86266 AUTHOR ACCEPTED MANUSCRIPT FINAL PUBLICATION INFORMATION Is Protectionism on the Rise? Assessing National Trade Policies during the Crisis of 2008 The definitive version of the text was subsequently published in Review of Economics and Statistics, 95(1), 2013-03 Published by MIT Press THE FINAL PUBLISHED VERSION OF THIS ARTICLE IS AVAILABLE ON THE PUBLISHER’S PLATFORM This Author Accepted Manuscript is copyrighted by the World Bank and published by MIT Press. It is posted here by agreement between them. Changes resulting from the publishing process—such as editing, corrections, structural formatting, and other quality control mechanisms—may not be reflected in this version of the text. You may download, copy, and distribute this Author Accepted Manuscript for noncommercial purposes. Your license is limited by the following restrictions: (1) You may use this Author Accepted Manuscript for noncommercial purposes only under a CC BY-NC-ND 3.0 Unported license http://creativecommons.org/licenses/by-nc-nd/3.0/. (2) The integrity of the work and identification of the author, copyright owner, and publisher must be preserved in any copy. (3) You must attribute this Author Accepted Manuscript in the following format: This is an Author Accepted Manuscript of an Article by Kee, Hiau Looi; Neagu, Cristina; Nicita, Alessandro Is Protectionism on the Rise? Assessing National Trade Policies during the Crisis of 2008 © World Bank, published in the Review of Economics and Statistics95(1) 2013-03 http://creativecommons.org/licenses/by-nc-nd/3.0/ © 2014 The World Bank NOTE IS PROTECTIONISM ON THE RISE? ASSESSING NATIONAL TRADE POLICIES DURING THE CRISIS OF 2008 Hiau Looi Kee, Cristina Neagu, and Alessandro Nicita* Abstract—This paper quantifies trade policy changes and the associated Many recent papers have studied the trade impact of the global trade impacts for about 100 countries between 2008 and 2009. Results crisis in 2008 (see Baldwin & Evenett, 2009, and Baldwin, 2009). show that there has been no widespread increase in protectionism. Only a few countries, including Russia, Argentina, Turkey, and China, have While consensus has yet to emerge among researchers, the two lead- increased tariffs on major imported products. The United States and the ing explanations for the large and synchronized drop in trade are the EU, by contrast, rely mainly on antidumping duties to shield domestic role of international supply chains (Yi, 2009) and the lack of trade industries. Overall, while the rise in tariffs and antidumping duties may credits and finance during the crisis period (Amiti & Weinstein, 2009; have jointly caused global trade to drop by US$43 billion, it explains less Chor & Manova, 2009). In a unified framework, Eaton et al. (2010) than 2% of the collapse in world trade during the crisis period. merges an input-output framework with a gravity trade model of the world and shows that changes in demand play the most significant role I. Introduction in explaining the large drop in trade-to-GDP ratio during the crisis, while trade frictions, which include trade policies and trade credits, W ITH the dramatic collapse of world trade in the wake of the biggest global recession in recent history, many have feared that governments may respond by increasing tariffs and other trade pol- explain at most about 15% of the collapse in trade. Thus, trade pol- icy as a protectionist device has not been seen to play a substantial role in the global collapse of trade—neither as a cause nor a conse- icy barriers to protect their domestic economies, which may indirectly quence. Nevertheless, anecdotal evidence suggests that some countries prolong the recession and lead to domestic unrest. In fact, in December are actively tinkering with their trade policies. For example, during 2008, the first of the crisis-related demonstrations erupted in several the crisis period, Bolivia, Ecuador, and Turkey have altered their tar- cities in Russia over the increase in car tariffs (Levy, 2008). iffs on a large share of their imported products, and India increased Has protectionism been rising since fall 2008? To answer this ques- its use of antidumping (AD) duties. How important are those changes tion, we compare the overall trade restrictiveness Indices (OTRI) of a in explaining or prolonging the collapse in world trade? The objec- wide range of countries in 2008 and 2009. The OTRI summarizes the tive of this paper is to carefully compare the trade policies of a wide trade policy stance of a country by calculating the uniform tariff that range of countries over the crisis period and assess the extent of the will keep its overall imports at the current level when the country in fact fall in trade due to the increase in tariffs and AD duties of these has different tariffs for different goods. Unlike trade-weighted average countries. For the purpose of this paper, we narrowly define trade poli- tariffs, the OTRI takes into account the importance of each good in total cies to include only tariffs and AD duties. Due to data limitations, imports, as well as the responsiveness of the import of each good with we do not look at other policies that may affect trade, such as gov- respect to its tariff. Thus, not only are the weights proportionate to the ernment bailouts and buy-national requirements, which could play a import value of the goods, but goods that have a larger fall in imports much larger role than tariffs and AD in affecting trade during the crisis when tariffs are imposed (those goods that are highly elastic in demand) period. are also given larger weights. The empirical methodology of the OTRI To achieve our objective, we obtained the most favored nations was first developed by Kee, Nicita, and Olarreaga (2008, 2009), based (MFN) applied tariff schedules and the bilateral tariff schedules for on the theoretical underpinning of Anderson and Neary (1994, 1996, a wide range of countries in 2008 and 2009.1 The MFN-applied tariffs 2003). Irwin (2010) also uses a similar methodology to study the his- tend to overestimate the level of protection because they do not account toric protection level of the United States from 1867 to 1961. A major for the existence of bilateral or regional tariff preferences. Hence, it benefit of looking at the changes in the OTRI over the crisis period is is important for us to construct the OTRI based on the bilateral tariff that it allows us not only to measure the changes in trade policy but schedules. This significantly complicates the calculation of the OTRI also to quantify the drop in trade due to those changes. This is the point because each country may have up to 200 trading partners and each bilat- of departure of our paper from the previous literature, which tends to eral tariff schedule consists of nearly 5,000 Harmonized System (HS) focus on only average tariff increases or the percentage of tariff lines 6 digit products. To capture the effect of antidumping, we also merge that have increased during the crisis period. the bilateral tariff schedules with the World Bank Global Antidump- ing Database. Thus, changes in the OTRI reflect trade policy changes related to both the changes in applied tariffs and antidumping duties Received for publication May 25, 2010. Revision accepted for publication during the crisis period. June 8, 2011. In addition, we need bilateral import demand elasticities and bilateral * Kee and Neagu: World Bank; Nicita: U.N. Conference on Trade and Development. trade flow data to properly weigh these bilateral tariffs. We modify the We are grateful to the International Trade Center in Geneva for kindly shar- multilateral import demand elasticity estimates in Kee et al. (2008) to ing the data with us and to Richard Newfarmer and Elisa Gamberoni for obtain bilateral import demand elasticities. Bilateral trade flow data are facilitating the request. We are indebted to Chad Bown for sharing his data from Comtrade. Finally, to make sure that changes in the OTRI period on antidumping duties and all the stimulating discussions and comments. We also thank Ann Harrison for feedback on a previous draft. Feedback purely capture changes in trade policies, we use the 2008 bilateral trade from Daniel Lederman, Caglar Ozden, and participants of the World Bank flows and elasticities as fixed weights. As such, changes in trade or DECRG Crisis Workshop in January 2010 is acknowledged. The findings, elasticity due to demand shocks will not affect our OTRI measures. interpretations, and conclusions expressed in this paper are entirely our own and should not be attributed to the World Bank or the United Nations Con- ference on Trade and Development. A supplemental appendix is available online at http://www.mitpress 1 As a robustness check, we also compare the tariffs of 2007 to those of journals.org/doi/suppl/10.1162/REST_a_00241. 2009. The results are very similar and available on request. The Review of Economics and Statistics, March 2013, 95(1): 342–346 © 2013 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology NOTES 343 Going through the schedules of all countries in our data set, we II. Change in the Overall Trade Restrictiveness Index found that overall, there has been no widespread increase in tariffs. Although many countries have increased tariffs on imported products, The OTRI summarizes the impact of each country’s trade policies on its aggregate imports. The OTRI’s conceptual framework was first the trade impact has generally been minimal. However, for a handful of proposed in Anderson and Neary (1994, 1996, 2003), it was simplified countries, tariff increases on important items in agriculture and manu- in Feenstra (1995), and it was empirically estimated in Kee, Nicita, facturing pushed up their OTRI and significantly affected trade. Russia, and Olarreaga (2008, 2009). The OTRI answers the following question: Malawi, and Argentina all increased tariffs on manufacturing products What is the uniform tariff that, if imposed on home imports, would leave that caused their OTRI to increase by 0.9 to 1.2 percentage points and the aggregate imports at their current level? In a partial equilibrium, their trade flows to drop by US$4.8 billion, US$29 million, and US$914 when we ignore the substitution between products and the potential million, respectively. Turkey increased tariffs on a wide range of agri- income effect due to tariff revenue redistribution, the OTRI is just a cultural products, which raised its OTRI by 0.8 percentage points and more sophisticated way to calculate the weighted average tariff of a caused its trade flow to decrease by US$2.2 billion. With the removal country, with the weight of a good set equal to the product of the good’s of a temporary tariff reduction on palm oil and the introduction of some import demand elasticity and its share in total import. Irwin (2010) also antidumping duties, India had a large increase in the level of protection- applies the same approach to study the historic level of protection of ism on agriculture products (8.3 percentage points), even though this the United States. was offset by tariff liberalization in the manufacturing sector such that We refer readers to Kee, Nicita, and Olarreaga (2008, 2009) for the India’s OTRI increased only by 0.1 percentage points. Other countries formal derivation of the OTRI of a country. In this paper, we adopt that had large drops in trade due to increases in tariffs include China a fixed-weight method to compare the OTRI of a country across two (US$5 billion), Canada (US$1.8 billion), and Brazil (US$991 million). years, where the bilateral trade flow data and elasticity estimates of the Finally, for the United States and the EU, although the tariff schedules base year, 2008, are used as weights:2 remained roughly the same throughout our period of analysis, spikes n p εncp tncp − tncp 2008 2008 2009 mncp 2008 in antidumping duties caused their OTRI to increase by 0.5 percent- OTRI2009 c − OTRI2008 c = 2008 ε2008 , (1) age points, and 0.1 percentage points, respectively. Jointly, if we add n p mncp ncp up all the decreases in trade for all countries during the crisis period where mncp is the bilateral import value of country c for good n from due to changes in tariffs and antidumping duties, in the worst-case sce- partner country p, tncp is the ad valorem tariff of country c on good n nario, the total decrease in imports is about US$43 billion, which is from country p, and εncp is the import demand elasticity of country c for less than half a percent of the world’s imports in 2008. According to good n from country p. Superscripts indicate the year of the variables. the latest estimate of the World Trade Organization (2010), the world’s In this way, the difference in the OTRI of a country between 2008 and imports decreased by 24% from its precrisis level. Thus, trade policies 2009 captures only trade policy changes and does not reflect the collapse can explain at most 2% of the sharp drop in world trade. This sug- of trade during the crisis period. gests that protectionism was not the main culprit behind the collapse of While the trade policy of a country could also consist of other nontar- world trade and the collapse of world trade did not cause protectionism iff measures, here, due to data limitations, we focus mainly on tariffs. to increase. However, unlike the earlier papers, we use the bilateral tariffs between Countries have been restrained in terms of raising their tariffs and AD country pairs at the HS 6 digit good level in our calculation of the for several reasons. First, most countries are part of bilateral, regional, or OTRI. Moreover, we also employ the bilateral import demand elastic- multilateral trade agreements, which may have significantly restricted ity at the same level of aggregation as the tariffs. Finally, when possible, their ability to adjust tariffs during the crisis period. The limitations we include any antidumping duties that were imposed during the crisis in policy space due to multilateral obligations are more relevant for period. developed countries, such as the EU and the United States, where Once the change in the OTRI of a country is calculated, some back- the difference between bound and applied rates is generally small. of-the-envelope calculations can be made to figure out the impact on This is not the case for most developing countries. Those are gener- trade-flows. One way is to use the change in the OTRI multiplied by ally more constrained by obligations within regional and preferential the trade-weighted import demand elasticities of the country. Then, trade agreements. Second, countries may recognize the adverse long-run Change in trade using the OTRI impact of those protectionist policies in the context of an increas- ingly globalized economy. This is particularly the case if the exports = OTRIc 2009 − OTRIc 2008 εncp . 2008 2008 mncp (2) n p of the countries depend heavily on imported materials: higher tariffs will severely affect exports, thus further hindering economic recov- Note that there is no bound for the calculation of change in trade in this ery. Similarly, global production chains and foreign direct investment formula, and thus it is possible for the change in trade to be higher than (FDI) that span national borders have made it harder to distinguish the existing level of trade. An alternative approach would be to calculate domestic from foreign. Thus, many multinational firms find that tra- the change in tariff at the tariff line level for each product from each ditional forms of protectionism are contrary to their interests. This partner country, multiply that by the bilateral import demand elasticity could explain why most countries continued to liberalize their tar- 2 For the purposes of this paper, we also calculated bilateral import demand iff policies during the crisis rather than raising tariffs. In this regard, elasticities, which vary across countries, products, and partners. For each carefully targeted AD may well be the more suitable policy choice. product n imported by country c from partner country p, we rely on the However, the modest increase in AD cases during the crisis suggests following formula and on estimates of the GDP function parameter, ann , that firms may also have found this instrument inadequate to protect from Kee et al. (2008) to construct bilateral import demand elasticities, their interests. where snc is the share of trade in product n in the GDP of country c in 2008 and sncp is the share of trade in product n from partner country p in the GDP This paper is organized as follows. We first briefly discuss the method- of country c in 2008 (refer to Kee et al., 2010, for a proof): ology behind the OTRI calculation in section II. Section III presents the ann data coverage. Section IV discusses the results, and section V concludes. εncp = + sncp − 1. snc 344 THE REVIEW OF ECONOMICS AND STATISTICS Figure 1.—Comparing Average Bilateral Tariffs, 2008 versus 2009 Table 1.—Antidumping Duties Affected Imports in 2008–2009 Simple Average Bilateral Tariffs in 2008 vs. 2009 Share in Share of AD Import Country Value (US$000) Total Import (%) with AD Data (%) .25 Argentina 336,499 0.59 32.33 Australia 50,931 0.03 100.00 .2 Brazil 657,543 0.38 76.14 Average Tariff in 2009 Canada 578,787 0.14 100.00 Chile 350 0.00 100.00 .15 MWI CHN BLR China 990,444 0.10 100.00 KOR ARG BRA Colombia 21,919 0.06 100.00 BOL European Union 8,560,695 0.38 100.00 PYF .1 India 1,405,095 0.44 23.35 Japan 27,417 0.004 1.00 GTM Mexico 3,171 0.00 100.00 .05 CAN Turkey 361,681 0.18 2.03 United States 3,538,908 0.16 100.00 Data retrieved from the Global Anti-Dumping Database of the World Bank. For India, the actual AD- 0 affected trade is US $2.2 billion; however, only $1.4 billion is matched to tariff reclassifications. 0 .05 .1 .15 .2 .25 Average Tariff in 2008 crisis. As long as these duties were not removed from the second quarter of 2008 onward, they do not affect the change in the level of protection- to obtain the change in trade at tariff line level, and then constrain the ism. Only the new cases and the removal of old duties are factored into fall in trade to be no more than the level of imports in 2008. Summing the calculations. all changes in trade at the tariff line level across all partners gives the Table 1 presents some summary statistics on the countries that have total change in trade: added antidumping duties since the second quarter of 2008.5 For the most part, changes in antidumping duties affect less than 1% of imports, Change in trade using tariffs ranging from US$8.5 billion in the EU to US$350,000 in Chile. Nev- ertheless, given that some countries cannot unilaterally increase their = max mncp εncp tncp − tncp 2008 2008 2009 2008 , −mncp 2008 . (3) tariffs without violating WTO agreements, AD may well be one of n p those few legitimate channels to increase trade protection during the crisis period. In addition, given that AD duties are imposed at the tariff III. Data line level, which for many countries is represented by eight- or ten- digit HS codes, we first need to identify the share of these goods in the We obtained tariff data for 135 countries from the International Trade bilateral trade of the corresponding HS 6 categories, and impose AD Center (ITC) in Geneva. For India, Japan, and South Korea we supple- duties only on the goods affected. In doing so, we avoid imposing AD mented the ITC data with MFN schedules from other sources.3 Figure 1 duties on all tariff-line goods within an HS 6 category, even though we summarizes the raw data by plotting the average bilateral tariff of 2008 are still making the assumption that AD duties affect all bilateral trade against that of 2009, along with a 45 degree line. Only ten countries within HS 8 goods and are not distinguishable among different firms have higher average tariffs in 2009 than those of 2008. These countries that export. For some countries, such as Turkey and India, only a por- are labeled in the figure and include Korea, Brazil, Argentina, China, tion of AD cases have information on the actual AD duties imposed (see and Canada. For Canada, the average bilateral tariff was 3.2% in 2008 table I, last column). For the missing AD duties, we use the inverse of and 4.0% in 2009.4 our bilateral import demand elasticity estimates to infer the minimum Data for antidumping duties are retrieved from the publicly available prohibitive AD duties. Global Antidumping Database of the World Bank, which is maintained by Chad Bown (2009a). The data set provides detailed information on IV. Results antidumping cases. While data can be traced back as far as the early 1990s, given that our focus is the changes during the 2008–2009 period, Figure 2 plots the level of OTRI constructed using bilateral tariffs in we use only those cases that are initiated between June 2008 and Sep- 2008 against the level of OTRI constructed using bilateral tariffs in 2009 tember 2009, net of antidumping duties that were removed during the and AD imposed during the 2008–2009 period.6 Most are located below same period. By doing so, we capture only the change in antidump- ing duties during the two-year period and not the level of antidumping 5 In addition to the thirteen countries listed in table 1, the global antidump- for each of the two years. This is an important point, because many ing database also has information for five more countries of the 135 in our antidumping duties in 2008 and 2009 are due to cases filed before the data set. Pakistan is not included because we have no data on its 2009 tariff schedules; and we also have no trade flow data for South Korea and South Africa at the tariff line level; and we fail to match the AD data with trade 3 India’s 2008 and 2009 MFN schedule as well as Japan’s 2008 data for Indonesia and Peru due to tariff reclassification. MFN schedule come from TRAINS. Japan’s 2009 MFN schedule was obta- 6 In the working paper version of this paper, we also compare the OTRI ined from http://www.customs.go.jp/english/tariff/2010/index.htm. South constructed using MFN tariffs with the OTRI constructed using bilateral Korea’s 2009 MFN schedule comes from http://english.customs.go.kr tariffs. We found that MFN tariffs tend to overestimate the level of protec- /kcsweb/user.tdf?a=user.customtariff.CustomTariffApp&c=1001&mc tion of a country by 75%. On the other hand, allowing for bilateral import =ENGLISH_INFORMATION_KOREA For these three countries, we demand elasticities marginally increases the overall level of protection, as lacked ad valorem equivalents of 2009 specific tariffs; hence, we used the bilateral elasticities tend to be larger than multilateral elasticities that are 2008 values. common across all trading partners within an imported product. At the sam- 4 A detailed data summary table is available on request, and it is also ple mean, the OTRI constructed using bilateral elasticities is 2% larger than included in the working paper version of this paper (Kee, Neagu, & Nicita, the OTRI constructed using multilateral elasticities. Finally, our results are 2010). robust to the exclusion of AD from the calculation. NOTES 345 the 45 degree line, indicating that most have further liberalized during Figure 2.—Comparing the OTRI in 2008 and 2009 the crisis. However, there are quite a few exceptions, notably Malawi, OTRI in 2008 vs. 2009 Russia, Turkey, China, Argentina, Canada, and Brazil. Malawi’s OTRI .25 for 2008 is 7.1%, and for 2009 it is 8.3%, which implies an increase of 1.2 percentage points. Russia increases its OTRI from 9.6% to 10.8%. For Turkey, the increase in tariffs of agricultural products pushes up its .2 OTRI from 2% to 2.7%. The OTRI of China, Argentina, and Canada each increases by 0.3 percentage points. Although small, such increases OTRI_2009 .15 in the overall level of tariff protection could significantly disrupt trade if imports are very elastic.7 RUS For most countries, adding AD does not change their OTRI in any sig- .1 IND MWI BRA nificant way, with the exception of the United States, the EU, and India. CHN CHL Incorporating AD duties increases the OTRI of the United States by .05 ARG QAT SAU half a percentage point. This seemingly small number in fact prompted JPN BOL ARE TUR BLR EUN trade to decrease by US$24 billion, if we allow AD to affect more than CAN USA the existing level of preAD trade (see equation [2]), or by US$3 bil- 0 lion if we assume the maximum effect of AD and other tariff increases 0 .05 .1 .15 .2 .25 cannot exceed the existing trade in 2008 (see equation [3]). Likewise, OTRI_2008 for the EU, incorporating AD duties causes its OTRI to increase by 0.1 percentage point. As a result, imports of the EU drop by US$2 billion. This exercise shows that while antidumping may not increase the overall level of protection by much, it has been in fact the main instrument used Argentina, and China are mainly driven by increases in tariffs in the by the United States and EU. Another heavy user of AD is India. With- manufacturing sector (the rise in the car tariffs of Russia and textile out AD duties, the OTRI of India decreases by 0.2 percentage points tariffs of Argentina).10 from 2008 to 2009.8 Once AD duties are included, the change becomes Overall, if we sum up all the negative trade impacts due to increased positive 0.1 percentage points, indicating that AD duties have made tariffs and AD duties, total world’s imports may have decreased by the overall level of trade restrictiveness of India worse. The net trade US$43 billion. In 2008, the value of world imports was about $11 tril- effect of the changes in tariff and AD duties for India is about US$306 lion, which implies that the changes in trade policy may have decreased million.9 Nevertheless, on a global scale, such duties hardly explain the the world’s imports by 0.4%. According to the latest estimate of the huge collapse in trade, which further suggests that this global collapse in WTO (2010), the world’s imports contracted by 24% in 2009. Thus, trade is probably not because countries are becoming more protectionist our results show that the trade policy changes we have discussed can but instead relates to factors such as demand shocks. explain less than 2% of the collapse in the world’s imports. Detailed analysis further shows that for most countries, most of the changes in the OTRI are driven by increased protection in the agricul- V. Conclusion tural sector. For example, in India, the removal of a temporary tariff reduction on palm oil and the introduction of some antidumping duties The fear that countries may raise tariffs to protect their domestic on agriculture products in 2009 resulted in an increase in protectionism markets in the wake of the largest global recession since the Great in the agricultural sectors by about 8.3 percentage points. Likewise, Depression has not materialized. Comparing the published 2008 and Turkey increased tariffs on a wide range of agricultural products, which 2009 tariff schedules of a wide range of countries shows that only a hand- pushed its OTRI for agricultural goods from 21.2% to 31.4%. Canada ful of countries have raised their tariffs significantly. These countries and Malawi also have large increases in their OTRI on agricultural prod- include Russia, Malawi, Argentina, Turkey, and China. The increase ucts. On the other hand, the overall increases in the OTRI of Russia, in motor vehicle tariffs in Russia not only restricted imports; it also caused one of the first reported crisis-related demonstrations. For some other countries, such as the United State and the EU most of the policy 7 Detailed OTRI estimates for each of the countries are available on request actions during the crisis are not about tariffs but antidumping duties. and are also available in the working paper version of this paper (Kee et al., 2010). Nevertheless, even after taking antidumping duties into account, evi- 8 For Chile, India, and Japan, we use their MFN tariffs and AD to calculate dence provided in this paper suggests that the trade impact due to trade their OTRI, since 2009 bilateral tariff schedules are not available. policy changes can explain no more than 2% of the collapse in the world 9 Our estimated changes in trade are not directly compatible to Bown trade. (2009b). For example, for the worst-case scenarios, Bown’s estimates of the AD impact in the United States, EU, and India are US$7 billion, US$8 billion, and US$4 billion, respectively. The differences can be attributed to 10 The detailed analysis can be found in the working paper version of this the following. First, our estimates are based on tariff line (HS 8 digit) data paper (Kee et al., 2010) or are available on request. rather than HS 6 digit data. In other words, within an HS 6 digit category, only the HS 8 digit goods that are affected by AD are included in the calculation, whereas Bown’s estimates use HS 6 digit trade flows. Second, REFERENCES we use 2008 trade value in our calculation, whereas Bown’s estimates are based on 2007 trade value. Third, our AD coverage is from June 2008 to Amiti, Mary, and David Weinstein, “Exports and Financial Shocks,” NBER September 2009, and Bown’s estimates are from the first quarter of 2008 to working paper 15556 (2009). the first quarter of 2009. Fourth, we take into account the bilateral import Anderson, James E., and J. Peter Neary, “Measuring the Restrictiveness of demand elasticities in the calculation of trade impact due to AD. Finally, Trade Policy,” World Bank Economic Review 8 (1994), 151–169. we include tariffs and AD in our calculation of trade changes, while Bown’s ——— “A New Approach to Evaluating Trade Policy,” Review of Economic estimates focus on only AD. For the EU and India, the negative impacts on Studies 63 (1996), 107–125. trade flows due to AD are partially offset by their overall tariff reduction ——— “The Mercantilist Index of Trade Policy,” International Economic during the two-year period. Review 44 (2003), 627–649. 346 THE REVIEW OF ECONOMICS AND STATISTICS Baldwin, Richard, The Great Trade Collapse: Causes, Consequences and Irwin, Douglas, “Trade Restrictiveness and Deadweight Losses from U.S. Prospects (London: Centre for Economic and Policy Research, Tariffs.” American Economic Journal: Economic Policy 2 (2010), 2009). 111–133. Baldwin, Richard, and Simon Evenett, The Collapse of Global Trade, Kee, Hiau Looi, Cristina Neagu, and Alessandro Nicita, “Is Protectionism Murky Protectionism, and the Crisis: Recommendations for the G20 on the Rise? Assessing National Trade Policies during the Crisis of (London: Center for Economic and Policy Research, 2009). 2008,” World Bank Policy Research working paper 5274 (2010). Bown, Chad P., “Global Anti-dumping Database” [Version 5.1] World Bank Kee, Hiau Looi, Alessandro Nicita, and Marcelo Olarreaga, “Import (2009a). Demand Elasticities and Trade Distortions,” this review 90 (2008), ——— “The Global Resort to Anti-Dumping, Safeguards, and Other Trade 666–682. Remedies amidst the Economic Crisis” (pp. 91–118), in Simon J. ——— “Estimating Trade Restrictiveness Indices,” Economic Journal 119 Evenett, Bernard M. Hoekman, and Olivier Cattaneo (eds.), Effective (2009), 172–199. Crisis Response and Openness: Implications for the Trading System Levy, Clifford J., “Protests Erupt in Rusia over Raising of Car Tariffs,” New (London: CEPR and World Bank, 2009b). York Times, December 21, 2008. Chor, Davin, and Kalina Manova, “Off the Cliff and Back? Credit Con- World Trade Organization, “Consultation with Ecuador,” Background ditions and International Trade during the Global Financial Crisis,” Document by the Secretariat, WT/BOP/S/15/Rev.1 (2009). unpublished manuscript (2009). ——— “World Trade 2009, Prospects for 2010,” press/598 (2010). Eaton, Jonathan, Samuel Kortum, Brent Neiman, and John Romalis, “Trade Yi, Kei-Mu, “The Collapse of Global Trade: The Role of Vertical Specializa- and the Global Recession,” unpublished manuscript (2010). tion,” in Richard Baldwin, and Simon Evenett (eds.), The Collapse Feenstra, Robert, “Estimating the Effects of Trade Policy” (vol. 3, pp. 1553– of Global Trade, Murky Protectionism, and the Crisis: Recommen- 1595), in Gene Grossman and Kenneth Rogoff (eds.), Handbook of dations for the G20 (London: Centre for Economic Policy Research, International Economics (Amsterdam: Elsevier, 1995). 2009).