Jan Paul Acton
Natural Resources and Commerce Division
Congressional Budget Office
Domestic Costs of Sanctions
on Foreign Commerce
Committee on International Relations
U.S. House of Representatives
June 3, 1998
Mr. Chairman and Members of the Committee, I am pleased to appear here today. My testimony is based on research that the Congressional Budget Office (CBO) has conducted at the Committee's request concerning the impact that sanctions on foreign commerce can have on the United States economy. Three principal findings have emerged from that research:
o Sanctions that restrict trade and foreign investment impose costs on the U.S. economy. Ultimately, those costs show up in the form of higher costs for U.S. consumers and businesses and lower national income. Although the immediate cost appears to be reduced output and employment in the industries directly affected by the sanctions, the actual net loss for the economy is always much smaller and more diffuse than that initial impact.
o To date, the cost of existing sanctions to the overall economy has been quite modest. CBO's review of research indicates that the net cost may be less than $1 billion annually of lost national income. That compares with $6.6 trillion of total national income in 1997. Gross exports may be lower by $15 billion to $19 billion annually as a result of sanctions, compared with nearly $1 trillion of total exports of goods and services in 1997.
o Although sanctions have not been very costly to the overall economy, they can result in sharp disruption to and dislocation of specific U.S. firms and workers. Those effects will generally be limited to the short run. In the long run, those firms and workers can shift to other productive uses, but the transition cost can be substantial in some cases.
The main reason that sanctions have had such a small effect on the overall U.S. economy is that the principal targets of sanctions have been countries with which the United States conducts relatively little trade. Large, industrialized countries are the dominant trading partners of the United States. They are nations that share many foreign policy and security concerns with the United States. The few sanctions that have touched business in those nations have had a negligible effect on the level of trade.
Multilateral sanctions can result in a greater loss of national income than unilateral sanctions because they can cut off trade more effectively. Yet without the full participation of other countries, sanctioned goods (or the inputs to sanctioned goods) can be rerouted through third countries, foiling the sanction. Similarly, unilateral sanctions on investment can lead the target countries to seek investors from third-party nations, who then turn to the United States to finance other projects.
The effects of sanctions have been small so far, but that does not mean they will not increase in the future. In general, analysts assume that the overall cost that sanctions impose on the U.S. economy will grow along with the volume of trade. Future costs will also depend on the type of economies (and hence, the composition of commerce) targeted and the participation of other nations.
Sanctions are one of several policy options for influencing the behavior of foreign governments. CBO's analysis does not evaluate the cost of other options, such as military action, or the effectiveness of any of the options. Although it is important for the Congress and the Administration to consider the benefits and costs of various approaches for achieving foreign policy objectives, it is beyond the scope of either CBO's research or this testimony to discuss those alternatives.
BACKGROUNDThe federal government uses sanctions on foreign commerce to influence the activities of foreign governments. Sanctions have included restrictions or prohibitions on exports and imports, private foreign investment, foreign assistance, and lending by international financial institutions. The goal of sanctions is to promote particular foreign policies or enhance national security. Some believe that the Congress and the executive branch are resorting more and more to the use of sanctions in the conduct of foreign policy. The Congressional Research Service recently identified more than 130 separate statutes that restrict some aspect of foreign commerce for foreign policy reasons.
It is easier to tally the number of statutes that impose sanctions than it is to identify the number of sanctions in place and whether they actually affect trade. Sanctions might be announced by the Congress or the Administration, for example, but be subject to delays or exemptions that mute their effect. Some sanctions might overlap existing restrictions or codify executive actions already in place. Even defining a sanction is subject to argument. Some policymakers would consider eliminating U.S. assistance a sanction, whereas others would not. Though it is hard to know exactly the number of sanctions or their effect, it is clear that sanctions--including restrictions on assistance--have become an important tool of U.S. foreign policy.
Regardless of their impact today, sanctions could become more costly for the United States over time. First, new sanctions could affect future trade opportunities by extending restrictions to additional countries and commodities. Second, existing sanctions on fast-growing economies, such as China, could affect future trade opportunities by excluding the United States from those markets and, if the sanctions actually harm the target's economy, denying the United States the indirect benefits of economic growth in unsanctioned markets. And third, the extraterritorial features of certain sanctions on small economies could have spillover effects on some of the United States' major trading partners. The cost of sanctions could spiral upward if those trading partners chose to retaliate with sanctions on the United States or to withdraw from other trade negotiations.
SUMMARY OF CBO'S RESEARCHFree trade benefits the economy. Trade expands the size of the market and allows domestic resources to be used where they can be most productive, thus increasing overall U.S. output and profits. Sanctions disrupt trade and cost the economy by causing domestic resources to move to less productive activities in which profit levels and wages are lower. The net loss to the economy is much smaller than the initial loss of trade because of that movement of resources.
The parties most directly affected by sanctions are U.S. businesses and workers producing and using goods and services that trade internationally and U.S. consumers of those goods and services. In the long term, the effects of reduced trade are widely diffused throughout the economy.
Several features of the trade that is subject to sanctions can reduce the economic costs of those sanctions. The negative effects of sanctions on U.S. consumers are blunted by the availability of substitute sources of goods and services, both domestically and internationally. Similarly, the availability of alternative investment opportunities means that sanctions that restrict particular foreign investments will result in only an incremental loss of return on investment. U.S. foreign assistance programs subsidize U.S. exports; ending those subsidies and the taxes that pay for them could benefit U.S. consumers.
The difficulties in totally restricting trade and investment with any particular country also diminish the cost of sanctions, especially when the United States acts unilaterally--that is, without the participation of other nations. For example, as a result of sanctions on exports, foreign competitors will most likely pick up the business that U.S. exporters lose; but those third-party countries may be unable to increase production without shifting resources away from other markets (for the sanctioned goods or unrelated goods). U.S. businesses can then move into those new markets with their excess capacity as a result of the sanctions. Such shifting of trade flows can cost the economy, but far less than the gross value of trade involved.
Despite data that indicate that the overall cost of sanctions to the U.S. economy may be small, the cost to individual groups of consumers, workers, and business owners can be significant. The fact that other sectors of the economy benefit from changes in trade patterns will be of little consolation to the losers. Workers in industries that lose sales may face job loss and the need to relocate or learn new skills. The losses can extend to businesses that support the affected industries and local economies. However, most of the cost will be transitory if individuals and resources can shift into other economic activities.
In its research, CBO has reviewed existing studies to identify a range of the most likely costs of sanctions to restrict a given amount of trade under various circumstances. The cost of sanctions depends on three factors: whether the sanctions are unilateral or multilateral; the type of goods or services being sanctioned (for example, agricultural commodities or high-technology products); and the volume of trade affected. Of the three, the amount of trade is probably the most important. Thus, the overall cost to the U.S. economy would generally be:
o Small for sanctions on small, developing economies, which currently account for little U.S. trade;
o Medium for sanctions on big, emerging economies, such as China, which are likely to account for an important share of trade in the future; and
o Large for sanctions on industrialized economies, which already account for significant U.S. trade.
The lowest cost would be for a unilateral sanction imposed against a small, developing economy. Countries in that category--in Latin America, Africa, and Eastern Europe--in total buy about 15 percent of U.S. exports. Some indication of the cost of disrupting trade with those countries comes from a study by Gary Hufbauer and others at the Institute for International Economics that looked at the effects of all current sanctions. Current sanctions disproportionately target developing economies, and many of those sanctions--but not all--represent unilateral actions by the United States. The results of the Hufbauer study are consistent with a national income loss of 5 cents for each $1 of decrease in exports because of sanctions.
The highest cost to the economy may be from multilateral and comprehensive sanctions. To gain insight into what such actions might cost, CBO looked at research on the gains from trade liberalization--research that modeled the effects of worldwide initiatives to reduce trade barriers, such as the Uruguay Round. If trade contractions produce costs that are symmetric with those gains, research indicates that multilateral trade sanctions may decrease national income by between 15 cents and 35 cents for each $1 decrease in U.S. exports. Those estimates probably overstate costs because multilateral sanctions applied to a single country are different from a general contraction of trade.
The income lost from cutting off trade with a big, emerging economy, such as China, would be likely to fall somewhere between the two--from 5 cents to 35 cents of income for each $1 drop in trade. Trade with fast-growing regions such as Mexico, China, and the newly industrialized economies of Asia (Hong Kong, Korea, Singapore, and Taiwan) already accounts for about 20 percent of U.S. exports.
As those figures indicate, the direct cost of sanctions for particular U.S. export industries greatly overstates the overall loss to the nation. It can be helpful to keep the reasons in mind. At one level, exports represent payments for imports. A disruption of exports will be offset by some combination of a rise in other exports and a drop in imports. To the extent that total exports and imports do not match, the resulting trade imbalance is influenced more by the imperatives of national saving and investment than by trade policy. Net imports means the country is not saving enough to pay for investment.
THE RECENT SANCTIONS AGAINST INDIA AND PAKISTANThe recent events in India and Pakistan provide a specific example of many of the important features of using sanctions to attempt to influence another country's policies. On May 11 and May 13, India detonated five nuclear devices. Pakistan responded about two weeks later with its own tests. U.S. law requires the imposition of sanctions against countries that conduct such tests. President Clinton announced the immediate application of sanctions against each country under section 102(b) of the Arms Export Control Act as amended, without invoking the 30-day delay that the act permits. Although the full extent and details of the sanctions are not yet known (such as which U.S. exports to the countries will fall under the sanctions), and there has been little time to analyze them, several significant things can be said.
Exports of goods to India in 1996 were $3.3 billion. That was about 0.5 percent of all U.S. exports and 0.04 percent of U.S. gross domestic product (GDP). Even if the sanctions ended all U.S. exports to India--and they will not--the cost to the United States would not be high. Using the formula of 5 cents to 35 cents for each $1 of reduction in exports, the cost would be between $0.17 billion and $1.16 billion, or between 0.002 percent and 0.015 percent of U.S. GDP.
The cost to the United States of the trade sanctions on Pakistan would be even lower. Exports of goods to Pakistan in 1996 were $1.3 billion, about 0.2 percent of all U.S. exports and 0.02 percent of U.S. GDP. Again, if one assumes a cost of between 5 cents and 35 cents for each $1 reduction in exports, even completely eliminating all U.S. exports to Pakistan would cost only $64 million to $447 million, or between 0.0008 percent and 0.0059 percent of U.S. GDP. Those low numbers and the corresponding numbers for India illustrate the generalization that most U.S. sanctions cost little because they do not affect much trade.
Any contraction of trade with those countries would have a small effect on the U.S. economy. U.S. imports from India in 1996 were $6.5 billion, and imports from Pakistan were $1.3 billion. Both values are a very small proportion of the $817.8 billion in total U.S. imports. Goods imported from India and Pakistan are also available from other sources--foreign and domestic--and the cost to U.S. business and consumers most likely will be modest.
The sanctions also withhold nonhumanitarian aid to India and Pakistan. In the case of Pakistan, there is little aid to eliminate because almost all of the aid to Pakistan was eliminated after 1990 as a result of its nuclear program. That fact illustrates the generalization that many U.S. sanctions have neither effect nor cost because they are imposed on goods or investments that are already being sanctioned.
The sanction package also includes voting against loans to the two countries from international financial institutions (IFI) such as the World Bank and the International Monetary Fund and halting U.S. commercial loans to the two countries' governments, except to finance purchases of food and agricultural commodities. Ending such loans from international financial institutions would impose costs on the United States only to the extent that it would cause net exports from the United States to fall, which is not likely to happen. CBO has not determined the significance of the halt in commercial loans.
Although CBO's analysis has concentrated on the effects of sanctions on the U.S. economy and not on the target country's economy, some effects may rebound through target countries to the United States. Imports from the United States constitute 8.8 percent of India's total imports of goods, and constitute 0.88 percent of India's GDP. Since most other countries have not announced trade sanctions against India and not all U.S. trade will be sanctioned, that restriction is likely to have very little effect on India. U.S. exports of goods to Pakistan account for 10.6 percent of that country's total imports and are equal to 2.14 percent of Pakistan's GDP. More important for Pakistan, that country is struggling to meet its current obligations to international lenders. If sanctions contribute to a default in Pakistan's servicing of international debt, they could exacerbate economic troubles in a part of the world that is already experiencing volatility. If the sanctions fueled the "Asian turmoil" or prolonged the recovery in that region, they could have wider economic effects, including negative effects on the United States. It is not possible at present to judge either the likelihood of such a development or its cost to the U.S. economy.
CONCLUSIONConsideration of the cost of sanctions, whether large or small, would most likely be only one part of any decision to impose sanctions. Other actions the government may take to address foreign policy or security concerns will have costs, too. CBO's analysis of income loss attributable to sanctions may provide the Congress with the ability to compare the cost of alternative actions.