Advocates of sanctions regard them as an important weapon in the foreign policy arsenal. Skeptics question whether sanctions are an effective stand-alone instrument and whether the costs to the users of sanctions are worth the benefits derived. To put these issues in perspective, we have delved into the rich history of the use of sanctions in the twentieth century in order to identify circumstances in which economic sanctions can succeed in attaining foreign policy goals.

Of 115 cases of economic sanctions between World War I and 1990, we judged 34 percent to be at least partially successful. These cases include instances of multilateral sanctions and they include sanctioners other than the United States. The objective is to provide as comprehensive an analysis of sanctions as possible. Comparing the economic and political circumstances across these episodes, we found that sanctions tend to be most effective when:

• The goal is relatively modest (thus lessening the importance of multilateral cooperation, which often is difficult to obtain, and reducing the chances a rival power will bother to step in with offsetting assistance).
• The target is much smaller than the country imposing sanctions (the average sender's economy was 187 times larger than that of the average target), economically weak and politically unstable.
• The sender and target are friendly toward one another and conduct substantial trade (the sender accounted for 28 percent of the average target's trade in success cases but only 19 percent in failures).
• The sanctions are imposed quickly and decisively to maximize impact (the average cost to the target as a percentage of GNP in success cases was 2.4 percent and 1 percent in failures, while successes averaged 2.9 years in duration and failures 8 years).
• The sender avoids high costs to itself.

There is an important caveat to this general story. Economic sanctions proved far more useful in contributing to foreign policy goals prior to about 1973, when they had a 44 percent success rate. Of 59 cases initiated between 1973 and 1990, only 14 (24 percent) resulted in at least a partial success even though the number of cases involving modest policy goals soared. Among other things, this can be attributed to the declining dominance in the world economy of the United States, which has been by far the most frequent user of economic sanctions (77 of the 115 cases).

Methodology

We define economic sanctions to mean the deliberate, government inspired withdrawal, or threat of withdrawal, of customary trade or financial relations. "Customary" does not mean "contractual"; it simply means levels of trade and financial activity that would probably have occurred in the absence of sanctions. We define foreign policy goals to encompass changes expressly sought by the sender state in the political behavior of the target state.

Sanctions also serve important domestic political purposes in addition to sometimes changing the behavior of foreign states. The desire to be seen acting forcefully, but not to precipitate bloodshed, can easily overshadow specific foreign policy goals. Indeed, domestic political goals increasingly appear to be the motivating force behind the imposition of many recent sanctions. Nevertheless, in judging the success of sanctions, we confine our examination to changes in the policies, capabilities, or government of the target country.

The Success of an Episode

The "success" of an economic sanctions episode--as viewed from the perspective of the sender country--has two parts: the extent to which the foreign policy outcome sought by the sender country was in fact achieved, and the contribution made by the sanctions (as opposed to other factors, such as military action) to a positive outcome. We have devised a simple index system, scaled from 1 (failed outcome; zero or negative sanctions contribution) to 4 (successful outcome; significant sanctions contribution), to score each element. By multiplication, the two elements are combined into a "success score" that ranges in value from 1 to 16. We characterize a score of 9 or higher as a "successful" outcome.

Success does not mean that the target country was vanquished by the denial of economic contacts, or even necessarily that the sanctions decisively influenced the outcome. Success is defined against more modest standards. A score of 9 means that sanctions made a modest contribution to the goal sought by the sender country and that the goal was in part realized; a score of 16 means that sanctions made a significant contribution to a successful outcome. By contrast, a score of 1 indicates that the sender country clearly failed to achieve its goals or may even have left the sender country worse off than before the measures were imposed.

Factors influencing success or failure

In evaluating the success of economic sanctions, the first step is to distinguish between the types of foreign policy objectives sought in different sanctions episodes. The nature of the objective may be the most important variable of all: sanctions cannot stop a military assault as easily as they can free a political prisoner. We have found it useful to classify the case histories in this study into five categories, according to the foreign policy objective sought by the sender country:

• Change target-country policies in a relatively modest way (modest in the scale of national goals if often of burning importance to participants in the episode); this type of goal is illustrated by cases involving human rights, terrorism, and nuclear nonproliferation.
• Destabilize the target government (including, as an ancillary goal, changing the target country's policies); this category is illustrated by the US campaigns against Fidel Castro and Manuel Noriega, and the Soviet campaign against Marshal Tito.
• Disrupt a relatively minor military adventure, as illustrated by the League of Nations' sanctions against Italy in 1935-36 over its aggression in Abyssinia.
• Impair the military potential of the target country, as illustrated by the sanctions imposed during World Wars I and II and the COCOM sanctions against the Soviet Union and its allies.
• Otherwise change target-country policies in a major way (including the surrender of territory), as illustrated by the US and UN campaign to force Iraq to withdraw from Kuwait, restore the legitimate government, and release all hostages.

In addition to the objective sought, we have identified several political and economic factors that we would expect to affect the outcome of a sanctions effort. The political variables we have analyzed (by no means an exhaustive list) include:

• Companion policies used by the sender country, namely, covert maneuvers, quasi military activity (such as massing troops on the target's border or stationing naval forces off its coast), and regular military activity.
• The number of years economic sanctions were in force.
• The extent of international cooperation in imposing sanctions, scaled from 1 (no cooperation) to 4 (significant cooperation).
• The presence of international assistance to the target country.
• The political stability and economic health of the target country, scaled from 1 (a distressed country) to 3 (a strong and stable country).
• The warmth of prior relations (i.e., before the sanctions episode) between sender and target countries, scaled from 1 (antagonistic) to 3 (cordial).
• The economic variables that we have analyzed (again not an exhaustive list) include:
• The cost imposed on the target country, expressed in absolute terms, in per capita terms, and as a percentage of its gross national product (GNP).
• Commercial relations between sender and target countries, measured by the flow of two-way trade between them expressed as a percentage of the target country's total two-way trade.
• The relative economic size of the countries, measured by the ratio of their GNPs.
• The type of sanctions used, namely, an interruption of exports from the sender country, an interruption of imports to the sender country, or an interruption of finance.
• The cost to the sender country, expressed as an index scaled from 1 (net gain to sender) to 4 (major loss).

General Conclusions

Sanctions have been successful--by our definition--in 34 percent of all cases. However, the success rate importantly depends on the type of policy or governmental change sought. Episodes involving destabilization succeeded in 52 percent of the cases, usually against target countries that were small and shaky. Cases involving modest goals and attempts to disrupt minor military adventures were successful 33 percent of the time. Efforts to impair a foreign adversary's military potential, or otherwise to change its policies in a major way, succeeded infrequently (see table 1).

Of course, some sanctions fail because they were never intended to succeed, in the sense of producing a real change in the target's behavior. As one analyst has noted, when sanctions have been used primarily for domestic political or other rhetorical purposes, "'effective' sanctions [in an instrumental sense] were not a primary policy goal, and such sanctions were not imposed." Sanctions may also be imposed timidly, and hence ineffectively, if conflicting goals are not weeded out.

Table 1 The Sanctions Record

(a) The figures include five instances of cases included under two different policy goals: 49-1: US v. China; 60-3: US v. Cuba; 63-1: US v. UAR; 63-3: US v. Indonesia; and 80-1: US v. USSR (Afghanistan). Since these cases are generally failures, double counting them adds a small negative bias to the success ratio.

Nine Lessions We have found that sanctions sometimes bear fruit, but only when planted in the right soil and nurtured in the proper way. We therefore offer nine propositions for the statesman who would act as a careful gardener.

I. "Don't Bite Off More Than You Can Chew."
Policymakers often have inflated expectations of what sanctions can accomplish. Sanctions are seldom effective in impairing the military potential of an important power, or in bringing about major changes in the policies of the target country. Of the 30 cases involving these "high" policy goals, success was achieved in only 7 (23 percent), and 4 of the 7 involved military conflict: the two world wars and two civil wars (between India and Hyderabad in 1948, and Nigeria and Biafra in the late 1960s).

II. "More Is Not Necessarily Merrier."
For the period studied here, ending just as the Cold War did, the greater the number of countries needed to implement sanctions, the less likely it was that they would be effective (table 2 summarizes our findings for the commandments that follow).

Without significant cooperation from its allies, a sender country stands little chance of achieving success in cases involving high policy goals. However, international cooperation does not guarantee success even in these cases, as evidenced from the long history of US and COCOM strategic controls against the Soviet Union and COMECON, and by the Arab League's largely futile boycott of Israel. When a sender country has found it necessary to seek cooperation from other countries, it was probably pursuing a sufficiently difficult objective that the prospects for ultimate success were not bright. It should also be noted that even when its goals were modest, the effectiveness of US unilateral sanctions has dropped sharply over the past 25 years (see below).)

On the other hand, active noncooperation can sabotage a sanctions effort. Offsetting assistance given to the target country by a third country erodes the chances of sender-country success, particularly in cases where the policy goal is destabilization of the target government or disruption of a military adventure. Such cases often occurred in the context of East-West rivalry. With the end of the Cold War, such "black knights" may be less likely to appear on the sanctions scene to rescue target countries.

III. "The Weakest Go to the Wall."
For our case sample as a whole, there seems to be a direct correlation between the political and economic health of the target country and its susceptibility to economic pressure. Countries in distress or experiencing significant problems are far more likely to succumb to coercion by the sender country.

This is most true of the destabilization cases, where successes generally came against weak regimes. The average health and stability index was also lower in successful than in failed cases when disruption of military adventures and other major policy changes were at stake. In episodes involving modest policy goals and impairment of military potential, the results based on the health and stability of the target country are less clear-cut--in the former set of cases because a wide variety of countries have been targeted, and in the latter because countries only attempt military impairment when the target is strong enough to be a threat.

Senders' economies are also typically much bigger than those of their targets. Size may be a necessary condition for success, but it is clearly not sufficient. The relative size of the target economy is less important than other factors that come into play such as the extent of trade linkage, the economic impact of the sanctions, and the warmth of relations between sender and target prior to the imposition of sanctions.

IV. "Attack Your Allies, Not Your Adversaries."
Economic sanctions seem most effective when aimed against erstwhile friends and close trading partners. In contrast, sanctions directed against target countries that have long been adversaries of the sender country, or against targets that have little trade with the sender country, are generally less successful.

The higher compliance with sanctions by allies and trading partners reflects their willingness to bend on specific issues in deference to an overall relationship with the sender country. However, the preservation of political alliances and economic ties should be equally important to prospective senders as to intended targets.

Likewise, the trade linkage data suggest that success is more often achieved when the target country conducts a significant portion of its trade with the sender. Overall, successful cases exhibit a higher average trade linkage (28 percent) than do failed cases (19 percent). One corollary: a sender country enhances its potential leverage by minimizing restrictions on trade well before the need for sanctions arises.

V. "If It Were Done, When 'Tis Done, Then 'Twere Well It Were Done Quickly."
A heavy, slow hand invites both evasion and the mobilization of domestic opinion in the target country. Sanctions imposed slowly or incrementally may simply strengthen the target government at home as it marshals the forces of nationalism. Moreover, such measures are likely to be undercut over time either by the sender's own firms or by foreign competitors. The average successful case lasted just under three years, while failures typically dragged on for eight years.

However, it is not the passage of time alone that undermines economic sanctions. Other factors are correlated with the length of an episode. Episodes between erstwhile allies are generally short, to the point, and often successful. Further, the target country is more likely to receive assistance from another major power if the episode continues for a number of years. Finally, the greater the latent likelihood of success, the shorter the sanctions period necessary to achieve results.

In any event, the inverse relationship between success and the duration of sanctions argues against a strategy of "turning the screws" on a target country, slowly applying more and more economic pressure over time until the target succumbs. Time affords the target the opportunity to adjust: to find alternative suppliers, to build new alliances, and to mobilize domestic opinion in support of its policies.

VI. "In For a Penny, In For a Pound."
Cases that inflict heavy costs on the target country are generally successful. The average cost to the target for all successful cases was nearly 2.5 percent of GNP; by contrast, failed episodes barely dented the economy of the target country, with costs averaging only 1 percent of GNP. The conclusion to be drawn from these findings is that if sanctions can be imposed in a comprehensive manner, the chances of success improve. Sanctions that bite are sanctions that work. However, there is a "black knight corollary" to this conclusion: sanctions that attract offsetting support from a major power may cost the target country little on a net basis and are less likely to succeed.

VII. "If You Need to Ask the Price, You Can't Afford the Yacht."

The more it costs a sender country to impose sanctions, the less likely it is that the sanctions will succeed. The average cost-to-sender index (scored from 1 to 4, with 1 representing a net gain and 4 a major loss to the sender), is generally lower in successful than in failed cases. The basic conclusion is clear: a country should shy away from deploying sanctions when the economic costs to itself are high. Countries that shoot themselves in the foot may not mortally wound their intended targets.

The sanctions episodes that are least costly to the sender are often those that make use of financial leverage--manipulating aid flows, denying official credits, or, at the extreme, freezing assets--rather than trade controls. Denial of finance may also compound the cost to the target country by inhibiting its ability to engage in trade even without formal trade controls being imposed. In fact, financial sanctions have been used alone more often and more effectively than trade controls alone.

When financial, export, and import controls are all used in a single episode, it is often because the goal is ambitious. A major reason for the better track record of financial sanctions alone is that they typically involve relatively modest goals, sought through the reduction, suspension, or termination of economic or military assistance flowing from richer nations (usually the United States) to smaller and poorer developing countries.

VIII. "Choose the Right Tool For the Job."
Economic sanctions are often deployed in conjunction with other measures, such as covert action, quasi military measures, or regular military operations. Companion measures are used most frequently in episodes involving destabilization and impairment of military potential. By contrast, companion policies are seldom used in cases involving modest policy changes, and were used in fewer than half the cases seeking disruption of military adventures or other major policy changes.

Though our findings reveal no correlation between the use of companion policies and increased effectiveness, this result is somewhat misleading. Our methodology only recognizes success in cases where sanctions made a positive contribution to the policy outcome. In several cases counted as failures--for example, the US sanctions against the Sandinistas in Nicaragua and against Noriega in Panama--the sender country achieved its goal but military or covert measures swamped the impact of the sanctions. It may also be unfair to say that sanctions "failed" in other cases--for example, the United States versus Grenada (Case 83-4)--where the military weapon was unsheathed before sanctions had been given a chance to work. Rather than buttressing a sanctions campaign, companion measures are frequently used when sanctions are perceived to be either wholly inadequate or simply too slow.

IX. "Look Before You Leap."
Sender governments should think through their means and objectives before taking a final decision to deploy sanctions. Leaders in the sender country should be confident that their goals are within their reach, that they can impose sufficient economic pain to command the attention of the target country, that their efforts will not prompt offsetting policies by other powers, and that the sanctions chosen will not impose insupportable costs on their domestic constituents and foreign allies. These conditions will arise only infrequently, and even then the odds are against success.

Sanctions imposed for symbolic purposes--for the benefit of allies or a domestic audience--should be just as carefully crafted. Although economic sanctions may be the best or even the only option in some cases where it is necessary to "do something," not just any sanction will do--the sanction chosen must be appropriate to the circumstances. Senders usually have multiple goals and targets in mind when they impose sanctions, and coercion is not always at the top of the list. Prudence argues that one carefully scrutinize unintended costs and consequences before choosing a particular measure. It makes sense to tailor sanctions carefully to the objective they are genuinely intended to achieve.

Prospects for the future Success in the use of sanctions has proved more elusive in recent years than in earlier decades. If one splits the case sample roughly in half, into those initiated before 1973 and those begun after that date, a striking difference emerges: 44 percent of the sanctions episodes in the pre-1973 period succeeded, whereas the success rate among the cases begun between 1973 and 1990 was just under 25 percent. Even more striking is the decline in the effectiveness of sanctions imposed in pursuit of modest goals, from 75 percent to 21 percent.

However, these trends need to be qualified: the increasing use of sanctions despite declining effectiveness can be attributed entirely to US experience. Other senders, including multilateral coalitions in which the United States played a relatively minor role, both reduced their reliance on sanctions and improved their record: from 10 successes in 28 attempts prior to 1973, to 6 out of 13 after that. In contrast, after posting a better than .500 average in the earlier period, the United States batted under .200 in the 1973-90 period.

Declining Success and Declining Hegemony

The most obvious and important explanation of the sharp decline in the effectiveness of US sanctions is the relative decline of the US position in the world economy. Unlike the early post-war era, the United States is no longer the major supplier of many goods and services, nor is it the only source of economic assistance for developing countries.

Since the 1960s, trade and financial patterns have grown far more diversified and new technology has spread more quickly. Recovery in Europe and the emergence of Japan have created new, competitive economic superpowers, and economic development has reduced the pool of potentially vulnerable targets. These trends are starkly illustrated by the declining average trade linkage between the United States and its targets (from 24 percent prior to 1973 to only 17 percent since), the lower costs imposed on targets (1.7 percent of GNP v. 0.9 percent of GNP), and the fading utility of manipulating aid flows. For example, the success rate for financial sanctions used alone (usually cases involving reduced aid to developing countries) declined from nearly 80 percent before 1973 to less than 20 percent after that.

The evidence from the cases suggests three other factors contributing to declining success. First, although détente allowed cases involving modest goals to multiply in the 1970s, concerns about Soviet influence or strategic position still claimed first priority in the strategic planning of the US government and frequently undermined the pursuit of less central goals. Also, while the goals in many cases may have been modest from the perspective of the United States, they were often of central importance to targeted regimes whose leaders believed that their survival depended on stifling domestic political opposition or keeping up with a regional rival thought to be pursuing a nuclear weapons option.

Second, in the past 25 years, Congress has increasingly forced the president's hand and constrained his discretion in various foreign policy situations by passing legislation requiring the use of economic sanctions. The confused signals sent by administrations that were forced to implement legislatively mandated sanctions may have led target countries to believe, often correctly, that the sanctions would not be sustained.

Finally, whereas financial measures were part of the sanctions package in more than 90 percent of episodes prior to 1973, they were present in only two-thirds of the cases after that. The favored type of financial sanction also changed. Economic aid was the dominant choice in the earlier period, whereas military assistance was prominent in the later period, especially in the human rights cases, where military governments were often the target. However, the amount of aid cut off in these cases was usually quite small; some cases also involved offsetting financial assistance from the Soviet Union or others.