Dumping of Exports and Antidumping Duties: Implications for the U.S. Economy

CRS Report for Congress
Dumping of Exports and Antidumping Duties:
Implications for the U.S. Economy
Updated November 23, 2004
Craig K. Elwell
Specialist in Macroeconomics
Government and Finance Division


Congressional Research Service ˜ The Library of Congress

Dumping of Exports and Antidumping Duties:
Implications for the U.S. Economy
Summary
Dumping in the United States is the selling of a product by a foreign producer
at a price that is below the product’s sale price in the country of origin, or at a price
that is lower than the cost of production. Under U.S. law such an action is
considered an unfair trade practice. If that action is found to cause “material injury”
to a competing domestic industry, an antidumping duty equal to the “dumping
margin” will be levied against the foreign good.
Until the 1990s antidumping actions were a protectionist device used almost
exclusively by a few rich countries: the United States, Canada, Australia, and Europe.
In the last decade, however, there has been an explosion of antidumping cases
brought by many other nations. Rising use by other nations has also meant that the
United States itself has become an ever more frequent target of antidumping
measures.
From an economic perspective, for dumping to be harmful to the U.S. economy
it must be part of a strategy of predatory pricing, aimed at monopoly control of a
market. Such predatory behavior is certainly possible, but likely to be relatively rare.
The practice of U.S. antidumping policy over the last 25 years has grown to
encompass foreign firm pricing behavior that very likely has no predatory intent or
effect. Indeed, it is comparable to behavior undertaken legally by domestic firms in
the home market every day.
Several economic studies confirm that U.S. antidumping duties can be costly
to the overall U.S. economy. Protected domestic producers gain, but consumers and
the wider economy lose more. Using trade barriers, like antidumping duties, as
preservers of “fairness” in international trade has merit if that trade and the foreign
practices that support it undermine longstanding domestic social norms. But
dumping is very unlikely to meet this criterion as similar behavior is widely practiced
by domestic firms in the home market as well.
Antidumping laws are often politically popular and any significant
circumscribing of their use would most likely hinge on finding more generous and
equitable ways of dealing directly with those hurt by international trade that are less
harmful to the efficient functioning of the economy.
There are also concerns that perpetuation and growth of antidumping initiatives
may become a significant impediment to advancing a new round of multilateral trade
liberalization.
This report will be updated as events warrant.



Contents
In troduction ......................................................1
Prevalence and Proliferation of Antidumping Measures....................2
Prevalence of U.S. Measures.....................................2
Proliferation Worldwide........................................3
Economic Conception of Antidumping.................................4
Predatory Pricing..............................................5
New Trade Theory Effects.......................................5
U.S. Antidumping Practice..........................................6
Other Reasons for Price Differences and Below Cost Sales.................8
Effect of Antidumping on U.S. Economic Welfare........................9
Dumping and Fairness.............................................10
Trade and Domestic Social Norms...............................11
Conclusion ......................................................13



Dumping of Exports and Antidumping
Duties: Implications for the U.S. Economy
Introduction
Dumping in the United States is the selling of a product by a foreign producer
at a price that is below the product’s sale price in the country of origin or at a price
that is lower than the cost of production. Under U.S. law such an action is considered
an unfair trade practice. If that action is found to cause “material injury” to a
competing domestic industry, an antidumping duty equal to the “dumping margin”
will be levied against the foreign good. The dumping margin is an estimate of the
difference between the actual price and an estimate of a “fair price;” that is, it is a
measure of the amount the product’s price in the U.S. market would have to rise to
offset the “unfair” price advantage.
The United States has been a frequent user of antidumping duties over the last
20 or more years and the use of antidumping remedies abroad has increased greatly
over the last decade. This trend runs counter to the U.S.’s otherwise strong support
for and leadership in the steady removal of trade barriers in the post-World War II
era. The considerable support for antidumping measures among Members of
Congress was evident in 2002 when the Trade Promotion Authority (TPA) bill (H.R.

3009) was passed by the Senate with an amendment to exclude any changes in U.S.


antidumping laws, arising from any future trade negotiations, from the expedited
ratification process that TPA established.
Antidumping would seem to be an area where the goal of economic efficiency
conflicts with notions of “fairness.” On the other hand, some question whether in
most circumstances dumping is evidence of unfair economic behavior but rather a
way of shielding domestic industries from the competitive rigors of expanding
international trade that is generally condoned by the current rules governing
international trade.
This report looks at the economic impact of U.S. antidumping policy, providing
a view of the rising incidence of antidumping actions by and against the United
States. It examines the conceptual economic basis for taking antidumping action
against imports, a practice appraised relative to that conceptual basis. Evidence on
the domestic economic impact of current antidumping measures is presented and1


arguments for dumping based on the notion of “fairness” are considered.
1 For a discussion of the legal and legislative issues on antidumping and other trade remedy
laws, see CRS Report RL30461, Trade Remedy Law Reform in the 108th Congress, by
William Cooper. For a discussion of Trade Promotion Authority (formerly called Fast
Track), see CRS Report RL31844, Trade Promotion Authority (Fast-Track Authority for
(continued...)

Prevalence and Proliferation of Antidumping
Measures
A 2001 study by the Congressional Budget Office (CBO) examined the level
and use of antidumping actions through 1999 by the United States and other
countries.2 That study found that before the Uruguay Round Agreement went into
effect in January, 1995, only a few countries used antidumping laws to any great
extent, and the United States was the heaviest user by virtually every indicator.3
Subsequent to 1995 the United States became a less frequent, but still a very heavy
user of antidumping actions.
Prevalence of U.S. Measures
The CBO study examines several indicators of the prevalence of U.S.
antidumping measures. One indicator is measures initiated. Prior to the Uruguay
Round Agreement the U.S. led the world in antidumping cases initiated, averaging
over 53 cases per year from 1990 to 1994, or about 25% of the average annual
number of cases initiated worldwide. From 1995 to 1999 cases initiated by the
United States fell to an average of about 26 per year. This was third after the EU,
which initiated an average of 39 cases per year, and South Africa, with an average of

27 case initiations per year. ( In the period subsequent to the CBO study, 2000-2003,4


the United States averaged 37 initiated cases per year.)
A second indicator of antidumping use is the number of measures actually
imposed. Prior to 1995 the U.S. led the world with an average of 23 antidumping
measures imposed per year, or about 25% of the average total worldwide. The EU
was second in this category with an average of about 21 measures imposed. From
1995 to 1999, by this standard, the United States reduced the number of new
antidumping measures imposed to an average of about 16. This put the United States
in second place behind the EU, which averaged 21 new measures imposed per year.
The most striking indicator, in CBO’s view, of U.S. “dominance” of
antidumping action worldwide is the stock of active measures in place. At year end
1994 the United States had 326 active measures in effect, representing over 30% of
all active measures by all countries. In this category the EU was a distant second with

148 active measures. On the first day of 2000 the U.S. reported 267 active measures,


a sizable fall, although still accounting for over 26% of the worldwide total of active
antidumping measures, and far more than the 148 active measures reported by the


1 (...continued)
Trade Agreements): Background and Developments in the 107th Congress, by Lenore Sek.
2 U.S. Congressional Budget Office, Antidumping Actions in the United States and Around
the World: An Update (Washington: June 2001).
3 The Uruguay Round Agreement made moderate revisions to antidumping law that could
be expected to restrain U.S. antidumping activity somewhat.
4 See U.S. International Trade Commission, Annual Statistics.

second place EU.5 (In 2003, the United States had 320 active antidumping measures
in place, 53 of which were against China alone.)6
A final indicator of the prevalence of U.S. antidumping activity is a comparison
of U.S. cases against other countries with those of other countries cases against the
United States. In 1994, the United States had 281 antidumping measures against
other countries compared with 79 measures against it, or a ratio of 3.6 U.S. measures
for every foreign measure. By the end of 1999, the United States had 267 measures
against other countries compared with 107 against it, or a ratio of 2.5.
The CBO study also assessed the magnitude and duration of antidumping duties.
Before the Uruguay Round the U.S. average antidumping duty rate was 56.2%,
among the highest for active users. U.S. measures also tended to be long lived,
remaining in effect for an average of more than seven years, with about 20% of its
active antidumping measures in effect for 10 or more years. Other country’s measures
typically had durations of three to four years. After the Uruguay Round the U.S.
average duty rate fell to 46.7%, placing it more in the middle of the pack; but the
average duration of the duty increased to over eight years, continuing to significantly
exceed the average duration of all other countries’ antidumping measures.
The CBO study shows that the United States continues to be a very active user
of antidumping measures and those measures continue to create a significant and
enduring impediment to trade in the targeted products. This is in sharp contrast to
the nation’s otherwise strong advocacy of free trade.
Proliferation Worldwide
Until the 1990s, antidumping actions were a protectionist device used almost
exclusively by a few rich countries: the United States, Canada, Australia, and Europe.
In the last decade, however, there has been a proliferation of antidumping actions in
numerous developing nations. Very active new users over the last decade include
Mexico, Brazil, Argentina, India, Turkey, and South Africa. A study by the Cato
Institute documents the spread of antidumping measures from “traditional users” to
“non-traditional users” over the last decade. The Cato study observes that in the
1990s, 2,483 antidumping investigations were launched worldwide, representing
more than a 50% increase over the already high activity level of the 1980s.7 In
addition, over this period the pattern of use of antidumping measures between
traditional users and non-traditional users changed significantly. Of the 1,254
antidumping measures initiated worldwide between 1990 and 1994, only 37% were
brought by developing countries. In contrast, of the 1,229 measures initiated between

1995 and 1999, non-traditional users accounted for 59% of the worldwide total.


5 The decline in the stock of active cases for the U.S. is primarily due to the improved sunset
review provisions for active antidumping measures imposed by the Uruguay Round
agreement.
6 See U.S. International Trade Commission.
7 Brink Lindsey and Dan Ikenson, “Coming Home to Roost: Proliferating Antidumping
Laws and the Growing Threat to U.S. Exports,” Cato Trade Studies, no. 14, July 30, 2001.

Notable non-traditional users in this recent period include Venezuela, Peru, Egypt,
Israel, Malaysia, and the Philippines
Rising use of antidumping measures by many other nations has also meant that
U.S. exports have become an ever more frequent target. Over the last half of the

1990s, the United States was the third most popular target (behind Japan and China)


of antidumping measures worldwide, with 81 investigations by 17 counties, and with
duties imposed in 51 of those cases. Antidumping measures in force against the U.S.
in this period were up 41% over the number in force over the previous five years.
Canada and Mexico remain by far the most frequent users of antidumping measures
against the United States, but in recent years many other nations have more
frequently taken aim at U.S. exports. Other prominent recent non-traditional users
of antidumping measures against the United States were Argentina, Brazil, India,
South Korea, South Africa, Taiwan, and Venezuela. The most frequently targeted
U.S. industries have been chemicals and metals, but plastics, vegetable products,
wood pulp and paper, machinery, textiles, and transportation equipment have also
been recent targets. As the world’s largest exporter, this is a trend that the United
States may well expect to escalate8.
The widening spread of antidumping measures is a perhaps not surprising
consequence of the sizable degree of trade liberalization that has occurred across the
world economy. For many developing countries, this has in recent years meant
having to live with substantial reductions in the level of their trade barriers. While
generally beneficial, more open trade will also cause economic disruption and
generate political tension. Antidumping measures provide developing countries with
an easily exploitable device for ameliorating the problems trade can bring and a
device already widely used by the United States and other developed economies.
Nevertheless, the proliferation of antidumping action is viewed by trade
economists as a disturbing trend, a form of backdoor protectionism that runs counter
to the post-World War II trend of reducing barriers to trade. But antidumping is legal
under the rules of the World Trade Organization (WTO) and remains politically
popular. There is concern of a vicious cycle where “legal” antidumping measures by
one country beget (retaliatory or emulatory) antidumping action by other countries.
Economic theory and evidence suggest that this may be a costly trend for the U.S.
and for the world economy.
Economic Conception of Antidumping
Dumping in standard economic analysis is defined as an international form of9
price discrimination. That is, dumping is the sale of an exported product for a price
that is less than the price charged for the same product in the home market. Such a


8 Lindsey and Ikenson, op cit.
9 The discussion in this section draws extensively from Joseph E Stiglitz, “Dumping on
Free Trade: The U.S. Import Trade Laws,” Southern Economic Journal, 1997, vol. 64(2),
pp. 402-424.

price difference is not necessarily a cause of economic harm to the importing
economy. The low price may well place competitive pressure on domestic import
competing industries, but that low price is also a clear benefit to consumers of the
imported product, and will also induce efficiency gains in other parts of the economy.
On balance, one can most often expect a gain in overall economic well-being from
lower priced imports. There are, however, two possible exceptions to this generally
positive impact on the importing economy of international price discrimination:
predatory pricing and new trade theory effects.
Predatory Pricing
Predatory pricing is behavior that uses low prices to drive all other producers out
of the market, with the ultimate intent of establishing monopoly control. During the
period of low prices consumers and the economy would experience an improvement
in economic welfare. But once the monopoly power is secured prices can be
expected to rise and overall economic welfare reduced.
Moreover, a foreign predator could have an even greater deleterious effect than
would a domestic predator. There are two reasons for this. First, the foreign predator
stands to capture economic gains that would only be reallocated within the domestic
economy with a home based predator. Second, because domestic labor is not
internationally mobile, there are likely to be more adjustment problems and more
employment mismatches than if the output source had shifted within the domestic
market rather than abroad.
Mainstream economic analysis has generally attached a very low probability to
the occurrence of a successful predatory pricing strategy. With the prospect of a long
and costly period of predation and the likelihood of limited ability to deter
subsequent entry by new rivals, the chances for actually earning full monopoly
profits and generating an acceptable long-run rate of return from the endeavor
seemed remote. However, some recent thinking on the subject has developed
possible strategies for deterring entry by potential foreign or domestic competitors
that would improve the chances of successful predatory pricing.
Thus while dumping for predatory ends is still likely a rare event, it could occur
and it is still an appropriate target for economic policy. But evidence of price
differentials across borders, or prices set below average total cost, are not conclusive
evidence of dumping for harmful predatory purposes and can certainly be consistent
with beneficial competitive behavior. Further, such competitive behavior is more
likely to be present in the market place than is predatory behavior.
New Trade Theory Effects
The new trade theory or what is also called “strategic” trade theory is a body of
relatively recent research that has shifted the focus of trade theory from the exchange
of a good between nations to the rivalry between a few firms in international
competition. In this environment there can arise a strategic interdependence between
firms such that pricing, investment, and output decisions by one firm will strongly
affect similar decisions by others in the group. The firm that can move first to exploit



sale economies and learning curve advantages can find itself in position to earn
extra-normal returns to the benefit of itself and the wider economy.10
In this framework it is theoretically possible for government promotion of the
domestic firm to induce behavior of foreign firms that leads to a shifting of profits
to the domestic firm and increases national economic well-being. Antidumping
actions could help capture these benefits as they work to help the home firm and
deter the foreign competitor.
As research has evolved in this area, however, it has also become clear that the
desirable outcomes of “strategically” deviating from free trade may be elusive. A
positive outcome is highly sensitive to just the right conditions being present in the
relevant markets. In practice, accurate identification of a true strategic industry
would be difficult and the accrual of sizable returns doubtful. Such policy efforts
could do more harm than good. For these reasons most economists would think it
unlikely that “new trade theory” effects provide a credible justification for most
antidumping actions.
U.S. Antidumping Practice
U.S. antidumping laws are administered by the Department of Commerce
(DOC) and the International Trade Commission (ITC). A dumping case can be
initiated by a domestic industry or by the DOC. The DOC determines the existence
and degree of dumping, using one of several possible methods for estimating whether
the import’s U.S. market price is below an estimate of “fair” price. The ITC
determines whether “material injury” has occurred based on consideration of factors
such as employment and volume of imports, but what constitutes material injury is
not formally specified and the injurious effects of factors other than imports are not
systematically disentangled, leaving a good deal of discretion to the ITC in making
its determination. If both agencies’ findings are affirmative, the DOC will direct the
Customs Service to levy a duty equal to the estimated “dumping margin.” Critics
observe that DOC findings are almost always positive, affirming that dumping has
occurred. Positive findings of material injury by the ITC are less frequent.
The form and practice of U.S. antidumping policy has changed over the years,
moving well away from trying to counter harmful predatory behavior by importers.
The original antidumping statute — the Antidumping Act of 1916 — was aimed at
the prevention of predatory pricing. Although this law is still in effect, it receives
little use. It was effectively superceded by a 1921 law — the Antidumping Act of

1921 — which gave much greater scope for the bringing of an antidumping action.


Under that act there was no need to establish predatory intent or effect, so that any
price discrimination could be grounds for imposing antidumping duties on an import.


10 For an elaboration of “new” or “strategic” trade theory and policy, see Paul Krugman, “Is
Free Trade Passe?,” Journal of Economic Perspectives, 1987, vol.1, pp.131-144.

However, most economic analysts argue that a series of amendments to U.S.
antidumping law during the 1970s facilitated a major increase in the likelihood of
achieving a successful antidumping finding. Changes of significance included use
of sales below cost as a measure of dumping, establishment of strict and shortened
time limits on cases, a shift of investigation power from the Department of Treasury
to the Department of Commerce, and a lowering of the threshold for assessing
imports’ role in harming a domestic industry. In the years that followed these
changes, the level of successful antidumping cases rose substantially.11
The key determination in an antidumping case is what constitutes a “normal”
price for the imported product. The excess of “normal” price over actual price in the
U.S. market establishes the “dumping margin” and becomes the basis for levying the
antidumping duty. U.S. antidumping law specifies a hierarchy of methods for
determining “normal” price for an alleged dumped import. The first preference would
be to establish a home market index that measures the price the product sells for in
the producer’s home market. However, if there are no home market sales or what
home sales do occur are less than 5% of sales in the U.S. market, then a third country
index that measures the product’s price in another of the exporter’s foreign markets
is used. If that can not be done, then a constructed value index that estimates the
exporter’s average total production costs, including some amount of profit, would be
used.
In practice, many direct price comparisons are rejected because U.S. rules
preclude making comparisons to the producer’s home market or third market prices
if more than 20% of such sales are deemed to be below total cost. Therefore
comparisons are very often based on a “constructed value” index. However,
“constructed value” indexes are thought to be subject to some significant
measurement problems. For example, in many cases the index is not made from data
provided by the foreign producer, but relies instead on “facts available,” where such
facts are often obtained from the domestic industries’ antidumping petition. In
addition, constructed value indexes incorporate a somewhat arbitrary measure of a
“normal” profit margin into the cost estimate. It is often the case that low profit sales
in the home market will be excluded from comparison, thereby inflating the
computed profit margin and, in turn, the imputed dumping margin.
Because of these and other problems many economic analysts argue that the
U.S. system of determining fair value has a built-in asymmetry that biases it toward
a finding that dumping has occurred.12 Moreover, it is argued that these measures are
unlikely to reveal price discrimination and are highly unlikely to discern the presence
of predatory pricing. An extensive study was undertaken by the OECD of
antidumping cases in Australia, Canada, the European Union, and the United States.13


11 See Thomas J. Prusa, The Trade Effects of U.S. Anti-Dumping Actions, National Bureau
of Economic Research, Working Paper no. 5440, 1996, pp. 3-5
12 Biases in DOC antidumping determinations are reviewed by Robert Boltuck, and Robert
Litan, Down in the Dumps: Administration of the Unfair Trade Laws (Washington: The
Brookings Institution, 1991).
13 The results of this study are published in: Robert Z. Lawrence, (ed.). Brookingss Trade
(continued...)

That study found that 90% of the cases where imports were determined to be dumped
under existing rules would not have been questioned as posing a predatory threat
under these same countries’ antitrust (or competition) laws. In other words, the
behavior of the importers, if undertaken by a domestic firm, would not have been
questioned as predatory or otherwise generally harmful. A more recent study by
Brink Lindsey looked at 141 dumping determinations by the DOC between 1995 and
1998.14 This study found that the DOC methods were ineffective in identifying price
discrimination or harmful predatory behavior. Nor were these methods likely to be
a reliable guide to the presence of other unfair practices on the part of foreign sellers,
Lindsey concluded. It is also found that overstated profit assumptions in constructed
value calculations often biased the process toward finding a positive dumping
margin.
As one observer notes, under current U.S. rules imports can be deemed unfairly
dumped “even if foreign firms charge higher prices to their export market than they
do at home and even if foreign firms earn healthy profits on each and every foreign
sale.”15 In general these results raise considerable doubt as to whether most U.S.
antidumping determinations are shielding the U.S. economy from economic harm.
Other Reasons for Price Differences
and Below Cost Sales
There are many reasons other than predation why prices may differ across
national borders and why sales might occur at a price below production cost. Each
is rooted in reasonable business practice and none are economically harmful.
Outside of agricultural and commodity markets, modern firms selling
differentiated products have some degree of market power, not enough to extract
monopoly profits but sufficient to adapt price to differing market conditions from
country to country. Because of this the price charged in a less competitive home
market may be higher than the price charged in a highly competitive foreign market.
Similarly, firms that are new entrants to a market often have a competitive
disadvantage relative to established firms. In this circumstance it is not unusual for
a new business to offer its product at a lower price in an attempt to offset this
disadvantage and gain market share.
Pricing below cost is also often part of a reasonable business practice.
Economists have long recognized that a profit maximizing (or loss minimizing) firm
will find it prudent to price below average total cost. In times of recession or other
market weakness, a price that covers variable cost (e.g., labor, energy, and materials)


13 (...continued)
Forum 1998 (Washington: The Brookings Institution, 1998).
14 Brink Lindsey, “The U.S. Antidumping Law: Rhetoric versus Reality,” Cato Trade Policy
Studies, no. 14, 1999, pp. 1-36.
15 Bruce Blonigen and Thomas Prusa, Antidumping, NBER, Working Paper, no. 8398, July

2001, p. 3.



and only a portion of fixed cost (e.g., plant, equipment, and r&d) is a practice that
does not earn profits but will minimize losses until business conditions improve.
Below cost pricing may also occur with the introduction of a new product. In
this case a firm may possess a productive process that presents a steep “learning
curve” that promises improved efficiency and lower unit production costs as time
passes and the output level increases. In this case the firm will often “forward price”
in relation to the expected long run marginal cost, which will be much lower than
the current marginal cost. Similar long run pricing behavior can occur when new
firms must initially incur large initial investments in physical plant and research and
development. Until sufficient market share is secured by these new producers, price
will fall below average total cost.
We can also expect in firms that produce multiple products with extensive cost
sharing that it will be difficult to allocate costs on a product specific basis. A single
product’s price may seem unprofitable if examined in isolation but in the context of
the full product line can still contribute to overall profitability. Multi-product firms
may also use “loss leaders.” These are products sold at low price to attract customers
in the hope that they will also purchase other higher price products from their line.
None of these pricing tactics are necessarily indicative of any predatory intent
and are frequently undertaken by domestic sellers to the benefit of consumers and the
wider economy. Given the likely rarity of predatory behavior and the relative
prevalence of these other reasons for keeping prices low, it can be argued that many
U.S. antidumping determinations are actually responding to this benign or beneficial
behavior rather than harmful predatory behavior.
Effect of Antidumping on U.S. Economic Welfare
Economic analysis strongly indicates that trade restrictions reduce national16
economic well-being. Protected industries and the collector of a tariff may gain,
but consumers of the protected good and the wider economy typically lose more.
Restrictions on imports such as antidumping duties raise the cost of acquiring the
targeted product. Because of this, consumers are hurt, and so is the productive
efficiency of the economy, since the protected industry is induced to produce more
than what is economically optimal. The extra productive resources used to generate
this added output comes at the expense of producing other goods that have greater
economic value. This inefficiency necessarily lowers the total value of the economy’s
output. The damage caused by the restriction (i.e., antidumping duty) will be greater
still if the targeted countries retaliate against the restriction and dampen U.S. export
sales.
The average antidumping duty imposed by the U.S. between 1980 and 1995 was
about 40%, with many cases of duties exceeding 100%. The estimated impact of
these duties on trade was also substantial, with targeted imports typically falling 50%


16 For a more formal development of this conclusion, see Gregory N. Mankiw, Principles
of Economics (Fort Worth, TX: The Dryden Press,1998), pp. 181-184.

to 70% over the first three years of protection.17 One would expect that such a large
market distortion would lead to substantial costs for the U.S. economy, and a number
of economic studies confirm this expectation.
The U.S. International Trade Commission published a study in 1995 that
evaluated data from all antidumping cases initiated between 1980 and 1993 and
assessed the overall impact on the U.S. economy.18 That study found that
antidumping duties caused a net economic loss of about $1.6 billion per year. Profits
and wages in protected industries increased about $658 million, but unprotected firms
and workers lost about $1.85 billion.
Another study by economist James DeVault looked at data from 30 U.S.
antidumping actions between 1987 and 1992 and found that those duties reduced
U.S. economic welfare by $275 million annually.19 Further, for each $1.00 protected
producers gained from the trade barrier, U.S. consumers lost $3.20.
Finally a 1999 study by Gallaway (et al.) focused more closely on the dynamics
of foreign product pricing when faced with dumping duties.20 It was found that
when presented with an adverse finding there was a sizable tendency for foreign
producers to raise prices to avoid the antidumping duty. This was detrimental to
consumers who incurred the higher price, and it also worked to redirect the benefit
of potential tariff revenue from the U.S. government to foreign producers. With this
effect included, the estimated welfare loss to the U.S. economy was estimated to be
in a range of from $2 billion to $4 billion annually.
There may well be other welfare reducing effects not directly related to
imposing an antidumping duty. The act of initiating an antidumping action — well
short of reaching a finding — may have significant effects. Staiger and Wolack, for
example, found that even suspended cases where no duty was applied caused a
significant reduction of imports and increases of domestic production.21
Dumping and Fairness
Even though economic evidence shows that removing most antidumping duties
would likely improve the overall economic well being in the United States, there


17 Thomas J. Prusa, The Trade Effects of U.S. Antidumping Actions, NBER, Working Paper
no. 5440, 1996, p.12.
18 U.S. International Trade Commission, “The Economic Effects of Antidumping and
Countervailing Duty Orders and Suspensions,” no. 2900, June 1995.
19 James M. DeVault, “The Welfare Effects of Antidumping Duties,” Open Economies
Review, vol. 7, 1996, pp. 19-33.
20 Michael P. Gallaway, Bruce A. Blonigen, and John E. Flynn, “Welfare Costs of U.S.
Antidumping and Countervailing Duty Laws,” Journal of International Economics, vol.49,

1999, pp. 201-223.


21 Robert W. Staiger, and Frank A. Wolak, “Measuring Industry-Specific Protection:
Antidumping in the United States,” Brookings Papers: Microeconomics,, 1994, pp. 51-118.

would also be industries and workers adversely affected by removing this
impediment to free trade. Their plight gives rise to concerns about trade with
countries that do not play by the same economic and social rules as the United States,
whose “unfair” practices can undermine the economic position of U.S. workers.
Worse, in this view, they can undermine important social conventions and
institutions that frame the terms for acceptable economic competition.
Differences in labor and environmental standards have been prominent concerns
in trade policy debates about efficiency versus fairness, but arguments for
antidumping measures are also often advanced as a fairness issue, with the
implication that dumped products arise out of various unfair business practices.
Unless offset, the cost advantages afforded by these different standards and practices,
it is argued, will steadily put competing American industries and, most often, low-
skilled American workers at a major disadvantage as well as erode established
domestic social relations.
On the other hand, a cheaper product, regardless of how it was made cheaper,
is an overall gain to the importing economy. It is certainly possible for those who
gain to compensate those who lose and still be better off. Adequate compensation,
in practice, is most often problematic, however. Of course, the loss to the displaced
worker in the import competing industry is the same whether caused by “unfair”
foreign practices or “state of the art” technological prowess. Yet, the former is often22
seen to be far less acceptable.
Under what circumstances do the concerns about “fair-trade” represent a valid
argument against the welfare raising effects of free trade? In the case of imports
there are two possible situations that need to be distinguished. While the economic
case argues against restriction in both situations, the degree to which other
constraining “social requirements” are met varies.
Trade and Domestic Social Norms
For many people the acceptance of the gains from trade will hinge on whether
those gains emerge from a process where all trading parties adhere to social norms
of “fair play.” For example, different child labor rules can provide a basis for trade.
The United States, with long held restrictions on child labor in its domestic practices,
can find it economically beneficial to trade with a country that has a low level or no
restriction against child labor.
Lower child labor standards could give a production cost advantage to the
foreign producer. If the exports of the low labor standards country compete directly
with U.S. industries, the price advantage of those lower standards will encourage a
higher level of imports to the United States. These imports will come at the expense
of domestic production. U.S. workers in the affected industry will lose jobs. With
time these workers may find new jobs but most likely at a lower wage. In the
aggregate, both nations are economically better off through this exchange, but the


22 The discussion in this section draws extensively on Dani Rodrik, Has Globalization
Gone too Far (Washington: Institute for International Economics,1997).

distribution of income has been changed in the United States as the income of the
import competing workers falls.
Is this an acceptable outcome, or should trade policy (tariffs, quotas, etc) be used
to protect the affected domestic workers from this outcome? While such trade is
certainly an efficient outcome, it is also an outcome that likely violates a prevailing
social norm (i.e., domestic adult workers should not have to compete against child
labor). If it is unacceptable to have child labor in a purely domestic context, why
would it be acceptable to have domestic workers’ living standard reduced by
competing indirectly through trade with countries with more lax child labor laws?
The overriding sentiment in this circumstance is that there is a difference
between gains from trade generated by a comparative advantage based on factor
endowments or consumer preferences, and trade generated by a comparative
advantage based on institutional choices in the exporting country that conflict with
the norms of the importing country.
In this circumstance, where trade with a country with different social standards
inflicts economic harm on domestic workers, the case can be made that trade
liberalization cannot be treated as an end in itself, without regard to how it affects
broadly shared values at home. Economic activity occurs in a social and moral
context which tells us that there are unacceptable ways of imposing a burden on
fellow citizens.
In concept, a trade restriction can be justified in this case to protect the strongly
held social norm. In practice, however, under the rules of the World Trade
Organization (WTO) it would be improper to use trade policy to curtail imports from
countries using child labor because WTO rules, while prohibiting trade in products
made by prisoners, generally do not allow discrimination on the basis of the mode of
production. Of course, WTO rules do provide safeguards under the so called “escape
clause” mechanism that allows domestic industries to seek relief from damage caused
by certain types of import surges. We can imagine an elaboration of the “escape
clause” mechanism that could be used for providing relief from damage done by
unacceptable modes of foreign production. In its current form, however, WTO rules
governing the use of such safeguards provide a very limited scope for using trade
restrictions and have been little used by member nations.
Dumping, unlike labor or environmental standards, is not so easy to see as a
violation of a domestic social norm. As discussed above, other than the rare case of
predatory pricing, there are a variety of business practices that can lead to price
differences between markets and pricing below average production cost. Such
pricing behavior is often a basic element of competition in the market place that
serves efficiency and overall economic well-being, and is widely practiced in the
domestic economy. If most dumping is largely reflective of similar behavior on the
part of foreign firms, then it is not violating any domestic social norm. In this
circumstance pursuing antidumping actions on “fairness” grounds would not appear
to be credible.



Conclusion
It is doubtful that the majority of antidumping actions initiated by the United
States are countering any harmful predatory pricing by importers. Nor is it likely that
the majority of importer behavior targeted by antidumping duties is “unfair” in the
sense that it deviates from widely accepted domestic practices. While certainly a
benefit to the domestic industries that are protected by antidumping levies, those
duties impose a greater and substantial cost on the wider U.S. economy. The
enormous proliferation across the globe of antidumping measures of recent years
suggests that the economic burden of antidumping on the United States will grow as
more U.S. exports are targeted.
It may be that because it is relatively easy to invoke the antidumping laws, they
are being used as a surrogate for the less easy to use “escape clause” mechanism to
provide domestic industries some safeguard from surging imports. But many
economists would argue that it is being overused. And that overuse is not only
costly but also, as one economist sees it, “subverts the trade regime, gives safeguards23
a bad name, and crowds out an effective outlet for legitimate concerns.”
Yet antidumping duties remain a popular political response to the costs that
international trade imposes on many domestic industries and workers. This is not
surprising given that the benefits of antidumping policies are very focused and accrue
to a well defined group, while the cost of these policies is widely dispersed over the
population among people with less natural cohesion and a more diluted political
voice. Another consideration, however, is that perpetuation and growth of
antidumping initiatives may become a significant impediment to advancing a new
round of multi-lateral trade liberalization.
In the United States, the pressure to maintain this costly trade barrier might be
allayed if workers harmed by international competition were more confident of
receiving equitable compensation and other adjustment assistance. Insurance against
adverse labor market outcomes has been part of a social bargain in the United States
and other countries that has enabled those nations to make fuller use of the wealth
creating potential of the market economy, including expanding free international
trade. Examination of social insurance policies and practices might be a useful
option in addressing the antidumping issue.
From an economic perspective, reform of the antidumping laws might be
sensibly directed toward merging them with domestic predatory pricing (i.e., antitrust
policy) laws, placing the policy focus on behavior that is clearly economically
harmful.24 Another, option would be for the antidumping injury test to encompass25


effects on the whole economy, producers and consumers alike.
23 See Rodrik, op. cit., p 82.
24 See Stiglitz op. cit., p 420.
25 J. Michael Finger, Antidumping: How it Works and Who Gets Hurt (Ann Arbor, MI:
University of Michigan Press, 1993).