairly current on any new episodes.
214 NOTES

13. This recent work is an outstanding study of public choice

theory, along with microeconomics applied to sanctions, but
(like many other public choice papers on sanctions) circumvents
empirical work.
14. A reviewer suggested these sanctions are not focal, but in fact aimed
to dismantle apartheid directly while minimizing the damage to South
Africa’s black citizens. This is a root problem with many studies on
South Africa. Starting with Porter (1979) on through Major and
McCann (2005), the belief that South African consumers and busi-
nesses were irreplaceable because of either the diamond industry or
financial investments in the Kruggerand is hyperbole. South African
sanctions finally worked because there was a general consensus in the
United Nations that institutions such as apartheid were detrimental to
worldwide movements toward economic amnesty and democracy for
all. Regardless of the study or model, a true international consensus
was needed for South African sanctions to become effective, and
part of that effectiveness was over a decade of economic stagnation,
which hurt the poor that the sanctions were attempting to help more
than the elite they were seeking to depose. This is the classic sanction
paradox when dealing with human rights.
15. Of course, there was a lack of conclusive evidence concerning the
stockpiling of such weapons by the Iraqi government as alluded to on
many occasions by President George W. Bush as the cause for Iraq’s
invasion. Iraq’s constant and protracted problems with facilitating UN
inspectors did not help their cause.
16. In 1996, the United States passed legislation allowing firms within its
borders and without to be economically punished as a result of not
complying with sanctions. The Iran–Libya Sanctions Act of 1996 pro-
hibited investment from flowing to Iran and Libya from U.S. firms,
and also set the conditions under which firms in other countries could
also be sanctioned for their financial involvement in these countries.
Libya was dropped in this act’s renewal on April 23, 2004, as it finally
conformed to the conditions of UN sanctions after the downing of
Pan Am Flight 103.
17. See the OFAC website for more details on this agency inside the U.S.
Treasury Department: www.treas.gov/offices/enforcement/ofac.
18. The black knight idea was first applied in Galtung (1967) implicitly as
a sanction buster; HSE (1990) used it as the name for the former
Soviet Union and its satellite nations when helping a communist
country or government during U.S. sanctions against new govern-
ments and rebellious forces in Central America and elsewhere during
the 1970s and 1980s. In many empirical studies, the existence of a
black knight is a control variable, as discussed in chapter six.
19. The carrot–stick approach also applies here, as it does when examin-
ing smart sanctions. See Drezner (1999) for more on this approach,
its pitfalls and possibilities.
 NOTES 215

20. See the Stockholm Process’ website, http://www.smartsanctions.se/

under the link “UN Sanctions.” There are regular updates at this site.
Bondi (2002) also lists recent arms embargoes (111).
21. See Naylor (1999), chapter sixteen, for an interesting and disturbing
history of the relationship between Iran and U.S. arms manufacturers
leading up to the Iran-Contra Affair.
22. See Goldstein and Turner (2004) for more. This text is regarded as
the best on currency crises. See Stein and Litan (2006) for a detailed
overview of the conclusions and their perceived pitfalls of this text.

Chapter 5 Open Economy Macroeconomics

and Sanctions
1. The NOEM model appears in many forms and forums as the initial
study. The textbook version in Obstfeld and Rogoff’s model (1996) is
somewhat accessible; Mark (2001) is more accessible.
2. The IS-LM model and its components can be found in a low-tech
form in many intermediate macroeconomics texts. See Blanchard
(2002), chapter six, as a good intermediate example, whereas Romer
(2005), chapter five, has a more advanced explanation.
3. In certain cases, I will refer to equations in chapter five A. However,
for the non-economist, the layout of this chapter is meant to mix
intuition directly with the mathematics to focus on the conclusions for
purposes of policy making.
4. Since 1990, the United States has engaged in many financial
sanctions. Asset seizures are default sanctions for the United States
through the Office of Foreign Asset Control (OFAC).
5. Throughout the analysis here, it is important that the non-sanction
world collapse to the baseline model for reasons of consistency.
6. The public choice theorist may argue that the income of special
interest groups may outweigh the deadweight losses of sanctions.
Following Bonetti (1997b) and his arguments, the public choice
framework is intuitively sound, its lack of an empirical model makes it
difficult to leap from macroeconomic theory where sources and uses
of income are explicit or just do not exist.
7. For more on Lerner symmetry, see Bhagwati (1999), chapter twelve.
8. It may be more aesthetically pleasing to sanction scholars to have the
target as the domestic economy, especially since the focus of almost
every model to date is on the target’s dynamics. However, this model
begins with policy initiation, and thus with the sender.
9. Recent advances in the NOEM change this assumption to allow
differentiation between the countries and their view of import substi-
tution for the domestic good. See Bergin (2003) for more.
10. The distinction between nominal bonds denominated in the sender’s
currency and real bonds denominated in local goods is a big one.
216 NOTES

However, in the case of sanction analysis, it seems more appropriate to

use nominal bonds, as the sender sanctions income and the market for
new loanable funds. The effects on the model are trivial given other
simplifications made here. See Mark (2001), chapter nine, for more
on this distinction in NOEM models.
11. Since 1990, the United States has engaged in many financial sanc-
tions. Asset seizures are default sanctions for the United States
through the Office of Foreign Asset Control (OFAC).
12. There are recent studies in the NOEM literature that include capital
costs. See Kollmann (2001).
13. The logarithmic utility function is a special case of the class of utility
functions called “Constant Relative Risk Aversion” or CRRA, where
C  C1  /(1  ), and the logarithmic case is when   1. Many
other studies have generalized and allowed  to vary from 1. See
Blanchard and Fischer (1989, 43–44) for a great overview.
14. Seminal models include Cagan (1956) and others. The money
demand function here is a “Cagan-style” model, similar to the CRRA
models of consumption.
15. This comparison of utility is important for thinking about sanctions in
the context of public choice models from chapter four.
16. We should expect that, much like other “labor supply” conditions, the
individual will only work a maximum number of hours, regardless of
those conditions. The labor supply curve will bend backward at some
number of hours. We will assume that in the log-linear form given
later, the backward bend is not a concern.
17. The terms of trade and their effects on the economy when changing
are generally the centerpiece in arguing against trade barriers. See
Pugel (2003) for an accessible approach to this issue. Equations 5A.70
and 5A.71 derive the terms of trade.
18. See HSE (1990) for more data and details.

Chapter 5A Mathematical Derivations

of NOEM Sanctions Model
1. Chapter six uses exchange rate data to estimate the sanction shock
econometrically. In that estimation, the target is the focus, as the
exchange rate shock is used to observe human and political changes
from economic coercion.
2. While it is more intuitive to think of the exchange rate in terms of the
sender currency, since most of the NOEM studies and model versions
followed for this text use an exchange rate in terms of the foreign
country, sender currency per target here, the same convention is
used. In chapter six, this definition is reversed because it is more
intuitive.
 NOTES 217

3. It is likely a version is imminent. See Cavallo and Ghironi (2002) for a

version of the NOEM model with productivity shocks, alluding to a
need for capital accumulation to be added in future research.
4. By choosing bond holdings, the household is really choosing how it
intends to trade off current consumption for future consumption once
the levels of work effort and money demand are determined. Once
bond holdings are decided, they imply an optimal consumption level,
derived in equation 5A.16.
5. This assumption is later relaxed, as in other pricing-to-market models.
See Mark (2001), chapter nine, for more details.
6. I used MathCad based on MapleTM technology to solve for these sys-
tems instead of doing the algebra by hand.
7. The amount of money held by the household is a clearinghouse for non-
interest bearing asset demand after consumption and work effort are
determined. This allows a focus on the real versus nominal variables in
this analysis. See Obstfeld and Rogoff (1996), chapter ten, for more.

Chapter 6 Empirical Analyses of

Sanction Effectiveness
1. See Kim and Pagan (1995) for basic calibration techniques, especially
for those followed in Eyler (1998).
2. Eyler (1998) was based on the target economy and did not provide any
estimates of effects on senders.
3. See Rose and Speigel (2004) for a recent addition to the literature.
While their model concentrates more on the reaction of lending
economies to sovereign default, the use of gravity models is well
showcased in their work.
4. A simple sanction model using gravity equations is shown here, for
countries i and j and taking natural logarithms: ln(Tradeij)    1·
ln(GDPi  GDPj)  2 · ln(distanceij)  3 · Sanctionij, where 1 is
expected to be positive and both 2 and 3 are expected to be
negative. Financial flows and the effects of smart sanctions or other
financial curtailments can be tested in similar ways. Askari et al. (2004,
123–90) provides an expansive look at gravity modeling and sanction
efficacy.
5. Rose and Speigel suggest that common borders are not a determining
factor in the level of international financial flows between adjacent
nations (2004, 55).
6. Using the exchange rate as the initial signal of sanction effects, as both
this text and Sobel (1998) do, brings with it some possible empirical
issues. First, the exchange rate must be floating at the time of the
sanction to provide the necessary independent variation to make an
econometric exercise worth doing. Second, we assume that the U.S.
exchange rate with the target is simply a reflection of sanctions. If the
218 NOTES

United States is sanctioning a country, and therefore by choice

changing specific balance of payments with the target, there may be
exchange rate effects. Sobel (1998) does not attempt any impulse
response estimations for the U.S. economy.
7. Sobel (1998) acknowledges the importance of this distinction, and
suggests that whether the sanction causes permanent or temporary
shocks to the exchange rate is important in assessing the effectiveness
of sanctions. As a larger connection to our discussions in chapter four,
Sobel claims that the lack of permanent effects on the exchange rate
suggests that the United Nations was driven more by special interest
groups, a la public choice theory, to ramp up sanctions against
South Africa rather than trying to impose sanctions for human rights
violations.
8. In an autoregressive model with p lags, the lagged dependent variable
represents feedback, or how much the past explains the dependent
variable’s contemporaneous outcomes. See Dickey and Fuller (1979)
for the original unit root test methodology and Hamilton (1994,
501) for an advanced but general discussion.
9. See Sims (1980). His idea was simple. Is there a way to forecast how a
monetary shock, which may or may not be anticipated, affects certain
macroeconomic variables? Using the past of each policy variable as the
exogenous variables, multiple equations are used to connect a change
in each variable to contemporaneous values. In the Sims model, the
natural logarithm of the money supply and real GDP in the U.S. econ-
omy was used to show how a percentage change in money supply may
affect real GDP growth. Sims also hypothesized a feedback loop,
where real GDP growth may influence monetary policy. Engel and
Granger (1987) is seen as the seminal article on cointegration and
uses linear combinations of variables with similar time series properties
to defeat potential unit root problems. Also see Studenmund (2003),
chapter twelve, for a less technically intensive explanation.
10. Drezner (2003) and Drury (2005) discuss the problems in the success
score used by HSE (1990). However, it is important to note that the
economic and political success scores, respectively, suffer from subjec-
tivity problems, but no one can blame HSE for making logical claims
with an ordinal variable.
11. Bergin (2003) provides the latest work in econometrically testing the
claims of the general NOEM framework. However, because of the
VAR model’s flexibility, it is possible to envision other models using
the BOP directly, through the current and capital accounts. The terms
of trade, as discussed at the end of chapter five, is also an important
component of exchange-rate pass through measurement in NOEM
models. Examining Gross Domestic Product (GDP) in per capita
terms or in growth rates also may be illustrative of economic damage
caused by sanctions. What any variation on this theme must do is link
economic shocks to humanitarian and political effects from economic
 NOTES 219

statecraft. Using the exchange rate, as shown in previous chapters and

suggested by Sobel (1998, 6), encompasses a large amount of macro-
economic information.
12. The impulse response idea takes previous information into account
naturally through the autoregressive process.
13. This is done by visualizing the autoregressive portion of the VAR as a
relationship simply between the current values of the exchange rate
and its lags. The estimated coefficients in this regression provide a
weighting structure from each successive lag. For example, suppose
is the matrix of coefficient estimates from the autoregression. The
weights are multiplied by the VAR coefficients to provide a moving
average. In period one, the weight is full. In period two, the weight is
distributed between the period one and period two coefficients. Since
the autoregression is simply an error process, this error process is now
weighted to distribute the amount of lagged period weights as a
moving average.
14. Mark (2001), chapter two, provides a great background on the tran-
sition from a VAR to an impulse response function. Greene (2003)
provides a statistical theory breakdown that is somewhat less intuitive
than Mark’s summary.
15. IASC 2004 discusses causal models at length. There are few, robust
causal models in economics, as most relationships are correlative rather
than causal due to endogenity problems that cannot be easily resolved.
This Handbook should be viewed as an empirical foundation of the
smart sanction literature to come. See http://www.humanitarianinfo.
org/sanctions/handbook/docs_handbook/iascsanchb.pdf for the handbook.
16. See Garfield (2000) for another source beyond the IASC study con-
cerned with measuring the sociopolitical and socioeconomic
impacts of sanctions. Much of the IASC study is drawn from
Garfield’s work.
17. Jorge Luis Borges, the famous Argentine poet and author, gave this
quote as his assessment of the Falkland Islands war and subsequent
embargoes of 1982 between the United Kingdom and Argentina.
I cannot take credit for that beautiful description.
18. The economic conditions are changed by the sanction shock. In IASC
2004, they recommend only one data set, “Proportion of households
with access to secure tenure” (55), which was not available in the
latest version of the World Development Indicators (2006) data.
19. See IASC 2004 (55) for more, from which these labels are drawn
directly. The data on the governance and education indicators, as
shown in table 6.3, were few and far between. The final indicator is
really the sum of the assessed changes in the first four indicators.
20. The World Development Indicators (2006) data did not list Taiwan
separately from mainland China or at all. Since these sanctions last
only one year, it seems unlikely there would be much change in social
conditions unless anomalous. In this way, short-term sanctions can be
220 NOTES

seen as “effective” in maintaining human conditions. The only case

that has all missing data is Taiwan, which has a final result of zero.
21. The logit and probit models of discrete choice analysis are predicated
upon the same basic idea: defeat the pitfalls of the linear probability
model predicting values outside the bounds of probability space. The
small differences are concerned with the assumption that the error
term is normally distributed (probit) versus logistically distributed
(logit). In each regression, the data have been adjusted to eliminate
heteroskedasticity, or potential non-constancy of the error term’s
variance, which leads to inconsistent estimators. See Studenmund
(2003), chapter nine, for an accessible explanation of heteroskedastic-
ity. See Greene (2003), chapter sixteen, for more advanced details or
Studenmund (2003), chapter thirteen, for an accessible approach for
the non-statistician on probit and logit models of ordered choice.
22. IASC 2004 makes some strange choices concerning defining causality.
See Studenmund (2003) for a simple causality explanation linked to
VAR analysis.
23. In ordered probit models such as this, the global statistic of choice is
the likelihood ratio, which is the family of statistics concerned with
“goodness-of-fit” in regression. The F-statistic and R2 summarize the
global properties of linear regression, analysis-of-variance (ANOVA)
results. When we leave linear models behind, as in the ordered probit,
classic ANOVA no longer is summarized by the F-stat and R2, and is
replaced by likelihood ratio (LR) statistic. For more on these issues,
see Kennedy (2004).
24. Others, such as Drury (2005), Martin (1992, 1993), and Drezner
(1999, 2003), use either the same dependent variable or an ordinal
(multinomial) measure instead. The major difference is a probability pre-
diction of a specific level of effectiveness among ordinal values, rather
than a binary prediction of either success or failure. Lam (1990) suggest
that HSE (1990) has selection bias within the success variable as well as
arbitrariness. However, the HSE (1990) data is still as good a data set as
this literature is likely to ever see concerning sanction episodes to 1990.
25. These results are free of heteroskedasticity. Specification tests provided
evidence to use the probit initially and the logit for the cumulative
model using the AIC.

Chapter 7 Conclusions and Policy

Recommendations
1. Fidel Castro has been out of the public eye in 2006 due to an illness.
Many speculated that his condition was worse than the Cuban
government was willing to divulge and has reportedly worsened. This
began a flurry of reports about sanction’s end with his death and
“new” government, likely run by Fidel’s brother, Raul.
 NOTES 221

Appendix 1 Brief Cases Histories of

Selected Sanction Episodes
1. Some cases stated here have not been written yet by HSE, though
their third edition is to be released in 2007. I have tried, using their
previous methodology and tendencies, to forecast what they would
conclude about each case they have yet to publish. Many thanks go to
them in readily supplying newly identified cases before they were pub-
lished at the Institute for International Economics (IIE) website in
the past: www.iie.com.
2. This is HSE (1990) case 50–1, a case that has been downplayed in the
literature for the most part, but is likely to get a lot more attention in
the years to come.
3. The United States was politically opposed to military rule in Brazil;
unfortunately, that is exactly what America received after Goulart was
overthrown.
4. The Hickenlooper Amendment effectively made sanctions mandatory
for any country that attempted to expropriate American assets.
5. See http://www.treas.gov/offices/enforcement/ofac/programs/cuba/cuba.
shtml for the latest statement of U.S. restrictions on trade, financial
and travel restrictions with Cuba. Recent events in Cuba may lead to
the end of sanctions, as Fidel’s health is reportedly waning. Until a
plebiscite takes place in Cuba, regardless of her leader, the sanctions
are likely to stay put.
6. In many sanction cases involving human rights abuses, the United
States and United Nations imposed sanctions due to pressure from
international lobby groups, such as Amnesty International. When
these groups brought reports of changes in human rights, the United
States and United Nations historically reacted by ending or reducing
sanctions.
7. The Haitian sanctions were UN devised and pursued, with America as
the key sender economy. This is a recurring theme in cases involving
the United Nations.
8. These numbers are comparable to the U.S. growth and inflation
numbers of the mid-1990s.
9. Meghan O’Sullivan (2003) assembled exhaustive case studies on Iran,
Iraq, Libya, and the Sudan. Her analyses are among the best to date.
10. HSE (1990) claim that the movements in the oil price at the end of
the 1970s helped Libya over the initial problems, and Libya’s econ-
omy is directly related to oil price movements. The United States
abstained from the 2004 UN vote on lifting these measures.
11. HSE (1990) cites a similar sanction case in Poland in 1981. Martial
law seemed to be the only reason for the sanction’s imposition.
12. Military problems and nuclear threats have continued between
Pakistan and India since the mid-1970s, and look nowhere near
resolution as of 2006.
222 NOTES

13. This is, however, an atypical case. HSE gave the sanctions an overall
score of four, which derived from a political result of four (Noriega’s
regime crumbling) and an economic success score of one (economi-
cally, the sanctions contributed very little). See HSE (1990, v. 2,
249–67) for details.
14. South Africa is the focus of many public choice studies on sanctions.
See Kaempfer and Lowenberg (1986) for the first explicit study link-
ing South African sanctions and interest groups within South Africa
and the United States.
15. This amount of aid was 1/1000 of GDP in military aid in 1975
according to HSE (1990).
16. For an expansive case study of the U.S. grain embargo, see Lundborg
(1987).
 Bibliography

Amuzegar, Jahangir. 1997. Adjusting to sanctions. Foreign Affairs 76: 31–41.

Askari, Hossein A. et al. 2003. Economic Sanctions: Explaining Their
Philosophy and Efficacy. Westport, CT: Praeger.
Balasubramanyam, V. N. 1984. The Economy of India. Boulder, CO: Westview
Press.
Baldwin, David. 1985. Economic Statecraft. Princeton, New Jersey: Princeton
University Press.
Becker, Gary S. 1983. A Theory of competition among pressure groups for
political influence. Quarterly Journal of Economics 98: 371–400.
Bergin, Paul. 2003. Putting the “new open economy macroeconomics” to a
test. Journal of International Economics 60(1): 3–34.
Bessler, Manuel, Richard Garfield, and Gerard McHugh. 2004. Sanctions
Assessment Handbook. New York: United Nations Inter-Agency Standing
Committee. Accessed at http://www.humanitarianinfo.org/sanctions/
handbook/docs_handbook/iascsanchb.pdf.
Betts, Caroline and Micheal B. Devereux. 2000. Exchange rate dynamics in a
model of pricing-to-market. Journal of International Economics 50(1):
215–44.
Bhagwati, Jagdish. 1998. Lectures in International Economics. Cambridge,
MA: MIT Press.
Black, P. A. and H. Cooper. 1987. On the welfare and employment effects of
economic sanctions. South African Journal of Economics 5: 297–99.
Blanchard, Olivier. 2002. Macroeconomics, 3rd Edition. New York: Prentice-
Hall.
Blanchard, Olivier and Stanley Fischer. 1989. Lectures on Macroeconomics.
Cambridge, MA: MIT Press.
Bolks, Sean M. and Dina Al-Sowayel. 2000. How long do sanctions last?
Examining the sanctioning process through duration. Political Research
Quarterly 53: 241–65.
Bondi, Loretta. 2002. Arms embargoes: In name only? Smart Sanctions:
Targeting Economic Statecraft, ed. David Cortright and George A. Lopez.
109–24. Lanham, MD: Rowman and Littlefield Publishers.
Bonetti, Shane. 1994. The persistence and frequency of economic sanctions.
The Economics Of International Security: Essays In Honour Of Jan
224 BIBLIOGRAPHY

Tinbergen, ed. Manas Chatterji, Henk Jager, and Annemarie Rima.

183–93. New York: St. Martin’s Press.
———. 1997a. A test of the public choice theory of economic sanctions.
Applied Economics Letters 4: 792–32.
———. 1997b. The analysis and interpretation of economic sanctions.
Journal of Economic Studies 24: 324–48.
———. 1998. Distinguishing characteristics of degrees of success and failure
in economic sanction episodes. Applied Economics 30: 805–13.
Bosworth, Barry P., Rudiger Dornbusch, and Raúl Labán, eds. 1994.
The Chilean Economy: Policy Lessons and Challenges. Washington, D.C.:
The Brookings Institution.
Brozska, Michael. 2002. Putting more teeth into UN arms embargoes.
Smart Sanctions: Targeting Economic Statecraft, ed. David Cortright and
George A. Lopez. 125–44. Lanham, MD: Rowman and Littlefield
Publishers.
Bruno, Michael and Jeffrey Sachs. 1985. Economics of Worldwide Stagflation.
Cambridge, MA: Harvard University Press.
Buchanan, James M. and Gordon Tullock. 1962. The Calculus of Consent.
Ann Arbor, MI: University of Michigan Press.
Burns, E. Bradford. 1993. A History of Brazil, 3rd Edition. New York:
Columbia University Press.
Cagan, Philip. 1956. The monetary dynamics of hyperinflation. Studies in the
Quantity Theory of Money, ed. Milton Friedman. 25–117. Chicago, IL:
University Of Chicago Press.
Carbaugh, Robert and Darwin Wassink. 1988. International economic sanctions
and economic theory. Rivista Internazionale di Scienze Economiche E
Commerciali 35: 217–25.
Carbaugh, Robert and John Olienyk. 1998. Do economic sanctions work?
The case of Iraq. Rivista Internazionale di Scienze Economiche E
Commerciali 45: 583–97.
Cavallo, Michele and Fabio Ghironi. 2002. Net foreign assets and the
exchange rate: redux revised. Journal of Monetary Economics 49:
1057–97.
Central Intelligence Agency. World Factbook, 1990–2006, Washington, D.C.
Chan, Steve and A. Cooper Drury, eds. 2000. Sanctions As Economic
Statecraft: Theory and Practice. International Political Economy Series.
New York: St. Martin’s Press.
Chubin, Shahram and Charles Tripp. 1988. Iran and Iraq at War. Boulder,
CO: Westview Press.
Close, David. 1988. Nicaragua: Politics, Economics, and Society. Marxist
Regimes Series, London: Pinter Publishers.
Collier, Simon and William F. Sater. 1996. A History of Chile: 1808–1994.
New York: Cambridge University Press.
Collins, Susan M. 2004. International financial integration and growth in
developing countries: Issues and implications for Africa. Journal of
African Economies 13, AERC Supplement 2: ii.55–94.
 BIBLIOGRAPHY 225

Congressional Record. 1996. The Iran–Libya Sanctions Act of 1996. June 18,
1996. Accessed at http://www.fas.org/irp/congress/1996_cr/ h960618b.
htm, October 1, 2006.
Conniff, Michael L. 1992. Panama and the United States: The Forced
Alliance. Atlanta, GA: University of Georgia Press.
Connor, John M. 2004. Global antitrust prosecutions of modern interna-
tional cartels. Journal of Industry, Competition and Trade 4: 239–67.
Corbo, Vittorio and Stanley Fischer. 1994. Lessons from the Chilean
stabilization and recovery. The Chilean Economy: Policy Lessons and
Challenges, ed. Barry P. Bosworth, Rudiger Dornbusch, and Raúl Labán.
29–68, 76–80. Washington, D.C.: Brookings Institution.
Corradi, Juan E. 1985. The Fitful Republic: Economy, Society and Politics in
Argentina. Latin American Perspectives Series, no. 2. Boulder, CO:
Westview Press.
Cortright, David and George A. Lopez, eds. 1995. Economic Sanctions:
Panacea or Peacebuilding in a Post-Cold War World? Boulder, CO:
Westview Press.
———, eds. 2000. The Sanctions Decade: Assessing UN Strategies in the 1990s.
Boulder, CO: Rienner Publishers.
———, eds. 2002. Smart Sanctions: Targeting EconomicStatecraft. New York:
Rowman and Littlefield.
Craven, Mathew. 2002. Humanitarianism and the quest for smarter sanctions.
European Journal of International Law 13: 43–61.
Crawford, Neta C. and Audie Klots, eds. 1999. How Sanctions Work: Lessons
from South Africa. International Political Economy Series. New York:
St. Martin’s Press.
Cronin, Richard P. 1994. The United States and Asia in 1993: Year of Asia
and the Pacific. Asian Survey 34: 98–109.
Dashti-Gibson, Jaleh, Patricia Davis, and Benjamin Radcliff. 1997. On the
determinants of the success of economic sanctions: An empirical analysis.
American Journal of Political Science 41(2): 608–18.
Davis, Brian L. 1990. Qaddafi, Terrorism, and the Origins of the U.S. Attack
on Libya. New York: Praeger Press.
Dehejia, Raheev H. and Bernard Wood. 1992. Economic sanctions and
econometric policy evaluation: A cautionary note. Journal of World Trade
26: 73–84.
Denktash, R. R. 1982. The Cyprus Triangle. London: George Allen and Unwin.
Di Tella, Guido and Rudiger Dornbusch, eds. 1988. The Political Economy of
Argentina. Pitt Latin American Series. Pittsburgh, PA: University of
Pittsburgh Press.
Dickey, David A. and Wayne A. Fuller. 1979. Distribution of the estimators
for autoregressive time series with a unit root. Journal of the American
Statistical Association 74: 427–31.
Dobson, Alan P. 2002. US Economic Statecraft for Survival, 1933–1991: Of
Sanctions. Embargoes And Economic Warfare. Routledge Advances in
International Relations and Global Politics. London: Routledge.
226 BIBLIOGRAPHY

Dornbusch, Rudiger. 1976. Expectations and exchange rate dynamics.

Journal of Political Economy 84: 1161–76.
Dornbusch, Rudiger and Sebastian Edwards. 1994. Exchange rate policy and
trade strategy. The Chilean Economy: Policy Lessons and Challenges, ed.
Barry P. Bosworth, Rudiger Dornbusch, and Raúl Labán. 81–104,
112–15. Washington, D.C.: Brookings Institution Press.
Dow Jones News Services. 1996a. Colombia/U.S. sanctions-2: Accord sub-
ject of fierce talks. Emerging Markets Report, January 9. Retrieved from
Factiva.
———. 1996b. “Clinton/Colombia—2: Cites lack of cooperation on Drugs.”
Dow Jones News, March 1. Retrieved from Factiva.
Doxey, Margaret. 1971. Economic Sanctions and International Enforcement.
London: Oxford University Press.
———. 1980. Economic Sanctions and International Enforcement.
2nd Edition. London: Oxford University Press.
———. 1987. International Sanctions in Contemporary Perspective. New
York: St. Martin’s Press.
Drezner, Daniel W. 1999. The Sanctions Paradox. New York: Cambridge
University Press.
———. 2003. The hidden hand of economic coercion. International
Organization 57: 643–59.
Drury, A. Cooper. 1998. Revisiting economic sanctions reconsidered.
Journal of Peace Research. 35: 497–509.
———. 2000. How and whom the U.S. president sanctions: A time-series
cross-section analysis of U.S. sanction decisions and characteristics.
Sanctions As Economic Statecraft: Theory and Practice, ed. Steve Chan and
A. Cooper Drury. 17–36. International Political Economy Series. New
York: St. Martin’s Press.
———. 2005. Economic Sanctions and Presidential Decisions. New York:
Palgrave/Macmillan.
Dunning, Thad. 2005. Resource dependence, economic performance, and
political stability. Journal of Conflict Resolution 49(4): 451–82.
Eaton, Jonathan and Maxim Engers. 1992. Sanctions. Journal of Political
Economy 100: 899–928.
———. 1999. Sanctions: Some simple analytics. American Economic Review
89: 409–14.
Economist, The. With friends like these: Venezuela’s foreign policy. September 2,
2006. Accessed from Academic Source Premier.
El Warfally, Mahmoud G. 1988. Imagery and Ideology in U.S. Policy Toward
Libya, 1969–1982. Pittsburgh, PA: University of Pittsburgh Press.
Elliott, Kimberly Ann. 2002. Analyzing the effects of targeted sanctions. Smart
Sanctions: Targeting Economic Statecraft, ed. David Cortright and George A.
Lopez. 171–82. New York: Rowman and Littlefield.
———. 2003. Economic leverage and the North Korean nuclear crisis.
International Economics Policy Briefs. No. PB03–3, April. Washington,
D.C., Institute for International Economics. Accessed from www.iie.org.
 BIBLIOGRAPHY 227

Engel, Charles and Kenneth D. West. 2005. Exchange rates and fundamentals.
Journal of Political Economy 113: 485–517.
Engel, Robert W. and C. W. J. Granger. 1987. Co-integration and error
correction: Representation, estimation, and testing. Econometrica 55:
251–76.
Evenett, Simon. 2002. The impact of economic sanctions on South Africa
exports. Scottish Journal of Political Economy 49: 557–73.
Eyler, Robert. 1998. Consumption Effects of Economic Sanctions. PhD
Dissertation. Davis, CA: University of California, Davis.
Fadopé, Cece M. 1997. Nigeria. Foreign Policy in Focus, International
Relations Center, 2: http://www.fpif.org/briefs/vol2/v2n12nig.html.
Accessed August 1, 2006.
Farouk-Sluglett, Marion and Peter Sluglett. 1990. Iraq Since 1958: From
Revolution to Dictatorship. London: I.B. Taurus and Co. Ltd.
Fass, Simon M. 1988. Political Economy in Haiti: The Drama of Survival.
London: Transaction Books.
Fender, John and Chong K. Yip. Tariffs and exchange rate dynamics redux.
Journal of International Money and Finance 19(5): 633–55.
Finch, M. H. J. 1982. The Political Economy of Uruguay Since 1870.
New York: St. Martin’s Press.
Fishelson, Gideon, ed. 1989. Economic Cooperation in the Middle East.
Boulder, CO: Westview Press.
Fisk, Daniel W. 2000. Economic sanctions: The Cuba sanctions revisited.
Sanctions As Economic Statecraft: Theory and Practice, ed. Steve Chan
and A. Cooper Drury. International Political Economy Series. 65–85.
New York: St. Martin’s Press.
Fitzgerald, E. V. K. 1979. The Political Economy of Peru, 1956–78. New York:
Cambridge University Press.
Fleming, Marcus J. 1962. Domestic financial policies under fixed exchange
and under floating exchange rates. IMF Staff Papers 9: 369–79.
Foreman-Peck, James. 1983. A History of the World Economy: International
Economic Relations Since 1850. Brighton, Sussex, U.K.: Wheatsheaf Books.
Frenkel, Jacob A. and Assaf Razin. 1997. Fiscal Policies and the World
Economy. 3rd Edition. Boston: MIT Press.
Gaisford, James D. and Sharad Sood. 1996. Are prohibitive trade sanctions
optimal? International Economic Journal 10: 1–20.
Galtung, Johan. 1967. On the effects of international economic sanctions.
World Politics 19: 378–416.
Garfield, Richard. The public health impact of sanctions: contrasting
responses of Iraq and Cuba. Middle East Report Summer 2000: 16–19.
Garoupa, Nuno R. and João E. Gata. 2002. A theory of international conflict
management and sanctioning. Public Choice 110: 41–65.
Goldstein, Morris and Philip Turner. 2004. Controlling Currency Mismatches in
Emerging Markets. Washington, D.C.: Institute for International Economics.
Greenberger, Robert. 1994. U.S. and China reach accord on M-11 missile.
Wall Street Journal—Eastern Edition 224: A18.
228 BIBLIOGRAPHY

Greene, William. 2003. Econometric Analysis. 5th Edition. New York:

Prentice Hall Publishers.
Grieco, Paul L. E. and Jeffrey J. Schott. 2006. Colombia’s readiness for a free
trade agreement. Trade Relations Between Colombia and the United States.
Washington, D.C.: Institute for International Economics Press, no. 79,
Policy Analyses in International Economics.
Haass, Richard. 1997. Sanctioning madness. Foreign Affairs 76: 74–85.
———. 1998. Economic Sanctions & American Diplomacy. New York: Council
on Foreign Affairs Press.
Hamilton, James. 1994. Time Series Analysis. Princeton, NJ: Princeton
University Press.
Harris, George S. 1985. Turkey: Coping with Crisis. Boulder, CO: Westview
Press.
Haugaard, Lisa. 1997. Nicaragua. Foreign Policy in Focus, International
Relations Center. 2: http://www.fpif.org/briefs/vol2/v2n32nic.html.
Hazlett, Thomas W. 1992. The Effect of U.S. Economic Sanctions on South
African Apartheid. Davis, CA.: Institute of Governmental Affairs Press,
University of California, Davis.
Holtfrerich, Carl-Ludwig. 1989. Interactions in the World Economy.
New York: New York University Press.
Howse, Robert. The road to Baghdad is paved with good intentions.
European Journal of International Law 13: 89–92.
Hufbauer, Gary C., and Barbara Oegg. 2000. Targeted sanctions: A
policy alternative? February 23: 1–8. Washington, D.C.: Institute for
International Economics. Accessed at http://www.iie.com/TESTMONY/
targsanc.htm.
———. 2001. Using sanctions to fight terrorism. Policy Brief 01–11.
Washington, D.C.: Institute for International Economics. Accessed at
www.iie.org.
———. 2002. Capital-market access: New frontier in the sanctions debate.
Policy Brief 02–6. Washington, D.C.: Institute for International
Economics. Accessed at www.iie.org.
———. 2003. Beyond the nation-state: Privatization of economic sanctions.
Middle East Policy 10: 126–134.
Hufbauer, Gary Clyde, Kimberly Ann Elliott, Tess Cyrus and Elizabeth
Winston. 1997. US Economic Sanctions: Their Impact on Trade, Jobs, and
Wages. Working Paper, April 1997, Peterson Institute for International
Economics., Accessed at http://www.iie.com/publications/wp/wp.cfm?
ResearchID149.
Hufbauer, Gary C., Schott, Jeffrey, and Elliott, Kimberly Ann. 1990. Economic
Sanctions Reconsidered, 2nd Edition, volumes 1 and 2. Washington, D.C.:
Institute for International Economics Press.
———. 2007. Economic Sanctions Reconsidered. 3rd Edition. Washington,
D.C.: Institute for International Economics Press.
Hurd, Ian. 2005. The strategic use of liberal internationalis: Libya and
the UN sanctions, 1992–2003. International Organization 59: 495–526.
 BIBLIOGRAPHY 229

Jack, D.T. 1940. Studies in Economic Warfare. Westminster, U.K.: P.S. King
and Son, Ltd.
Kachala, Francis F. C. 2002. Unfolding events leading to political pluralism in
Malawi: Drawing lessons from the archives. Conference Paper, XXVII
Biennial Eastern and Southern Africa Regional of Branch of the
International Council on Archives (ESARBICA) General Conference on
Archives, Society and Good Governance. Mozambique, Maputo, July
22–26, 2002.
Kaempfer, William and Anton Lowenberg. 1986. A model of the political
economy of International investment sanctions: the case of South Africa.
Kyklos 39: 377–96.
———. 1988. Determinants of the economic and political effects of trade
sanctions. South African Journal of Economics 56: 270–77.
———. 1992. International Economic Sanctions. Boulder, CO: Westview Press.
———. 1999. Unilateral versus multilateral international sanctions.
International Studies Quarterly 43: 37–58.
Kaempfer, William, Anton Lowenberg and William Mertens. 2004.
International economic sanctions against a dictator. Economics and Politics
16: 29–51.
Karp, Jonathan. 1999. U.S. proposes lifting sanctions on India—if New Delhi
signs treaty banning nuclear tests, loans will be unfrozen. Wall Street
Journal February 2, 1999, A19.
Karunatilake, H. N. S. 1971. Economic Development in Ceylon. Praeger
Special Studies in International Economics and Development. New York:
Praeger Press.
Kennedy, Peter. 2004. A Guide to Econometrics. 5th Edition. Cambridge,
MA: MIT Press.
Khan, Ali Haider. 1989. The Political Economy of Sanctions Against
Apartheid. London: Rienner Publishers.
Kim, K. and A. R. Pagan. 1995. The econometric analysis of calibrated macro-
economic models. Handbook of Applied Econometrics: Macroeconomics, ed.
M. Hashem Pesaran and M. R. Wickens. 356–90. London: Blackwell
Publishers.
Knorr, Klaus. 1975. The Power of Nations: The Political Economy of
International Relations. New York: Basic Books.
Kodikara, Shelton, U. 1982. Foreign Policy of Sri Lanka. Delhi: Chanakya
Publications.
Kofas, Jon V. 1996. Foreign Debt and Underdevelopment. New York: University
Press of America.
Kollmann, Robert. 2001. The exchange rate in a dynamic-optimizing busi-
ness cycle model with nominal rigidities: A quantitative investigation.
Journal of International Economics 55: 243–62.
Lam, San Ling. 1990. Economic sanctions and the success of foreign policy
goals. Japan and the World Economy 2(3): 239–48.
Lane, Philip R. 2001. The new open macroeconomy: a survey. Journal of
International Economics 54: 235–66.
230 BIBLIOGRAPHY

Levy, Philip I. 1999. Sanctions on South Africa: What did they do? American
Economic Review 89: 415–20.
Lewis, Paul. 1992. Security council votes to prohibit arms exports and flights
to Libya. April 1: A1. New York: The New York Times.
Lewis, Stephen R., Jr. 1990. The Economics of Apartheid. New York: Council
of Foreign Relations Book.
Leyton-Brown, David, ed. 1987. The Utility of Economic Sanctions. London:
Croom-Helm Publishers.
Lipton, Merle. 1988. Sanctions and South Africa: The Dynamics of Economic
Isolation. London: The Economist Intelligence Unit.
Losman, Donald L. 1979. International Economic Sanctions: The Cases of
Cuba, Israel, and Rhodesia. University of New Mexico Press: Albuquerque.
———. 1998. Economic sanctions: An emerging business menace. Business
Economics 33: 37–42. Accessed from Expanded Academic ASAP.
Lundahl, Mats. 1992. Apartheid in Theory and Practice: An Economic
Analysis. Boulder, CO: Westview Press.
Lundborg, Per. 1987. The Effect of Export Embargoes: The Case of U.S.–Soviet
Grain Suspension. London: Croom Helm Publishers.
Major, Solomon and Anthony J. McGann. 2005. Caught in the crossfire:
Innocent bystanders as optimal targets of economic sanctions. Journal of
Conflict Resolution 49: 337–59.
Malloy, James M. 1988. Revolution and Reaction: Bolivia, 1964–1985. New
Brunswick, NJ: Transaction Books.
Mark, Nelson C. 2001. International Macroeconomics and Finance. Oxford,
U.K.: Blackwell Publishers.
Martin, Curtis H. 2000. The U.S.–North Korean agreed framework: Incentives-
based diplomacy after the cold war. Sanctions As Economic Statecraft: Theory
and Practice, ed. Steve Chan and A. Cooper Drury. International Political
Economy Series. 86–109. New York: St. Martin’s Press.
Martin, Lisa L. 1992. Coercive Cooperation: Explaining Multilateral
Economic Sanctions. New York: Princeton University Press.
———. 1993. Credibility, costs and institutions: ccooperation on economic
sanctions. World Politics 45: 406–32.
McCain, Roger A. 2004. Game Theory: A Non-Technical Introduction to the
Analysis Of Strategy. Mason, Ohio: Thomson/South-Western Publishers.
Medlicott, W. N. 1952. The Economic Blockade. London: HMSO and
Longmans Green.
Miller, Norman and Rodger Yeager. 1994. Kenya: The Quest for Prosperity.
2nd Edition. Boulder, CO: Westview Press.
Miyagawa, Makio. 1992. Do Economic Sanctions Work? New York: Macmillan.
Moorsom, Richard. 1986. The Scope for Sanctions: Economic Measures Against
South Africa. London: Catholic Institute for International Relations Press.
Morgan, T. Clifton and Valerie L. Schwebach. 1997 Fools suffer gladly: The
use of economic sanctions in international crises. International Studies
Quarterly 41: 27–50.
 BIBLIOGRAPHY 231

Mundell, Robert A. 1963. Capital mobility and stabilization policy under

fixed and flexible exchange rates. Canadian Journal of Economics and
Political Science 29: 475–85.
Naylor, R. T. 1999. Economic Warfare: Sanctions, Embargo Busting and Their
Human Costs. Boston, MA: Northeastern University Press.
Ndlovu, Lindani B. 1994. The System of Protection and Industrial
Development in Zimbabwe. Brookfield, VT: Avebury Press.
Newcomb, R. Richard. 2002. Targeted financial sanctions: The U.S. model.
Smart sanctions: Targeting economic statecraft, ed. David Cortright and
George A. Lopez. 41–63. Lanham, MD: Rowman and Littlefield
Publishers.
New York Times. 1990. U.S. recalls its ambassador in Guatemala over slayings.
March 6: A10. New York: The New York Times.
Noman, Omar. 1990. Pakistan: Political and Economic History Since 1947.
London: Kegan Paul International.
O’Connell, Mary Ellen. 2002. Debating the law of sanctions. European
Journal of International Law 13: 63–79.
O’Sullivan, Meghan. 2003. Shrewd Sanctions. Washington, D.C.: Brookings
Institute Press.
Obstfeld, Maurice and Kenneth Rogoff. 1995. Exchange rate dynamics
redux. Journal of Political Economy 103: 624–60.
———. 1996. Foundations of International Macroeconomics. Cambridge,
MA: MIT Press.
Oette, Lutz. 2002. A decade of sanctions against Iraq: Never again! The end
of unlimited sanctions in the recent practice of the UN security council.
European Journal of International Law 13: 93–104.
Office of Foreign Asset Control. 2006. http://www.treas.gov/offices/
enforcement/ofac/.
———. 2006. Cuban sanctions. Accessed at http://www.treas.gov/offices/
enforcement/ofac/programs/cuba/cuba.shtml.
Orkin, Mark, ed. 1989. Sanctions Against Apartheid. London: Catholic
Institute for International Relations Press.
Osbourne, Martin J. 2004. An Introduction to Game Theory. New York:
Oxford Press.
Pape, Robert A. 1996. Bombing to Win: Air Power and Coercion in War.
Cornell University Press, Ithaca, NY.
———. 1997. Why economic sanctions do not work. International Security
22: 90–136.
———. 2005. Dying to Win: The Strategic Logic of Suicide Terrorism. Random
House, New York.
Perthes, Volker. 1995. The Political Economy of Syria Under Asad. London:
I.B.Tauris.
Pigou, A. C. 1941. The Political Economy of War. New York: Macmillan Press.
Ponnambalam, Satchi. 1981. Dependent Capitalism in Crisis: The Sri Lankan
Economy, 1948–1980. London: Zed Press.
232 BIBLIOGRAPHY

Porter, Richard. 1979. International trade and investment sanctions. Journal

of Conflict Resolution. 23: 579–612.
Pugel, Thomas A. 2003. International Economics. 12th Edition. New York:
McGraw-Hill/Irwin Higher Education.
Reid, Natalie, Sue E. Eckert, Jarat Chopra, and Thomas J. Biersteker. 2002
Targeting financial sanctions: Harmonizing national legislation and
regulatory practices. Smart Sanctions: Targeting Economic Statecraft, ed.
David Cortright and George A. Lopez. 65–86. Lanham, MD: Rowman
and Littlefield Publishers.
Remmer, Karen L. 1989. Military Rule in Latin America. London: Unwin
Hyman Press.
Rennack, Dianne E. 2003. North Korea: Economic Sanctions. Report for
Congress, Congressional Research Service, January 24. Washington D.C.:
the Library of Congress.
Renwick, Robin. 1981. Economic Sanctions. Center for International Affairs.
Cambridge, MA: Harvard University Press.
Rogers, Elizabeth S. 2002. Targeted financial sanctions: sanctions that do
work. Smart Sanctions: Targeting Economic Statecraft, ed. David Cortright
and George A. Lopez. 23–40. New York: Rowman and Littlefield.
Romer, David. 2005. Advanced Macroeconomics. 3rd Edition. New York:
McGraw-Hill/Irwin Publishers.
Rose, Andrew and Mark M. Speigel. 2004. A gravity model of sovereign lend-
ing: Trade, default, and credit. IMF Staff Paper, Special Issue 51: 50–63.
Rottenberg, Simon, ed. 1993. Costa Rica and Uruguay: The Political
Economy of Poverty, Equity, and Growth. A World Bank Comparative Study.
New York: Oxford University Press.
Rubinstein, Ariel. 1982. Perfect equilibrium in a bargaining model.
Econometrica 50: 91–110.
Russell, Philip L. 1984. El Salvador in Crisis. El Paso, TX: Colorado River
Press.
Saine, Abdoulaye. 2002. Post-coup politics in The Gambia. Journal of
Democracy 13: 167–72.
Salazar-Carrillo, Jorge and Roberto Fendt, eds. 1985. The Brazilian Economy
in the Eighties. Oxford, U.K.: Pergamon Press.
Sharp, Jeremy M. et al. 2006. Lebanon: The Israel–Hamas–Hezbollah Conflict.
Report for Congress, Congressional Research Service, September 15.
Washington D.C.: the Library of Congress.
Simons, Geoff. 1996. The Scourging of Iraq: Sanctions, Law and Natural
Justice. London: Macmillan Press.
———. 1999. Imposing Economic Sanctions: Legal Remedy or Genocidal Tool?
London: Pluto Press.
Sims, Christopher A. 1972. Money, income and causality. American Economic
Review 62: 540–52.
———. 1980. Macreconomics and reality. Econometrica 48: 1–48.
Smith, Adam. 1776. An Inquiry Into The Nature And Causes Of The Wealth
Of Nations. Reprint 1994. New York: Modern Library.
 BIBLIOGRAPHY 233

Smith, Hazel. 1992. Nicaragua: Self-Determination and Survival. London:

Pluto Press.
Sobel, Russell S. 1998. Exchange rate evidence on the effectiveness of United
Nations policy. Public Choice 95: 1–25.
Sonntag, Heinz R. 2001. Regression: Ecuador, Paraguay, Peru, Venezuela.
Democracy In Latin America: (Re) constructing Political Society, ed.
Manuel Antonio Garretón and Edward Newman. 126–57. New York:
United Nations University Press.
Spindler, Zane H. 1995. The public choice of superior sanctions. Public
Choice 85: 205–26.
Sponeck, H. G. Graf. 2002. Sanctions and humanitarian exemptions: A prac-
tioner’s commentary. European Journal of International Law 13: 81–87.
Steil, Benn and Robert E. Litan. 2006. Financial Statecraft. New Haven, CT:
Yale University Press.
Studenmund, A. 2003. Using Econometrics. 4th Edition. New York: Addison
Wesley.
Thorp, Rosemary. 1991. Economic Management and Economic Development
in Peru and Colombia. Pitt Latin American Series. Pittsburgh, PA:
University of Pittsburgh Press.
Toronto Star. 1992. Malawi loses West’s aid over rights. May 13. Reuter’s
News Services. Accessed from Lexis-Nexus.
Tullock, Gordon. 1987. Public choice. The New Palgrave: A Dictionary of
Economics 3: 1040–44.
U.S. International Trade Commission (USITC). 1998. Overview and
Analysis of Current US Unilateral Economic Sanctions. Washington, D.C.:
Government Publications Office, Pub. 3124.
United Nations. 2006. Chapter 7 of UN Charter. Accessed from United
Nations website: http://www.un.org/aboutun/charter/chapter7.htm.
United States Congress. 1979. Joint Committee Report on Iran Revolution.
November 19 Washington D.C.: Government Publishing Office.
United States Department of State. 1994a. Cameroon Human Rights
Practices. January 31. Accessed from U.S. Dept. of State website:
http://dosfan.lib.uic.edu/ERC/democracy/1993_hrp_report/93hrp_
report_africa/ Cameroon.html.
———. 1994b. Nicaragua Human Rights Practices, 1993. January 31.
Accessed from U.S. Dept. of State website: http://dosfan.lib.uic.edu/
ERC/democracy/1993_hrp_report/93hrp_report_ara/ Nicaragua.html.
van Bergeijk, Peter A. 1989. Success and failure of economic sanctions. Kyklos
42: 385–404.
Vanhoose, David D. 2004. The new open economy macroeconomics: A
critical appraisal. Open Economies Review 15: 193–215.
Varian, Hal R. 2002. Intermediate Microeconomics. 6th Edition. New York:
W.W. Norton.
Véganzonès, Marie-Ange and Carlos Winograd. 1997. Argentina in the
20th Century: An Account of Long-Awaited Growth. Paris: Organization of
Economic Cooperation and Development.
234 BIBLIOGRAPHY

Walker, Thomas W., ed. 1987. Reagan Versus the Sandinistas: The Undeclared
War on Nicaragua. Boulder, CO: Westview Press.
Wall Street Journal. 1994. U.S. and China reach accord on M-11 missile.
October 5: A18 New York: The Wall Street Journal.
Watson, Joel. 2001. Strategy: An introduction to game theory. New York:
W.W. Norton & Company.
Wesson, Robert G. 1981. The United States and Brazil: Limits of Influence.
New York: Praeger Press.
Wintrobe, Ronald. 1990. The tinpot and the totalitarian: An economic
theory of dictatorship. American Political Science Review 84: 849–72.
Yang, Jiawen et al. 2004. U.S. economic sanctions: an empirical study.
International Trade Journal 18: 23–62.
Yousef, Tarik. 2004. Development, growth and policy reform in the Middle
East and North Africa since 1950. Journal of Economic Perspectives 18:
91–115.

Data Sources
International Currency Analysis, Inc. 1984–89. World Currency Yearbook.
Brooklyn, NY.
International Monetary Fund. 1979–2006. International Financial
Statistics: Yearbook. New York: International Monetary Fund Publications.
Pick World Currency Report. 1955–77/79. Pick’s Currency Yearbook. New
York: Pick Publishing Corp.
Survey of Current Business. 1951–2007. Department of Commerce, Bureau
of Economic Analysis. Monthly publication. Accessed at http://www.
bea.gov/scb/index.htm for 1994 to current, accessed at http://fraser.
stlouisfed.org/publications/SCB/ for 1951–1993.
United Nations. 1951–1959. Statistical Yearbook. New York: UN
Publications.
———. 1960a–2004a. Statistical Yearbook. New York: UN Publications.
———. 1960b–2004b. International Trade Yearbook. New York: UN
Publications.
———. 2005. International Trade Yearbook. New York: UN Publications.
World Bank. 1976–2004. World Tables. New York: World Bank Publications.
———. 1978–2006. World Development Report. New York: World Bank
Publications.
———. 1978–2006. World Development Indicators. New York: World Bank
Publications. Internet resource at http://devdata.worldbank.org/
dataonline/.
 Index

1974 Trade Act: economic Drury, A. Cooper, 3, 5, 7, 26,

sanctions, 202, 204; Section 29–30, 32–3, 49, 57, 67,
301, 202 76, 125–7, 133, 145, 150,
Allende, Salvador, 27, 174–5 151–8, 166, 190; study of
Argentina, 171–2 HSE, 29
article 41, U.N. Charter, 22–3, 56
article 42, U.N. Charter, 22–3 Eaton, Jonathan, 6, 37, 47–8, 49,
52, 56–8
Baldwin, David, 5, 11, 23–4, 27, economic effectiveness: of sanctions,
31, 33, 52, 70 162, 167
blockade, 45 El Salvador, 18, 178–9, 190
Brazil, 172–3 Elliott, Kimberly Ann, 5, 11, 21, 24,
Burma, see Myanmar 26, 57, 78–9, 126, 129, 169,
Bush, G. H. W., 180 171
Bush, G. W., 187 Engers, Maxim, 6, 37, 47–9, 52,
56–8
Cameroon, 174 Ethiopia, 177, 179–80
Ceylon, see Sri Lanka
Chile, 174–5 Fleming, Marcus, 85
China, 2, 28, 42, 53, 66, 133,
160–1, 168, 171, 175–6, 210, Galtung, Johan, 10–11, 27, 33, 49,
219 52, 54; the “rally around the
Colombia, 176–7 flag effect”, 70; U.N. Charter
Cuba, 14, 23, 27, 32, 48, 54, 68, Articles 41 and 42, 23
160–2, 167, 177–8, 199; Gambia, The, 44, 201
public choice models, 66; travel game theory: coadjustment games,
sanctions, 16 53; coercion games, 53;
Cyprus, 51, 202–3 coincidence games, 53;
commitment games, 41;
Dashti-Gibson, J., 151 cooperative games, 41; mixed
Dornbusch, Rudiger, 85, 87, 117, strategy games, 44; repeated
175 games, 43; sanctions, 37;
Doxey, Margaret, 24–5, 33, 46, 52, subgame equilibrium, 48; the
54, 62, 74 prisoner’s dilemma, 41; two-
Drezner, Daniel, 6, 35, 46–7, 52, stage games, 43; extensive
57, 126–7, 156 form, 39; normal form, 38
236 INDEX

gravity models, 128–9, 158 Lowenberg, Anton, 19, 59–61, 65,

Guatemala, 130, 180–1 67, 69–70, 83

Haiti, 56, 163, 181–2 Malawi, 48–9, 189

Hickenlooper Amendment, 175, Martin, Lisa L., 6, 31–2, 49, 52–4,
179, 198 57–8, 126, 156
Hufbauer, Schott and Elliott Mundell, Robert A., 85
(HSE), 5, 29, 30, 33, 72, multilateral sanctions, 2, 4, 10–11,
126–34, 151–2, 156, 23, 29, 31–2, 40, 45, 49–50,
169–75, 178–9, 181, 183–4, 52–5, 77, 103, 150, 158, 164,
186, 190–1, 193–5, 198–9, 169, 187
201, 203–6; 1990 volumes, Myanmar (Burma), 189–90
26; success scores, 26; trade
and financial independence, Nash equilibrium, 41, 44, 48, 50,
27–8 52, 55–6; economic statecraft,
humanitarian Effects: of sanctions, 40; sanctions, 36, 38
139, 162, 167 new open economy
macroeconomics, 85–123,
India, 19, 145, 176, 182–3, 193, 130, 165; beggar thy
196, 198, 200, neighbor effects, 86;
Indonesia, 184 beggar thyself effects, 86;
interest groups, 2, 6, 31, 59–73, pricing to market (PTM)
75–6, 83, 153, 191 assumptions, 86
Iran, 14, 17, 32, 44, 50–1, 54, 60, Nicaragua, 18, 133, 135,
76–8, 176, 184–8, 200 190–2
Iran-Libya Sanctions Act of 1996 Nigeria, 78, 160–1, 192
(ILSA), 51, 77, 185, 188
Iraq, 14, 27, 32, 53, 58, 60, 66, 67, Obstfeld, Maurice, 85
71, 75, 80, 129, 150, 160, Office of Foreign Asset Control
186–7 (OFAC), 77
Israel: Hamas, 1; Hezbollah, 1, 46 ordered logit, 145, 148; estimating
humanitarian effects, 149;
Kaempfer, William, 19, 59–61, 65, estimating political effects,
67, 69–71, 83 155
knights, black, 37, 51, 54, 57, 76–8,
89, 146, 153 Pakistan, 176, 182–3, 193, 200
knights, white, 37 Panama, 193–4
Knorr, Klaus, 5, 11, 23–4, 27, 31, 33 Pape, R., 49, 154
Korea, North, 1–2, 27, 30–2, 47–8, Paraguay, 194–5
54–5, 66, 77, 81, 159–61, Peru, 18, 195–6
167–71, 197 political economy, 22; of sanctions,
3–7, 9
Lebanon, 1–2, 129, 200 public choice, 2, 6, 31–3, 37–8,
Libya, 14, 27, 32, 51, 60, 77, 160, 81–3; interest groups, 63–73;
185, 188–9, 199, 202 the sender’s political market, 63;
 INDEX 237

the target’s political market, Sobel, Russell S., 7, 129–31

63–5; theory basics, 58–60 South Africa: apartheid, 11, 19;
sanction case history, 196–7;
rally around the flag effect, 70, 82, sanctions against, 11, 129,
154, 162, 167 149, 162; smart sanctions,
Renwick, Robin, 24–5, 46 74; VAR analysis of sanctions,
Rhodesia, 23–5, 70, 196–7, 205 129
Rogoff, Kenneth, 85 South Korea, 171, 197–8
Soviet Union, 68, 77, 133, 177,
sanctions, aid, 19, 23, 176, 189 179, 190, 204–5
sanctions, arms, 2, 6, 43, 78; as Sri Lanka (Ceylon), 198–9
smart sanctions, 17 Sudan, The, 19, 199
sanctions, export: arms sanctions, Surplus, Consumer, 13–14
17; definition, 16–17; effects in Surplus, Producer, 13–14
NOEM models, 101; welfare Syria, 200
effects, 18
sanctions, financial: credit sanctions, Taiwan, 200–1
92; debit sanctions, 92; effects Terms of Trade (TOT): sanction
in NOEM models, 101–2; effects, 89, 99–100, 102,
freezing assets, 20 122–3, 158
sanctions, import: changing Thailand, 201–2
elasticities, 14–15; definition, Turkey, 18, 51, 202–3
12; effects in NOEM models,
101; welfare effects from, 14 Union of Soviet Socialistic Republics
Sanctions, Physical Capital, 19 (USSR), see Soviet Union
Sanctions, Smart, 6, 8, 16–17, 24, United Nations Security Council
31, 58–62, 71, 73–84, 126, Resolutions: SCR 1695, 2; SCR
139, 145, 150, 165; arms 1701, 1; SCR 1718, 2; SCR
sanctions as, 78; definition, 4, 748, 188; SCR 883, 188
60, 75; financial sanctions as, Uruguay, 203
79; legal sanctions, 81
Sanctions, Technology, 19 Vector Autoregression (VAR),
Sanctions, Travel: as import 129–30, 132
sanctions, 16
Sancton Effectiveness Continuum: Zimbabwe: sanction case history,
figure, 7 19, 205–6; as Southern
Selected cases: table, 170 Rhodesia, 205