Sixty years of GATT rounds have resulted in low levels of tariff protection, especially for developed countries. Constrained by these commitments, many countries have switched to other instruments to wield protection. Among these, antidumping duties are some of the most important. Although originally devised to combat unfair trade in the form of export prices below normal value (i.e., dumping)1, their wide and arbitrary applicability has led economists to conclude that antidumping “has nothing to do with keeping trade ‘fair.’ ... It is simply another form of protection” (Blonigen and Prusa, 2003). Still, because of the stated objective of fairness, antidumping duties are legal provided they respect the WTO antidumping agreement.
While in the 1980s only the EU, Australia, Canada, and the United States were active and heavy users of antidumping, many more countries – especially developing countries – joined this group during the 1990s, as illustrated in Figure 1. In fact, the WTO’s most recent report lists India as the country with the largest number of new investigations and final measures applied during the first six months of 2007.2 Other new frequent users include Brazil, China, and South Africa. This shift follows various trade liberalisation efforts by developing countries and their increased role in international trade. The EU and other traditional users of antidumping now increasingly find themselves on the receiving end of antidumping investigations –according to WTO statistics, the EU was the second most frequent target in the first semester of 2007.
Figure 1: Evolution of the number of countries with antidumping laws
Source: Vandenbussche and Zanardi (2008).
Economists have documented various strategic effects that derive from the use of antidumping laws. On most occasions, these effects imply a net loss for the country using antidumping duties, although the protected industries usually gain.3 The wave of new adopters of antidumping laws illustrated in Figure 1 has revived economists’ interest in this type of trade protection in order to understand what the effects of such widespread use of antidumping may be on the international trading system.
At the heart of the matter is why countries adopt antidumping laws. The seemingly obvious answer is to combat unfair trade. However, our recent work (Vandenbussche and Zanardi 2008) shows that other reasons are at play. Retaliatory motives are an important determinant of the adoption and subsequent use of antidumping laws. It is not by chance that some of the new users were heavily targeted by the EU and US in the 1980s. Also worrying is the finding that the adoption of these laws has been prompted by past trade liberalisation. Figure 2 clearly suggests a substitution effect between tariffs and antidumping. It is therefore doubtful that antidumping duties are limited to cases of unfair trade practices. Moreover, antidumping laws can be abused by special interests once they are in place. For instance, we have found that the aggregate effect of antidumping use on total trade can be substantial enough to reverse the gains from past trade liberalisation efforts (Vandenbussche and Zanardi 2006).
Figure 2: A substitution effect between tariffs and antidumping
Notes: annual caseload of antidumping initiations by countries that adopted an antidumping law after 1980; average applied tariff rates of countries that do not have an antidumping law or adopted one after 1980.
Source: Vandenbussche and Zanardi (2008).
Time for reform?
Given this consensus on the negative effects of antidumping, economists have long called for substantial reforms, but political will has been largely absent among traditional users, notably the EU and the United States. For many years, developing countries have been insisting on a change of the antidumping rules which they felt were inadequate and were in many cases unjustly hurting their interests. Paradoxically, the diffusion of antidumping laws and the capacity of developing countries to retaliate offer a unique opportunity for reforms. The traditional users, now the main targets of antidumping, should be more willing to discuss the topic.
The most recent wave of globalisation provides another reason for reforms. Increased offshoring is changing the organisation and definition of firms in ways that current antidumping practices cannot handle. When parts of the production process take place abroad, the intermediate products being traded may be subject to antidumping actions, which amount to shooting oneself in the foot.
These considerations led the EU to be a frontrunner in considering reforms. In December 2006 the EU Trade Commissioner, Mr Peter Mandelson, released a Green Paper calling for a “reflection on the application of the EU Trade Defence Instruments in light of emerging new realities in the global economic context.”4 One prominent reason was the ‘leather shoe’ antidumping investigation against China and Vietnam in 2005. This case revealed that antidumping duties were hurting, not benefiting, a large number of EU producers.5 This resulted in large opinion differences between member states on whether or not to impose antidumping duties. In the end, the 25 member states compromised to impose antidumping measures but for a substantially shorter period than usual. A similar situation arose in 2007, when a case for the imposition of antidumping measures on lightbulbs from China brought by the German firm Osram, which produces most of its bulbs in Europe, was resisted by the Dutch firm Philips, which manufactures large quantities of bulbs in China.
If the time is ripe for reform and the Commission has recognised it, political willingness seems missing in many European capitals. Mr Mandelson recently had to announce that a proposal for reforms of the EU trade defence instruments, due in January 2008, would not be put forward before July. Thus, we are now facing the prospect of no reforms at all in spite of mounting evidence and economic arguments to the contrary.
Even leaving aside legal details, our recent work and other studies suggest some obvious starting points for any reform of the EU antidumping system. Antidumping instruments that are simply another form of protection leave too much discretion in the hands of the administrating authorities. Duties are thus used in a retaliatory fashion or to compensate for lost tariff protection resulting from trade liberalisation. The solution is relatively simple: prescribe (and enforce) tighter economic criteria identifying dumping, injury, and the causal link between the two. Here, the extensive economic literature and legal doctrine on competition policy can be of great help.
Two related points are the need for more transparency in the work of the authorities and more sound quantitative methods. Transparency, which leads to accountability, provides a natural check on the use of discretion and is part of any system of checks and balances. A rigorous quantitative analysis also limits abuse of discretionary powers and would provide evidence, if any, of the relation between dumping and injury to a domestic industry. But the simplest of all reforms requires minimal legal changes for the EU. Its antidumping law, differently from many others (e.g., the US antidumping law), already includes a ‘public interest clause’ by which antidumping measures should only be imposed if they are in the interest of the EU community at large. A strict interpretation of this test is all that economists would call for, once consumers’ and producers’ (of intermediate and final goods) interests are correctly measured! This criterion would also reconcile the conflict between firms that offshore part of their production and those that do not. Unfortunately, this criterion has rarely been given fair consideration while it should be the first and ultimate test of the desirability of antidumping measures. In a sense, the antidumping system’s problems have been solved since 1968 (when the EU antidumping law was first introduced), if only the law were put into practice.
For all the blame that (some) European governments share if EU antidumping reform is abandoned, it is still laudable that such an agenda was put forward by Mr Mandelson. Needless to say, but important to remember, that such reforms should best be pursued at the multilateral level since many countries are now actively using this protectionist tool. In particular, an individual country may not want to restrain its use of antidumping (through serious reforms) if other countries do not follow suit. The WTO would be the ideal place but the already troubled Doha Round does not make it likely that substantial reforms of antidumping could be put on the negotiating table. The disagreement amongst the EU member states does not improve the odds while a different outcome may have even helped the Doha Round. This is really a missed opportunity.
Blonigen, B. and Prusa, T.J. (2003): “Antidumping”, in: Choi, E.K. and Harrigan, J. (eds.): Handbook of International Trade, Oxford, U.K. and Cambridge, MA: Blackwell Publishers.
Kommerskollegium (2007): “Adding value to the European economy”, available at http://www.kommers.se/upload/Analysarkiv/Arbetsomr%C3%A5den/EUs%20yttre%20handelspolitik/AddingvaluetotheEuropeaneconomy.pdf
Vandenbussche, H. and Zanardi, M. (2006): “The Global Chilling Effects of Antidumping Proliferation”, CEPR Discussion Paper n. 5597.
Vandenbussche, H. and Zanardi, M. (2008): “What Explains the Proliferation of Antidumping Laws?”, Economic Policy 23(53), 93-138.
1 Dumping occurs when exporting prices are below cost or below prices in the country of origin.
2 See http://www.wto.org/english/news_e/pres07_e/pr497_e.htm for details.
3 See Blonigen and Prusa (2003) for a survey of the literature.
4 http://ec.europa.eu/trade/issues/respectrules/anti_dumping/comu061206_en.htm. The EU’s trade defence instruments also include countervailing duties (used when imports receive illegal subsidies by the government of the exporting country) and safeguard measures (used against sudden surges of imports that threaten the existence of an industry). Over the period 1995-2006, the EU initiated more than 350 antidumping cases but only around 45 countervailing duty cases and four safeguard cases.
5 See Kommerskollegium (2006) for details.