Skip Navigation

Antidumping Action in the United States and Around the World: An Analysis of International Data June 1998

SUMMARY

Dumping is the selling of an import at a price below its cost of production or below the price at which the manufacturer sells the good in its own domestic market. U.S. antidumping law views such imports as being sold at less than fair value. Under the law, duties are imposed on dumped imports that cause "material injury" to the competing U.S. industry. Almost any injury that is not negligible is considered to be material. The duties are set equal to the difference between the market price and the administratively determined fair value. Many other countries have similar laws.

The first U.S. antidumping law was very similar to a prohibition on predatory pricing of imports. Predatory pricing is the intentional selling of a good at a price below the cost of production with the intent of driving competitors out of business and increasing the market power of the predatory firm, allowing the firm to subsequently increase its prices above competitive market levels and thereby increase profits. The economic conditions under which predatory pricing can be successful and profitable are relatively rare, however. Consequently, the first antidumping law, which is still in effect, has never received much use.

Antidumping cases today are generally brought under another law with a more expansive definition of dumping. Under that law, no attempt is made to determine whether the pricing at issue is predatory or even whether successful predatory pricing is possible in the case at hand. All that is required to have duties imposed is a finding that the good has been sold below cost or below the price in the home market and that material injury has resulted. The vast majority of cases in which antidumping duties are imposed do not involve predatory pricing.

The change in the pricing behavior targeted by antidumping law is important. Predatory pricing is detrimental not only to the competing domestic industry but also to the economy as a whole. In cases for which predatory pricing is not an issue, however, imports priced below cost or their foreign price are generally beneficial to the economy. Thus, the original law was beneficial to both the competing domestic industry and the economy as a whole, whereas the most frequently used current law helps the competing domestic industry but hurts the economy as a whole. U.S. law places no restrictions on the pricing behavior of domestic firms in the U.S. market that are comparable with those placed on foreign firms by the antidumping law. For those and other reasons (not the least of which are charges of bias in U.S. administrative procedures and methodologies), antidumping law has been a continuing center of controversy and deliberations in multilateral trade negotiations and in the Congress.

In such deliberations, it is useful to know how U.S. antidumping practices compare with those of other countries and how the practices of other countries--especially those of the major U.S. trading partners--affect U.S. firms. The usefulness of such knowledge is demonstrated by many of the claims of participants in the public debate. For example, critics say that the United States is the foremost user of antidumping laws, that other countries are following the U.S. lead and are beginning to make more use of such laws themselves, which is hurting U.S. exporters. Some further argue that countries are aiming their use of antidumping laws at U.S. firms in retaliation for U.S. use against the firms in their own country. Many critics say that the use of antidumping laws has been increasing around the world as other protectionist practices have been systematically and progressively prohibited to more and more countries over the years by the General Agreement on Tariffs and Trade (GATT) and, subsequently, the World Trade Organization (WTO).

The claims are usually supported by at most a cursory reference to statistics to back them up, most likely because such statistics are difficult to come by. The best sources of the data needed for deriving such statistics are the semiannual reports made by signatories to the Antidumping Codes of the GATT and the WTO. Drawing summary statistics from those reports is difficult and time consuming for several reasons: they are not in a readily usable computer format, information about each antidumping case is scattered among several tables in several reports, various countries have failed at one time or another to file reports for given reporting periods, and the reports have many errors and omissions.

The Data Set and Its Limitations

To throw some light on some of the major claims and relevant factors in the debate over U.S. policy, the Congressional Budget Office (CBO) has taken information from the GATT/WTO reports to construct a usable computer database of the antidumping cases, duties, and other measures of the United States and most of the countries with which the United States conducts a significant volume of international trade. To the extent feasible, errors and omissions in the reports have been corrected. The database extends from July 1979 through December 1995, with more countries covered in later portions of that interval than in earlier portions. Using the database, CBO has calculated and analyzed statistics relevant to the claims and issues surrounding antidumping practices.

Aggregate antidumping statistics are imperfect indicators of the economic significance of antidumping activity because no two cases are identical. Even if the same antidumping duty rate is imposed in two cases, the cases may involve different products, different quantities of imports, different source countries, and so on. Furthermore, one country might have a tendency to bring cases against more narrowly defined products than another country. In that case, a larger number of cases by the former could have a milder economic effect than a smaller number by the latter. Despite those qualifications, however, the statistics provide useful information that can be used to draw a number of important conclusions.

The Prevalence and Significance of Antidumping Activity

Only a few countries make significant use of antidumping laws, and the United States is the most active user among them. Over three-quarters of U.S. exports from 1991 through 1995 went to countries that averaged fewer than half as many antidumping case initiations as did the United States. On December 31, 1995, it had 294 antidumping measures in effect; no other country had half that many. Under a reasonable set of assumptions about the import market, the large quantities of U.S. imports and the large U.S. gross domestic product do not explain why U.S. antidumping activity is so much greater than that of other countries. U.S. antidumping activity against other countries is much greater than their antidumping activity against the United States, both one on one and in the aggregate.

Antidumping duty rates are high enough to be significant impediments to trade, especially the duties imposed by the United States and a few small, mostly developing, countries. The average rate imposed by the United States from 1991 through 1995 was 56.8 percent. With the exception of Mexico, the most active users of antidumping laws impose substantially lower average rates of duty than does the United States, although their rates are still high enough to be significant impediments to trade. Among the most active users, Canada had the next highest average rate--36.1 percent. The United States progressively and substantially increased the initial duty rates it imposed over the years covered by the data set. The average initial rate imposed from 1993 through 1995 was almost triple the average from 1981 through 1983.

U.S. antidumping measures tend to last much longer than those imposed by any other country, and a large fraction of them last so long as to be effectively permanent--10.6 years on average. That difference in longevity at least in part reflects the fact that a number of other countries have had provisions for automatically reviewing and sunsetting their antidumping measures whereas the United States has not. The new WTO agreement requires reviewing and sunsetting. That requirement did not become effective immediately, however, and as of the end of 1995, it had not yet had any effect on the statistics for the duration of U.S. measures. CBO cannot say how much effect the requirement will have.

The United States tends to impose the most antidumping measures on the countries that export the largest quantities of goods to the United States. It also tends to impose measures on developing countries and countries that have (or recently had) nonmarket economies.

The Increasing Use of Antidumping Laws Around the World

Statistics from the data set are consistent with the claim that the United States has been a leader in the aggressive enforcement of antidumping laws, and they lend some credence to fears that U.S. policy may be starting to come back to haunt U.S. exporters as other countries follow its lead. The statistics also appear to be broadly consistent with the notion that most countries use antidumping enforcement as a substitute for other means of protecting their domestic industries from international competition and that antidumping enforcement is consequently rising as the GATT/WTO increasingly proscribes more countries from using those other forms of protection.

Almost alone among industrialized countries, the United States has increased its antidumping activity fairly consistently and substantially throughout the 16 years covered by the data set. In recent years, increased antidumping activity has been spreading among developing countries, in which such activity has historically been least prevalent. Most industrialized countries have not increased their activity (and some have decreased it), but many of them were already large users of antidumping laws at the beginning of the time period covered by the data set and remained so at the end.

The increasing activity by developing countries has boosted the number of active antidumping measures that they maintain against the United States, with most of the increases coming fairly recently. The increases for most of them were small, however. As of December 31, 1995, the total increase for all developing countries was less than the total decrease by some of the larger users in the industrialized world, primarily Australia and the European Community/Union. If trends among developing countries continue, however, that could change.

A stronger form of the claim of harm to U.S. exporters--that other countries are singling out U.S. firms for antidumping enforcement in retaliation for U.S. antidumping enforcement against their own firms--does not appear to be supported by the data. For 16 of the 18 countries for which data are available, the percentage of the countries' imports coming from the United States from 1991 through 1995 was larger than the percentage of all active antidumping measures at the end of 1995 that were against U.S. firms. Although retaliation may have occurred in certain individual cases, there is no widespread pattern of retaliation. Most countries seem to avoid imposing antidumping measures on the United States rather than single it out.

Consistent with the proposition that countries use antidumping enforcement as a substitute for other protection for their industries, the United States has been a leader in lowering other forms of protection and, correspondingly, a leader in increasing antidumping enforcement. Developing countries as a group have more recently come under GATT/WTO restrictions on their use of a number of other protectionist practices and, correspondingly, only recently have started to become significant players in antidumping enforcement.

Bias in Procedures for Nonmarket Economies

Determining dumping margins (the amount by which the selling price is below cost or below the home-market price) on imports from nonmarket economies requires different procedures from those used for market economies. Although not conclusive, some evidence suggests that U.S. procedures may be biased toward finding dumping margins for nonmarket economies that are higher than the margins actually are. Since antidumping duty rates are set equal to the dumping margin, the bias would result in higher duty rates being imposed on goods from nonmarket economies. The statistics indicate that initial U.S. duty rates imposed on imports from nonmarket economies tend to be higher than the rates imposed on other countries, and the ratio of the former to the latter is higher than the same ratios for duties imposed by most other countries.

Injury to Downstream Industries

Antidumping measures against upstream goods (that is, goods that are in turn used as inputs in producing other, downstream goods) can in some cases put downstream users of those goods at a competitive disadvantage relative to their foreign competition. Some parties have therefore proposed a so-called short-supply provision for U.S. antidumping law to reduce or eliminate antidumping duties on individual goods in specified conditions of domestic shortages.

Most U.S. antidumping activity--approximately four-fifths of active measures and approximately two-thirds of the products covered by the active measures--is against upstream goods. The average duty imposed on upstream goods--over 52 percent on raw and processed materials and over 32 percent on intermediate goods--is high enough to create a significant disadvantage to downstream users if certain other conditions are in place, but the average market share of the dumped imports is probably not large enough for that to occur. A short-supply provision, however, would not be intended for the average case but for exceptional cases. The market shares for a number of individual goods covered by U.S. antidumping measures are indeed large enough that antidumping duties could disrupt markets, thus harming downstream users.

---