65939 POVERTY THE WORLD BANK REDUCTION AND ECONOMIC MANAGEMENT NETWORK (PREM) Economic Premise DECEMBER 2011 • Number 72 JUN 010 • Numbe 18 Taking Stock of Trade Protectionism Since 2008 Mohini Datt, Bernard Hoekman, and Mariem Malouche Following the onset of the financial crisis in September 2008 and the subsequent “Great Trade Collapse� (Baldwin 2009), many countries actively used trade policy instruments as part of their response to the global recession. Governments pursued a mix of trade liberalization, trade promotion, and trade restrictions. The choice of trade policy has varied, with limited use of tariff hikes or antidumping and safeguard actions. Sector-specific support to industries dominated initial responses to the crisis, and there has been increasing resort to nontariff measures. Recent research suggests that vertical specialization—the growth in global supply chains—has played a significant role in limiting the use of traditional protectionist instruments. Pressures on governments to support domestic economic activity may increase, given current gloomy economic prospects and more binding macroeconomic policy constraints, and the number of protectionist measures has recently risen. Open trade cannot be taken for granted, thus the need for monitoring persists. Active Use of Trade Policy Measures and are only “discovered� over time. The number of new pro- tectionist actions peaked in the first quarter of 2009 and bot- According to monitoring reports issued by the World Trade Or- tomed in the third quarter of 2010. GTA data suggest that new ganization (WTO), a total of 1,243 new trade measures have been implemented since the onset in late 2008 of the current protectionist measures in 2011 are running at levels similar to financial crisis. Some 900 of these measures were trade restric- what was observed in 2010, and protectionist actions in the tive (figure 1), while 327 reduced the level of import protection third quarter of 2011 alone are as high as in the worst periods (figure 2). Each year since the crisis erupted has seen about 300 of 2009 (Evenett 2011c). G-20 countries account for the bulk new trade restrictions imposed. Data up to July 2011 had indi- of all measures according to both sources, and their share has cated a drop in this annual tally, but data for Q3 2011 indicate increased from 60 percent of the total in 2009 to about 80 per- the tally has risen again.1 cent in 2011—implying that the largest countries are the most The Global Trade Alert (GTA), which covers a larger spec- active users of trade policy instruments (figure 4). trum of actions that may affect trade, reports a total of 1,593 Many countries actively pursued liberalizing measures, par- measures implemented between November 2008 and Novem- ticularly in the second and third years after the onset of the cri- ber 2011, of which 1,187 discriminated against foreign suppli- sis (figure 2). In fact, tariffs were more frequently lowered/lib- ers and 406 were liberalizing (figure 3).2 The total number of eralized than they were hiked, reflecting efforts to lower prices new trade measures has consistently increased over the 10 sum- (costs) of goods used by domestic industries and/or consumed mary monitoring reports the GTA has issued to date, reflecting by households (mostly food items). The number of all trade lib- the fact that many measures are not made public immediately eralizing measures increased by two-thirds between October 1 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise Figure 1. Trade Restrictive Measures Figure 2. Trade Liberalizing Measures Oct 2010 – 141 12 Oct 2010 – Oct 2011 99 53 Oct 2011 26 123 8 Oct 2009 – Sep 2010 87 145 24 Oct 2009 – Sep 2010 57 85 3 Sep 2008 – Sep 2009 60 8 173 Sep 2008 – 53 0 20 40 60 80 100 Sep 2009 86 trade facilitation tarif f reductions NTMs terminated 0 50 100 150 200 Source: Authors’ calculations using WTO data from 2009, 2010b, 2011b. Note: Total = 327; NTMs = nontariff measures. TTB initiations tarif f s NTMs Source: Authors’ calculations using WTO data from 2009, 2010b, 2011b. Note: Total = 916; TTB = temporary trade barriers (antidumping, countervailing duties, safeguards); NTMs = nontariff measures. Figure 3. Global Trade Alert, Nov. 2008 – Nov. 2011 Figure 4. Number of Total G-20-Imposed Restrictive Measures by Country Income Group 1,027 250 678 200 406 150 265 160 100 103 50 worldwide G-20 0 Sep 2008 – Oct 2009 – Oct 2010 – trade-liberalizing measures Sep 2009 Sep 2010 Oct 2011 measures that may discriminate G-20 high income G-20 middle income measures that are discriminatory Source: Authors’ calculations using WTO data from 2009, 2010b, 2011b. Source: Global Trade Alert. Note: Total = 1,593. 2009 and September 2010 compared with the previous year. Differentiated Use of Import Tariffs One-fifth of the restrictive nontariff measures implemented The scope that exists to raise tariffs is a function of the extent to since the crisis were subsequently removed. which countries have bound their tariffs in the WTO. Many Countries can be grouped into trade policy activists—pursu- developed countries have bound their tariffs at the actual ap- ing a mix of trade-restricting and trade-liberalizing actions—and plied levels, whereas many developing countries have not trade policy passivists. The most active users of trade policy since bound tariffs in the WTO or have made so-called ceiling bind- 2008 include Brazil, China, India, and the Russian Federation— ings that are much higher than their applied tariff rates. This the BRICs.3 These countries—as did others such as Pakistan and implies that developing countries often have significant scope Indonesia—imposed both new trade restrictive measures and to raise tariffs in response to a recession, whereas high-income took actions to reduce the prices of certain imports (figure 5). nations do not. India initiated the greatest number of restrictive trade measures, Excluding antidumping and other forms of contingent pro- followed by Argentina, the European Union, Brazil, Indonesia, tection, tariffs were more frequently lowered than raised. No- the United States, China, and Russia. Among the non-G-20 mid- table examples of tariff liberalization by G-20 members in- dle-income economies, Vietnam, Pakistan, Ukraine, Belarus, clude a decision by Mexico to reduce tariffs on some 5,000 Kazakhstan, and Ecuador were the most active (figure 5). tariff lines (97 percent of all imports), with average tariffs fall- 2 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise Figure 5. Number of Newly Initiated Trade Measures, 89 Countries, September 2008 – July 2011 Yemen, Rep. of Vietnam, (21) Uzbekistan United States, (52) Ukraine, (20) UAE Ukraine, 6 Turkey, (26) Turkey, 5 Tanzania Switzerland Sri Lanka liberalizing Sierra Leone restrictive Saudi Arabia Russian Federation, (49) Russian Federation, 33 CU of Blr, Kzk, Rus, (24) CU of Blr, Kzk, Peru Rus, 29 Pakistan, (21) Pakistan, 11 Nigeria Morocco Moldova Malaysia Malaysia, 7 Kyrgyz Rep. Kazakhstan, (10) Jamaica India, (119) Indonesia, (53) Indonesia, 16 India, 39 Honduras Gabon European Union, (62) El Salvador Dominican Republic Colombia China, (50) China, 21 Chile Cameroon Brazil, (55) Brazil, 36 Bosnia & Herzegovina Bangladesh Australia, (30) Australia Argentina, (69) Algeria (120) (100) (80) (60) (40) (20) 0 20 40 Source: Authors’ calculations using WTO data from 2009, 2010b, 2011b. Note: CU of Blr, Kzk, Rus = Customs Union of Belarus, Kazakhstan, and the Russian Federation. ing from over 10 percent to 4.6 percent; the U.S. Miscellaneous Most tariff decreases seen after the 2008 crisis appear to Tariff Bill, which temporarily suspended import tariffs on myr- have aimed at lowering prices of food staples, consumer goods, iad intermediate goods through end-2012; and the phasing and industrial inputs such as machinery, parts, and compo- down of tariffs to 5–10 percent by 2015 on 23 HS 2-digit lines nents. Production of manufactured goods is increasingly orga- by Australia, affecting hundreds of products, including textiles, nized through global value chains, with goods being processed clothing, footwear, and a range of industrial goods. Canada (value being added) in multiple countries that are part of the eliminated tariffs on over 1,500 products—mostly machinery, chain. Plants in each country specialize in specific processing equipment, and industrial inputs. Non-G-20 countries also activities that make up a final product. As a result of this verti- implemented instances of tariff liberalization, although mostly cal specialization, a significant share of the value of any export modest in nature and limited to just a few products (figure 6). reflects imported inputs. For the world as a whole, the import Of course, the number of tariff hikes or reductions does not say content of exports is about 30 percent (Daudin, Rifflart, and much about the political intent or the impact on trade volume. Schweisguth 2011). Using the data compiled by Daudin, Rif- Determining the net impact of trade policy responses on trade flart and Schweisguth, Gawande, Hoekman, and Cui (2011) flows and the effective levels of protection for affected indus- show that the intensity of vertical specialization helps to ex- tries requires detailed research. plain the observed pattern of changes in tariffs postcrisis. There 3 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise Figure 6. New Tariff Measures, G-20 /Non-G-20, September 2008 of production. Brazil is an example of a country that is both low- – October 2011 ering and hiking tariffs to support its domestic industry. For 180 example, Brazil reduced tariffs to 2 percent on hundreds of 160 capital goods and equipment in 2010 and again recently in Au- 140 tarif f increases gust 2011,4 with the objective of supporting domestic down- 120 tarif f reductions stream industries. At the same time, it also increased taxes by 100 80 30 percent on vehicles (including trucks) that have less than 65 60 percent local content and do not originate in Mercosur coun- 40 tries5 or Mexico. Brazil is also expected to implement a 3 per- 20 cent tax rebate to exporters of manufactured products.6 Other 0 economies have also intervened in trade to support domestic G-20 non-G-20 industry. For example, Indonesia eliminated tariffs earlier this Source: Authors’ calculations using WTO data from 2009, 2010b, 2011b. year on 182 raw materials (including soybean oil, crude petro- leum, and numerous inorganic chemicals) and capital goods; are two channels: the first that higher tariffs are a tax on down- raised tariffs on imports of prepared food; and imposed export stream processing parts of the chain, so importing governments taxes on cocoa. have an incentive to keep tariffs low. The second is that these In a few instances, countries hiked tariffs substantially, firms (chains) have an interest in the governments of the coun- mostly on food products. In 2009, Russia increased tariffs to tries producing the inputs they use to keep trade costs low—in- 50–80 percent on certain meat products, some of which were cluding through low or zero tariffs. This benefits the exporting subsequently removed. Turkey increased tariffs on certain countries that are further down the chain, and those that pro- grains to 130 percent in May 2009, and then eliminated them duce the inputs through higher overall exports (sales of the final in February 2011. Since 2008, tariff reductions on food prod- product). One or both of these vertical specialization effects are ucts have been pursued by governments seeking to contain found to be statistically significant determinants of reductions food price increases in their domestic markets. The number of in 2009 tariffs. tariff reductions on food products (80+) since late 2008 great- There are also incentives for lowering trade costs for indus- ly exceeds the number of tariff hikes on such products (20). tries that are not part of global value chains. Reducing input Half of all such tariff reductions pertained to grains, sugar, costs through either tariff reduction on imports or taxing ex- and edible oils. For example, Morocco lowered tariffs on dif- ports of locally produced raw materials can increase the effec- ferent types of wheat from highs of 135–170 percent to 80– tive protection of domestic industries. Conversely, if there is a 90 percent in 2010; Kenya reduced tariffs on rice from 75 to significant parts and components industry and a country is less 35 percent. integrated into global value chains, governments may have an The tariff hikes that were imposed since the crisis have not incentive to raise tariffs on inputs and/or final products. Thus, reversed the trend of steadily declining average applied tariffs in incentives continue to exist to engage in traditional protection- the last 15 years (figure 7). The average most-favored-nation ism, with the balance of incentives a function of the structure (MFN) tariff of developing country G-20 members fell from 14 percent in 1999 to about 10 per- Figure 7. Average G-20 Tariffs, 1996–2010, Simple MFN Applied Average cent in 2010. As a result of this lib- average tarif f G-20 high-income economies eralization, the tariff gap between average tarif f G-20 middle-income economies emerging/developing and advanced 20 average tarif f by G-20 in agriculture G-20 members has narrowed from 18 average tarif f by G-20 in manuf acturing 10 to about 5 percentage points. Ac- 16 average G-20 tarif f tually applied tariffs on trade are 14 less than the average MFN rate as a 12 result of preferential trade agree- percent 10 ments and trade preference pro- 8 grams. 6 4 Limited Use of Anti- 2 dumping and Safeguard 0 Actions 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Countries may use temporary trade Source: WITS data (TRAINS + WTO-IDB), OECD; tariff data missing for Russia 1998–2000, Saudi Arabia 1996–98, 2010, Indonesia 1997, India 2010, European Union 2010, and South Africa 1998. barriers (TTBs) such as antidump- 4 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise Figure 8. Newly Initiated TTBs, Selected Most Active Users Among Figure 9. A Steady Reduction in Protectionism, 1995–2010 G-20, Non-G-20 Countries 400 number of new AD initiations CU of Blr, Kzk, Rus 350 United States 300 250 Ukraine 200 Turkey 150 South Af rica 100 Russian Federation Oct 2010 – Oct 2011 50 0 Peru 1995 1997 1999 2001 2003 2005 2007 2009 Pakistan Oct 2009 – Sep 2010 Source: Data from WTO Web site, http://www.wto.org/english/tratop_e/adp_e/adp_e.htm. Note: AD= antidumping Mexico Korea, Rep. of Sep 2008 – Sep 2009 isfy required criteria (for example, pertaining to injury caused Israel by imports, dumping, or subsidization). Just under half of all Indonesia new TTB investigations led to duties, whether provisional or India definitive. There is considerable heterogeneity in the average European Union level of AD duties imposed, varying by imposing country and Dominican Rep affected trade partner. AD duties imposed on Chinese prod- China ucts tend to be substantially higher than those imposed on products from other exporters (Bown 2010a). For example, the Canada average ad valorem duty imposed by the United States on Chi- Brazil nese exports in the postcrisis period was 149 percent, com- Australia pared to 45 percent on the products of other exporters.8 Argentina However, taking a longer-term perspective, the use of AD 0 10 20 30 40 has been falling steadily since the late 1990s, with the crisis leading to only a small increase. Overall, the number of AD Source: Authors’ calculations using WTO data from 2009, 2010b, 2011b. Note: CU of Blr, Kzk, Rus = Customs Union of Belarus, Kazakhstan, and Russian actions is still far below the average levels observed a decade ago Federation. (figure 9). To date, fewer TTBs have been put in place than what would have been expected judging by previous reces- ing (AD), countervailing duty (CVD), and safeguard (SFG) sions. The stock of products covered by TTBs imposed by de- measures instead of tariffs; the use of such instruments gener- veloping G-20 members has increased, rising from 1.71 per- ally peaks during recessions when domestic industries invoke cent in 2007 to 2.55 percent in 2009 (Bown 2010b),9 but in them to reduce foreign competition and keep more of the the aggregate, TTBs have not had a significant impact on trade market for themselves (Leidy 1997). Most countries have leg- volumes. One reason for the low coverage is that three-quarters islation allowing for TTBs to be used when imports put too of all TTBs are AD, which is product and firm specific, as op- much pressure on domestic industries. Such instruments of posed to safeguards, which tend to affect broader industries contingent protection are permitted under WTO rules and many preferential trade agreements. A key feature of Figure 10. TTB Database—Antidumping Initiations by Developed/ these TTBs is that they require an investigation to deter- Developing Countries, 2007–10 mine whether the preconditions for taking action have 40 been satisfied. 35 TTBs were a frequently used policy instrument in the 30 post-2008 period, with the WTO reporting a total of some 25 370 new investigations (282 AD, 45 CVD, and 43 SFG) 20 into alleged foreign dumping, subsidization, or requests for 15 temporary protection by G-20 members.7 Frequent users 10 of TTBs include India (68); the European Union (56); Ar- 5 gentina (45); the United States (45); Brazil (34); Indonesia 1Q 3Q 1Q 3Q 1Q 3Q 1Q 3Q (22); Australia (21); and China (20) (figure 8). One-fifth of 2007 2007 2008 2008 2009 2009 2010 2010 postcrisis TTB investigations were subsequently dropped, developed developing many within a year of initiation, as a result of failing to sat- Source: Bown 2011. 5 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise and all sources of imports. The product coverage of AD actions were quantitative import restrictions, quotas, and bans.10 Im- has remained similar to precrisis coverage—chemicals, plastics, port licensing and discriminatory government procurement metals, paper, and machinery—suggesting AD was not seen by accounted for another 25 percent (figure 11). The number of firms as an effective instrument to help them deal with crisis- “buy national� measures (including local content/national induced market pressures. preference incentives) increased significantly in 2011, espe- While the aggregate impact of TTBs on global trade vol- cially in emerging market G-20 members, affecting sectors umes has been limited to date, it is important to recognize that such as energy, telecom, and motor vehicles. As is the case AD action is long lasting (generally more than five years) and with tariffs, the direction of policy has not been uniform. can give rise to retaliatory measures that generate further pro- Since January 2009, some 44 NTMs that predated the crisis tectionism. Developing countries have been taking over from were removed, half by India, followed by China (figure 12; high-income economies (the traditional users of AD), reflect- WTO 2011a). One-fifth of NTMs imposed since 2008 have ing both increased institutional capacity and the rise of China been removed or are explicitly time bound. as an export powerhouse (figure 10). AD is increasingly a The rise in the relative importance and use of NTMs by de- south–south phenomenon, with China the main target (Bown veloping countries does not reflect WTO constraints limiting 2011). their ability to use tariffs, given the substantial room for ma- neuver that exists between bound tariff rates and actual ap- Increased Use of Less Transparent Forms plied rates. More important may be the constraints imposed by of Industry Support membership in regional trade agreements. For example, Argen- While the post-2008 use of tariffs and TTBs has been con- tina has increased the use of nonautomatic import licensing, a tained, nontariff measures (NTMs) are being used more fre- process under which import approvals are discretionary (WTO quently. The number of new trade-restricting NTMs averaged rules require a decision within 60 days), and that applies to all 85 per year in the first two years after the onset of the crisis, sources of imports, including from Mercosur.11 The Argentine and increased considerably in 2011 to over 120 (figure 1). De- government has also introduced technical requirements for veloping G-20 countries have been the main users of NTMs, myriad imported products that act as trade barriers. Brazil has including India (38), China (25), Indonesia (25), Argentina also introduced nonautomatic import licensing for over 40 (23), and Russia (15). NTMs are generally less transparent product groups at the HS 4-digit level, including vehicles, car than tariffs and often generate greater distortions because parts and consumer goods (confectionary), and is giving tax they limit trade volumes. A little less than half of all NTMs credits to manufacturers of cars and trucks that use locally pro- duced inputs. Almost one-third of all NTMs are imposed on exports, espe- Figure 11. G-20 NTMs, September 2008 – October 2011 cially of agricultural goods, with a clear upward trend in the use price of export restrictions (WTO 2011b), underscoring the need support for stronger WTO disciplines (Martin and Anderson 2011). measures 5% NTMs on exports also take different forms. Half of all export restrictions were quantitative in nature rather than price based (taxes or duties). Countries imposing restrictions include Ar- export incentives import gentina, India, China, Ukraine, Indonesia, Bolivia and Pakistan, 8% restrictions with many of the measures affecting grains and raw materials. export 23% India has been a frequent user of export restrictions for grains restrictions 7% and food products as well as for exports of textiles and leather. China has restricted the export of certain minerals and raw ma- export bans 5% terials and increased rebates on a gamut of industrial goods. export China also recently lost the panel stage of a WTO dispute quotas 5% import bans brought by several countries against export taxes on a variety of other 15% import natural resources. Overall, export taxes mostly affected raw ma- measure terials such as ores, metal scrap, minerals, cotton, yarn, and raw 2% import cocoa. A number of countries also increased export incentives quotas national import licensing for specific products (for example, dairy products). Examples 5% pref erence 15% 10% include the United States, European Union and Switzerland, as well as India, Brazil, and China. NTMs have become more prevalent partially as a response Source: Authors’ calculations using WTO data from 2009, 2010b, 2011b. Note: Total=166 to public demand for traceability of goods and protection 6 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise Figure 12. Rollbacks of NTMs by the G-20 three-year period since late 2008, the GTA reported a total of 273 actions supporting national industries (Evenett 2011c). United States The sectors most frequently targeted by ongoing support pro- CU Blr, Kzk, Rus grams are agriculture and motor vehicles. Russian Federation Impacts on T rade: Limited Overall, But Mexico Significant for Some Countries and Products Korea, Rep. of There is considerable debate regarding the impact of the post- Indonesia 2008 trade policy actions. Trade flows collapsed spectacularly in India 2009, but recovered almost as rapidly when economic activity European Union picked up (figure 13). The magnitude and rapidity of the bounce back reflects the limited impact of protectionist mea- China sures. The WTO estimates that G-20-initiated trade measures Canada currently affect a little over 2 percent of world trade (WTO 2011b), up from about 1 percent in 2009. Evenett and Fritz Brazil (2010) estimate that a subset of 15 “jumbo� discriminatory Australia measures affected more than 10 percent of world merchandise Argentina imports in 2008. Such calculations do not tell us anything about the ad valorem tariff equivalent of the measures concerned or 0 5 10 15 20 25 30 their incidence, a matter that requires further research. Source: Authors’ calculations using WTO data from 2009, 2010b, 2011b. Clearly the impact of trade restrictions on specific products Note: Total=78 (44 originally imposed precrisis); CU of Blr, Kzk, Rus = Customs Union of can be significant. Henn and McDonald (2011) match “red� Belarus, Kazakhstan, and the Russian Federation. measures from the GTA database (that is, trade restrictive, dis- criminatory measures), for which the GTA could identify af- against health and environmental hazards. While technical fected trade partners and product categories, with monthly HS regulations and product standards are appropriate policy in- 4-digit level bilateral trade data from the Global Trade Informa- struments to ensure public safety and achieve other policy tion Services. Henn and McDonald conclude that trade flows objectives, they can also be a disguised form of protection. that were affected by trade restrictions saw between a 5 and 8 While analysis of the trade restrictiveness of NTMs has long percent decrease relative to trade flows of the same product been constrained by a lack of data, a recent multiagency effort among partners that were not affected by protectionist mea- (involving the African Development Bank, the International sures. Exports of poorer developing economies were hit harder, Trade Centre, the United Nations Conference on Trade and with an 8 percent drop, compared to a 5 percent reduction in Development, and the World Bank)—the Transparency in upper-middle-income countries.12 Against these negative im- Trade Initiative—is improving the cross-country availability pacts, one must consider the effects of the liberalizing measures of information on NTM use. Moreover, a toolkit has recently taken by many countries, as well as the general fiscal and mone- been developed by the World Bank’s Trade Department that tary stimulus measures taken by many countries that generated recognizes the complexity of NTMs and can be used by gov- demand for imports. Henn and McDonald argue that bailout ernments and stakeholders to minimize restrictive impacts on and stimulus programs put in place by high-income countries trade (Cadot, Malouche and Saez 2012). adversely affected developing-country exports by less than the Greater transparency and monitoring of both NTMs and border measures imposed by developing countries themselves. industry-specific support measures is a priority. Many govern- ments responded to the crisis by complementing fiscal and Concluding Remarks monetary stimulus measures with industry-specific support, While there has been a marked increase in the use of trade pol- such as loans and guarantees for classes of firms (such as SMEs) icy since 2008, with developing countries particularly active, or specific firms (bailouts, equity infusions) and financial sup- measures to restrict trade have been complemented by actions port for certain activities such as the development of green to reduce barriers for specific products. Average levels of import technologies. These measures can distort competition and dis- tariffs remain much lower than they were 10–15 years ago. criminate against foreign firms because they often explicitly TTBs such as AD were used less intensively than in past slow- target national firms. The GTA reported over 100 state aids downs, especially by OECD nations. There has been a long- and bailout measures with a discriminatory trade effect as of term shift toward greater use of AD by developing countries, December 2009, a year after the onset of the crisis. Over the but here again, the global number of cases remains much lower 7 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise Figure 13. Trade Collapse and Recovery, Jan. 2008 - Oct. 2011 world economy, high unemployment rates across the globe, (Goods Export Volume, % Change 3m/3m, seasonally adjusted) and inflation and currency appreciation in a number of 15 emerging economies will continue to pressure governments high income low income to support domestic economic activity. Open trade cannot 10 middle income be taken for granted, thus continued monitoring of trade 5 policy measures is necessary. Notes 0 1. WTO data references reflect authors’ calculations, which -5 are based on a compilation of measures reported in WTO - 10 (2009, 2010b, 2011b). 2. The GTA is an initiative that brings together a network of - 15 think tanks and institutes from around the globe that col- Jan 08 May 08 Sep 08 Jan 09 May 09 Sep 09 Jan 10 May 10 Sep 10 Jan 11 May 11 Sep 11 lect information on trade-related policy measures taken by Source: Authors’ calculations using data from Datastream. governments in regions they are covering (see http://www. GlobalTradeAlert.org). 3. The utility of grouping this set of countries together is fre- than it was 15 years ago. Many trade policy measures were of- quently criticized because there are so many differences be- ten aimed at lowering prices of consumer goods (for example, tween them, but it appears that they are similar when it comes food) and inputs used by domestic industries. The more limit- to the use of trade policy. ed use of traditional protection measures—tariffs and TTBs—re- 4. CAMEX (Chamber of Commerce, Brazil) December 2009 flects the reality of a more intricately linked global trading envi- and 2010; tariff reductions on Mercosur Common Nomencla- ronment, where countries are increasingly part of global value ture (NCM) chapters 39, 40, 68, 73, 76, 82, 84, 85, 86, 87, 89, chains (Baldwin and Evenett 2009; Dadush, Ali, and Odell 90, and 94. In August 2011, tariff eliminations took place on 2011). This has changed the traditional political economy dy- 3,779 capital goods and 111 integrated systems (NCM chap- namics of trade policy, where domestic industries and workers ters 73, 81, 84, 85, 87, 90, and 94). lobby for import protection. The changing nature of global pro- 5. Argentina, Brazil, Paraguay, and Uruguay. duction and trade—increasingly, intermediate inputs and re- 6. The Wall Street Journal, “Brazil Files Currency-Dumping Pro- exports—seems to have supported open trade. How robust this posal in WTO to Protect against Cheap Imports,� November new constellation of trade interests will be is something that 15, 2011 (http://online.wsj.com/article/BT-CO-20111114- only time will tell. 718925.html). The Brazilian government blames the sharp ap- Disentangling the multitude of trade policy measures and preciation of its currency, the real, against the U.S. dollar since their determinants (the underlying government objectives) the end of 2008 for undermining the competitiveness of its will be a rich area for research in the years to come, helping industry in export markets and leading to a flood of cheaper stakeholders to better understand the political economy of imports, a situation that is argued to be exacerbated by inter- trade policy in a world that is increasingly characterized by ventions of other countries to undervalue their currencies to vertical specialization. The same applies to assessing the net give their firms an export advantage. impact of the policy responses on global trade. The fact that 7. What follows focuses only on new initiations of TTBs; duties bailouts and stimulus packages—even if they were designed to imposed as a result of investigations launched before the crisis be discriminatory—were significant in size and number, and are excluded. WTO numbers are comparable with those in the most likely had beneficial effects on global trade, is a factor World Bank’s TTB Database, which provides more detail than that must be considered in any assessment of the trade effects the WTO on the specifics of each case, including product of 2008–9 policies. codes, countries affected, and the level of duties imposed (see Less heartening is that the apparent slowdown in post-2008 http://data.worldbank.org/data-catalog/temporary-trade-barri- protectionist measures observed in the first half of 2011 was ers-database). reversed in the summer of 2011 as a result of renewed econom- 8. Where data are available, dumping margins and ad valorem ic uncertainty and a decline in already weak growth. There are duties imposed by other countries on China show the same dis- increasing signs of pressure to support domestic industries. criminatory pattern. Global growth prospects are likely to remain subdued for some 9. Number of products measured at the 6-digit level of the Har- time to come and will impact trade volumes. The WTO recent- monized System (HS) of product classification. ly revised its forecast for 2011 trade growth downward from 10. Import bans were applied mainly for food products, espe- 6.5 percent to 5.8 percent. The scope for new stimulus mea- cially meat and livestock, in response to natural disasters or ani- sures is now much more limited than it was in 2008. A sluggish mal diseases (for example, the H1N1 virus). 8 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise 11. The Economist, “Keep Out,� September 24, 2011, (http:// ———. 2011a. “Crisis-Era Commercial Policymaking by the BRICs: www.economist.com/node/21530136?fsrc=scn/fb/wl/ar/ Evidence from the Global Trade Alert.� Mimeo. keepout). ———. 2011b. Resolve Falters as Global Prospects Weaken: The 9th GTA Report. London: CEPR. 12. Henn and McDonald (2011) note that having data at just ———. 2011c. Trade Tensions Mount: The 10th GTA Report. London: the HS 4-digit level (as reported by the GTA) rather than HS CEPR. 6-digit level means that their estimates likely underestimate Evenett, S., and J. Fritz. 2010. “’Jumbo’ Discriminatory Measures product-level effects. and the Trade Coverage of Crisis-Era Protectionism.� In Un- equal Compliance: The 6th GTA Report, ed. S. Evenett. London: References CEPR. Baldwin, R., ed. 2009. The Great Trade Collapse: Causes, Consequenc- Gawande, K., B. Hoekman, and Y. Cui. 2011. “Determinants of es and Prospects. London: Centre for Economic Policy Research Trade Policy Responses to the 2008 Financial Crisis.� World (CEPR; http://www.voxeu.org/index.php?q=node/4297). Bank Policy Research Paper 5862 (http://www.voxeu.org/index. Baldwin, R., and S. Evenett, eds. 2009. The Collapse of Global Trade, php?q=node/7249). Murky Protectionism, and the Crisis: Recommendations for the Henn, C., and B. McDonald. 2011. “Protectionist Responses to the G-20. VoxEU.org; London: CEPR. Crisis: Damage Observed in Product-Level Trade.� IMF Work- Bown, C. 2010a. “China’s WTO Entry: Antidumping, Safeguards, ing Paper WP/11/139, Washington, DC. and Dispute Settlement.� In China’s Growing Role in World Kee, H. L., C. Neagu, and A. Nicita. 2010. “Is Protectionism on Trade, ed. Robert C. Feenstra and Shang-Jin Wei. NBER/Univer- the Rise? Assessing National Trade Policies during the Crisis sity of Chicago Press. of 2008.� World Bank Policy Research Paper 5274, Washing- ———. 2010b. “Taking Stock of Antidumping, Safeguards, and ton, DC. Countervailing Duties, 1990–2009.� World Bank Policy Re- Leidy, M. 1997. “Macroeconomic Conditions and Pressures for search Working Paper WPS5436, Washington, DC. Protection under Antidumping and Countervailing Duty Laws: Bown, C., ed. 2011. The Great Recession and Import Protection: The Empirical Evidence from the United States.� IMF Staff Papers Role of Temporary Trade Barriers. Washington, DC: World Bank 41(1): 132–52. and CEPR. Martin, W., and K. Anderson. 2011. “Export Restrictions and Cadot, Olivier, Mariem Malouche and Sebastian Saez. 2012. Stream- Price Insulation during Commodity Price Booms.� World Bank lining Nontariff Measures: A Toolkit for Policy Makers. Washing- Policy Research Paper 5645, Washington, DC. ton, DC: World Bank. WTO (World Trade Organization). 2009. Overview of Developments Dadush, U., S. Ali, and R. Odell. 2011. “Is Protectionism Dying?� In in the International Trading Environment. WTO/TPR/OV/12, The Carnegie Papers, 3–5. Washington, DC: Carnegie Endow- Geneva (November). ment for International Peace. ———. 2010a. Report to the TPRB from the Director-General on Trade- Daudin, G., C. Rifflart, and D. Schweisguth. 2011. “Who Produces Related Developments. WTO/TPR/OV/W/3, Geneva (June). for Whom in the World Economy?� Canadian Journal of Eco- ———. 2010b. Overview of Developments in the International Trading nomics 44(4): 1403–37. Environment. WTO/TPR/OV/13, Geneva (November). Evenett, S. 2010a. Managed Exports and the Recovery of World Trade: ———. 2011a. Report on G-20 Trade Measures. Geneva (May). The 7th GTA Report. London: CEPR. ———. 2011b. Report on G-20 Trade Measures. Geneva (October). ———. 2010b. Tensions Contained…for Now: The 8th GTA Report. ———. 2011c. “WTO scales back its trade forecast to 5.8% as down- London: CEPR. side risks build.� Press Release 641, Geneva (23 September). The Economic Premise note series is intended to summarize good practices and key policy findings on topics related to economic policy. They are produced by the Poverty Reduction and Economic Management (PREM) Network Vice-Presidency of the World Bank. The views expressed here are those of the authors and do not necessarily reflect those of the World Bank. The notes are available at: www.worldbank.org/economicpremise. 9 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise