TTheWorld Bank 1 5 8 S68 9g M M g | E | 111 S | 111 l ~~~~~~~~~~~~~~A P R i L ;. S 7 7 | | | | 10 | l ~~~~1998 NUMBER 3 ECONOMIC POLICY Trade and the East Asian crisis Trade shocks appear to have played a role in the East Asian crisis. A key part of the recovery will be growth in net exports from the region, which will require supportive trade, structural, and macroeconomic policies both in the region and in its major export markets. The recent financial crisis in East Asia was intraregional trade. Japan is both a major preceded by a sharp global slowdown in export market for other East Asian pro- exports and substantial realignments in ducers and a competitor in many export the region's trade patterns. One important markets. Japan's imports from the question is whether trade shocks played a Republic of Korea and Hong Kong fell in Changes in trade majorroleinthecrisis.Asecondiswhether 1996 (by 8.5 percent and 6.6 percent), its the crisis has revealed long-run structural imports from China rose by more than 10 flows were both a problems. A third relates to policy choices. percent, and from the Philippines by 23 High-growth East Asian economies percent. Overall, East Asia's imports cause and a have become important players in world dropped almost 4 percent, and imports trade, making their trade adjustment fromJapan dropped 18 percent. consequence of the more important than was previously the In recent years there appears to have case. The region's nine main industrial- been an inverse relationship between the East Asian crisis izing economies increased their share in evolution of the yen-dollar exchange rate global exports of manufactures, and in and the growth in exports from ASEAN total exports, from 12 percent in 1990 to countries and Korea (figure 1). Korea's 17 percent in 1996 (table 1). Within the exports are more similar to Japan's than region, export shares of some high-growth are those of other East Asian countries, economies grew substantially faster. Table 1 Shares in global exports of Malaysia and the Philippines, for example, manufactures, ingo-96 more than doubled their shares in global (percent) exports of manufactures between 1990 and 1996. ~~~~~~~~~Economy 1990 1992 1994 1996 and 1996. Developing EastAsia 11.9 13.6 16.2 17.2 Recent shocks China 1.9 2.5 3.3 3.5 Hong Kong 1.2 1.1 0.9 0.7 East Asian exports slowed significantly in Indonesia 0.4 0.6 0.7 0.7 1996, following a period in which exports Korea, Rep. of 2.6 2.7 2.9 3.1 boomed. The region was not unique in this Malaysia 0.7 1.0 1.4 1.6 regard-world export growth fell by the Philippines 0.2 0.3 0.3 0.5 same proportion as in East Asia (table 2). Singapore 1.6 1.9 2.7 2.9 Btteps Taiwan, China 2.6 2.8 2.8 2.9 But the pattern of regional trade flows Thailand 0.6 0.8 1.1 1.1 compounded the negative impacton many Japan 11.9 12.4 12.7 10.9 East Asian economies. Causing particular United States 12.3 12.7 12.9 12.9 damage were the depreciation of the yen World 100.0 100.0 100.0 100.0 relative to the dollar and a sharp drop in Source: UN COMTRADE. FROM THE DEVELOPMENT ECONOMICS VICE PRESIDENCY AND POVERTY REDUCTION AND ECONOMIC MANAGEMENT NETWORK Table 2 Export growth in developing East arrangements with Central and Eastern Asia, Japan, and the world, 1995-97 European countries. Combined with poli- (percentage change, in dollar terms) cies pegging nominal exchange rates to an 1995 1996 1997Q2a appreciating U.S. dollar and the possible Developing East Asiae 21.6 4.5 8.2 emergence of domestic supply constraints, Japan 12.1 -7.2 6.3 these may have caused serious competitive- World 20.0 4.0 4.0 ness problems. a. Changes relative to the second quarter of 1996. Some analysts also believe that China's b. China, Hong Kong, Indonesia, Republic of Korea, 1994 devaluation of the yuan contributed Malaysia, Philippines, Singapore, Taiwan (China), and to East Asia's woes. But that devaluation Trade patterns Thailane.IDirectionofTrade, simply unified official and secondary mar- ket rates at the secondary market rate. within the region making the country more vulnerable to Because most trade was based on market changes in the yen-dollar rate (see table 4). exchange rates, the devaluation raised have changed effective export prices by just 10 percent. Structural fctors Import prices were unchanged. Indeed, significantly. Overall, These trade shocks appear to have been since unification China's real exchange largely cyclical in nature, although struc- rate has appreciated against the (strength- East Asia's irnports tural changes also may have played a role. ening) dollar by more than 20 percent. One "structural" argument is that the terms Although China's trade policy reforms dur- dropped almost 4 of trade of some Asian economies deterio- ing the 1990s have resulted in dramatic rated sharply in 1996, driven by excess export growth, import growth was also percent, ancd imports capacity and overinvestment in certain sec- quite high until 1997 (when exports grew tors. For example, prices of semiconductors much faster than imports, leading to a large from Japan dJropped and computer components fell dramatically expansion of the trade surplus). in 1996. Other potential sources of difficul- iS percent ties include the rise of new competitors Shifting terms of trade (such as China), the revival of major Latin Available (but incomplete) data suggest American countries, and trade diversion widely varying changes in the region's resulting from the North American Free terms of trade. Only Korea has suffered a Trade Agreement (NAFTA) and from the substantial decline in terms of trade-a 27 European Union's preferential trade percent drop in the three years ending in the third quarter of 1997. In 1995-96 the Figure 1 As dollar prices fell, East Asia's exports slowed income effect of this decline was about -4.3 Year-on-year percentage change Yen/dollar percent of GDP. This significant shock 35 Korean exports 140 clearly contributed to Korea's recent crisis, K0 reYen-dollarparticularly when compounded by contin- 30 , Yen-dollar / > uing declines in 1997. 25 'rae Still, growth in export volumes 20 Jf7 AS \ \ 8 120 remained high in Korea and Indonesia- aided by a slight real exchange rate depre- 1 5 exports I ciation in Korea and despite a larger real appreciation in Indonesia (table 3). Since 100 early 1994 export volumesfrom Korea have 5 more than doubled. Export growth fell in o ' ; \ / Hong Kong, perhaps because of a substan- tial real appreciation in its exchange rate. a5 80 The modest real appreciation in Malaysia January Jarnuary January January January January Sept 1992 1993 1994 1995 1996 1997 may have contributed to the decline in its Nate: Calculated as 12-month moving average in current dollars. export volume, but it does not explain it Source: World Bank and IMF data, completely. Export growth in Taiwan (China) and Thailand fell to low levels footwear) and higher-income economies despite only modest real appreciations. expanding exports of machinery and elec- tronic products. This has required contin- NVew competition uing investments in physical capital and in Has East Asia become less competitive? skills upgrading. One way to find out is to determine whether the region's economies have been Prospects for export growth losing market share in major markets. They Given fiscal constraints and the recent down- have not. Except for Thailand, East Asian turn in private consumption and investment, developing economies expanded their export growth will be a major vehicle for shares of world markets in 1995 and 1996, short-term economic expansion in EastAsia. Given fiscal with the five major exporters increasing If not offset by inflation, the sharp currency their share of world exports from 7.3 per- devaluations that have occurred will lead to constraints and the cent in 1994 to 7.8 percent in 1996-a sig- substantial real depreciations. This will cause nificant increase, but about half the rise in a shift in the internal terms of trade from recent downturn 1990-92. Greater competition from new nontraded to traded goods. Accompanied by suppliers may have been a factor in the reductions in domestic absorption, this shift in private declining rate of growth. will facilitate the improvements in the trade For example, since 1990 Central and balance that are needed to service foreign consumption and Eastern Europe has doubled or tripled its debt. Demand for imported consumer share of the EU market for products (such goods will decline, while demand for inter- investment, export as electrical equipment) traditionally sup- mediate inputs used in the production of plied by East Asia. Export structures of traded goods will increase. growth will be a these countries are similar to those of higher-income East Asian economies (cor- Intraregional export similarities major vehicle relation coefficients in the 0.8 range). East Asia's ability to expand trade will However, Central and Eastern Europe's depend partly on the composition and pat- for short-term market share increased only from 1-2 per- tern of trade flows. Economies in the cent to 6-7 percent. And, except for Korea, region trade heavily with one another and economic East Asia's share in this market continued often have similar export structures. to rise in 1995-96. Concern has been expressed that as these expansion in East In general, there has been a strong pat- economies seek to expand their net exports tern of shifting specialization among East to the world, they will create "excessive" Asia Asian economies, with lower-income competition in export markets. economies expanding exports of labor- Although some East Asian economies intensive goods (such as clothing and have very similar export structures, corre- lations for many country pairs are relatively Table 3 Annual change in export volumes low (table 4). Low correlations reflect dif- and real exchange rates ferent endowments of natural resources (percent) and patterns of specialization. Correlations Data Real tend to be higher in major export markets, for year Export exchange such as the European Union, which sug- Economy ending volume rate gests that some East Asian economies are Hong Kong 1997Q3 4.4 10.6 competing in similar product lines. For Indonesia 1996Q4 16.0 5.8 electrical equipment, for example, correla- Korea, Rep. of 1996Q4 26.8 -1.0 tions of East Asian export shares in the Malaysia 1996Q2 -5.5 6.1 Taiwan (China) 1997Q1 6.4 5.5 European Union are often 0.9 or higher, Thailand 1996Q1 3.9 6.1 reflecting exports of computing equip- a. A positive number indicates that the exchange rate ment. But such narrow product lines appreciated. account for a major share of exports only Source: IMF International Financial Statistics. for Singapore and Taiwan (China). Table 4. Export share correlations in world and EU markets, 1996 high tariffs-Thailand has more than 650 Above the diagonal: correlations in world market; below diagonal: correlations in EU market tariffs above 60 percent-and nontariff bar- JPN CHN HKG IDN KOR MYS PHL SGP TWN THAI riers. Simple average tariffs in the region are JPN 1 0.11 0.17 0.02 0.68 0.52 0.18 0.53 0.57 0.40 high:morethan22percentinThailandand CHN 0.59 1 0.91 0.43 0.43 0.31 0.21 0.21 0.41 0.69 about 15 percent in Indonesia, Korea, and HKG 0.48 0.79 1 0.50 0.46 0.39 0.26 0.31 0.44 0.71 the Philippines. Temporary entry and other IDN 0.14 0.43 0.42 1 0.23 0.32 0.32 0.08 0.18 0.33 KOR 0.92 0.72 0.59 0.23 1 0.75 0.69 0.61 0.71 0.60 exemption schemes imply that collected MYS 0.76 0.74 0.62 0.33 0.92 1 0.27 0.77 0.72 0.61 tariffs are only 3-7 percent of the value of PHL 0.68 0.77 0.69 0.39 0.87 0.70 1 0.20 0.18 0.27 imports in most economies, but all SGP 0.81 0.62 0.53 0.15 0.82 0.80 0.70 1 0.83 0.64 economies have hundreds of tariffs above TWN 0.85 0.64 0.52 0.17 0.83 0.77 0.66 0.98 1 0.72 THAI 0.75 0.71 0.64 0.30 0.81 0.81 0.76 0.88 0.88 1 30 percent. High tariffs are often applied to Note: The two data sets are not strictly comparable because of different levels of product goods such as automobiles and compo- disaggregation. nents, where inward-oriented production is Source: UN, 3-digit SITC (174 product categories); EUROSTAT, 2-digit CN (100 product a recipe for falling behind the rest of the categories). world. The recent devaluations increase the Conventional indexes of effective competitiveness of traded goods industries, The boost to exchange rates ignore developments in providing scope for trade liberalization. competing suppliers that can be particularly To continue the process of product tradable goods important. The Development Research upgrading and diversification that is cen- Group is developing customized procedures tral to maintaining high growth rates will industries provided for assessing the impacts of devaluations in require heavy investments in education competing countries. and in skills upgrading. by recent Disciplined macroeconomic policies Intraregional trade will, of course, be needed to capitalize on devaluations greatly Intraregional exports accounted for almost the initial boost to competitiveness pro- 40 percent of EastAsia's exports in 1996, up vided by the recent devaluations in the increases these from 32 percent in 1990. If Japan is region. Unless policies are suitably sup- included, the figure rises to 51 percent. portive, these devaluations will be dissi- industries' High levels of intraregional trade reflect pated in higher inflation. specialization between economies in the Rising net exports to the rest of the competitiveness and region. About half the intraregional trade world from EastAsia may give rise to adjust- is in raw materials and intermediates, ment pressures in OECD markets and, may provide which suggests that a significant portion of especially if gross exports increase strongly, the trade is complementary. Thus intrare- to pressures for protection. OECD govern- opportunities to gional trade is probably more resilient to ments must understand the inevitability of domestic demand shocks than if it were ori- such adjustments and avoid raising barri- reduce protection ented mostly toward final goods. However, ers. Doing so would harm their export the composition and importance of intrare- industries and hinder East Asia's recovery. further gional trade make these countries suscepti- This note was written by Bernard Hoekman and ble to adverse output shocks in the region. Will Martin. It draws on analysis of trade-- related dimensions of the East Asian crisis pre- Policy implications pared by Francis Ng, Maurice Schiff, L. Alan Although most East Asian economies are Winters, and Alexander Yeats of theDevelopment relatively open, some of their trade regimes Research Group's trade team, as well as contri- harbor severe distortions. These include butionsfrom Dipak Dasgupta and Ying Lin. . M This note series is intended to summarize good practice and key policy findings in PREM-related topics. PREMnotes are distributed widely to Bank staff and will also be hEE available on the PREM website (http://prem). If you are interested in writing a PREMnote, send your idea by email to KEim Murrell. For additional copies of this s:u PREMnote please contact the PREM Advisory Service at extension 87736. Prepared for World Bank staff