51635 The World Bank notes OctOber PREM TRADE POLICY 2009 NUMBER 144 The Pattern of Antidumping and Other Types of Contingent Protection Chad P. Bown, Development Economics Research Group* I. Newly Initiated Trade level investigations in response to domestic Remedy Investigations industry requests for the imposition of new Many of the major economies in the multilat- import restrictions under national trade eral, rules-based trading system find them- remedy laws such as antidumping (AD), selves in a situation in which their applied global safeguards (SG), countervailing duties tariff rates are quite close to the tariff binding (CVD), and China-specific safeguards (CSG). levels that form their legal commitments These 44 new investigations in 3Q 2009 were at the WTO. This implies that they cannot an increase of 52.6 percent compared to the simply raise applied tariff rates to respond third quarter in 2008.1 to domestic industry demands for additional The cumulative number of such new trade barriers to protect them from imports. requests for protection throughout the One of the fundamental and potentially first three quarters of 2009 is 30.3 percent WTO-legal ways in which national govern- higher than the number of requests that ments can respond to domestic industry calls took place in the first three quarters of 2008. for additional protection from imports is by As illustrated in figure 1, the 2009 increase resorting to trade "remedy" policy instru- continues an upward trend, as the 30.3 per- ments such as antidumping, safeguards, and cent increase in the first three quarters of countervailing duty (antisubsidy) policies. 2009 compared to 2008 builds upon a 36.4 This note, which describes newly collected percent increase in 2008 compared with the data made available through the World same period in 2007.2 Bank-sponsored Global Antidumping Database, While the imposition of a preliminary reports on the combined use of such policies, import restriction occurs typically within a comprehensively collected across the major couple of months of initiation of the inves- WTO member economies. tigation, the historical data on the use of In the third quarter (3Q) of 2009, WTO these trade policies, especially in the case member governments initiated 44 new product- of antidumping, indicate that the majority * This PREM Note is based on the Global Antidumping Database, which is part of the International Trade Department's effort in PREM to monitor and analyze trade policy responses to the crisis. Aksel Erbahar was the lead research assistant for this report. Yubing Cui, Laura Gutowski, Ludmila Cieszkowsky Elias, and Sharon Kim also provided outstanding research assistance. Correspondence: Bown, DECRG, The World Bank, 1818 H Street, NW; Mailstop: MC 3-303, Washington, DC 20433 USA, Tel: +1.202.473.9588, Email: cbown@worldbank.org FROM THE POVERTY REDUCTION AND ECONOMIC MANAGEMENT NETWORK Figure 1. Newly Initiated Trade Remedy Investigations, 1Q 2007­3Q 2009 (nonredundant AD, SG, CSG, CVD at the product level) 50 45 Initiated by Initiated by developing developed Number of investigations 40 economies economies 35 30 25 20 15 10 5 0 1Q-2007 2Q-2007 3Q-2007 4Q-2007 1Q-2008 2Q-2008 3Q-2008 4Q-2008 1Q-2009 2Q-2009 3Q-2009 Fiscal year (quarter) Source: Global Antidumping Database. of new investigations ultimately result in the WTO of 21 new import-restricting global the imposition of new "definitive" import safeguard investigations, 6 of which were restrictions, typically with a 12- to 18-month initiated in 3Q 2009 alone. The 21 new lag. One implication of the 2008­09 surge investigations already constitute the third- in new investigations is the likelihood that highest yearly total for any year since 1995, they may result in an increase in 2010 in following only 2002 (35 initiations) and newly imposed definitive import restricting 2000 (27 initiations). The 21 new initiations measures. This is an issue to which we return in 2009 build upon the 8 newly initiated in sections II and III below. safeguard investigations taking place in the second half of 2008 amid the spread of the Continued Use of Global Safeguards financial crisis. One of the most striking features in the 2009 As figure 2 illustrates, a continuation data to date is that members have notified of the 2009-to-date pace of new safeguard Figure 2. Global Safeguard Initiations by WTO Members, 1995­2008 and 2009 (projected) 40 SG initiations 35 over steel Number of initiations 30 products 25 20 SG initiations over nonsteel 15 products 10 5 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009... Year Source: Global Antidumping Database compiled from reports to the WTO Committee on Safeguards. 2 PREMNOTE OCTOBER 2009 investigations through the fourth quarter financial crisis. The surge was 42.6 percent would make 2009 the second most pro- higher than the average of 24 new product- lific safeguard-using year since the WTO's level antidumping investigations initiated 1995 inception, following only the "steel per quarter during the crisis period of 1Q safeguard year" of 2002. What differenti- 2008 through 2Q 2009. ates 2009 from 2002 is the fact that a large share of the 2002 safeguard use was trig- Policy-Imposing Countries gered by a single industry event, which is Seventeen of the thirty-three different WTO not the case in the year 2009 data.3 In fact, members for which the Global Antidump- the projected use of safeguards in 2009 for ing Database tracks information on trade nonsteel products would be the most ever remedy use initiated at least one new prod- initiated in one year. uct-level investigation in 3Q 2009 under one of these trade remedy laws. As figure 1 Resurgence of the Dominant illustrates, developing countries dominated Antidumping Policy use by initiating 68.2 percent of these new Despite the increase in global safeguards use investigations, compared to developed and the recent media attention paid to the economies, which initiated 31.8 percent. China-specific safeguard policy,4 antidump- In 3Q 2009, 56.8 percent of the new ing was still the "preferred" trade remedy investigations were undertaken by only of choice by national governments and four economies: India and Argentina each their industries: 37 of the 44 product-level initiated seven new product-level investiga- investigations initiated in 3Q 2009 occurred tions, followed by the United States (six) under a national antidumping law. and the European Union (five). Turkey ini- As figure 3 indicates, there was a sharp tiated three investigations; Brazil, Canada, increase in initiated antidumping investi- Pakistan, and Peru each initiated two; and gations in 3Q 2009, even compared with eight other WTO members (Australia, Chile, earlier data from the period of the global China, Croatia, Mexico, Morocco, Ukraine, Figure 3. Newly Initiated Antidumping Investigations, 1Q 2007­3Q 2009 (product level) 40 Initiated by Initiated by 35 developing developed Number of investigations 30 economies economies 25 20 15 10 5 0 1Q-2007 2Q-2007 3Q-2007 4Q-2007 1Q-2008 2Q-2008 3Q-2008 4Q-2008 1Q-2009 2Q-2009 3Q-2009 Fiscal year (quarter) Source: Global Antidumping Database. OCTOBER 2009 PREMNOTE and Vietnam) initiated one product-level Exporters from the European Union or its investigation each. member states and Thailand faced three investigations each, while Brazil, Indonesia, Affected Sectors and the Republic of Korea each faced two The new trade remedy investigations cov- new product-level investigations during 3Q ered a number of different sectors in 3Q 2009. 2009. The dominant sectors were chemicals, China was specifically named in 62.1 with 13 total investigations, followed by iron percent (23 of the 37) of the newly initiated and steel with 8. Other sectors targeted with product-level investigations under (AD, multiple investigations include agriculture CVD, CSG) laws that require the investigat- (four), textiles (four), machinery (four), ing country to specifically name at least one materials (four), plastics and rubber (three), exporting country.5 However, the 3Q 2009 and other miscellaneous manufacturers figure is lower than that in 2Q 2009, in which (two). 82.6 percent of the newly initiated prod- Associated with problems in global dairy uct-level investigations included China's markets in 3Q 2009, both Chile and Croatia exporters in the petition. Finally, the 23 new initiated global safeguard investigations over product level investigations in 3Q 2009 is cheeses and other dairy products. only a slight increase over the number of new investigations facing China's exporters Targeted Exporters in 1Q 2009 (20 new investigations) and 2Q China continued to be the exporting country 2009 (21 new investigations). most frequently targeted by new investi- When examined on a year-by-year basis, gations in 3Q 2009, facing 23 new prod- as figure 4 indicates, industry demands for uct-level investigations. Other frequently new import restrictions against China under investigated exporters in 3Q 2009 include these policies were up 22.7 percent in 2008, the United States (five); and Taiwan, India, and are on the path to another 7.8 percent and Japan, with four investigations each. increase in 2009. It is notable that this in- Figure 4. WTO Members' Newly Initiated Investigations of Imports from China, 2001-09 (nonredundant AD, CVD, CSG at the product level) 90 80 investigations per year 70 Newly initiated 60 50 40 30 20 10 0 2001 2002 2003 2004 2005 2006 2007 2008 2009 (projected)* Year Source: Global Antidumping Database. 4 PREMNOTE OCTOBER 2009 crease in trade remedy use targeting China's As figure 5 indicates, WTO members im- exports in 2008-09 in the midst of the crisis posed 20 new product-level definitive import is somewhat smaller than the overall growth restrictions in 3Q 2009 under national trade in new product-level investigations (see remedy laws, a slight reduction from 3Q again figure 1 and the discussion above). 2008, when 21 new measures were imposed. A contributing explanation is that, relative The year-to-date count of newly imposed to the set of cases that WTO members have measures through the first three quarters initiated against all exporters (including of 2009 is 21.4 percent higher than the rate under the global safeguards policy), the set at which definitive new measures were im- of cases against China during the pre-crisis posed during the same time period of 2008, period is much larger, forming a much larger though it is notable to point out that 2008 base from which additional growth has been does cover the low point in the post-1995 slower. Indeed, WTO member use of trade period for new measures imposed.7 remedies to target China's exports is not a new, crisis-related phenomenon, as it con- Policy-Imposing Countries tinues a trend dating back to China's WTO Eight different WTO members imposed at accession in 2001 and even earlier.6 least one definitive import-restricting trade remedy in 3Q 2009. As figure 5 indicates, II. Newly Imposed Import- developing countries imposed 75 percent Restricting Trade Remedies of the definitive new measures, compared In addition to the newly initiated investiga- to developed economies, which imposed 25 tions, WTO members imposed a number percent. Multiple new product-level import of new definitive import-restricting trade barriers were imposed by India (six), Argen- remedies in 3Q 2009. Most of these new im- tina (three), United States (three), Brazil port restrictions were imposed after months (two), European Union (two), and Turkey of consideration for investigations initiated (two). Indonesia and the Philippines each only in 2008. imposed one new definitive import-restrict- ing trade remedy. Figure 5. Newly Imposed Trade Remedies, 1Q 2007­3Q 2009 (nonredundant AD, SG, CSG, CVD at the product level) 30 Imposed by Imposed by Number of imposed remedies 25 developing developed economies economies 20 15 10 5 0 1Q-2007 2Q-2007 3Q-2007 4Q-2007 1Q-2008 2Q-2008 3Q-2008 4Q-2008 1Q-2009 2Q-2009 3Q-2009 Fiscal year (quarter) Source: Global Antidumping Database. OCTOBER 2009 PREMNOTE Affected Sectors (figure 5), the following question arises: Why Newly imposed trade remedies covered a haven't more definitive trade barriers been number of different sectors in 3Q 2009. The imposed during the crisis? This question dominant sector was machinery, as well as is especially relevant for the most recent plastics and rubber, with four new barriers quarters, as a sufficient period of time has each. Other sectors targeted with multiple elapsed for a number of initiated investiga- new definitive import restrictions include tions to reach the stage at which govern- textiles and apparel (three), chemicals ments come to a final decision on whether (three), and iron and steel (two). Agricul- to impose new trade barriers. ture, materials, wood, and miscellaneous One possible contributing explanation manufactures each faced one newly imposed to this phenomenon is that governments measure in 3Q 2009. may be imposing final measures less fre- quently at the completion of investigations Targeted Exporters than they have in the past. Figure 6 provides China was the exporting country most fre- an initial examination of this potential ex- quently targeted by imposition of new im- planation. The figure plots, for each quarter, port-restricting trade remedies in 3Q 2009. the share of all completed investigations China's exporters were targeted in 56.3 that result in the imposition of final trade percent (9 of the 16) of new product-level barriers.9 The figure covers the 1Q 2007­3Q import-restrictions imposed under the trade 2009 period and provides two series of data: remedy laws (AD, CVD, CSG) that require the investigations undertaken by developed the investigating country to name at least economies and those undertaken by devel- one exporting country.8 While high, this is oping economies. Note that that historical lower than the 100 percent (17 out of 17) of research examining the pre-crisis data on cases in 2Q 2009 in which the imposition of trade remedy investigations finds that de- a product-level trade barrier included China veloping countries have traditionally had as one of the exporting countries. a higher share of completed investigations Other countries whose exporters faced resulting in the imposition of measures. So the imposition of more than one newly the difference in the levels of the two series imposed trade remedy in 3Q 2009 include of data is not what is potentially interesting. Indonesia (four), Malaysia (three), and Thai- What is important is how each series of data land (three). The only developed economy may be changing over time. exporter facing imposition of a country- The movements in the two data series specific trade remedy was the United States in figure 6 track relatively closely during the at the conclusion of the European Union crisis period, beginning in 2Q 2008. During investigation of Biodiesel. the three consecutive quarters of 3Q 2008, 4Q 2008, and 1Q 2009, the share of com- III. The Share of Investigations pleted investigations resulting in imposition Resulting in Trade Barriers of final trade barriers was consistently above Given the sharp increase in newly initiated 80 percent for both developed and develop- investigations in 2008 (figure 1), combined ing countries. However, beginning in 2Q with the more moderate increase in imposi- 2009, this number has fallen on average and tion of definitive measures to date in 2009 for each group separately. This provides at PREMNOTE OCTOBER 2009 Figure 6. Share of Completed Investigations Resulting in Definitive Import Barriers, by Investigating Country Category, 1Q 2007­3Q 2009 100 Completed investigations (%) 90 80 70 60 50 40 Developed 30 20 Developing 10 0 1Q-2007 2Q-2007 3Q-2007 4Q-2007 1Q-2008 2Q-2008 3Q-2008 4Q-2008 1Q-2009 2Q-2009 3Q-2009 Fiscal year (quarter) Source: Global Antidumping Database. least some (weak) evidence that each group import barriers, but broken down by the is imposing new trade barriers over a smaller category of exporting country being inves- share of completed investigations in 2Q and tigated, and separating developing country 3Q 2009 than they were in the immediately exporters into two subcategories: (i) China preceding quarters. And across all imposing and (ii) all other developing countries. The countries, only 50 percent of the investiga- share of completed investigations in which tions completed in 3Q 2009 resulted in the China's exporters end up facing definitive imposition of new trade barriers.10 trade barriers is consistently higher than for Figure 7 presents the same data on the other developing country exporters. Never- share of completed investigations that have theless, as was the case in figure 6, figure 7 resulted in the imposition of final, definitive indicates a general downward trend for all Figure 7. Share of Completed Investigations Resulting in Definitive Import Barriers, by Investigated Exporting Country Category, 1Q 2007­3Q 2009 100 90 Completed investigations (%) 80 70 60 50 40 30 20 China Developing economies (non-China) 10 Developed economies 0 1Q-2007 2Q-2007 3Q-2007 4Q-2007 1Q-2008 2Q-2008 3Q-2008 4Q-2008 1Q-2009 2Q-2009 3Q-2009 Fiscal year (quarter) Source: Global Antidumping Database. OCTOBER 2009 PREMNOTE exporters in 2Q and 3Q 2009 in the share of the Republic of Korea, Taiwan, Trinidad and completed investigations that have resulted Tobago, Turkey, Uruguay, the United States, in the imposition of final trade barriers. and República Bolivariana de Venezuela. Any decrease in the share of completed According to data from the WTO, for the investigations that end with trade barriers economies that still control use of their own being imposed is likely due to a number of trade policies in 2009--that is, subtracting factors. One contributing explanation may out newly acceded member states to the be that national governments are taking the EU that used AD prior to their own ac- WTO's legal and evidentiary criterion for cession--these 25 members initiated 91.5 implementing new trade barriers seriously.11 percent of all antidumping investigations by For example, while the global economic the WTO membership during 1995­2008. crisis may have led to circumstances (and Thus tracking data from these economies thus evidence) in which many industries pe- serves as a relatively comprehensive sample titioning for import protection under trade likely to reflect general trends in the WTO remedy laws are "injured," this is only one of membership. the conditions necessary for governments to The antidumping data provided above impose new trade barriers. If governments are collected from each country's national do not also find evidence that this injury is government publications and made publicly caused by dumped, surging, or subsidized available on their Web sites. imports, depending on the trade remedy The 17 WTO members from whom law under which the investigation is taking the countervailing duty data derive are: place, the WTO rules mandate that govern- Argentina, Australia, Brazil, Canada, Chile, ments refuse to implement the new trade Colombia, the European Union, India, barriers. While mainly speculative at this Mexico, New Zealand, Peru, South Africa, point, if governments are rejecting calls for the Republic of Korea, Taiwan, Turkey, the new trade barriers with higher frequency, it United States, and República Bolivariana could be because they are adhering more de Venezuela. According to data from the closely to WTO guidelines. WTO, these 17 members initiated over 90 percent of all countervailing duty investi- Sources gations by the WTO membership during Version 5.0 of the publicly available Global 1995­2008. Thus, this serves as a relatively Antidumping Database can be found at http:// comprehensive sample likely to reflect gen- www.brandeis.edu/~cbown/global_ad/. The eral trends in the WTO membership. The fully updated version 5.1, complete with countervailing data provided are collected detailed data on activity through the third from national government publications, quarter of 2009, will be available in late the WTO's semi-annual reports, and news October 2009. agency Web sites. The 25 WTO members from whom the Data from WTO members' use of global antidumping data derive are: Argentina, safeguards and China-specific safeguards are Australia, Brazil, Canada, Chile, China, Co- taken from the WTO and national govern- lombia, Costa Rica, Ecuador, the European ment publications. Union, India, Jamaica, Mexico, New Zea- land, Pakistan, Paraguay, Peru, South Africa, PREMNOTE OCTOBER 2009 Notes import tariffs under the China-specific safe- 1. Detailed data on cases can be found on guard provision, that is, USTR (2009) "Kirk: the Global Antidumping Database Web site. White House Fulfilling Trade Enforcement The 82 different trade remedy investigations Pledge With Announcement of Remedies in during 3Q 2009 can be reduced to the 44 Chinese Tire Case," USTR Press Release of (nonredundant) product-level investigations 11 September. illustrated in the figure. To make comparable 5. Of the 44 product-level newly initiated the data on policy use across different (AD, investigations in 3Q 2009, 6 did not name CVD, SG, CVD) trade remedy laws, AD or any exporting countries because they were CVD investigations (measures) against mul- global safeguards, and China initiated one tiple exporting countries are treated as one investigation itself. product-level investigation (measure). For 6. For a discussion and analysis, see Chad example, Brazil's two antidumping investi- P. Bown (forthcoming) "China's WTO En- gations of "Polypropylene Resin" from India try: Antidumping, Safeguards, and Dispute and the United States are treated as one Settlement," in Robert Feenstra and Shang- product-level investigation. Furthermore, to Jin Wei, eds., China's Growing Role in World ensure that they are not redundant, a WTO Trade. Chicago, IL: University of Chicago member's simultaneous AD and CVD inves- Press for NBER. tigations (measures) over the same product 7. The 31 different trade remedy investiga- are treated as one investigation (measure). tions that resulted in the imposition of defini- For example, Canada's simultaneous AD and tive new import restrictions during 3Q 2009 CVD investigations of "Certain Oil Country can be reduced to the 20 (nonredundant) Tubular Goods" from China are treated as product-level trade barriers illustrated in the one product-level trade remedy investigation. figure. Indeed, the low point for newly im- Finally, associated with its terms of acces- posed definitive trade remedies occurred in sion to the WTO agreement in 2001, WTO 2Q 2008 (figure 5), that is, four quarters after members were granted access to a transitional the low point for newly initiated trade remedy (until 2014) China-specific safeguard (CSG) investigations in 2Q 2007 (figure 1). with which they can implement new China- 8. Of the 20 product-level newly imposed specific import restrictions if there is evidence trade barriers in 3Q 2009, 4 did not name of injury (or a threat thereof) to a domestic any exporting countries because they were industry associated with increased imports global safeguards. from China. 9. We define a "completed investigation" 2. The year 2007 was the low point in the as any initiated investigation that terminates global use of trade remedies during the pe- in a given quarter because of any of the fol- riod since 1995. lowing: (i) definitive measures were imposed; 3. Following the U.S. initiation of a global (ii) the final decisions in the investigation safeguard investigation covering billions of were made and the government decided dollars of imported steel in June 2001 (result- that no measures would be imposed; (iii) the ing in a March 2002 imposition of definitive preliminary decisions in the investigation safeguard measures), other WTO members were made and the government decided followed on by initiating 10 safeguard inves- that no measures would be imposed; or (iv) tigations over similar steel imports during the domestic industry withdrew the petition 2002 alone. requesting new import restrictions. Note 4. See, for example, the United States that for figures 6 and 7, an investigation is decision in September to impose 35 percent defined at the product-exporter level (except OCTOBER 2009 PREMNOTE 9 for global safeguard investigations), to allow ferent export sources of "Hot Rolled Steel for the data to determine whether definitive Products," a set of cases initiated in 4Q 2008. trade barriers may be imposed on one set However, this does not explain the decline of named exporters but not others. Global in the developed economy share of completed safeguard investigations are dropped from investigations that have resulted in imposed consideration in figure 7, which focuses on measures. exporting countries. 11. See Chad P. Bown (2009), "Confronting 10. An important caveat to the 3Q 2009 the Protectionism Spawned by the Crisis," data is recognition that a significant deter- in G20 Summit: Recovering from the Crisis. minant of the developing country decline is Brookings Institution, Global Economy and termination of a single product-level case in Development Program, September. India, which had been investigating 15 dif- This note series is intended to summarize good practices and key policy findings on PREM-related topics. The views expressed in the notes are those of the authors and do not necessarily reflect those of the World Bank. PREMnotes are widely distributed to Bank staff and are also available on the PREM Web site (http://www.worldbank.org/prem). If you are interested in writing a PREMnote, email your idea to Madjiguene Seck at mseck@worldbank.org. For additional copies of this PREMnote please contact the PREM Advisory Service at x87736. PREMnotes are edited and laid out by Grammarians, Inc. Prepared for World Bank staff