Using Economic Sanctions to Prevent Deadly Conflict

Using Economic Sanctions to Prevent Deadly Conflict

Elizabeth S. Rogers

96-02                      May 1996

Prepared for the Carnegie Commission on Prevent Deadly Conflict.

Working Draft: Do not quote or cite without expressed, written permission from the author.

CITATION AND REPRODUCTION

This document appears as Discussion Paper 96-02 of the Center for Science and International Affairs. CSIA Discussion papers are works in progress. Comments are welcome and may be directed to the author in care of the Center.

This paper may be cited as: Elizabeth S. Rogers. "Using Economic Sanctions to Prevent Deadly Conflict." CSIA Discussion Paper 96-02, Kennedy School of Government, Harvard University, May 1996.

The views expressed in this paper are those of the authors and publication does not imply their endorsement by CSIA and Harvard University. This paper may be reproduced for personal and classroom use. Any other reproduction is not permitted without written permission of the Center for Science and International Affairs, Publications, 79 JFK Street, Cambridge, MA 02138, telephone (617) 495-3745 or telefax (617) 495-8963.

I. INTRODUCTION

What are the prospects for using economic sanctions to prevent deadly conflict? This paper assesses the academic writing related to this question. I also offer my own assessment of the efficacy of using sanctions for preventing deadly conflict based on my reading of relevant data.1

A tide of worldwide violence lends this question urgency. Civil and interstate violence around the globe has continued since the cold war's end, dashing hopes for a peaceful post-cold war world.. Since the fall of the Berlin wall in 1989 new wars have erupted in the former Yugoslavia, the Caucasus, Tajikistan, the Persian Gulf, Algeria, Rwanda, Burundi, Liberia, and Sierra Leone, and older conflicts continued or intensified in Somalia, Sudan, Afghanistan, Guatemala, Peru, Colombia, Sri Lanka, Burma, and elsewhere. In 1992, the first year of the post-Soviet era, there were twenty-nine wars, more than in any year since 1945. These wars killed some 460 thousand people in 1992 alone, and tamed millions into refugees. Some 18.2 million refugees were in UN care and some twenty-five million people were categorized as "internally displaced" in 1992, for a world total of 43.2 million displaced people— a new all-time high.2 Nearly all were displaced by war. From 1980 to 1992 the number of new refugees, nearly all fleeing from war, increased by a million per year .3

These wars threaten the interests of the industrialized states as well as the developing world. They also comprise a humanitarian disaster that the advanced industrial countries and international organizations have a moral responsibility to address. They impose the cost of providing for refugees on the UN, neighboring states, the U.S. and western Europe. They threaten Western commerce in the Third World. Finally, they pose the risk that their

violence could widen to engulf others, eventually ensnaring states far from the area of the initial conflict. Wars are like fire, sometimes spreading out of control in unpredictable ways. This gives all members of the international community an interest in preventing regional wars before they expand.

Increasingly, the use of force does not offer a practical remedy for these conflicts. U.S. support is important to the success of international military operations because the U.S. is the world's greatest power. However, since the Vietnam War, the U.S. public has grown increasingly unwilling to support overseas military deployments that might injure or kill its soldiers .4 In the future this public feeling will probably preclude most uses of U.S. forces to dampen regional conflicts. Hence alternatives to force must be explored, and economic sanctions are the most plausible alternative because next to force they are the strongest available policy instrument.

The historical record contains more than 125 uses of economic sanctions by states and international organizations since 1914. Most of the literature analyzing these economic sanctions focuses on the general question of sanctions' effectiveness. There is no writing that specifically evaluates economic sanctions as a tool for preventing deadly conflict. Hence, I survey the general literature on economic sanctions and assess its conclusions. I then evaluate the extent to which these general conclusions can be applied to the specific case of using economic sanctions to prevent conflict in the post-cold war era. I also identify gaps in the literature where further research is needed.

In this paper, I make one general argument about the effectiveness of sanctions for preventing deadly conflict and two specific arguments about using financial sanctions (freezing assets and multilateral political aid conditionality) for that purpose. The general argument is that economic sanctions will enjoy some success at preventing deadly conflict and should be used for that purpose. This argument rests on conclusions drawn from the sanctions literature, an assessment of the post-cold war environment, an evaluation of the costs associated with imposing economic sanctions, and the nature of the task. I also assess two factors that work against successfully using sanctions to prevent deadly conflict: the problem of identifying and isolating targets of sanctions and the difficult nature of the conflict prevention task.

The first specific argument is that freezing assets is an under-used and under-studied sanction. Freezing assets has the advantage of being a "precision guided" sanction that can be crafted to cause the target state's political leaders and elites to suffer economic pain while not similarly injuring innocent civilians. This distinguishes asset freezes from other more common types of economic sanctions such as trade embargoes and aid cut-offs.

Second, I argue that political aid conditionality by international financial institutions (IFIs), such as the IMF and the World Bank, is a promising tool whose potential is only just starting to be realized. Political aid conditionality means that IFIs would only provide economic assistance to states that meet certain political criteria. The conditions could include respect for human rights, democracy, military spending levels, etc.5 Many countries depend on assistance from IFIs. Thus, those states have a clear incentive to obey the conditions set by the IFIs including conditions that may help to prevent deadly conflict.6

The paper also offers recommendations for policy makers about how to implement economic sanctions in order to maximize their effectiveness. Important steps include gaining widespread international cooperation for the sanctions effort, imposing sanctions quickly, using the full range of available economic tools from the outset, setting clear conditions for the repeal of the sanctions, and clearly communicating to the target that the sanctioners have the political will to keep the sanctions in place until the conditions are met.7

The next section of the paper examines the sanctions record and literature to determine what is known about the effectiveness of economic sanctions generally. Section three analyzes the prospects for using sanctions for preventing deadly conflict, concluding that conditions are reasonably auspicious. This section also includes a brief assessment of financial sanctions— freezing assets and political aid conditionality— since these are especially promising policy instruments, and are also under-utilized. Section four presents conclusions.

II. THE SANCTIONS RECORD AND LITERATURE

The historical record shows that economic sanctions imposed before and during the cold war failed more often than they succeeded, and the literature on economic sanctions concludes that sanctions are a weak policy instrument. However, both the record and the literature underestimate the likely effectiveness of post-cold war economic sanctions.

Thus, the U.S. is the best choice to lead sanctions efforts. Other candidates for leadership are the other members of the Group of Seven (G-7) and the permanent members of the UN Security Council.

Some might argue that the United Nations plays a major role in leading sanctions efforts, and that UN sanctions therefore deserve study separate and apart from sanctions efforts led by the great powers. However, this view rests on an overestimate of the UN's autonomy from the great powers (and hence from the United States.) In fact when the UN acts, it always acts as an arm of the great powers. Concurring with regard to economic sanctions are Hufbauer, Schott, and Elliott, who argue that "instances of the collective use of sanctions-the League of Nations against Italy in 1935-36, the United Nations against Rhodesia from 1965 to 1979, the allies against Germany and Japan in World War II, the United Nations against Iraq in 1990-are in fact usually episodes of major powers enlisting their smaller allies." Economic Sanctions Reconsidered, 1:10-11. In short, the concept of "UN-led sanctions" is a myth. Instead, we have great power-led sanctions implemented by the UN. For an analysis of the problems and possibilities associated with UN sanctions see John Stremlau, "Sharpening International Sanctions For Collective Security and Preventing Deadly Conflict," January 1996, Draft Report to the Carnegie Commission on Preventing Deadly Conflict.

Moreover, the cost of imposing economic sanctions is quite low but often overestimated.8 Overall the conditions for using sanctions now seem auspicious, and this augurs well for using sanctions to prevent deadly conflicts.

A.  Economic Sanctions, 1914-1989.

The Record. The record reveals, 115 instances from 1914 to 1989 in which economic sanctions were used or threatened by the United States, other states, and international institutions.9 The vast majority of the 115 efforts were partial measures that omitted important forms of punishment. For example, freezing assets, a strong sanctions measure, was very rare. Often sanctions were confined to cuts in bilateral foreign aid, which seldom inflict much harm. Many sanctions were imposed unilaterally, especially by the United States during the 1970s and 1980s. Many of these unilateral efforts were undertaken largely for cosmetic or symbolic reasons. Given their unilateral and partial nature, there was little reason or expectation that they would succeed in causing the target state to change its behavior.

Many sanctions efforts were undercut by states acting as "black knights," or spoilers, who helped replace target states' lost trade and aid with their own.10 This was especially common during the cold war: the Soviet Union and United States regularly countered each other's sanctions efforts. For example, the Soviet Union undercut U.S. sanctions against Cuba (1960) by providing offsetting foreign aid and shipping key commodities such as oil to Cuba.11 Similarly, the United States undercut Soviet sanctions against Yugoslavia (1948) by providing offsetting aid.12

In none of the cases from 1914 to 1989 were sanctions used primarily to prevent conflict; rather they were used largely to contain communism, punish support for terrorism, prevent nuclear proliferation, support human rights, and punish the nationalization of foreign owned property. In several cases the success of sanctions would have ended a conflict. Examples include the 1980 U.S. grain embargo imposed on the Soviet Union to punish its invasion of Afghanistan, and the 1950 U.S. sanctions imposed on North Korea to compel its withdrawal from South Korea. However, there are no cases in which sanctions were employed expressly to prevent a civil or international war. Hence we must deduce the utility of sanctions for these purposes from our experience of their use for other purposes.

The Literature. The literature on economic sanctions is dominated by the view that sanctions are ineffective.13 In Economic Sanctions Reconsidered Hufbauer, Schott, and Elliott conclude that 34 percent of the sanctions undertaken between 1914 and 1990 were effective. They further conclude that the success rate is declining. While 44 percent of the sanctions efforts undertaken before 1973 were classified as successes, only 24 percent of post-1973 sanctions efforts (roughly half of the total) were successful. 14

The overall 34 percent success rate is often quoted to show that sanctions are ineffective. However, this success rate is only low if measured against a severe standard of expected performance. In fact, since sanctions cost the United States (their biggest user) relatively little to impose, involve little risk, and have not been proven less successful than other foreign policy instruments such as the use of force or covert action, 15 a 34 percent success rate is respectable.

This interpretive question aside, the pessimistic sanctions literature suffers from three biases that cause it to underestimate sanctions effectiveness. First, it defines a successful sanction' too narrowly, so that some successful efforts are coded as failure.16 Sanctions can have three major foreign policy purposes, all of which are relevant to preventing deadly conflict. First, sanctions can be imposed to persuade the target to change its behavior. Second, sanctions can be imposed to weaken the target, thereby leaving it less able to make trouble. Third, sanctions can be imposed to punish the target, thereby deterring it and others from making trouble in the future.17 The sanctions' literature judges sanctions' success according to only the first of these purposes, target compliance. It codes as failures cases where sanctions successfully weakened or punished the target, and since many sanctions are undertaken largely for these purposes, it underestimates the power of sanctions to achieve results. 18

Second, the literature focuses on the ability of imposed sanctions to compel the target to change its behavior, but does not systematically study whether the threat of sanctions can deter the target from taking a certain action. Hence the literature asks if sanctions can achieve compellence (which is difficult) without asking if they can achieve deterrence (which is easier). Deterring an untaken action is easier than compelling a policy reversal because leaders do not face the higher political costs that accompany reversing course.19

Sanctions can deter in two ways. Target states can be threatened with sanctions directly,20 or sanctions can deter indirectly, by example. States seeing the economic damage sustained by sanctioned states may decide to avoid actions that will make them the targets ofsimilar sanctions. Press accounts imply that this logic was part ofthe rationale for sanctioning Haiti in 1991 and Niger in 1996 after military coups in those states.21

Third, the literature generalizes largely from cases when partial sanctions were imposed half-heartedly, so it underestimates the possible effectiveness of forceful sanctions. As noted above, and contrary to common impression, most past sanctions efforts have been partial. In fact total economic sanctions have rarely been imposed.22 Instances of partial sanctions include the many unilateral sanctions against countries that violated human rights, supported terrorist activities, and worked to acquire nuclear weapons imposed by the United States starting in the 1970s.23 These three categories account for 74 percent of the forty-six U.S. economic sanctions imposed between 1973 and 1990 and largely account for the poor 17 percent sanctions success rate that Hufbauer, Schott and Elliott report for the U.S. during this period. The failure of partial sanctions is a poor predictor of the potential of forceful sanctions.

The sanctions literature is biased in one way that may lead it to overstate the likely future effectiveness of sanctions for preventing deadly conflict. The target's resistance is probably smaller in some of Hufbauer, Schott, and Elliott's cases than it would be when sanctions are imposed to prevent conflicts. This task require that states be dissuaded from adopting war policies. War policies are seldom made for small reasons and are not easily overturned. Hence conflict prevention sanctions may often be imposed on targets that are more willful than "typical" sanctioned states, and so may fail more often. However, it seems unlikely that this bias is large enough to offset the other biases noted above.

The Costs of Imposing Sanctions. Some skeptics argue that sanctions are a poor foreign policy instrument because their use imposes exorbitant costs on sender states. In this view, the sanctions game is not worth the candle for senders: the price of success is excessive even for sanctions that eventually succeed .24 Other skeptics argue that the high cost of sanctions to senders makes them less likely to succeed because the expense weakens the sender's resolve. Seeing this, targets are emboldened to hold out, further weakening the senders' will and eventually leading to the collapse of the sanctions effort. Finally, some critics argue that the moral cost of sanctions outweighs their value.

There are no systematic studies analyzing the costs of different types of sanctions to different states over time. However, there is evidence suggesting that sanctions seldom impose large economic costs on sender states. The case that costs are high has four shortcomings. First, even figures used by sanctions skeptics do not indicate that absolute economic costs to sender states are high. The National Foreign Trade Council (NFTC), an anti-sanctions organization, has estimated that in 1987, U.S. economic sanctions cost the United States $7 billion in lost exports to the target states.25 Even if this is true, $7 billion is only a tiny fraction of the U.S. 1987 gross domestic product (GDP) and total U.S. exports.26 Moreover, the actual losses from these sanctions were certainly lower because some portion of the exports not sent to sanctioned states were almost certainly sent to other markets. Second, many economic sanctions are economically cost-free to the senders.

Reducing or terminating bilateral foreign aid payments costs little or nothing and usually provides a net economic and domestic political gain. Similarly, financial sanctions, such as freezing assets and slowing or halting assistance by IFIs impose no significant costs on senders. Even trade sanctions can sometimes help domestic businesses by providing protection from foreign competitors. Hufbauer, Schott, and Elliott concluded that 85 percent of U.S. and U.S.-led economic sanctions produced either a net gain to the sender states or had little effect on them.27 Third, when sanctions are imposed with broad international cooperation, they avoid the cost that businesses are most averse to paying: relative loss of market share to foreign competitors. With export restrictions, the pain felt by businesses and the level of their complaints varies depending on whether the costs are relative or absolute.28 When a sanctions' effort has widespread support, such as the 1990 sanctions against Iraq, the costs are absolute. Under these conditions firms lose business but they do not lose market share to their foreign competitors because they are also participating in the embargo. When a country imposes unilateral sanctions, or when only a small number of other states participate, the costs are relative. Under these conditions businesses are likely to lose market share to foreign competitors.29 However, since broad cooperation on sanctions efforts is more likely in the post-cold war era, this problem will likely arise less often than in the past.

Fourth, the costs to senders of imposing sanctions are far smaller than the cost of threatening or using force, a prime policy alternative. For example, while sanctions cost the United States $7 billion in lost exports, the defense budget— a crude measure of the cost of preparing to use force and of using force— cost a far larger $283.5 billion in 1987.30 Moreover, sanctions have the obvious advantage of not risking the lives of U.S. soldiers.

A full comparison of the costs of economic sanctions and force requires an assessment of the cost of preparing to use each instrument and the actual cost of their use. Economic sanctions are usually far cheaper at both stages. Preparation to impose economic sanctions costs very little, requiring at most some stockpiling of goods. Preparation to use force includes the cost of recruiting, training, maintaining, and arming the military. Like sanctions, the cost of using force varies from event to event. For example, preparation for the Gulf War cost the United States $68 billion and the war itself another $52 billion. Because of unique circumstances, the entire $52 billion cost of the war itself was paid by allies.31 However, the United States paid its own preparation cost. Hence, even this relatively cheap war cost the United States nearly ten times the $7 billion total cost of all the economic sanctions in place during the year 1987.

Finally, some critics condemn sanctions because they injure innocent civilians in the target country, and thus violate a moral proscription on injuring political innocents.32 Two main points are relevant to this criticism. First, while the injury to innocents is a drawback to sanctions use, the cost should be compared to the benefits that sanctions provide. The value of avoiding harm to innocents is not absolute, and should give way if the benefits of sanctions are greater .33 Second, the likely alternative policy instrument— force— usually also risks harm to innocent civilians.

B. Sanctions Efforts Since the End of the Cold War

The previous section established the limitations of directly extrapolating from the past sanctions record and literature to predict the effectiveness of economic sanctions in the post­-cold war era. This section explores a more revealing indicator of future sanctions' success: the effectiveness of U.S.-led economic sanctions since 1989. Post-1989 cases are more comparable to possible future sanctions efforts than pre-1990 cases because they occurred in an international environment free of Soviet-American competition. Arguably, the end of the cold war brightens the prospects for success, since the perennial cold war black knight problem has disappeared. On the other hand, without the alarming specter of an enemy superpower, it may be more difficult to build and maintain coalitions that can effectively impose sanctions.34

Which view is right? The post-1989 record is brief but it does shed light on the answer, and supports the optimistic view.  In the eighteen sanctions efforts the United States has initiated since 1990 (see Table I), no country has tried to frustrate international sanctions efforts; nor have any new problems disarrayed sanction coalitions.  Forcefully applied U.S.-led sanctions have succeeded at a high rate.

Three of the post-cold war sanctions efforts – Iraq (1990-), Haiti(1991-1994), and Yugoslavia (1992-1995) – are examined below.  These cases were selected for two reasons.  First, these cases represent forceful rather than half-hearted efforts to bring sanctions to success.  The breadth of the tactics reveals the senders’ seriousness of purpose in these three cases.  In each case a broad international coalition was organized in support for the sanctions effort, used all available economic sanctions (trade cutoffs, aid cutoffs, and financial sanctions), and used sanctions in conjunction with other policy instruments (for example, force or threat of force).  Hence these are the cases that show what can be accomplished when the sender states are serious.

Second, I focus on Iraq, Haiti, and Yugoslavia because the goals included dampening conflict.  One involved conflict resolution and the other two, conflict prevention.  Yugoslavia is an example of sanctions used to resolve a deadly conflict. 35 In Iraq, Saddam Hussein has previously attached two neighboring states –Iran and Kuwait.  Through scud missile attacks on Israel he also demonstrated his willingness to widen conflicts.  In addition, Iraqi weapons of mass destruction would likely be targeted at Israel.  Thus, while the sanctions against Iraq are primarily an effort at conflict prevention, they may also have long term conflict containment goals. Haiti is the most difficult case to classify because conflict dampening was not an immediate priority of the sanctions. However, since civil war was a long term danger in Haiti had the military junta remained in power, there was a conflict prevention aspect to these sanctions. Hence, these three efforts shed light on the ability of forceful sanctions to control conflict.

Absence of New Spoilers. No spoiler state (or black knight) emerged in these three cases, or in any of the other fifteen cases. In both the Yugoslav and Iraq cases, one might have expected Moscow to play this role. Russia and Serbia have historic ties based on culture and ethnicity. However, while Russia was more supportive of the Yugoslav position on Bosnia-Herzegovina than were Western states, it neither broke the sanctions nor took other counter-measures as the Soviet Union did to support Cuba during the cold war. The Soviet Union was also intermittently friendly with Iraq during the cold war, but Moscow supported the 1990 UN sanctions on Iraq and has not undercut them.

Since the end of the cold war, China has come the closest to becoming a spoiler but has not crossed the line. China has sold M-11 missiles, or at least the technology to make them, to Pakistan; helped Algeria build a nuclear power plant; and assisted Iran's nuclear and chemical weapons programs.36 China also blocked an agreement to impose UN economic sanctions on North Korea to compel it to halt its nuclear program.37 However, only in the Pakistani case is there a question of violating a U.S. economic sanctions effort. China has abided by all multilateral efforts. Thus, no powerful black knight has emerged since 1990. If this situation continues it should markedly increase the effectiveness of sanctions.38

Continued Coalition Cohesion. The emerging post-cold war evidence does not support the proposition that international coalitions will be harder to form and maintain in the absence of a superpower enemy to rally against. When the United States has been serious about gaining and maintaining cooperation for economic sanctions efforts to control conflict— in the cases of Iraq, Haiti, and Yugoslavia— it has succeeded. The UN sanctions against Iraq achieved an unprecedented level of international cooperation.39 The United States, the Organization of American States (OAS), and the UN cooperated effectively in sanctioning Haiti once the United States became serious about sanctions. In the case of Yugoslavia the United States and its NATO allies have disagreed about the use of NATO forces, whether to maintain the arms embargo, and about peace plans, but not about economic sanctions policy.

While these cooperation successes are encouraging; other evidence suggests that cooperation will not be automatic or easy in the post-cold war world. The United States was unable to build a coalition in support of two of its sanctions efforts, those against Cuba and Iran. Also, efforts in 1995-96 to impose tough sanctions against Nigeria failed due to a lack of cooperation. The Cuban case is an anomaly left over from the cold war. However, the Iran and Nigeria cases are more problematic, suggesting the limits of post-cold war cooperation. These cases indicate that it will be difficult to build coalitions when the goal is not directly connected to conflict control and the proposed sanctions carry high costs for powerful states.40

The post-cold war conditions for sanctions seem auspicious, but what about the results? Contrary to common opinion, international sanctioning coalitions have achieved quite good results in Iraq, Haiti, and Yugoslavia.

Iraq.  Iraq's invasion of Kuwait in August 1990 triggered an immediate response from the United States and the UN in the form of comprehensive economic sanctions which remain in place. After Iraq was forcefully expelled from Kuwait in the Gulf War, the primary goal of the sanctions shifted to dismantling Iraq's weapons of mass destruction and creating a monitoring system to ensure that these weapons programs are not restarted.

The economic sanctions have done massive damage to the Iraqi economy. Iraq's GNP fell by more than 50 percent after sanctions were imposed. The value of the Iraqi dinar has plummeted. Before the Gulf War, one dinar bought three U.S. dollars. In January 1996, a dollar was worth 3,000 dinar.41 A large portion of Iraq's commercial aviation and marine fleets which were outside Iraq in August 1990, remain outside.42 Iraqi President Saddam Hussein has been forced to exhaust his large cash reserves secretly held in foreign banks .43

This economic damage has had several positive results. Economic pain has slowed the rebuilding of the Iraqi military by making it difficult to purchase spare parts and replace aging equipment.44 It has helped compel Iraq to grudgingly comply with U.N. efforts to prevent it from acquiring weapons of mass destruction.45 It contributed to Iraq's decision to recognize Kuwait. Increasing economic misery was also a factor in Saddam's decision to negotiate with the UN regarding limited oil sales for the purpose of purchasing humanitarian assistance.46 Finally, the vast economic harm that sanctions have inflicted on Iraq warns other potential aggressor-states that the United States can assemble and maintain a sanctions coalition capable of gravely injuring an aggressor's economy.

Critics make two main observations to support the claim that sanctions have failed in Iraq. First, Saddam remains in power. Second, sanctions failed to induce Iraq to withdraw from Kuwait during the 1990-91 crisis. The first observation measures sanctions against an unduly high standard of performance. Firmly entrenched leaders, like Saddam in 1990, are very hard to overthrow. This is perhaps the hardest task we could demand of sanctions. While economic sanctions did not overthrow Saddam, neither did military force or covert action. The failure of sanctions to overthrow Saddam means that they were only partially successful in Iraq, but this should not obscure the success they have achieved.47 

The second observation, that sanctions failed to remove Iraq from Kuwait, is a weak, indictment because the sanctions were not left in place long enough to accomplish the task. Sanctions do not work overnight. During the pre-war Gulf crisis, sanctions advocates forecast that sanctions would take at least a year to force an Iraqi policy change,48 but U.S. President George Bush decided to move ahead with the military option six months into the crisis.

Haiti. A September 1991 military coup against the elected government of Haitian President Jean-Bertrand Aristide triggered economic sanctions against Haiti. The sanctions effort was led by the United States and the Organization of American States (OAS) and later included the UN. The primary goals of the sanctions were to punish the Haitian military junta and to restore Aristide to power. The sanctions were lifted in October 1994 when Aristide resumed the presidency of Haiti.

The United States achieved the outcome it wanted: President Aristide was restored to power. Did economic sanctions contribute to this result? Some argue that the sanctions, which were imposed for three years, were ineffective, and that the military junta stepped aside only under the imminent threat of U.S. military invasion.

In fact, the sanctions were initially ineffective because they were imposed in a tentative, half-hearted manner. The sanctions imposed in 1991 were partial and were not targeted at the Haitian elites until January 1994.49 This was a serious omission since the coup that ousted Aristide was sponsored by the Haitian elites. Thus, sanctions would have to be targeted against these elites in order to bring about a policy change. Moreover, the early sanctions contained loopholes that allowed U.S. businesses to continue operating in Haiti, were prematurely loosened during the crisis before Haiti fully complied with U.S. and UN demands, and were accompanied by weak official American rhetoric and signs of U.S. irresolution.50 Hence, it is not surprising that the sanctions were initially ineffective.51

In the end, however, the sanctions did make a positive contribution to the outcome in Haiti, largely because they were finally toughened and targeted at Haiti's economic and military elites. No doubt the junta's September 1994 decision to step aside was triggered by the U.S. forces airborne for Haiti, but economic sanctions helped set the stage. Once the sanctions were toughened to include a total trade embargo and asset freezing, they crippled the Haitian economy and caused economic suffering for the elites.52 This helped convince the elite to accept Aristide's return leaving the military rulers without their base of support, and in no mood to resist the U.S. invasion force.53 Hence, they were more willing to leave power peacefully when the United States threatened force.

Thus, despite being poorly implemented at first, the economic sanctions against Haiti eventually injured the Haitian elites and thereby eased Aristide's restoration to power. The damage to the Haitian economy also may deter other military leaders from considering anti­democratic coups.

Yugoslavia. The United States and the UN imposed aid, trade, and financial sanctions on Yugoslavia (Serbia and Montenegro) during May-June 1992 to persuade it to reign in its Bosnian Serb clients and thus bring an end to the war that erupted in Bosnia Herzegovina in April, 1992.54 These sanctions were lifted in late 1995 following the Dayton Accords that halted the war in Bosnia.

The sanctions devastated the Yugoslav economy and helped persuade Yugoslav President Slobodan Milosevic to end his war for a greater Serbia. In late 1995, Milosevic accepted a U.S.-brokered peace settlement reached at Dayton, Ohio and pressed the Bosnian Serbs to accept it as well. Without the sanctions it seems very unlikely that the Serb side would have considered the terms of the Dayton accord, let alone accepted them.55

It is always difficult to end a war using economic instruments, but in this case it was especially difficult. First, the sanctions were indirect. They targeted Yugoslavia, which was not one of the warring parties in Bosnia. Second, for three years the United States sought an immoderate goal, consistently rejecting a partition settlement and instead insisting that Bosnia must be maintained as a unitary state.56 This demand required the Bosnian Serbs to surrender sovereignty of all of their territory and abandon their goal of national independence— things they predictably refused to do. Not until late in the summer of 1995 did Clinton administration officials even hint that a partition solution might be acceptable.57 A combination of economic sanctions and diplomacy might have succeeded in achieving peace far earlier if the U.S. goal had been more modest.58

In short, the sanctions had a substantial impact on policy making in Yugoslavia. Their failure to bring peace to Bosnia sooner resulted partly from the U.S. pursuit of grandiose aims and partly from the difficulty of the task.

These three cases show that when an international coalition has had the will to impose strong sanctions it has achieved positive results. At a minimum, the economic damage done by these sanctions warns the world that the United States can assemble a sanctions' coalition and impose high economic costs on miscreant states and regimes. Beyond this, the sanctions did compel the target states to make clear positive changes in their behavior. Without sanctions it seems likely that Saddam Hussein would have withheld his concessions on weapons of mass destruction and further built up his military, the Haitian elites would not have accepted Aristide's return, and Yugoslav President Milosevic would not have pressured the Bosnian Serbs to make concessions for peace.

In sum, the post-cold war experience suggests that the future prospects for the success of serious multilateral sanctions are bright. Sanctions are not always effective, but they often produce substantial compliance if they are forcefully applied. In each of the three cases discussed above, the sanctions achieved marked positive results that are traceable to tough, total economic sanctions.

III. SANCTIONS AND PREVENTING DEADLY CONFLICT

This section first analyzes the prospects for using economic sanctions to prevent internal and international conflict, then presents recommendations for applying sanctions.

A. Preventing Internal Conflict

Outside powers could use economic sanctions to' prevent an internal conflict if they could foresee an impending civil war and could pressure one or both of the colliding parties to adopt more peaceful policies. For example, governments could be pressured to adopt reforms that would defuse impending rebellions. Such reforms might include implementing democratic changes, granting autonomy to or sharing power with national minorities, land reform, redistribution of wealth, or refraining from disseminating hate propaganda. The sanctions that pressured South Africa to end apartheid could be considered a successful conflict prevention effort.

Sanctions efforts of this sort are worth attempting because it is far easier to prevent a conflict than to resolve it. If the parties are not yet politically mobilized for war, they will be more tractable.

There are, however, three serious impediments to using economic sanctions for conflict prevention. First, violent internal conflict is hard to predict. It is seldom clear where conflict-prevention measures are needed. Second, the causes and preventives of civil wars can be difficult to distinguish and if they are confused, attempts at preventing can make things worse. For example, Rwanda in the 1990s and Nicaragua in the 1970s are examples of outside pressure contributing to the outbreak of war.59 The successful use of sanctions to prevent conflicts requires good intelligence and a good understanding of the roots of internal conflict in order to avoid costly mistakes. Pressing regimes to respect human rights even at the risk of triggering violence can, of course, sometimes be the best choice. In the Nicaraguan case, the Somoza regime brutally violated human rights, and its removal achieved an important goal.

Third, it is especially difficult to impose sanctions against non-governmental civil parties. Often no clear geographic border separates the opposition from the government, making trade sanctions problematic.60 Also, these groups are generally not direct recipients of foreign economic aid. Finally, freezing assets is often not an option either because the opposition members are not wealthy enough to have extensive assets in foreign banks, or because secrecy makes it difficult to identify the individuals whose assets should be frozen.61

Thus, the prospects for using economic sanctions to prevent internal warfare are mixed. That conflict prevention involves deterrence rather than compellence bodes well for sanctions' success. The greatest difficulty with using sanctions to prevent civil conflict is not with the power of the tool, but the wisdom of its user. There are limits to our ability to recognize when civil war is imminent and to distinguish when reform pressures will prevent war (as in South Africa) and when such pressures will trigger war (as in Rwanda). To use sanctions effectively for conflict prevention, we will need excellent intelligence and a better understanding of the causes of war.

B. Preventing Interstate Conflict

Most regional violence is internal in nature. The years 1989-1992 saw 82 armed conflicts around the world in which at least 1000 people were killed. Of these, seventy-nine took place within country borders.62 Nevertheless, regional interstate conflict can pose serious costs and risks to world peace (witness the costs and hazards of the Arab-Israeli conflict). Hence the world's major powers should have an active strategy for its control. How much can economic sanctions contribute to such a strategy?

Sanctions have a fair chance of succeeding when applied to prevent regional interstate wars. Interstate wars can begin in two ways. Either a civil war widens to engulf other states, or war can erupt between two states that suffer no civil conflict. Outside powers can become involved in civil wars in two ways. First, they can intervene in a civil war, as Germany and Italy did in the Spanish civil war, and as the United States did in the Vietnamese civil war. Second, belligerents in the civil war can attack outside powers. Examples include Sandinista attacks on Honduras in the 1970s and Vietnamese communist intrusions into Cambodia in the 1960s and 1970s.

The threat of economic sanctions could help to avert both scenarios. Specifically, an international coalition could avert the first scenario by threatening tough sanctions against any outside powers that intervene, and could avert the second scenario by threatening to sanction either civil belligerent if it attacks surrounding states.

Economic sanctions and the threat of sanctions both seem well-adapted to prevent outside intervention in civil wars. First, success requires the deterrence of a contemplated action, rather than the more difficult task of compelling the reversal of actions already undertaken. Second, it is relatively easy to identify the outside powers who might intervene in a given civil war. Hence the target of the threat (and of the sanction) is clear. As noted above, this is not always true of internal conflict prevention efforts.

It may be more difficult to prevent belligerents from lashing out at neighboring states because the belligerents are highly motivated and therefore less likely to be swayed by economic sanctions or the threat of sanctions. However, even here a sanctioning coalition can be successful if its aid or trade is critical to a belligerent's war effort, or to its post-war rebuilding effort.63

Sanctions are likely to be more successful at preventing interstate war than civil conflict for several reasons. Many of the difficulties associated with using sanctions for preventing internal conflict are absent. Determining when and where the risk of war is rising is easier because unlike civil wars, interstate conflicts seldom erupt without warning. It is also easy to identify the parties that must be deterred to prevent an interstate war.64 With inter-state conflict, the targets of the sanctions (or threats) are nearly always states.65 Predicting which states are likely to be drawn into the conflict is relatively easy. Neighboring states, former colonial powers, and superpowers are the obvious candidates. Finally, the problem of distinguishing between causes and preventives of conflict is less of a problem.

C. How Should Economic Sanctions be Applied?

Policy makers should follow four rules to maximize the effectiveness of sanctions. Total and complete sanctions should be imposed. They should be imposed immediately. Gaining the cooperation of key states is necessary. Finally, members of the sanctioning coalition must demonstrate resolve.

First, the full range of economic instruments available should be used. A combination of aid, trade and financial sanctions is markedly more effective than any lesser combination. The total sanctions imposed on Iran (1979), Iraq (1990), Haiti (1991), and Yugoslavia (1992) show that total sanctions achieve results.

Second, total sanctions should be imposed as soon as any decision to impose sanctions is taken. A slow incremental tightening of sanctions is far less effective because it allows the target time to adjust by taking steps— such as stock piling goods and moving money— that would make future sanctions less effective. Slow incremental sanctions may also cause the target to question whether the coalition has the resolve to see the sanctions through to success.66

Third, the cooperation of the neighbors, major trading partners of the target, and major aid donors must be secured. Clearly it is desirable to have as many states as possible participate in the sanctions effort. However, the cooperation of these particular states is critical since they have the greatest economic interaction with the target. Gaining their cooperation may not be easy because the neighbors and trading partners are likely to suffer from the imposition of economic sanctions. To encourage cooperation, the UN or the U.S. could provide carrots, such as measures to compensate for the loss of revenue from trade with the target, and sticks such as threats to reduce aid or trade. Carrots and sticks will seldom induce perfect compliance. Even if neighboring governments cooperate,

 

some degree of smuggling is likely. Fortunately, perfect compliance is not required for sanctions' success. The results in Iraq and Haiti demonstrate that a little leakage does not prevent sanctions from devastating the target state's economy.

Fourth, the sanctioning coalition must convince the target that it will keep the sanctions in place until they achieve success. This is especially important because sanctions generally take years to produce results.67 To create the appearance of endurance, the U.S. must publicly commit itself in a highly visible way to maintaining sanctions until the target complies. Such a commitment from the world's most powerful state dampens target states' hopes of a retreat.68

D. Financial Sanctions: Under-studied and Under-used

Financial sanctions have been little used and little studied. Yet, two major types of financial sanctions— freezing assets and political aid conditionality by IFIs— have clear advantages as policy tools. Freezing assets has the advantage of being a precision guided weapon that can zero in on government officials and their supporting elites without causing extensive collateral damage to the largely innocent civilian population. Thus, freezing assets avoids the moral criticism made against economic sanctions, that they unfairly punish the ordinary people of a target state without causing serious suffering to the elites. IFI political aid conditionality also has a built-in advantage in that most of the countries where war is likely to break out are places that badly want assistance from multilateral banks.69 Hence IFIs have large potential leverage.

Given these advantages, why have these sanctions been so little used? In the case of IFI conditionality the answer is relatively simple. IFIs have always seen their mission primarily in economic, not political terms. They chose loan recipients based on economic criteria, not political criteria.70 This IFI culture is softening as IF Is have since the mid­1980s become more willing to consider environmental issues, poverty, government corruption, and military spending levels in making their decisions.71 The timing of recent announcements promising aid to Boris Yeltsin's Russia illustrates this changing attitude. But the IFIs apolitical cultural legacy limits their willingness to employ political conditionality widely.

Why have assets been frozen so infrequently? Trade embargoes and bilateral aid cuts have been the economic tools of choice, while assets of target regimes were seldom frozen.72 One possible explanation lies in the culture of the banking industry. Banking is an unusual industry in that it has virtually no fixed assets. It is an industry based almost solely on trust and relationships. Therefore, bankers are reluctant to take actions that damage their relationships with clients and place future business in jeopardy.73 Banking industry culture is important because bankers often have large influence on sanctions related policy decisions in the U.S.

Another possible explanation for the U.S. reluctance to freeze assets is the difficulty in gaining widespread international cooperation. Corrupt and thuggish rulers often spread their money around, placing deposits in many banks in many countries. For example, it was reported in late 1995 that the family of former Mexican President Carlos Salinas had ninety bank accounts in nine countries.74 This suggests that for asset freezes to have a significant impact, there must be broad international cooperation with the effort. If such cooperation seems unlikely, states may be reluctant to unilaterally freeze assets since this would impose costs on their own banking industry that would not be borne by foreign competitors. 

The concern that freezing assets will scare off future depositors may be overblown. First, it is likely that the only investors deterred will be other corrupt and thuggish rulers. Fortunately, there are only a few such individuals, so lost banking profits are small. Second, although the U.S. has frozen assets with increased frequency over the last fifteen years (e.g. Iran, Libya, Panama, and Iraq), no literature has emerged suggesting that the costs to the U.S. banking industry are high. Third, any costs in terms of future deposits lost must be weighed against the large benefits that an asset freeze can achieve. It need not be universal to produce some benefit. Even a unilateral or small scale multilateral effort will reduce the pool of money available to the target, thereby forcing unfrozen accounts to be drawn down. The sanctions against Iraq are an example. Years of sanctions (including an asset freeze), have forced Saddam to exhaust the funds he held in Swiss banks.75

Both freezing assets and political aid conditionality have been largely ignored in the sanctions' literature. Sanctions specialists have focused their studies on specific cases such as South Africa, COCOM, and Cuba, rather than specific types of economic sanctions. Since freezing assets has been so infrequent and IFI conditionality is a relatively new option, it is not surprising that neither has received a great deal of attention in the sanctions' literature .76 These are both clearly important areas for future research. 

What are the prospects for using asset freezes and IFI political aid conditionality to prevent deadly conflict? Both of these tools could be very helpful in that effort, and the general trends towards greater use of these instruments makes it more likely that they will be used for this purpose.

Freezing the assets of dictatorial governments and their key supporters could help to ease them from power or induce them to make policy changes that reduce the likelihood of deadly conflict. The recent decisions to freeze assets in Iraq and Haiti are encouraging signs of the willingness of the U.S. and other industrial democracies to use this tool. Also encouraging was the recent decision by the World Court ordering Swiss banks to release some of the assets of former Philippine ruler Ferdinand Marcos to compensate his victims in the Philippines. These decisions suggest that the anti-freezing banker culture is fading or giving way to the policy concerns of the industrial democracies and they raise the possibility of greater future cooperation on asset freezing. By raising the possibility that frozen assets could become forever lost, the court decision could also deter future dictators from taking any steps that might cause their overseas assets to be frozen.

IFIs clearly have an interest in preventing deadly conflict since their investments are lost in the devastation of war and they will be called upon to provide massive financial assistance to rebuild countries when wars end. Thus, assisting in prevention efforts could save them a great deal of money and effort. The recent willingness of the IMF and the World Bank to link high levels of military spending with a lack of economic development indicates a willingness to consider factors related to deadly conflict.77 However, the experience of the European Bank of Reconstruction and Development (EBRD) sounds a cautionary note. It illustrates the difficulties involved in moving IFIs away from their traditional banking mission towards an active consideration of political conditions in recipient countries.78

VI. CONCLUSIONS

Economic sanctions have not historically been used for the purpose of preventing deadly conflict. They should be. Conditions for using economic sanctions for this purpose are more auspicious today than in the past and should remain so for the foreseeable future. The Iraq, Haiti, and Yugoslavia cases show that when an international coalition applies immediate and total sanctions, the sanctions can pay policy dividends. The Haitian case shows that even badly designed sanctions, once corrected, can help sender states to achieve their policy goals.

Economic sanctions are more effective than most analysts suggest. Their efficacy is underrated in part because unlike other foreign policy instruments, sanctions have no natural advocate or constituency. Business leaders tend to dislike sanctions because they disrupt international commerce. Most governments have counterparts to the U.S. State Department, CIA, and Defense Department. These bureaucracies are natural advocates for the use of their own tools— diplomacy, covert action, and military force. Economic sanctions have no equivalent champion. As a result, their successes are widely unreported while their failures are exaggerated by those with an interest in either avoiding their use, or in using other instruments. However, economic sanctions are an important policy tool and deserve more attention from both scholars and policy makers.

The main impediment to successfully using economic sanctions to prevent deadly conflict is not the weakness of the instrument. Rather, it lies in the sanctioners' understanding of regional conflict and their inability to act decisively.. To use sanctions successfully, sanctioners must gain a better grasp of the roots of regional wars, learn to better predict their occurrence, and become better able to decide on effective solutions. The United States and other sanctioners must also recognize the need to commit publicly to maintaining sanctions for long periods. Finally, sanctioners must formulate a strategy for addressing the economic suffering of domestic interests and neighboring states that are injured by sanctions. These innovations will make economic sanctions an even more effective instrument.

FOOTNOTES

1 Portions of this paper are adapted from Elizabeth S. Rogers, "Economic Sanctions and Internal Conflict" in The International Dimensions of Internal Conflict, ed. Michael E. Brown (Cambridge: MIT Press, 1996) and Rogers, "Using Economic Sanctions to Control Regional Conflicts," Security Studies 5:4 (Summer 1996).

2 Data on the number of wars and war deaths are from Ruth Leger Sivard, World Military and Social Expenditures 1993 (Washington D.C.: World Priorities, 1993), 20-22. (War deaths are read from chart 14.) Data on refugees and internally displaced persons are from Myron Weiner, The Global Migration Crisis: Challenge to States and to Human Rights (New York: Harper Collins, 1995), 2 and 7.

 

3 Weiner, Global Migration Crisis, 5.

4 The limits of the U.S. public's tolerance for casualties were seen in Somalia, where a total of thirty U.S. combat deaths triggered a U.S. decision to withdraw. For U.S. deaths, see Donatella Lorch, "Last U.S. Marines Quit Somalia As Escorts for the U.N.," New York Times 3 March 1995, 3. On increasing U.S. unwillingness to take casualties in war see Harvey M. Sapolsky, "War Without Killing," in S. Sarkesian and J. Flanagin, eds. U.S. Domestic and National Security Agendas, (Westport, CT: Greenwood), 27-40.

5 The principle of conditionality also applies to membership in regional organizations such as NATO, the North American Free Trade Association (NAFTA), and the European Union (EU). Milada Anna Vachudova argues that the desire for membership in NATO and the EU has had a positive effect on policy making in the states of Eastern and Central Europe. See Vachudova, "Peaceful Transformations in East-Central Europe," in The International Dimensions of Internal Conflict, 69-105.

6 Conditioning bilateral aid on political criteria has long been a practice in the United States and other. countries. For example, during the Cold War, a government had to be anti-communist to receive U.S. aid. U.S. laws currently prohibit aid to countries sponsoring international terrorism, violating human rights, or developing nuclear weapons. However, the World Bank and the IMF long resisted conditioning aid on political criteria as their charters prohibited it.

7 Sanctions require leadership by a great power, and the United States is the only superpower in the post-cold war era.

8 This is especially true for the United States for which the most data exists. See the sub-section on costs below.

9 My inventory of sanctions efforts relies on Gary Clyde Hufbauer, Jeffrey J. Schott, and Kimberly Ann Elliott, Economic Sanctions Reconsidered, which offers the most thorough compilation of sanctions attempts. They list 116 sanctions efforts during 1914-1990. One of these 116 was a post-1989 effort (UN sanctions on Iraq following its 1990 invasion of Kuwait), leaving 115 efforts during 1914-1989. Economic Sanctions Reconsidered, 2nd ed., 2 vols. (Washington D.C.: Institute for International Economics, 1991).

10 Hufbauer, Schott, and Elliott use the term "black knight" to refer to a power that counters another's sanctions by providing offsetting aid and trade to the target state. Economic Sanctions Reconsidered, 1:12.

11 Hufbauer, Schott, and Elliott, Economic Sanctions Reconsidered, 2:109.

12 Ibid., 96-97.

13 David Baldwin begins his classic analysis of sanctions effectiveness by observing that "the two most salient characteristics of the literature on economic statecraft are scarcity and the nearly universal tendency to denigrate the utility of such tools of foreign policy. " David A. Baldwin, Economic Statecraft (Princeton: Princeton University Press, 1985), 51. Henry Bienen and Robert Gilpin have likewise noted "the nearly unanimous conclusion of scholars that sanctions seldom achieve their purposes and more likely have severe counter-productive consequences." Bienen and Gilpin, "An Evaluation of the Use of Economic Sanctions to Promote Foreign Policy Objectives, with Special Reference to the Problem of Terrorism and the Promotion of Human Rights," an unpublished report prepared for the Boeing Corporation, 2 April 1979, 3. Works concurring with this pessimistic view of sanctions effectiveness include James Barber, "Economic Sanctions as a Policy Instrument," International Affairs 55 (July 1979): 367-384; Bienen and Gilpin, "An Evaluation of the Use of Economic Sanctions"; C. Lloyd Brown-John, Multilateral Sanctions in International Law: A Comparative Analysis (New York: Praeger, 1975); Margaret P. Doxey, Economic Sanctions and International Enforcement, 2nd ed. (New York: Oxford University Press, 1980); Doxey, International Sanctions in Contemporary Perspective (New York: St. Martin's, 1987); Hufbauer, Schott, and Elliott, Economic Sanctions Reconsidered; James M. Lindsay, "Trade Sanctions as Policy Instruments: A Re-examination," International Studies Quarterly 30, no. 2 (June 1986): 153-173; Donald Losman, International Economic Sanctions: The Cases of Cuba, Israel, and Rhodesia (Albuquerque: University of New Mexico Press, 1979); Makio Miyagawa, Do Economic Sanctions Work? (New York: St. Martins, 1992); Miroslav Nincic and Peter Wallensteen, "Economic Coercion and Foreign Policy," in Dilemmas of Economic Coercion: Sanctions in World Politics, ed. Nincic and Wallensteen (New York: Praeger, 1983), 1-15. For a collection of statements from political leaders and the popular press arguing that sanctions do not work see M.S. Daoudi and M.S. Dajani, Economic Sanctions: Ideals and Experience (Boston: Routledge and Kegan Paul, 1983), Appendix II, 178-188.

For dissents from the dominant pessimistic view of sanctions' efficacy see Baldwin, Economic Statecraft; and David A. Deese, "The Vulnerability of Modern Nations: Economic Diplomacy In East-­West Relations," in Dilemmas of Economic Coercion: Sanctions in World Politics, 155-181.

A useful review of the sanctions literature is Baldwin, Economic Statecraft, 51-58.

14 Much of the change is accounted for by a drop in the success rate of U.S. efforts from over 50 percent before 1973, to just 17 percent after 1973. Hufbauer, Schott, and Elliott, Economic Sanctions Reconsidered, 1:93, 1:106-108. The general success rates are calculated from data presented in Table 5.12. (Ibid., 106) The post-1973 U.S. sanctions rate is calculated from data in Table 5.13. (Ibid., 108)

15 Baldwin argues that economic sanctions are often judged by a higher standard than other foreign policy instruments. Economic Statecraft, chapter 7, esp. 138-144.

16 See Baldwin, Economic Statecraft. He argues persuasively that economic sanctions nearly always have multiple goals, and that all goals should be considered when judging sanctions' success.

17 Sanctions can also have other goals that are ancillary to conflict prevention. Sender states may also wish to reassure their allies that the sender will stand by them in future crises; inflict punishment as vengeance; raise an issue to prominence on the international agenda; and bolster the sender government's domestic public image as tough and decisive. I do not consider these purposes here since they are related only tangentially, if at all, to conflict prevention.

18 Sanctions' success should be measured by asking: would senders be closer to their policy goals, farther from them, or at the same point if sanctions had never been imposed? Thus sanctions should be judged at least partially successful if they bring senders closer to policy goals.

19 Documenting the effectiveness of sanctions use for deterrence is extremely difficult. Leaders rarely advertise that they have been deterred. Such an admission could make them look weak or expose their vulnerabilities. Thus, they have every incentive to conceal the fact that they have been deterred. Also, successful deterrence often results in a non-event, that by its very nature is difficult to verify. Thus, the small number of known cases of sanctions used for deterrence are likely a substantial underestimate of the actual deterrent affect of economic sanctions.

20 The Hufbauer, Schott, and Elliott data set includessix such cases. However, these six cases are not analyzed separately from the remaining 110 efforts at compellence. The cases are: League of Nations v. Yugoslavia (1921), League of Nations v. Greece (1925), United States v. Israel (1956), Federal Republic of Germany v. German Democratic Republic (1961), and United States v. El Salvador (1987).

21 See Thomas L. Friedman, "Preventing A Reprise," New York Times 4 October 1991, A6; and Howard W. French, "Coup in Africa Puts Western Nations in Quandary," New York Times 3 February 1996, 3.

22 The U.S. imposed economic sanctions seventy-four times before and during the cold war era, but only five of these sanctions were complete and total: those against North Korea (1950), North Vietnam (1954), Cuba (1960), Kampuchea (1975), and Iran (1979). Hufbauer, Schott and Elliott, Economic Sanctions Reconsidered, 1:9, 1:67. The U.S. has imposed complete sanctions three times since the end of the cold war (against Iraq in 1990, Yugoslavia [consisting of Serbia and Montenegro] in 1992, and Haiti in 1991).

Sanctions succeeded in only the four most recent cases (Iran, Iraq, Yugoslavia, and Haiti), for a 50 percent success rate. However, in three of the four unsuccessful sanctions efforts (North Korea, North Vietnam, and Kampuchea) the U.S. had very little trade with its target, and none were receiving aid that could be severed. Moreover, black knights (the U.S.S.R. and China) gave North Korea and North Vietnam unusually large offsetting aid. These conditions were not typical of sanctions situations. Most targets of future comprehensive sanctions would find themselves in worse circumstances.

23 U.S. human rights sanctions prevented foreign aid from going to governments that engaged in human rights abuses, but imposed no trade or financial penalties. Anti-terrorism sanctions initially cut foreign aid and air travel, and later cut exports to and imports from countries supporting international terrorism, but no financial sanctions were imposed. Nuclear non-proliferation sanctions prevented U.S. foreign aid from going to countries working to acquire nuclear weapons, but did not impose trade or financial sanctions excepting only nuclear-related exports and imports. See Hufbauer, Schott, and Elliott, Economic Sanctions Reconsidered, 2:336-339, 2:327-329, and 2:353-358.

In each of these instances U.S. economic sanctions resulted from congressional legislation. Such sanctions are often implemented in a half-hearted manner because the executive branch, which handles sanctions implementation, is ambivalent about the sanctions policy.

24 On this argument see Mary H. Cooper, "Economic Sanctions," CQ Researcher 4, no. 40 (28 October 1994): 941, 943.

25 Gary Hufbauer, "The Impact of U.S. Economic Sanctions and Controls On U.S. Firms," A Report to the National Foreign Trade Council, April 1990, 23. The National Foreign Trade Council describes itself as an anti-sanctions organization. Ibid., iii.

26 Total U.S. exports for 1987 were $250.4 billion and GDP was $4.46 trillion. Thus, $7 billion is 2.8 percent of total exports and .16 percent of GDP. Further, two thirds of the $7 billion in losses resulted from the COCOM sanctions which have since been lifted. Hufbauer, "Impact of U.S. Economic Sanctions," 44; The Military Balance 1988-1989 (London: International Institute for Strategic Studies, 1988), 18; World Military Expenditures and Arms Transfers 1988 (Washington D.C.: U.S. Arms Control and Disarmament Agency, 1989), 107.

27 Of the seventy-three cases involving the United States, 48 percent resulted in a net gain, 37 percent had "little effect on [the] sender," 14 percent resulted in a "modest welfare loss to sender," and only 1 percent (a single case-Iraq in 1990) caused a "major loss to sender." The negative economic effects of the Iraqi case fell mainly on oil dependent U.S. allies rather than on the United States. Data calculated from tables in Hufbauer, Schott, and Elliott, Economic Sanctions Reconsidered, 1:84-90.

28 The concepts of relative and absolute costs are the inverse of Joseph Grieco's concepts of relative and absolute gains. He applies these concepts to states, while I apply them to firms. See Joseph M. Grieco, Cooperation Among Nations (Ithaca: Cornell University Press, 1990)

29   This dynamic helps to explain the growing U.S. corporate opposition to the COCOM sanctions in the 1980s. See Beverly Crawford, Economic Vulnerability in International Relations: East-West Trade. Investment, and Finance (New York: Columbia University Press, 1993), 31-36; and Kevin F.F. Quigley and William J. Long, "Export Controls: Moving Beyond Economic Containment," World Policy Journal 7, no. I (Winter 1989-90): 175-178.

30 The Military Balance 1988-1989, 18. A strong U.S. military could contribute to the effectiveness of sanctions in some cases. The comparison of the defense budget to the cost of sanctions is therefore problematic. However, it does provide a rough sense of the difference in cost between using economic sanctions and using force.

31 Cost and payment figures are from Ann Markusen, "Mixed Messages: The Effects of the Gulf War and the End of the cold war on the American Military-Industrial Complex," in John O'Loughlin, Tom Mayer, and Edward S. Greenberg, eds., War and Its Consequences: Lessons from the Persian Gulf Conflict (New York: Harper Collins, 1994), 165.

32 For more on this issue see Lori Fisler Damrosch, "The Civilian Impact of Economic Sanctions," in Enforcing Restraint: Collective Intervention in Internal Conflicts, ed. Damrosch (New York: Council on Foreign Relations Press, 1993), 274-315; and Jack T. Patterson, "The Politics and Moral Appropriateness of Sanctions," in Economic Sanctions: Panacea or Peace-building in a Post-Cold War World? ed. David Cortright and George A. Lopez (Boulder: Westview, 1995), 89-96.

33 Sanctions' harm to innocents is an injury that the innocents may welcome when they share the sanctions' goals. For example, Claudette Antoine Werleigh noted the paradox of "seeing the poor in Haiti, although hurt and deeply affected by the embargo, stand by it and exert pressure for strengthening its enforcement, while other groups in the Haitian society who were well-off and far less hurt by its impact press for its end." Werleigh, "The Use of Sanctions in Haiti: Assessing the Economic Realities," in Economic Sanctions, 161. See also, Pamela Constable, "Dateline Haiti: Caribbean Stalemate," Foreign Policy 89 (Winter 92/93): 175-176, 183; "Still Embargoed," The Economist, 10 October 1992, 54. Nevertheless, sanctions should of course be targeted as much as possible at the government and its primary supporters.

34 I thank Sam Huntington for suggesting this argument.

35 Ending a war (conflict resolution) is of course a different task from preventing the outbreak of fighting in the first place. However, success at conflict resolution means that lives will be saved as deadly conflict is stopped and future conflict prevented.

36 Nicholas D. Kristof, "The Rise of China," Foreign Affairs 72, no. 5 (November/December 1993): 71.

37 Nicholas D. Kristof, "China Opposes Sanctions in North Korea Dispute," New York Times, 24 March 1993, A8.

38 Some undercutting of sanctions will occur even in the absence of spoiler states but this undercutting will be on a much smaller scale. For example, neighbors of sanctioned states often fail to exert perfect control over their borders, allowing some leakage. However, this leakage is usually quite small in comparison to the assistance provided by a spoiler's concerted efforts to undercut sanctions. Spoilers and leakers differ in the magnitude of their efforts and often in their intentions. Unlike spoilers, states that leak often have an official public policy of honoring the sanctions. Sanctions can seldom survive undercutting by black knights, but usually retain their power despite leakage.

39 In late 1994 and early 1995, France, Russia, and China reportedly supported lifting the sanctions on Iraq. However, the U.S. was able to keep the coalition together and get the necessary votes in the UN Security Council to maintain the sanctions. See "France Starts to Open the Doors," The Economist, 14 January 1995, 41-42; and Barbara Crossette, "Iraq Hides Biological Warfare Effort, Report Says," New York Times, 12 April 1995, A8.

The defection to Jordan in August 1995 of Saddam Hussein's sons in law brought to light new information on the Iraqi development of biological weapons. This has, at least temporarily, silenced calls for repeal of the sanctions. "U.N. Extends Its Sanctions Against Iraqis," New York Times, 9 September 1995, 3.

40 Whereas the Iraq, Haiti, and Yugoslavia cases all involved violence and internal or external aggression, the sanctions against Iran were imposed to prevent the acquisition of weapons of mass destruction and the calls for sanctions against Nigeria were triggered by the execution of the dissident, Ken Saro-Wiwa.

Japan and Western European countries are the largest purchasers of Irani oil. The U.S. and its European allies also account for nearly 90 percent of Nigerian oil purchases. David E. Sanger, "Japan Tells U.S. It Will Not Join Embargo On Iran," New York Times 19 June 1995, Al; Christopher S. Wren, "U.S. Is Seeking Further Ways to Punish Nigeria for Executions," New York Times 17 December 1995, 11.

41 Iraq's GNP was $35 billion in 1989, $15 billion in 1991, and $17 billion in both 1992 and 1993. The Military Balance 1994-1995 (London:.Brassey's, 1994), 129.

The economic damage that sanctions inflict on targets is an indirect but nevertheless useful measure of their success. A target state may defy a sender's wishes even if it suffers large damage, but the likelihood that the target will comply generally increases with the severity of the damage it suffers.. Moreover, the damage sanctions inflict weakens the target (thereby reducing its capacity for mischief) and punishes the target (thereby deterring it and others from future mischief). As noted above, achieving these objectives can serve a valuable purpose.

42 David E. Reuther, "UN Sanctions Against Iraq," in Economic Sanctions, 125-126.

43 Youssef Ibrahim, "Iraq Said to Sell Oil In Secret Plan to Skirt U.N. Ban," New York Times, 16 February 1995, A6.

44 For the argument that "sanctions have had a serious effect in degrading Iraq's military" see Phebe Marr, "Iraq and Sanctions: What Lies Ahead. for the Future?" Testimony before the Senate Foreign Relations Subcommittee on Near East and South Asia, 3 August 1995, 5.

45 Prior to the August 1995 defection of Saddam's sons-in-law, UN monitor Rolf Ekeus described Iraq as having complied with UN demands regarding nuclear and chemical weapons, but not regarding biological weapons. Crossette, "Iraq Hides Biological Warfare Effort, Report Says," A8. Since the defections, new evidence in all three areas has emerged and is being examined. Crosette, "Iraq Seeking U.N. Talks On Oil Sales," New York Times, 17 January 1996, A8.

46 Stanley Meisler, "Iraq, U.N. Start Talks on Easing Ban on Oil Sales," Los Angeles Times 7 February 1996, A6; "Iraq and U.N. Open Talks on Oil Sales," New York Times 7 February 1996, A9.

47 It also bears mention that Saddam's overthrow is not a formal goal or condition of the sanctions. UN Security Council resolution 687 passed in April 1991 requires that Iraq renounce all weapons of mass destruction-chemical, biological, and nuclear-and missiles with ranges above 150 kilometers, forego terrorism, respect the Kuwaiti border, return Kuwaiti property, return missing foreign nationals, and pay compensation and war reparations from the proceeds of future oil sales. See Alan Dowty, "Sanctioning Iraq: the Limits of the New World Order," Washington Quarterly 17, no. 3 (Summer 1994): 180; Reuther, "UN Sanctions Against Iraq," 123. Overthrowing Saddam became an informal American sanctions goal shortly after the war.

48 See Gary C. Hufbauer and Kimberly A. Elliott, "Sanctions Will Bite-and Soon," New York Times, 14 January 1991, A17; and Les Aspin, "The Role of Sanctions in Securing U.S. Interests in the Persian Gulf," in U.S., Congress, House, Committee On Armed Services, Crisis In the Persian Gulf: Sanctions. Diplomacy and War. Hearings Before the Committee On Armed Services. 101st Cong., 2nd sess., 1990, 862-863.

49 Most strikingly, the assets of the Haitian military in the United States were not frozen until January 1994, more than two years into the crisis. "U.S. Extends Economic Sanctions on Haiti," New York Times, 28 January 1994, A7.

50 The half-hearted and tentative nature of U.S. sanctions is revealed by the slow and meandering chronicle of their imposition. Not until May/June 1994 were forceful sanctions imposed by a broad international coalition.

Shortly after Aristide's overthrow, in September 1991, the United States suspended aid, trade, and all transactions with the Haitian government. OAS sanctions followed shortly. But in February 1992 the Bush administration created exemptions for U.S. businesses operating in Haiti. Only in June 1992 was the embargo broadened to deny U.S. port access to ships engaged in commerce with Haiti. Only in June 1993, nineteen months into the confrontation, did the UN finally impose an oil embargo on Haiti. This embargo was suspended two months later in August after the Haitian junta signed the Governors Island Agreement, but before they complied with it. The oil embargo then was re-imposed in October 1993 when the junta reneged on that agreement. Also in October 1993-two years into the confrontation-­the United States finally froze the assets of forty-one government supporters. Only in January 1994 did the United States finally freeze the assets of members of the Haitian military and prohibit transactions with them. Commercial air service with Haiti was banned and financial transactions further restricted only in June 1994. Erin Day, Economic Sanctions Imposed by the United States Against Specific Countries: 1979 Through 1992 (Washington DC: Congressional Research Service, CRS Report for Congress, 10 August 1992, 513-514, 523-525; Paul Lewis, "U.N. Again Imposes Sanctions on Haiti After Pact Fails," New York Times, 14 October 1993, Al; "U.S. Extends Economic Sanctions on Haiti," New York Times, 28 January 1994, A7; Heather M. Fleming, "Give Sanctions Time To Bite, Gray Tells Lawmakers," CQ Weekly Report, 11 June 1994, 1540.

51 Arguing that U.S. mixed signals and lack of resolve unnecessarily prolonged the Haitian crisis is Werleigh, "The Use of Sanctions in Haiti," 170.

52 Howard W. French, "Haiti's Poor Feeling the Pinch As Sanctions Ruin Economy," New York Times, 15 November 1993, A7; and "Tightening the Stranglehold," The Economist, 6 August 1994, 35.

53 "Sanctions Work," The Economist, 4 September 1993, 41; Rick Bragg, "Many of Haiti's Elite Resign Themselves to Aristide's Return," New York Times, 25 September 1994, 16.

54 Day, Economic Sanctions Imposed by the United States Against Specific Countries, 425-428; "Wide-ranging Sanctions Imposed Against Yugoslavia," UN Chronicle 28, no. 3 (September 1992): 5-12.

55 "Feeling the Pinch," The Economist, 8 October 1994, 54-55; Roger Cohen, "An Imperfect Peace," New York Times, 22 November 1995, Al.

56 David Binder, "U.S. Policymakers on Bosnia Admit Errors in Opposing Partition in 1992," New York Times, 29 August 1993, 10; Chuck Sudetic, "Clinton Writes to Reassure Bosnian Government of Support," New York Times, 5 December 1994, A12; Steven Greenhouse, "U.S. Says Bosnian Serbs Face NATO Attack if Talks Stall," New York Times, 28 August 1995, A6.

57 The 1995 peace effort led by U.S. Under-secretary of State Richard Holbrooke tried to finesse the issue by placing a Serb republic within a Bosnian state. See Roger Cohen, "Serb Shift Opens Chance For Peace, A U.S. Envoy Says," New York Times, 1 September 1995, Al.

58 Arguing in 1993 that the United States should abandon the pursuit of a unitary Bosnia and pursue a partition plan for peace were John J. Mearsheimer and Robert A. Pape, "The Answer," The New Republic, 14 June 1993, 22-28. Noting that the Dayton agreement was in fact a veiled if incomplete partition were John J. Mearsheimer and Stephen Van Evera, "When Peace Means War," The New Republic, 16 December 1996, 16-19.

59 Arguing that international mediators and the use of economic leverage contributed to the genocide in Rwanda is Alan J. Kuperman, "The Other Lesson of Rwanda: Mediators Sometimes Do More Damage Than Good," SAIS Review 16, no. 1 (Winter-Spring 1996): 221-240.

Pressing regimes to respect human rights even at the risk of triggering violence can, of course, sometimes be the best choice. In the Nicaraguan case, the Somoza regime brutally violated human rights, and its removal achieved an important goal.

60 Pre-war Bosnia with its ethnic intermingling illustrates this problem. UN sanctions against UNITA held territory in Angola demonstrate that this issue does not apply once war is underway.

61 President Clinton's decision to freeze the assets of members of the Colombia-based Cali drug cartel, to threaten sanctions on those states that launder drug money, and to prohibit U.S. corporations from conducting business with members of the cartel or its front companies perhaps indicates a new willingness by the United States to identify and target non-state actors. Alison Mitchell, "U.S. Freezes Assets of Cartel in New Effort Against Drugs," New York Times 23 October 1995, All.

62 United Nations Development Program, Human Development Report. 1994 (New York: Oxford University Press, 1994), 47, quoted in Weiner, Global Migration Crisis, 17.

63 Sanctions will be more effective if married to threats to assist opponents either economically or militarily.

64 The 1995 U.S. threat to impose economic sanctions on Croatia should it use military force in eastern Slavonia is an example of using sanctions to prevent interstate conflict. Roger Cohen, "U.S. Cooling Ties To Croatia After Winking At Its Buildup," New York Times 28 October 1995, Al.

65 The exception would be cases of civil war in which a non-governmental belligerent attacks outward in hopes of widening the conflict.

66 For the opposite argument, that moderate sanctions with the threat of escalation are best, see Ivan Eland, "Economic Sanctions as  Tools of Foreign Policy," in Economic Sanctions, 29-42.

67 The average duration of sanctions coded as successes by Hufbauer, Schott, and Elliott was 2.9 years. Hufbauer, Schott, and Elliott, Economic Sanctions Reconsidered, 1:101.

68 If the target believes that the sanctions will be short lived, then its willingness to comply with the terms of the sanctions declines greatly.

69 In the 1990s there have been two major causes of war-the breakup of empires and the democratization of multi-ethnic states. All of the states in these categories need development assistance. This argument on the causes of war comes from Stephen Van Evera, "Future of War" seminar presentation, MIT, December 1995.

70 For example, the World Bank's Articles of Agreement state that, "the Bank and its officers shall not interfere in the political affairs of any member, nor shall they be influenced in their decisions by the political character of the member or members concerned." Cited in Wolfgang H. Reinicke, "Cooperative Security and the Political Economy of Nonproliferation," in Global Engagement: Cooperation and Security in the 21st Century ed. Janne E. Nolan (Washington, D.C.: Brookings, 1994), 200.

71 Wolfgang Reinicke notes that the World Bank and IMF now consider these issues to be an integral part of economic development where they were previously considered secondary or incidental. Reinicke, "Can International Financial Institutions Prevent Internal Violence? The Sources of Ethno-National Conflict in Transitional Societies," in Preventing Conflict in the Post-Communist World: Mobilizing International and Regional Organizations Abram and Antonia Handler Chayes, eds. (Washington, D.C.: Brookings, 1996), 291-92.

72 For example, a preliminary count reveals that the U.S. froze assets in less than ten percent of its sanctions efforts between 1914 and 1996. Hufbauer, Schott, and Elliott, Economic Sanctions Reconsidered.

73 A striking illustration of this culture in action is found in Britain's decision to transfer $28,349,440 of Czech gold from London to Berlin in the summer of 1939. Willimson Murray, The Change in the European Balance of Power, 1938-1939: The Path to Ruin (Princeton: Princeton University Press, 1984), 291.

74   Monitor Radio, December 1995.

75 Ibrahim, "Iraq Said to Sell Oil In Secret Plan."

76 This is especially true of freezing assets, a topic that has largely been ignored by political scientists and economists. With rare exceptions, the academic writing on this subject consists of a few highly technical articles in law journals on the mechanics of asset freezes. Two such exceptions, both focusing on the U.S. freezing of Iran's assets in 1979, are Mahysh Alerassool, Freezing Assets: The USA and the Most Effective Economic Sanction (New York: St. Martins, 1993); and Robert Carswell and Richard J. Davis, "The Economic and Financial Pressures: Freeze and Sanctions," in American Hostages in Iran: The Conduct of a Crisis, Warren Christopher et. al. (New Haven: Yale University Press, 1985), 173-200.

77 Reinicke, "Cooperative Security and the Political Economy of Nonproliferation," 199.

78 EBRD was founded in May 1990 to help the states of Eastern and Central Europe make the transition from command to market economies. Unlike other IFIs, its charter explicitly obligated the bank to lend money only to countries committed to "the principles of multiparty democracy, pluralism and market economics." However, over the last three years the bank has moved steadily away from active political conditionality to become "an operationally apolitical institution wholly focused on financial transactions." Cited in Melanie H. Stein, "Conflict Prevention in Transition Economies: A Role for the European Band for Reconstruction and Development?" In Preventing Conflict in the Post-Communist World, 341, 352.