Economic sanctions are fast becoming the United States' policy tool of choice. A 1997 study by the National Association of Manufacturers listed 35 countries targeted by new American sanctions from 1993 to 1996 alone. What is noteworthy, however, is not just the frequency with which sanctions are used but their centrality; economic sanctions are increasingly at the core of U.S. foreign policy.

Sanctions—predominantly economic but also political and military penalties aimed at states or other entities so as to alter unacceptable political or military behavior—are employed for a wide range of purposes. The United States, far more than any other country, uses them to discourage the proliferation of weapons of mass destruction and ballistic missiles, promote human rights, end support for terrorism, thwart drug trafficking, discourage armed aggression, protect the environment, and oust governments. To accomplish these ends, sanctions may take the form of arms embargoes, foreign assistance reductions and cutoffs, export and import limitations, asset freezes, tariff increases, import quota decreases, revocation of most favored nation (MFN) trade status, votes in international organizations, withdrawal of diplomatic relations, visa denials, cancellation of air links, and credit, financing, and investment prohibitions. Even U.S. state and local governments are introducing economic sanctions. Dozens have adopted "selective purchasing laws" that prohibit public agencies from purchasing goods and services from companies doing business with such countries as Burma and Indonesia.

With a few exceptions, the growing use of economic sanctions to promote foreign policy objectives is deplorable. This is not simply because sanctions are expensive, although they are. Nor is it strictly a matter of whether sanctions "work"; the answer to that question invariably depends on how demanding a task is set for a particular sanction. Rather, the problem with economic sanctions is that they frequently contribute little to American foreign policy goals while being costly and even counterproductive. A recent study by the Institute for International Economics concluded that in 1995 alone, sanctions cost U.S. companies between $15 billion and $19 billion and affected some 200,000 workers. Secondary sanctions, leveled against third-party states that do not support a particular sanctions regime, add to this cost by jeopardizing the United States' trade relations. Thus, policymakers need to give more serious consideration to the impact of a sanction and weigh alternative policies more carefully.


Economic sanctions are popular because they offer what appears to be a proportional response to challenges in which the interests at stake are less than vital. They are also a form of expression, a way to signal official displeasure with a behavior or action. They thus satisfy a domestic political need to do something and reinforce a commitment to a norm, such as respect for human rights or opposition to weapons proliferation. Reluctance to use military force is another motivation. As the National Conference of Catholic Bishops points out, "Sanctions can offer a nonmilitary alternative to the terrible options of war or indifference when confronted with aggression or injustice."

The frequency with which the United States uses sanctions is also a result of the increased influence, especially on Congress, of single-issue constituencies, notably those promoting human rights, environmentalism, or ethnic, religious, or racially oriented causes. The media, too, plays a part. The so-called CNN effect can increase the visibility of problems in another country and stimulate Americans' desire to respond. Sanctions offer a popular and seemingly cost-free way of acting. The end of the Cold War and the demise of the Soviet Union have also contributed to the sanctions boom. Sanctions can now usually be introduced without opposition from Moscow, which in the past meant a veto in the U.N. Security Council or a Soviet subsidy for a target of U.S. sanctions.

Some evidence supports the efficacy of economic sanctions. One influential study concludes from analysis of more than 100 cases that economic sanctions have worked to some extent about a third of the time. Other advocates are more selective in their views of history. For groups on the left, it is an article of faith that sanctions helped dismantle apartheid, just as the right argues that sanctions played a major role in the demise of the "evil empire."

Under the right circumstances, sanctions can achieve, or help achieve, various foreign policy goals ranging from the modest to the fairly significant. Sanctions introduced against Iraq after the Persian Gulf War have increased Iraqi compliance with U.N. resolutions calling for the elimination of its weapons of mass destruction. They have also diminished Baghdad's ability to import weapons and related technology. Iraq today is considerably weaker militarily and economically than it would have been without these sanctions.

Sanctions were one reason for the Serbs' decision to accept the Dayton agreement in August 1995 ending the fighting in Bosnia. The threat of sanctions may have also deterred several European firms from investing in Iran's oil and gas industry. Sanctions have burdened the economies of Iran, Cuba, and Libya, and may eventually contribute to change in those societies or in their behavior. U.S. sanctions against Pakistan, while having little discernible effect on that country's nuclear weapons program, have hurt Islamabad both economically and militarily, possibly influencing Pakistan's future actions as well as those of other would-be proliferators.


The limitations of sanctions are more pronounced than their accomplishments. Sanctions alone are unlikely to achieve results if the aims are large or time is short. Even though they were comprehensive and enjoyed almost universal international backing for nearly six months, sanctions failed to compel Saddam Hussein to withdraw from Kuwait in 1990. In the end, it took nothing less than Operation Desert Storm.

Other sanctions have also fallen short of their stated purposes. Despite sanctions against Iran, Tehran remains defiant in its support of terrorism, its subversion of its neighbors, its opposition to the Middle East peace process, and its pursuit of nuclear weapons. Fidel Castro still commands an authoritarian political system and a statist economy. Pakistan's nuclear program is well advanced; it now has enough material for at least a dozen bombs. Libya has refused to hand over the two individuals accused of destroying Pan Am Flight 103 over Lockerbie, Scotland. Sanctions did not persuade Haiti's junta to honor the results of the 1990 election that brought Jean Bertrand Aristide to power, nor did they convince Serbia and the Bosnian Serbs for several years to call off their military aggression. Unilateral sanctions are particularly ineffective. In a global economy, unilateral sanctions impose higher costs on American firms than on the target country, which can usually find substitute sources of supply and financing. Unilateral sanctions did, however, prove more costly for Haiti and Cuba, which were heavily dependent on trade with the United States. They also hurt Pakistan, which had been receiving substantial U.S. military and economic aid. Such cases are the exception, though; most unilateral sanctions will be little more than costly expressions of opposition except in those instances in which the ties between the United States and the target are so extensive that the latter cannot adjust to an American cutoff.

Generating international support for sanctions is often extremely difficult. In most instances, other governments prefer minimal sanctions, or none at all. They tend to value commercial interaction more than the United States does and are less willing to forfeit it. In addition, the argument that economic interaction is desirable because it promotes more open political and economic systems normally has more resonance in other capitals, although it has been used successfully by both the Bush and Clinton administrations to defeat Congress' attempts to revoke China's MFN status. Such thinking makes achieving multilateral support for sanctions more difficult for the United States. It usually takes something truly egregious, like Saddam Hussein's occupation of Kuwait, to overcome this anti-sanctions bias. Even with Iraq, generous compensation for third-party states affected by the sanctions, including Egypt and Turkey, was a prerequisite for their support.

Trying to compel others to join a sanctions regime by threatening secondary sanctions can seriously harm U.S. foreign policy interests. Congress is increasingly turning to secondary sanctions to bolster ineffective unilateral sanctions regimes, as with Cuba, Iran, and Libya; in all three instances, sanctions now apply to overseas firms that violate the terms of U.S. legislation like the Iran-Libya Sanctions Act and Helms-Burton Act. This threat appears to have deterred some individuals and firms from entering into proscribed business activities, but it has increased anti-American sentiment, threatened the future of the World Trade Organization (WTO), distracted attention from the provocative behavior of the target governments, and made Europeans less likely to work with the United States in shaping policies to contend with post-Cold War challenges.


Sanctions often produce unintended and undesirable consequences. Haiti is a prime example. Sanctions exacerbated the island's economic distress, causing a massive exodus of Haitians to the United States that proved life-threatening for them and expensive and disruptive for Florida. In Bosnia, the arms embargo weakened the Muslims, since Bosnia's Serbs and Croats had larger stores of military supplies and greater access to outside sources. This military imbalance contributed to the fighting and to the disproportionate Muslim suffering. Military sanctions against Pakistan may actually have increased Islamabad's reliance on a nuclear option because they cut off its access to U.S. weaponry and dramatically weakened Pakistan's confidence in Washington.

All this demonstrates that sanctions can be a blunt instrument. Most sanctions do not discriminate within the target country. There is a rationale for this: funds and goods can easily be moved around, and governments can often command what is in the hands of others. The problem with such a broad-brush approach is that sanctions tend to affect the general population, while those in the government and the military are able to skirt the sanctions.

Thus, the tendency to see economic sanctions as "below" the use of military force on some imagined ladder of foreign policy escalation must be revised. Sanctions can be a powerful and deadly form of intervention. The danger inherent in broad sanctions—beyond missing the true target—is both moral, in that innocents are affected, and practical, in that sanctions that harm the general population can bring about undesired effects, including strengthening the regime, triggering large-scale emigration, and retarding the emergence of a middle class and a civil society. Mass hardship can also weaken domestic and international support for sanctions, as with Iraq, despite the fact that those sanctions have included from the outset a provision allowing Iraq to import humanitarian goods and services.


"Smart" or "designer" sanctions, which penalize leaders while sparing the general public, are only a partial solution. It is possible that Haiti's military leaders were bothered by the fact their families could no longer shop in Florida. And executives who risk being denied access to the United States under the 1996 Helms-Burton act may think twice before entering into proscribed business deals. But opportunities to employ effective sanctions with precision are rare. Gathering the necessary information about assets, and then moving quickly enough to freeze them, can often prove impossible. Leaders and governments have many ways to insulate themselves, and designing "smart" sanctions to target only them is extraordinarily difficult, especially with a totalitarian or authoritarian state run by a few people.

In addition, authoritarian, statist societies are often able to hunker down and withstand the effects of sanctions. There are several possible reasons: sanctions sometimes trigger a "rally around the flag" nationalist reaction; by creating scarcity, they enable governments to better control the distribution of goods; and they create a general sense of siege that governments can exploit to maintain political control. This conclusion is consistent with literature suggesting that market economic reform reinforces the development of civil society; by reducing the scope for independent action, sanctions can work against forces promoting political pluralism.

Last, but far from least, sanctions can be expensive for American business. There is a tendency to overlook or underestimate the direct costs of sanctions, perhaps because, unlike the costs of military intervention, they do not show up in U.S. government budget tables. Sanctions do, however, affect the economy by reducing revenues of U.S. companies and individuals. Moreover, this cost is difficult to measure because it includes not only lost sales but also forfeited opportunities: governments and overseas companies can elect not to do business with the United States for fear that sanctions might one day be introduced, wreaking havoc with normal commercial relations.


A fundamental change in thinking and attitude is required. Economic sanctions are a serious instrument of foreign policy and should be employed only after consideration no less rigorous than for other forms of intervention, including the use of military force. The likely benefits of a particular sanction to U.S. foreign policy should be greater than the anticipated costs to the U.S. government and the American economy. Moreover, the sanction's likely effect on U.S. interests should compare favorably to the projected consequences of all other options, including military intervention, covert action, public and private diplomacy, or simply doing nothing. Broad sanctions should not be used as a means of expression. Foreign policy is not therapy; its purpose is not to make us feel good but to do good. The same holds for sanctions.

For pragmatic more than normative reasons, multilateral support for economic sanctions should typically be a prerequisite for the United States' imposition of them. Such support need not be simultaneous, but it should be all but certain to follow. Except when the United States is in a unique position to exert leverage based on its economic relationship with the target, unilateral sanctions should be avoided. Building international support will require intense, often high-level diplomatic efforts and even then may not succeed. Policymakers must then consider whether some alternative would not be better than weaker or unilateral sanctions.

International compliance with sanctions regimes can be increased by providing assistance to third parties to offset the economic cost of implementing sanctions. Greater use should be made of Article 50 of the U.N. Charter, which allows such states to approach the Security Council for redress. In addition, a fund for this purpose should be established within the U.S. foreign assistance budget. The cost would be more than offset by the benefits of multilateral cooperation.

By contrast, secondary sanctions are not a desirable means of securing multilateral support. They are not only an admission of diplomatic failure but they are also expensive. The costs to U.S. foreign policy, including the damage to relations with major partners and U.S. efforts to build an effective WTO, almost always outweigh the potential benefits of coercing unwilling friends to join sanctions regimes.

Sanctions should focus, as far as possible, on those responsible for the offending behavior and on limiting penalties to the particular area of dispute. Such limited sanctions would avoid jeopardizing other interests or an entire bilateral relationship. They would cause less collateral damage to innocents, and make it easier to garner multinational support. Sanctions designed to stem the proliferation of weapons of mass destruction are a prime example. Where there are transgressions, the United States should direct any sanctions toward nuclear or weapons-related activity, for example by cutting off associated technological cooperation or trade. Similarly, political responses such as event boycotts and visa denials might be the best way to signal opposition to objectionable behavior when no appropriate economic or military sanction is available or as a complement to something as specific as freezing an individual's assets. Political sanctions should not, however, extend to breaking diplomatic relations or canceling high-level meetings. Such interactions help the United States as much as the targeted party.

Sanctions should not hold major or complex bilateral relationships hostage to one or two issues. This is especially true with a country like China, where the United States has to balance interests that include maintaining stability on the Korean peninsula, discouraging any support for "rogue" states' weapons of mass destruction or ballistic missile programs, managing the Taiwan-China situation, and promoting trade, market reform, and human rights. Similarly, the United States has a range of interests with Pakistan that go well beyond nuclear matters, including promoting democracy, economic development, and regional stability. The principal alternative to broad sanctions in such instances is sanctions that are narrow and germane to the issue at hand. With Pakistan, for example, sanctions could focus on specific defense articles and technologies but exempt all economic assistance and military education and training.

Humanitarian exceptions should be part of any comprehensive sanctions regime. In part this is a moral judgment, that innocents should not be made to suffer any more than is absolutely necessary. But it is also pragmatic, since it is easier to generate and sustain domestic and international support for sanctions that allow the importation of food and medicine. Sanctions, however, should not necessarily be suspended if the humanitarian harm is the direct result of cynical government policy, such as Iraq's, that creates shortages among the general population in order to garner international sympathy.

Any imposition of sanctions should be swift. As with other forms of intervention, including military action, gradual escalation allows the target to adapt and adjust. Such an approach forfeits shock value and allows asset shifting, hoarding, and other arrangements to circumvent sanctions—as Libya and Iran found. This recommendation is easier said than done, since gaining international support for sanctions will in many cases require that the United States move slowly and gradually, further limiting the potential effectiveness of economic sanctions in today's world.


Restraint will not materialize by itself. Policymakers should be required to prepare and send to Congress a policy statement similar to the reports prepared and forwarded under the 1973 War Powers Act before or soon after a sanction is put in place. Such statements should clearly explain the sanction's purpose, the legal or political authority supporting its use, the expected impact on the target, retaliatory steps the target or third parties may take, the probable humanitarian consequences and what is being done to minimize them, the expected costs to the United States, prospects for enforcement, the expected degree of international support or opposition, and an exit strategy, including the criteria for lifting the sanction. In addition, policymakers should be able to explain why a particular sanction was selected over other sanctions or policies. If necessary, portions of this report could be classified to avoid providing information that would be useful to the target. Any sanction Congress initiates should be approved only after the relevant committees have carefully considered the matter; members being asked to vote on the proposal would then be able to refer to a report that addresses their questions.

All sanctions embedded in legislation should allow the president to suspend or terminate them in the interest of national security. Beyond being consistent with the Constitution's bias in favor of executive primacy in foreign affairs, such latitude is needed if relationships are not to fall hostage to one interest and if the executive is to have the flexibility to explore whether limited incentives could bring about a desired policy result. The benefits of this latitude outweigh any diminution of automatic sanctions' deterrent power. Current legislation that mandates sanctions in specific circumstances should be repealed or modified.

The federal government, together with affected firms, should challenge in court states' and municipalities' right to institute economic sanctions against companies and individuals operating in their jurisdiction. This practice is eliciting protests not just from the targeted countries but from the European Union, which argues convincingly that such sanctions violate commitments made by the U.S. government to the World Trade Organization. The Constitution may not settle the struggle between the executive and legislative branches in foreign affairs, but it limits it to the federal branch of government. State and local sanctions undermine the flexibility necessary for the executive branch to effectively carry out foreign policy. To paraphrase Justice Louis Brandeis, states may be laboratories of democracy, but not of diplomacy. Unfortunately, the Clinton administration—like the Reagan administration, which never challenged the more than 100 state and local sanctions targeting firms involved with South Africa—has chosen not to confront this issue. Beyond using the legal system, companies might consider deploying their economic power and avoid investing in states that have a history of supporting sanctions. Firms would also be wise to build broader coalitions (including labor unions) that would have a stake in opposing certain state and local sanctions.

Any sanction should be the subject of an annual impact statement, prepared by the executive branch and submitted in unclassified form to Congress, which would provide far more information and analysis than the pro forma documents written to justify many current sanctions. Like the report that would accompany a new sanction, the annual statement would introduce much-needed rigor into the sanctions decision-making process. A more careful calculation of economic costs would also provide a basis for determining payments to workers and companies being asked to bear a disproportionate share of the sanctions burden. Such seriousness has not been the hallmark of the American embrace of sanctions, which are often imposed and maintained with only cursory analysis of likely or actual effects.

The consequences of what is recommended here—less frequent and more modest use of economic sanctions—would risk creating something of a policy vacuum. In Washington it is difficult to beat something with nothing. So how does one beat economic sanctions?

Sometimes military force will be required. This was the lesson of Desert Storm and Bosnia. In Cuba, for example, instead of tightening sanctions—which increased the misery of the Cuban people—and going along with Congress' introduction of secondary sanctions against U.S. allies, the Clinton administration might have been wiser to launch a cruise missile salvo or use stealth aircraft to take out the MiGs that in 1996 shot down the unarmed plane flown by Cuban exiles from the group Brothers to the Rescue.

In other instances, focused sanctions could be useful. A more appropriate response to Pakistan's nuclear program would have been export controls designed to slow missile and nuclear bomb development. With Haiti, narrow sanctions aimed at the illegitimate leadership would not have triggered the human exodus that pressured the administration into an armed intervention that could well have proved far more costly than it did. China was stung by the U.S. decision to oppose Beijing's bid to host the Olympic games in the year 2000 and is bothered by being singled out in various international bodies for its treatment of its citizens.

The principal alternative to economic sanctions is best described as constructive or conditional engagement. Such an approach, involving a mix of narrow sanctions and limited political and economic interactions that are conditioned on specified behavioral changes, might be preferable, especially if the goal is to weaken the near-monopoly of an authoritarian leadership over a country like Cuba, Iran, or China. Such an approach is not as simple as imposing economic sanctions; nor does it yield as dramatic a sound bite. Its principal advantage is that it might have a more desirable impact at less cost to Americans and American foreign policy.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • RICHARD N. HAASS is Director of Foreign Policy Studies at the Brookings Institution. He was Chairman of a Council on Foreign Relations study group on economic sanctions. This article is adapted from his forthcoming book, Economic Sanctions and American Diplomacy.
  • More By Richard Haass