China - Foreign trade reform

-q i,;\ ne , 1 ':' S ' f k g 0 ' i ' ' t , t f liCy s; Ait tS a, 'S; ; t l t I ~ ~ t ;n 91:1s;;t;0';1; 0 0 ^if h ? w X ' a 0: i: ;'';,.:i:'001; :i; ;00: i pj S~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I fW C 'VX' ;8~~~~~~~~~~~~~~~~~~~~~~Ufi . . . ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Ir b___=_b_ - _ b ~~~~~~~~~~~~~~~~~~~~~~~~~~~~q  A W OR L D BANK CO U NT RY STUDY China Foreign Trade Reforn The World Bank Washington, D.C.  Copyright @ 1994 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing February 1994 World Bank Country Studies are among the many reports originally prepared for internal use as part of the continuing analysis by the Bank of the econoniic and related conditions of its developing member countries and of its dialogues with the governments. Somne of the reports are published in this series with the least possible delay for the use of governments and the academic, business and financial, and development communities. The typescript of this paper therefore has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. Some sources cited in this paper may be informal documents that are not readily available. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map m this volume do not imply on the part of the World Bank Group any judgment on the legal status of anur territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy portions for classroom use is granted through the Copyright Clearance Center, Inc., Suite 910,222 Rosewood Drive, Danvers, Massachusetts 01923, USA The complete bacldist of publications from the World Bank is shown in the annual Index of Publications, which contains an alphabetical title list (with full ordering information) and irLdexes of subjects, authors, and countries and regions The latest edition is available free of charge from the Distribution Unit, Office of the Publisher, The World Bank, 1818 H Street, N-W., Washington, DC- 20433, ULSA., or from Publications, The World Bank 66, avenue d'lIna, 75116 Paris, France. ISSN: 0253-2123 Library of Congress Cataloging-in-Publication Data China : foreign trade reform p. cm. - (A World Bank country study) ISBN 0-8213-2751-8 1. China-Commercial policy. 2. Exports-China. 3. China- Foreign econonic relations. L World Bank. IL Series. HF1604.C4513 1994 382.0951-dc2O 93-47692 CIP  CONTENTS Contributors ...................................... xi Currency Equivalents; Acronyms and Abbreviations . ................. xiii Executive Summary ............. ,v IL China's Merchandise Trade-Trends and Perspectives ............ 1 A. Introduction ..................................... 1 B. Trends in the Trade Balance ........................... 2 C. Perspectives on China's Export Performance ................. 4 The Changing Composition of China's Exports .... ........ 6 The Role of Assembly Operations .................... 11 Markets for China's Exports ........................ 12 The Role of the Nonstate Sector ..................... 13 The Role of Hong Kong .......................... 1S D. Trends in China's Merchandise Imports .15 The Structure of Merchandise- Imports .15 E. Trade in China's Economy: Some Concluding Observations .20 Endnotes .23 2 The Trade Plani and Foreign Exchange Systems-Managing Reforms ffirough a Period of Transition .24 A. Foreign Trade Planning, Foreign Trade Corporations, and Subsidies .24 The 1988 and 1991 Reforms .26 B. The Exchange Rate Regime .28 Foreign Exchange Plan .29 Retention Scheme .......... 29 Foreign Exchange Markets.31 C. Broad Impact of Recent Developments .34 Inpact of Developments in the Exchange Rate Regime .34 Impact of Reforms in Trade Planning .................. 38 D. Remaing Problems and Recommendations for Future Reform 39 Export Planning and the Foreign Trade Contract .... ....... 39 Import Planing ............................... 40 Exchange Rate Issues ............................ 41 Endnotes .... 45 ...i  - iv - 3 China's Systen of Foreign Trade ControL-: A Quantitative Evaluation ......................................... 47 A. Objectives and Instruments of China's System of Foreign Trade Controls ........... ....................... 47 B. Tariffs . ....................................... 48 Tariff Structure .......... ...................... 48 Conclusions ........... ....................... 56 C. Tariff Revenues and Exemptions ...... ................. 57 Administration and Revenue Raising: Implications of China's Import Duty System .57 Impact of Exemptions on Exports and Economic Performance . . . 60 Conclusions .62 D. Nontariff Barriers to Trade ............................. 63 Types, Administratzv.i and Coverage ................... 63 The Structure of N13 Protection ..................... 66 Conclusions ......... ................ 67 E. Export Controls and Taxes ............................ 67 F. Impact of Trade Controls ............................. 69 International Price Comparisons ...................... 69 Effective Rates of Protection ..... 73 Evidence from Sectoral Analyses .75 Conclusions ...... 78 Endnotes ...... 79 4 Priorities and Perspectives on Reforming China's Trade Regime ... ... 80 A. Trade Strategy and Trade Regime Orientation .80 B. Reform Priorities for the Immediate Term .82 COnalized Inports and Products Subject to Mandatory Import Planning .82 Import Licensing and Quotas .82 Import Controls .83 NTBs and Balance of Payments Management .84 Tariffs ..-.......... . ........ 84 Tariff Exemptions .85 Export Controls and Taxes .85 C. Priorities for the Medium Term ......................... 86 Reducing Protection on Consumer Goods .87 Reducing Protection on Intermediate and Capital Goods .87 Reducing the Bias against Raw Materials .88 D. Alternative Approaches to Reform .88 The Base Case: Effects of a 50 Percent Radial Cut in Effective Tariffs .91 Impact of the Other Reform Scenarios .94 Summary .................................... 97 Endnotes ........ 98  -V. S Toward a Program for Trade Liberalization ................... 100 A. Timing, Sequencing and Linkages with Other Reforms .... ...... 100 Trade Reform and Reform of the Planning System .... ...... 100 Trade Reform, Reform of the Exchange Rate Regime and Macroeconomic Policy .......................... 101 Trade Policy and Price Reform ...................... 102 Trade Liberalization and Enterprise Reform .............. 102 Trade Liberalization, Support for Exports and Industrial Policy . . 103 Trade Reform and the International Environment .... ....... 104 B. Recent Refcrm. Initiatives ............................. 104 Transparency ................................. 104 Liberalization of the Import Regimne ..............10..... C. Conclusions and Recommendations ........ ............... 106 Recommnendations for the Immediate Term ..... .......... 106 Recommendations for the Medium Term ..... ........... 107 Enduotes .................................... 109 6 Policies for Export Development: A Critical Evaluation ......... .. 110 A. Introduction ..................................... 110 B. Developing Buyer-Seller Links: The Role of Trading Companies ... 110 The International Experience ........................ 111 The Experience of China's FTCs ..................... 113 Lessons for FTC Policy in China ..................... 1!8 C. Developing Buyer-Seller Links: The Role of Public Support Services . 120 Public Support for Export Marketing .................. 120 Public Support for Quality Control .................... 122 Public Support for Other Nonfinancial Export Support Services . . 125 D. Export Financing .................................. 126 Recommendations .............................. 129 E. Product Targeting for Exports .......................... 129 . Conclusions .................................. 134 F. Geographical Targeting ...... ........................ 134 Policies for Coastal Development .136 Assessing the Performance of SEZs and Open Cities .137 Lessons and Recommendations .139 Endnotes .......................................... 141  - vi - 7 External Markets and China's Exports . ..................... 143 A. Protection Against China's Exports: The Facts ............... 143 B. Evolving External Markets and Implications for China's Exports .... 150 Implications of the Uruguay Round for China's Exports .... ... 150 The Potential Effects of Regionalism on Chie's Exports ...... 151 Implications of a Loss of China's Provisional MEN Status in the United States ................................ 155 C. Perspectives on Future Directions for China's Exports .158 Diversification of Markets .158 Product Diversification and Comparative Advantage .158 The Potential for Upgrading China's Major Exports .163 Opportunities for Diversification .164 Conclusions ...... 166 Endnotes ...... 167 Rderences ........ 169 ANNS 1.1 Differentating Labor- and Capital-Intensive Manufactures .178 2.1 Criteria for Access to FEACs ............................. 180 2.2 Operational Mechanism of FEACs ....................... 182 2.3 Understanding China's Foreign Exchange System ................. 184 3.1 Effective Rates of Proteceon Calculations: A Methodological Note ...... 193 4.1 A Computable General Equilibrium Model of the Chinese Economy ...... 196 6.1 Developing Buyer-Seller Links: The Intenational Experience .... ..... 208 6.2 China's SEZ Policy-An Evaluation ............... ..... 221 6.3 Economic Zones in China: A Taxonomy ...................... 246 7.1 Description of the SMART Trade Production Model used to Simulate the Effects of a 30 and 50 Percent Liberalization of Nontariff Barriers and Tariffs. 252 7.2 An Assessment of China's Changing Revealed Comparative Advantage in Labor-Intensive Manufactures . 254 Statistical Annex .260  - vii - TABLES IN TEXr 1.1 Structural Changes in China's Major Exports: Selected Years from 1965 to 1990 ....................................... S 1.2 The Value and Share of Major Three-Digit SITC Products in China's Exports-Selected Years from 1965 to 1990 .................... 7 1.3 The Value and Share of Various Types of Products in China's Manuu Exports: Selected Years 1965 to 1990. 9 1.4 Concentration of China's and Comparator Country Exports .10 1.5 Exports from Assembly Operations . ......................... 12 1.6 Markets for China Exports, 1990 . ........................... 13 1.7 Township and Vilage Enterprise Exports ........ .............. 14 1.8 China: Structurzl Change in China's Imports (CIF) Custows Basis. 16 1.9 Structure of China's Imports (CIF) Customs Basis (A Comparison with Other Importers for 1990) ........... s18 1.10 Origin of China's Imports: A Comparison with Odter Economies of East Asia (Selected Years) ............................... 19 1.1! Share of Merchandise Trade in GDP: Selected Countries .... ........ 20 1.12 China: The StructLre of Production, Imports and Exports, 1985 and 1990 ...................................... 21 2.1 Losses of Foreign Trade Corporations Financed by Central Government Budget ................................... 26 2.2 Foreign Exchange Retention Rates, 1991 ....................... 32 2.3 Exchange Rate and Trade Tax Relationships under China's Exchange Rate System .35 3.1 Average Tariff Levels by Broad HS Category ..................... 0 3.2 Average Taiff Levels by HS Section ......................... 51 3.3 The Tariff Systems of China and Other Large Developing Countries ..... 56 3.4 The Commodity Pattern of Tariff Systems in China and Other Large Developing Countries (Unweighted Average Tariff Rate) .... ........ 57 35 Imports by Import Duty Concession Category ................... 59 3.6 Value of Imports and Revenues from Import Duties ..... ........... 60 3.7 Coverage.of Nontariff Barriers by Sector and Type (1992) .... ........ 61 3.Sa Estimates of Protection to Importables based on International Price Comparisons (1992) .................................. 71 3.8b Estimates of Export Taxation (1992) ......................... 71 3.9 Effective Rates of Protection to Chinese Industry (1991) .... ......... 74 4.1 Effects of Reductions in Protection: Various Scenarios .............. 90 4.2 Effects of Reduction of Foreign Exchange Retention Ratios for the Machinery Sector .93 4.3 Nontrade Barriers before and after Import Liberalization Proposed by China .95 6.1 Foreign Trade Credit ................................... 127 6.2 Exports of Targeted Sectors, 1978-91 ......................... 131  - viii - TABLES im TExT (cont'd) 7.1 Average Level of Tariffs Chinese Exports Encounter in Nine Major OECD Markets .......................................... 146 7.2 Analysis of the Relative Importance of Nontariff Barriers on Exports from China and Other Developing Countries ................... 147 7.3 Sectoral Coverage of China's Exports by Nontariff Barriers in the United States, Japan and European Comununity .................. 148 7.4 Estimated Effect of a 50 Percent Liberalization in Trade Barriers by the EC, United States and Japan on Inports from China ................. 152 7.5 Estimates of Trade Diversion in the United States due to a Mexican- US Free Trade Arrangement ............................. 153 7.6 Aram of Potential Displacement of China's Exports to Canada due to the United States-Canada FTA .......................... 154 7.7 Projections of the Trade Effects of Applying United States General Tariffs on Major Chinese Export Products ........................... 157 7.8 Cross-Country Comparison of Trends in Average Labor Intensity of Exports ........... .............................. 160 7.9 Share of China and Other East Asian Countries in the World Exports of Selected Products .161 7.10 Relative Prices Received by China and Other Exporters to the EC, Japan and United States in 1990 .163 BOXES IN TEXr 3.1 The Construction of Three Smanl Polyester Factories in Chengdu .76 3.2 Supply Imbalances withia the Refrigerator Industry: Asymmetric Investmect Requirements .76 6.1 Foreign Trade Corporations in China: Two Examples .116 7.1 Can International Markets Accommodate a Major Chinese Trade Expansion? 144 7.2 Current Protectionism and China's Trade Prospects: Some Comparisons with an Earlier Period .145 FIGURES IN TErr 1.1 China: Trends in Trade, 1978-91 .3 1.2 Broad Money and Imports ................................ 6 2.1 Trends in the Official and Swap Market Exchange Rates ............. 36 2.2 Trends in Real Effective Official and Swap Market Rates .37 2.3 The Weighted Average Exchange Rate and Exports .38 3.1 Tariff Structure-Food, Beverage and Tobacco .53 3.2 Tariff Structure-Raw Materials .54 3.3 Tariff Structure-Manufactures .55 3.4 Structure of NTB Coverage, 1992 .66  - ix - FiGuRns IN TEr (cont'd) 7.1 Nominal Protection Against Clhina's Textile and Clothing Exports to the United States in 1988: Tariffs, Nontariff Barriers and Combined Protective Effects .................................... 149 7.2 Analysis of Cbina's Revealed Comparative Advantage in Broad Groups of Manufactured Products, 1979/80 and 1989190 . .162 7.3 Selected Labor-Intensive Manufactured Products in Which China May Have a Major Future Export Potential ........... .. ............... 165  CONTRIBUTORS This report and its annexes are based on the findings of a mission that visited China during May-June 1992. In addition to interviews and discussions in Beijing, the mission visited Jinan and Qingdao (Shandong Province) and Wuhan (Hubei Province). The mission consisted of Hoe Ee Khor from the IMF; Nicholas Lardy and David Wall (Consultants); Sonia Zhao and Chen Xingdong from the World Bank Resident Mission in Beijing; and Alexander Yeats, Will Martin, Peter Harrold and Rajiv Lall from World Bank headquarters. The report is based on contributions from mission members. Research assistance in Washington was provided by Teja Raparla. Rajiv Lall is the principal and coordinating author of the Report. The mission received valuable collaboration from a team headed by Mr. Yuan Wenqi, Director of the Research Department of China's Foreign Trade at the Institute of Finance and Trade Economics of the Chinese Academy of Social Sciences. This team was comprised of Messrs./Ms. Chen Jiaqin, Li Xiaoxi, Yang Shangxiang, Yu Lixin, Feng Yuan, Zhang Li and Lu Shengliang. Special thanks are owed to Mr. Ji Chongwei, Senior Research Fellow of the Development Research Center, Mr. Liu Xiangdong, Assistant Minister, Ministry of Foreign Trade and Economic Cooperation (MOFTEC), and Mr. Song Haipeng, Deputy Director General of the State Administration of Exchange Control (SAEC) for their guidance and suggestions. Acknowledgments are also due Mr. Qiu Xichun, Deputy Division Chief, Deparmnent of Policy and Development, MOFIEC, and Mr. Wang Qin Hua, Division Chief, Department of Foreign Trade, State Council Economic and Trade Commission (SCETC) for their comments. The mission is grateful to MOFTEC for facilitating its work in Beijing and for helping to make the visits to Shandong and Hubei provinces fruitful. Finally, the mission extends its gratitude to all the officials from the various government agencies, trading companies and enterprises with whom it met - xi -  CURRENCY EQUIVALENTS Currency Unit: Yuan (Y) $1.00 = Y 5.73 Y 1.00 = $0.175 FISCAL YEAR January 1 - December 31 WEIGHTS AND MEASURES Metric System ACRONYMS AND ABBREVITIONS ACP African, Caribbean and Pacific (Countries of the Lomd Convention) BESD (World) Bank Economic and Social Database BOC Bank of China CES Constant Elasticity of Substitution CET Constant Elasticity of Transformation CGE Computable General Equilibrium (Model) CCPIT China Council for the Promotion of International Trade CETDC China External Trade and Development Council CNC Computer Numerically Controlled GOEs Collectively Owned Enterprises COFERT Commission of Foreign Economic Relations and Trade cOMTRADE Commodity Trade (System) DIP de facto import promotion DLUC Domestic Letters of Credit EC European Community EP Export Promotion EPOs Export Promotion Office ERP Effective Rate of Protection ETCS Export Trading Companies ETDZs Economic and Technology Development Zones E1E Export Tax Equivalent FDI Foreign Direct Investment FEACs Foreign Exchange Adjustment ("Swap") Centers FFEs Foreign-Funded Enterprises FIEs Foreign-Invested Enterprises FT Free Trade FTA Free Trade Area FTCs Foreign Trade Corporations GATT General Agreement on Tariffs and Trade GDP Gross Domestic Product GNP Gross National Product GSP Generalized System of Preferences - xiii -  - xiv - GTCs General Trading Ccmpanies GVIO Gross Value of Industrial Output HKDTC Hong Kong Trade Development Council HS Harmonized System HTDZ High Techniology Development Zone IMF International Monetary Fund IS Import Substitution ISO International Standards Organization ITCs Import Trading Companies JETHRO Japanese Export Trading Organization KOTRA Korea Trade Promotion Corporation LF3 Linear Expenditure System LICs Large Trading Companies MFA Multifibre Arrangement MFN Most-Favored-Nation (status) MOFERT Ministry of Foreign Economic Relations & Trade MOFTEC Ministry of Foreign Trade and Economic Cooperation MPS Materials Product System NAFrA North American Free Trade Area NIEs Newly InLdustrialized Economies NTB Nontariff Barrier OECD Organization for Economic Cooperation and Development PBC People's Bank of China PEP Protected Export Promotion PNEs Production Networks for Exports PTA Phthalic Anhydride QCC Quality Control Circle RAS Richard A Stone (algorithm) RCA Revealed Comparative Advantage REER Real Effective Exchange Rate RMC Resident Mission in China SCEIMO State Council Machinery and Electronics Import Office SCETO State Council Economics and Trade Office SCETC State Council Economic and Trade Commission SAECs State Administration of Exchange Control SEZs Special Economic Zones SITC Standard International Trade Classification SMART Software for Market Analysis and Restrictions on Trade SOEs State-Owned Enterprises SPC State Planning Commission SRC System Reform Comnission STDB Singapore Trade Development Board TVEs Township and Village Enterprises UN United Nations UNCTAD United Nations Conference on Trade and Development US United States USSR Union of Socialist Soviet Republics XTB Xiznen International Bank  EXECUTIVE SUMMARY A. OvERvIEw Since the launching of the reform program in 1979, the promotion of external trade has been central to China's efforts to modernize its economy. The policy has met with remarkable success, with exports having increased ninefold and imports more than sevenfold over the period. This has been accompanied by rapid changes in the institutional support system for foreign trade and in the incentive framework. But China has a long way to go in replacing direct administrative intervention with indirect price-based instmments for managing its trade policy, and much is still unclear about the fumctioning of China's trade regime. Historically, China's approach to trade policy has been aimed at achieving export growth fcr the sake of generating foreign exchange without sufficient regard to its costs, while import policy has featured controls to regulate import growth. Although, as a result of the "open-door' policies of the 1980s, decisions concerning exports have become increasingly determined by the market rather than administrative flat, reform of the import regime has, by comparison, remained neglected and is now taking on some urgency. The report addresses key issues in import as well as export policy for accelerating the country's transformation into a market based economy. Four important conclusions emerge. First, while the role of planning in China's foreign trade sectr has been declining, several problems of transition remain. The report recommends reforms that would enable China to transform what remains of its trade and foreign exchange planning systems into instrunents for the indirect management of trade policy. Second, the report fids that despite a complex array of tariff and nontariff barriers to trade, China has become a relatively open economy, with evidence of considerable tariff redundancy. It demonstrates that the present is, not least for this reason, an- opportune time for China to embark on a substantive program of trade liberalization. Import competition would be an important instrument for helping China make its state-owned sector more responsive to market forces-an issue of mounting concern for the country. Import liberalization would also contnrbute to the growing momentum of China's drive to become a full member of the international trading community. Moreover, in view of its remarkable export performance so far and its comfortable reserves position, China would appear to be well positioned to launch such a program at this time. Thrd, notwithstanding the diminishing reliance on export planning, the Chinese government has been actively involved in export development. Aside from financial assistance and incentives, the goverrunent has directly provided a variety of other support services, including marketing and quality control. Such support should be maintained. However, as China continues its drive towards a miarket economy, the way in which export assistance is - xv -  -xvi - delivered will need to be refined. The report proposes a program of action for making the institutional support structure for exports, and public support for export development in general, more effective. Finally, the success of China's continued reforms in the area of foreign trade depends on its ability to sustain a healthy rate of export growth. The report finds that, even in today's uncertain global trading environment, China should be able to maintain its export performance through quality improvements and some product and market diversification. On the other hand, there is little doubt that the successful conclusion of the Uruguay Round would be very helpful for China, or that the discontinuation of China's Most-Favored-Nation (MFN) status in the United States would lead to significant trade dislocation. B. MAiN FEATuRES OF CEHNA's TRADE oRMANCE Over the decade of the 1980s, China's exports outperformed those of most other countries, including such newly industrialized economies (NIEs) as Malaysia. The only economies that registered average znnualized export growth rates higher than China were Thailand (13.2 percent), the Republic of Korea (12.8 percent), Taiwan, China (12.1 percent) and Portugal (11.7 percent). China is now the eleventh largest exporting nation in the world. It exports about 17.percent of the gross value of industrial output (GVIO) of its overall manufactmring sector. Institutional decentralization, foreign investment (especially from Hong Kong), depreciation of the real effective exchange rate, and duty-free access to imported inputs for export assembly all seem to have contributed to this strong performance. Whereas in 1978 all trade was monopolized by only 12 Foreign Trade Corporations (FTCs) and their branches, today over 3,600 FTCs compete increasingly fiercely for export business. From no foreign direct investment (FDI) in 1978, the total has risen to over 90,000 approved projects with a contracted value of $58.1 billion. Since 1985, the real effective official exchange rate has depreciated more than 100 percent, and almost two thirds of China's manufacturing exports are now based on processing activity that utilizes duty-free imports. Another development which has had a bearing on China's export performance over the decade of the 1980s has been the decliiing importance of trade planning. Direct subsidies for exports have been more or less phased out, and export targeting has become progressively more macro in nature (it now operates essentially only through the foreign trade contract system).. As a result, the composition of China's exports has been allowed to evolve increasingly along the lines of the country's comparative advantage, with the contnbution of nonstate and foreign-invested enterprises also growing very fast. Nonstate and foreign-funded enterprises now account for at least one third of China's exports. Manufactures as a whole constitute 80 percent of exports, with labor-intensive manufactures contributing almost three fourths of total exports. As in many other East Asian economies, clothing, toys, sporting goods and footwear have emniged as among the most dynamic of China's export sectors. Unlike exports, imports seem to have remained subject to much stronger government management. Over the 1980s, import trends closely followed trends in planned domestic economic activity and appear alSD to have been sensitive to changes in administrative controls such as foreign exchange retention and import licensing. Overall, the pattern of China's imports remained very stable and reflected the country's import strategy of assuring supply of key raw materials and acquiring embodied technology while minimizing imports of consumer goods. The degree of central government control over imports has been diminishing, but even  - xvii -. today, over 50 percent of the country's foreign exchange earnings are subject to central government control, and 50 percent of imports are subject to some form of nontariff barriers (NTBs). On the whole though, China's economy has become increasingly open. Most remarkably, the share of merchandise trade in China's gross domestic product (GDP) went from 10 percent in 1978 to 31 percent in 1991, with imports accounting for 15 percent.l/ Based on this measure, China appears to be more than twice as open as India and Brazil, and significantly more open than the United States or Japan. Moreover, import penetration in certain sectors is extremely high. For example, in 1990, imports amounted to an estimated 28 percent of GVIO in China's machinery and tansport equipment sector. While the growing openness of the Chinese economy has had a perceptible impact on the quality of a range of Chinese products, it is noteworthy that the structure of China's industry hardly changed between 1985 and 1990, despite a nonnegligible degree of import penetration and rising export ratios over the same period. Thus, the respective shares of light and heavy industries in total industrial output remained virtually unchanged at around 47 and 53 percent between 1985 and 1990. Likewise, the share of consumer manufctres (with such heavily export-oriented sectors as clothing, footwear and travel goods) has also stayed remarkably stable at around 8 percent of GVIO. This suggests that investment in China remained heavily directed, at least until the late 1980s, such that the allocation of investment across sectors went largely unaffected by the country's changing patterns of external trade. As open as the Chinese economy is today, trade still does not appear to play a sufficient role in domestic resource allocation. C. TRADE AND FOREGN ExCHANGE PLANNIG: THE REMABINIG 1SsuEs Trade Planing: What Next? China's trade system has moved from one in which, at the start of the reform period, almost all trade was planned and carried out through a handful of FTCs, to one in which the role of planning is much diminished. Similarly, pricing has moved from a position of wide- ranging rlbsidy and cross-subsidy, to one in which, in 1992, only import subsidies remain, and these are reaIly enterprise subsidies. Such export planning as exists now takes place through the foreign trade contract system in the form of value targets for export earnings. The present foreign trade contract system, although intended to be fixed on a "bottom-up' basis, still takes on a compulsory quality for several reasons. First, the value targets negotiated i- the contract for exports and the amount of foreign exchange to be remanded to the center are still mandatory. Second, fulfillment of the targets in the contract is a precondition for awarding bonuses to officials responsible for carrying out the export plan. Each province's contractually determined targets are in turn disaggregared and assigned as targets to 1/ This figue needs to be teated with caution. First, China's GNP is likey to be mesimated. Second, customs statistics on exports include the full value os exports based on processing of imported inputs, which tends to exaggerate the role of trade in the Chinese economy. Excluding the latter would reduce the value of China's exports in 1990 by about $11.9 billion, and the trade to GDP share to about 28 percent. Depending on what estimate of China's GDP is taken, the trade to GDP share could be anywhere between 18 and 26 percent.  - xviii - various provincial trading companies. Third, rebates for domestic taxes levied on export goods have now been linked to fulfilling targets for exports. The problem is that these targets have to be met by FTCs that are not entirely free to choose what goods they can trade and are no longer eligible for subsidies to cover their loss-making exports. As a result, FTCs cannot pay full attention to profitability while at the same time meeting their obligations under the foreign exchange contract. Available evidence suggests that the incidence of bad bank loans to FTCs nas gone up sharply since the contract responsibility system was introduced and subsidy payments began to be phased out in 1988. In order to address this problem, the authorities need to take the reforms of the export planning system to their logical conclusion as soon as possible. The trade contract responsibility system should be abolished and FTCs should be allowed to work towards maxinizing profits rather than foreign exchange earnings. Linking bonuses to profits, instead of to foreign exchange targets, would motivate FTCs to market only profitable exports, thereby generating export earnings for the country without concomitant domestic currency losses. Such measures will, however, not be effective without further institutional reform of the FTCs themselves. FTCs need to be granted greater autonomy (along the lines of the recent regulations pertaining to the operating mechanism of state-owned enterprises) so that they can function as truly independent profit centers. At a minimum, FTCs must be granted the right to choose their own product scope. On tht import side too, as noted above, the importance of planning has declined, with the coverage of the trade plan having fallen to under 20 percent of all imports (from 40 percent in 1988), and this trend can be expected to continue. However, a large proportion of nonmandatory plan imports continues to be subject to administrative regulation through tight control of foreign exchange allocations. The central government still controls 50 percent of all fioreign exchange earnings. Funding for key projects and for their associated imports is already allocated as part of the government's investment program and the state industrial policy. It is, therefore, redundant to also administratively assign foreign exchange for the purpose of procuring these imports (which account for more than 30 percent of all imports). All concerned state-owned enterprises (SOEs) or government agencies ought simply to purchase the foreign exchange they need in the foreign exchange adjustment ("swap') centers (FEACs) instead of having local enterprises surrender 30 percent of their foreign exchange earnings to the central government at the swap rate and then having these funds allocated to them administratively. The Exchange Rate Regime: Toward Convertibility Since the establishment in 1986 of FEACs, which created an official two-tier exchange rate system, the volume of FEAC transactions has grown rapidly; it reached $25 billion in 1992, or about one-half of all cash imports. The FEAC system has served two key functions: it has provided critical relief to exporters in maintaining export incentives, and it has forced the government to move the official rate to more market-determined levels. Nevertheless, important defects remain. For one, the existing system of quota retention is flawed. Quotas are monopolized by FTCs, with most local manufacturing enterprises being left with little or none of the foreign exchange they help generate. Since FTCs themselves do not have any direct import requirements, they have tended to hoard retention quotas for speculative purposes. More importandy, the system of trading predominantly in quotas instead of cash has denied the  - xix - People's Bank of China (PBC) an instrument for intervening in the foreign exchange market for purposes of stabilizing the exchange rate. Second, although 80 percent of foreign exchange earnings are now priced at the swap rate, the market for foreign exchange remains thin and fragmented, in large part because the government still does not purchase its foreign exchange requirements through the FEACs. Besides, the differential between the swap market and the official exchange rate, down to under 10 percent in 1991, has once again become significant. The differential exceeded 45 percent at the end of the first quarter of 1993 (despite amtempts, later abandoned, to enforce price ceilings in the FEACs). Strucural reform of the exchange regime has therefore become urgent. It is the stated objective of the authorities to unify the exdhange rates and make the renminbi a convertible currency. This process needs to be accelerated through the speedy implementation of the following recommended measures: (a) Replacement of the system of retention quotas with a system of cash retenton that would allow enterprises to retain their foreign exchange in resident bank accounts, reduce the incidence of hoarding and allow the PBC to intervene, if necessary, to stabilize the swap market rate; (b) Elimination of remaining restrictions on access to FEACs (reflecting either the status of the purchaser or the purpose of the transaction), and parallel creation of an integrated national swap market; (c) Widening of the scope of FEACs to cover all current account transactions (icluding nontrade transactions), and parallel abolition of all remaining surrender requirements, with the govermment having to purchase all its current account foreign exchange needs (mcluding for mandatory imports) through the swap market; and (d) Abolition of administered foreign exchange allocation for priority investment projects, and phasing out of that for mandatory imports. The above measures would make the renminbi convertible on the current account, strengthen the links between the monetary and the external sector of the economy, and thereby place a greater burden on monetary policy as an instrument for influencing the balance of payments outcome. It is important, particularly in light of the recent trends in monetary aggregates, that the authorities adopt a disciplined monetary policy stance while these measures are being implemented. Finally, as concems the question of moving towards full convertability on the capital account, the experience of other countries in general suggests that a measured pace is advisable, although there are some exceptions. Generalized opening of the capital account often leads to exchange rate instability. As such, the prudent course of action would be to leave liberalization of the capital account until after important remaining structural reforms, such as those relating to the import regime, have been implemented.  - xx - D. REFORmiNG CHNA'S TRADE REGIME Priorities and Perspectives on Reform Despite the declining importance of trade planning, China still operates a relatively complex trade regime. Apart from continued reliance on the mandatory import plan and the use of controls over the allocation of, and access to, foreign exchange, imports are regulated through tariffs, canalization (monopoly or limited import rights), licensing and direct controls. Overall, more than 50 percent of China's imports were subject to some form of nontariff administrative control in 1992, with imports under the mandatory plan covering 18.5 percent of imports. The import regime, therefore, bears the marks of considerable government management. In addition, licenses, quotas and taxes are maintained to regulate the exports of a variety of products. After seeing a substantial increase over the last few years, China's weighted average tariff in 1992 was back to its pre-1987 level, with a trade-weighted average of 32 percent On average, however, China's tariffs remain higher, more numerous and more dispersed than those of most other large developing countries, with 69 rates and a standard deviation of 30 percent, compared with, for example, 34 rates and 17 percent for Brazil at an identical, average ariff rate. The multiplicity of objectives seems to account for the high dispersion of China's tariff structure, with a desire both to protect sectors in which domestic production is significant and to penalize nonessential consumption. This has kept import penetration in certain sectors very low and has provided high margins of protection to local production. The most important method of nontariff control of imports is to assign inmport rights to one or a few FTCs, such as for timber, cement and fertilizers. This process we label canalization. An estimated 32 percent of total imports are subject to control through canalization. Of these, two thirds are imports under the mandatory trade plan. For the remaining 13.5 percent of imports, therefore, canalization is used as an instrument for controlling import demand for reasons that have nothing to do with the trade plan. In addition to canalization, import licenses are used to serve multiple objectives. On the one hand, licensing is used as an administrative device to allocate a fixed quantity of planned imports and centrallylprovincially controlled foreign exchange. Here it functions as a quota allocation mechanism. On the other hand, licensing is also used for protecting domestic economic activity as well as for regulating, for balance of payments purposes, the demand for imports financed through retained foreign exchange. In all, there are presently 53 broad categories of products subject tO import licensing. These accounted for 12 percent of all HS tariff lines in 1992 and covered 25.1 percent of China's total imports. Of these imports, however, more than half were also subject to canalization and there appears to be some redundancy. In 1992, those imports for which licensing requirements applied in a nonoverlapping manner accounted for an estimated 11.7 percent of China's total imports. Import controls (distinct from import licerLes) are primarily used to protect the machinery and electronics sector, through the State Council Machinery and Electronics Inport Control Office (SCEMIO), and such controls currently apply to about 7.7 percent of total imports. The combined effect of licensing and controls serves primarily to control three groups  - xxi - of products: agricultural raw materials subject to domestic price control, critical domestic production such as steel and textiles, and nonessential consumer goods. Export quotas/licenses covered 15 percent of China's exports in 1992. A large set of commodities subject to export licensing was agricultural goods, such as beef, pork and vegetables, exported to Hong Kong. Here, the objective of the licensing arrangements is to increase the prices received for these commodities by controlling supply. The same is true for export licensing in the case of such commodities as tungsten, in which China has a very large share of the international market (40 percent). On the other hand, export controls on such products as rice and maize have been used to ensure the adequate availability of these goods domestically. The scope of export controls was reduced somewhat this year. Even so, 38 broad categories of products still remain subject to export quotas/licenses. The use of export licenses to increase the domestic availability andlor depress the price of a variety of key planned commodities has, however, been more significant. The government's objective seems to be not only to fix official prices below international levels but also to maintain the secondary market price of selected exportables, such as coal, petroleum, maize and rice, below world parity by restricting exports through the widespread use of export licensing. The rationale for these controls will disappear as China phases out the implicit subsidies to consumers and industry that its policies of price control entail. In summary, China's import and export regimes appear to operate essentially to raise the price of final consumer goods relative to producer intermediates, mirroring basic biases in China's general industrial policvy Prices of many agricultural goods appear to be depressed through the use of implicit export taxes and their equivalents. Prices of basic producer inputs to manufacturing, such as coal, oil and timber, are likewise depressed. On the other hand, the prices of most intermediate and capital goods are maintained above import parity. The prices of some intermediate inputs, especially petrochemicals and textile yaws, that account for a significant proportion of China's total industrial output are exceptionally high. This no doubt penalizes the competitiveness of some downstream sectors, such as apparel and footwear, in which China has obvious comparative advantage and seems to contribute to the low domestic content of export-processing activity. Import licensing is also used to reinforce the price- increasing effect of even higher tariffs on a selection of "higher-tech" manufactured goods. Notwithstanding the formal regulatory system tat survives de jure, China's import regime is de facto more open than the above description might suggest- First, despite the 32 percent average nominal tariff rate (which is not dissimilar to average tariff lei Als in other developing countries), China's actual duty collection ratio is only 5.6 percent (more akin to the situation in industrial countries). The gap between nominal and effective rates indicates very high levels of duty exemption in China. China operates a relatively well-developed system of duty exemptions for exporters, and duty concessions of 50 percent are provided for foreign- funded enterprises. About half of all imports are treated as concessional in this sense. In addition, it seems that a range of imports for priority projects is also exempted. The rate of duty collection as a share of total imports has declined from 9.7 percent in 1986 to today's very low level, which is only about one-third of the average rate of collection of other developing countries. The rapidly declining duty collection ratios are in fact cause for sorme concern to the extent that they are caused by (a) increasing evasion on products for domestic consumption, or (b) increasing exemptions on imports for use on domestic (as opposed to export) production. On the other hand, the small revenue contribution of China's tariffs endows it with much greater  - xxii - flexibility than most other developing countries. In particular, it enhances China's options with regard to the sort of fundamental restructuring of its tariffs suggested by the report's analysis. Second, price comparison data suggest that such protective devices as high nominal tariffs and nontariff barriers (NTBs) are not binding for many products (particularly for a range of mature consumer manufactures), i.e., their domestic prices, although still higher than world prices, are nevertheless below the duty-inclusive prices of competing imports. For these products, import licenses are redundant as a protective instrument and there would also appear to be considerable 'water in the tariff" caused in part by smuggling. Examples of such products include automobile tires, small gasoline engines, cassette recorders, televisions and domestic refrigerators. In the case of cassette recorders, domestic prices are still close to two-thirds above import parity, although they are 35 percent below the duty-inclusive price of competing imports. Likewise, domestic prices of color and black-and-white televisions are between 80 and 40 percent above import parity but between one-fifth and two-thirds below the duty-inclusive price of imports. The foregoing analysis of the trade regime underlines the need for substantive reform. Aside from simplification and reduction in the number of instruments of control and the dispersion of tariff rates, significant overall lowering of tariffs (but with fewer exemptions, except for export production) would seem to be in order. There could be a number of strategies for such a reform effort, and the report examines six alternatives through a simulation exercise. The results of the simulations provide several useful pointers. First, they suggest that China should be able to undertake deep cuts in tariffs and NTBs without this resulting in any major contraction of even the most protected sectors such as textiles and machinery. In a country of China's size, domestic consumption relative to imports can be expected to remain large and domestic dislocations are likely to be limited because of the considerable differentiation between imports and domestic production. Import liberalization should improve China's export performnance, particularly of its machinery sectors. The conditions for successful liberalization, however, are that activities within each sector be allowed and able to switch to the more export-oriented segments and that macromanagement remain disciplined. Second, it seems that radial import liberalization (e.g., a 50 percent across-the- board reduction in nominal tariffs) is a superior option to selective liberalization limited to the currently most protected sectors. Nonetheless, among the possibilities for selective liberalization, the option of reducing import protection on only the machinery sector is likely to produce the best results for the least effort and dislocation. Third, notwithstanding possible dislocation to downstream industries, significant gains could be derived from the reduction of export controls and taxes in addition to the reduction of import protection. Timing, Sequencing and Linkages It is evident that trade reform on its own is unlikely to yield the desired results. For it to succeed, progress in and coordination with, other areas of policy and reform will be essential. This does not mean, however, that reform of China's trade regime should wait. While it is true that trade reform needs other reforms to make it fully effective, it is also true that progress in trade is likely to generate important momentum for reforms in other areas. Moreover, a number of factors relating to both the domestic situation and the international trading environment suggest that the time is indeed ripe for China to embark on a bold program of trade liberalization.  - XXIiI - Trade reform in China cannot proceed without the further dismantling of the country's trade and investment planning apparatus. In addition to the elimination of the current system of foreign exchange planning and allocation, the system of foreign exchange contracting needs to be replaced with one in which FTCs are free to pursue profits rather than foreign exchange targets. Further, it is imperative that investment (especially nonplan investment) be allocated in accordance with market signals. One step that should be taken immediately in this context is to substantially raise the minimum level of investment requiring central government approval. Trade policy is an integral part of overall macroeconomic policy. The evidence from other countries is clear about the importance of a real depreciation of the currency for the success of programs of trade liberalization. It is critical that the steps recommended above for unifying the exchange rate and making the currency convertible for all current account transactions be implemented as quickly as possible and certainly before trade liberalization proceeds too far, not least so that the impact of reduced protection on the domestic industrial sector can be moderated through an appropriate depreciation of the exchange rate. Action on reforming the exchange rate regime is all the more important now, given the most recent tendency of the gap between the official and the swap market exchange rate to widen again. The experience of other countries also indicates that trade reform will go much better if carried out during a period of relative macrostability. At the same time, by acting as a "safety valve," a more liberal trade regime can itself contribute to the management of aggregate demand. This is a perdnent consideration for China. With its comfortable level of international reserves and a current account surplus, China is well placed at present to use trade liberalization as a means of addressing the threat of overheating and the emerging shortage of essential industrial raw materials. An important objective of reforming the trade regime is to rationalize the structure of incentives for domestic economic activity and thereby improve resource allocation. If domestic price controls remain in place, however, the trade regime would be of little help in accomplishing this objective. This is not to say, in the case of China, that trade liberalization should await further price liberalization. China has already made very considerable progress with regard to price reform. The incidence of subsidies for imported commodities has declined substantially. Moreover, the widespread application of the two-tier pricing mechanism has meant that a large measure of price flexibility exists for much of nonplan domestic economic activity. Under the circumstances, import liberalization can be expected co be effective for an important segment of the economy even without removing such price controls as still remain. Enterprise efficiency is not only a micro but also a significant macro problem in China. With losses equivalent to almost 5 percent of GDP, and financed in good part through loans from the banking sector, SOEs are the largest contributors to the government's fiscal and quasi- fiscal deficit. Trade liberalization can be a valuable reform tool for altering the behavior of enterprises and improving the efficiency of resource allocation. To be successful, it requires enterprises to be able to shift patterns of production and investment in response to changing incentives. In this context, the growing volume and importance of nonstate enterprises and the increasing flexibility being accorded to SOEs are encouraging trends which suggest that the time is right for initiating trade liberalization. Indeed, import liberalization should not wait because, once initiated, it can play a key role in exerting competitive pressure on SOEs and in maintaining the momentum for enterprise reform.  - xxiv - It should be emphasized that rationalization of China's industry will require much more than just import liberalization. Parallel efforts will need to be made in such areas as support for export development, quality control, worker training, technology policy and competition policy. If China wishes to have a targeted program of support to its export sector and, indeed, if it wishes to have a meaningful industrial policy, then these two policies will have to become the mirror image of each other and be very closely coordinated, preferably within a single agency. Finally, as China's presence in global markets continues to grow, it must become increasingly responsive to the demands of trade diplomacy. In its bid to attain full membership status in the General Agreement on Tariffs and Trade (GATI), China has already made significant efforts to conform to the expectations of the international community in general and the United States in particular. However, the requirements for GATT membership are not precise and are a matter for some negotiation. If the treatnent of recent applications to the GAiT is any guide, China is likely to be called on to go further in relaxing its import regime. From the perspective of the emerging international trade environment, therefore, this appears to be an opportune time for China to pursue a substantive program of trade liberalization. Defining a Bold Program of Trade Liberalization Over the last decade and a half, China has achieved a phenomenal upsurge in exports and trade. Most recently, however, its success in raising investment, growth and trade surpluses has raised the specter of overheating and inflation. Besides, enterprise inefficiency remains an important micro and macro problem for China, and its dramatic entry into export markets has raised concerns amongst its major trading partners about its responsibility for opening up to imports. The launching of a program of trade liberalization could help China address all three of these issues. What is more, given (a) its comfortable reserves position, (b) the advanced state of price reform, (c) the expanding role of the nonstate sector and growing flexibility in domestic resource allocation in general, and (d) evidence of tariff redundancy and the de facto openness of the economy, China is particularly well positioned to implement bold measures in the area of import lit ..lization at this time. China has begun to move in the right direction. Recent import liberalization initiatives undertaken as part of China's bid to attain full membership status in the GATT and as part of its bilateral trade negotiations with the United States include publication of regulations, some reduction of tariffs (tariffs on 3,371 tariff lines were reduced an average of 7 percent in December 1992), abolition of import substitution lists, and limited removal of import licenses and controls, with a commitment to eliminate two-thirds of these over time. However, no aanouncements have yet been made on import planning, canalization or phased tariff reduction. Considering the opportunity that China has to liberalize trade at this time, these measures seem quite incomplete and need to be complemented by other measures over the immediate and medium term. The report makes the following recommendations: Phaseout of Canalization. The distinction between Category I and IH imports should be abolished immediately, such that a single list of only those products that are subject to mandatory import planning remain subject to canalization for an interim period. All other imports should be made open to any FTCs or enterprises with direct trading rights. Beyond this, as reliance on import planning declines, China should phase out the practice of canalization altogether.  - xxv - Phaseout of Licensing and Controls. Nonbinding NTBs (as in the case of a range of consumer manufactures) should be removed immediately. The discretionary element of remaining import licenses and controls should be reduced by ensuring that all decisions with regard to imnport licensing and controls are made only by central govermnent authorities according to criteria that are uniform and transparent. Thc Memorandum of Understanding (MOU) that China recently concluded with the United States is an important initiative intended to reduce NTB coverage significantly by 1997. China must implement this agreement on a multilateral basis. Only then would this initiative constitute significant progress. Tariff Simplification and Reduction. The number of rates and the level of tariFfs applying to consumer goods should be reduced right away. imnediate steps could be taken on a range of mature consumer products for which there is evidence of "water in the tariff" (i.e., a partially redundant level of protection). On the basis of available data, it appears that reductions in the order of 20 to 40 percent should be possible, depending upon the product, without resulting in any significant dislocations in domestic production. Where tariffs are currently being used as a way to discourage consumption, these tariff reductions should be accompanied by the imposition of an appropriate sales tax. In parallel with the implementation, on a multilateral (not merely bilateral) basis, of the program of NTB reduction agreed to in the MOU with the United States, China should pursue an equally bold program of tariff reduction. A reasonable target would be to implement a 50 percent radial cut in tariffs as soon as possible. Such a tariff cut would bring China's tariff structure into line with that for Korea, for example, with average rates on consumer goods declining to about 32 percent, those on intermediate and capital goods to around 14 percent, agricultural goods to 17 percent and mining to 10 percent. Given that trade diplomacy is likely to remain an important part of China's future trade strategy, the actual phasing of these tariff reductions could be linked to progress in negotiations with trading partners, most notably within the context of the GATT. Tariff Exeemptions. It is recommended that all tariff exemptions (including those for foreign-invested enterprises) on imports for use in domestic production not as yet on- stream should be abolished, while exemptions on imports for use in domestic production already on-stream should be phased out over the shortest period possible. Reduction in Export Controls. Wherever China's existingprice controlpolicies necessitaw.: the use of export regulation, export taxes should be used in place of licenses because the latter allow the few FTCs designated to handle such products to capture sizable rent which would more appropriately accrue to the Government budget Such export regulatory devices as remain should in any case be eliminated progressively and in tandem with ongoing price reform efforts. E. POICICES FOR EXPORT DEVELOPMENn PRIORrS FOR THE 1990S China's past export support structure, centered on national foreign trade corporations, a very active and interventionist Ministry of Foreign Economic Relations and  - xxvi - Trade (MOFERT),Z/ and some foreign partners, such as entrepreneurs from Hong Kong, that were allowed to play an active role in the export sector, has served it very well during the first decade and a half of reform. Its experience has provided another example of the lesson observed elsewhere in East Asia that appropriate public intervention in export development and provision of explicit public support to the export sector can yield substantial dividends. While successful reform and continued success in export markets may require a change in the way in which China provides this support, it does not imply the removal of this support. Rather, it requires reform in the way support is provided in such areas as marketing, quality control, export promotion, export credit and the fostering of appropriate trade intermediaries. International experience offers a rich set of options from which China can learn in drawing up future reform plans in this regard. The most important issue will be how to deepen the reform of FTCs, which are likely to remain at the core of China's export support structure, without losing the considerable body of expertise that has been accumulated over the past. In order to meet its diverse needs, China should encourage the development of a variety of trading firms, ranging from small, flexible Hong Kong-type trading companies (best suited to handle its rapidly growing export segments such as garments and light industrial goods) to large trading companies similar to the Japanese sogo sosha or Korean chaebol (suitable for developing efficient production systems or promoting the trade of afiTliated conglomerate groups). To help achieve this objective, the report offers the following suggestions: subject FTCs to greater competition by removing remaining barriers to cross-provincial transactions, granting more producing enterprises the right to trade directly, and allowing entry to foreign trading firms; make FTCs operate as independentprofit centers, with bonuses linked to profits rather than to foreign exchange targets; eliminate all restrictions on the product scope of FTCs and permit them to participate in domestic commerce; allow ailing FTCs to exit and permit mergers between FTCs and, for example, emerging enterprise groups. At the same time as FTCs are being converted into competitive enterprises, the government should also assist manufacturing enterprises to export directly if they so choose, given the well-accepted benefits of direct contact between producers and overseas buyers. There are four essential elements in such support: (a) Export Marketing. At present, direct trading rights are not often granted to domestic manufacturing enterprises themselves, on the grounds that they are inexperienced in matters pertaining to international trade. Public support for export marketing (similar to that provided in Hong Kong, for example) could help ensure not only that those domestic enterprises that wish to export directly learn how to establish direct contact with foreign buyers but also that FTCs compete more vigorously to earn the business of local enterprises. Given the economies of scale involved, MOFERT could easily develop, with the support and participation of the nonstate owned sector, an effective intelligence network worldwide and chanrel information services to small firms that would otherwise be denied the opporrunity to trade directly- (b) Quality Control. China is well aware of the importance of export quality control, and has devoted a lot of attention to developing the State Commission V Now called the Ministry of Foreign Trade and Economic Cooperation (MOFTEC).  - xxvii - for Import and Exporn Commodity Inspection, but despite this effort, it appears that rejection rates remain high. Success in this area will depend on more general economic reform, but progress could be achieved, as in Taiwan (China), by making the present inspection system less extensive while increasing the intensity of the inspections. At the same time, the government could encourage the creation of quality control institutions by business associations and by accredited quality control agencies, including foreign ones, and it could then focus its attention on inspecting these. In short, China should focus on the efficacy of its quality control apparatus, rather than its coverage. (c) Other Support Services. The general promotion of an efficient, competitive service sector, which is now a key element of China's development strategy, will do most of what is required. Care must be taken to avoid monopolies in services for exports, and restrictions on the use of foreign service suppliers should (as has already begun to happen) be dismantled. (d) Export Financing. The Bank of China already provides a large volume of credit to support China's exports. Four measures seem necessary to improve the effectiveness of the export financing system. First, steps must be taken to ensure that firms with direct trading rights enjoy the same access to trade credits as FTCs. Second, if FTCs are to take responsibility for their own profits and losses, they must not be allowed to benefit from a soft budget constraint by tapping without restraint into the banking sector. Third, export insurance should be made available to all local enterprises as a way of encouraging and enabling them to acquire and exercise direct trading rights. Product selection has been an inherent part of China's export strategy, with special schemes and incentives being used to promote the exports of specific sectors. The experience of Korea suggests that export selection is a risky business and success requires that it be accompanied by an active and focussed policy of assistance for industrial restructuring to the targeted sectors. Unlike Korea, however, where decision making on trade and industrial policy issues has been highly centralized, the focus of China's industrial policies has been dispersed. This is partly due to the involvement of a large number of agencies, and partly to differences in priorities across provinces. As a result, China's efforts at export selection do not appear to be adequately supported by matching initiatives in industrial assistance. Given the country's size and diversity, such a strategy may prove to be difficult to pursue at the national level, except perhaps for a few strategic sectors. In such cases, greater national coordination between trade policy measures and assistance for industrial restructuring would seem to be warranted. The establishment of the State Council Economic and Trade Office (SCETO) in early 1992 (and its subsequent elevation to the status of a Commission), and ongoing efforts to create enterprise groups free from multiple channels of supervision and control, could prove to be important initiatives in this regard. A fundamental featare of China's past success in attracting FDI and generating exports has been its policies towards geographical targeting in general and the Special Economic Zones (SEZs) in particular. A key lesson is the importance of the policy environment in attracting export-oriented FDI. In this context, Pudong is potentially an important new initiative. As long as Pudong can offer a policy environment that is as flexible as the SEZs, it can be  - xxviii - expected to attract more investment flows and inject an important measure of vitality in the greater Shanghai area. The policy environment, though important, has by no means been the only attraction for export-oriented FDI. Thus, despite their new open cities, inland provinces would seem to offer limited prospects for attracting FDI flows because of the relatively longer distance of these provinces from international markets. By the same token, provinces such as Liaoning and Shandong, should be actively encouraged to exploit their proximity to Korea and Japan along lines similar to what Guangdong has done with Hong Kong and Macao and Fujian is pursuing with Taiwan (China). The tax and import duty concessions of China's SEZs and open cities have served their purpose in helping generate a momentum for export-oriented EDI flows. Their application has now become counterproductive. The ubiquitous use of such incentives has resulted in serious resource misallocation, with numerous domestic firms changing location merely to reduce their tax burden. These incentives should be phased out at the soonest possible opportunity and a standard national corporate tax should be adopted with local and foreign- invested enterprises being accorded equal treatment in the fiture. Instead, SEZs should focus on expanding their role as economic laboratories. Experiments with market mechanisms in China are still at an early stage, and adequate rules and regulations to prevent the abuse of market power and rent seeking are conspicuously absent. SEZs should be at the fiorefront of experimentation with the introduction of such checks and balances. Specifically, the System Reform Commission could be asked to develop a program along these lines in cooperation with the SEZ authorities. Meanwhile, plans to increase the number of SEZs should be resisted, and proposals to develop Hainan, Shenzhen and Xiamen as free ports should not be pursued. The creation of free ports would only exacerbate tle problem of smuggling, which is already quite severe with regard to goods from Hong Kong and those being channeled through existing SEZs. F. THE INTERNATIONAL ENvIRoNmENT Am CHNA'S EXPORT PROSPEcTS In order for China to successfully implement the program of import liberalization proposed above, it is critical that world market conditions allow it to sustain the momentum of its export growth. With a sluggish world economy, growing protectionism and trends towards regionalism, it is clearly relevant to assess China's prospects for continued high export growth. In this regard, the analysis of the report shows that while China faces a relatively low set of tariffs in its export market, it does face a significant number of nontariff barriers, primarily because of the importance of products such as clothing, textiles and footwear. However, this does not imply that China's export prospects are poor, nor does it suggest that China needs to reorient its export structure rapidly toward higher technology or knowledge-based products. The report finds that even in today's global trade environment, there exist opportunities, in terms of both markets and of other products that China could pursue in order to sustain high export growth rates. Although it can in general be concluded that import penetration issues will not act as a constraint on China's exports, two external factors could have an enormous impact: (a) the outcome of the Uruguay Round of the GAIT; and (b) the possible loss of MFN status in the United States.  - xxix - Most observers anticipate that a successful Uruguay Round could reduce protection levels in the European Community (EC), the United States and Japan by up to 50 percent. In such an event, China's exports would increase by an estimated 38 percent, or $11.4 billion, in terms of 1988 prices. Moreover, China would fare considerably better than other developing countries, the exports of which are projected to rise by only about 15 percent from a Uruguay Round liberalization. This is because the Uruguay Round tariff cuts would erode the preference margins that exports from other countries currently enjoy (through schemes such as the Generalized System of Preferences) and that China does not receive. China would benefit from trade diverted away from those countries. In addition, China would gain a lot due to the relatively high share of textiles and clothing products in its total exports. China has thus much to gain from the successful conclusion of the Uruguay Round. Although China's MFN status in the United States has, by executive order, been renewed for another year, uncertainty remains about its future and the impact of its possible discontinuation. The report finds that complete MFN loss would lead to severe dislocation of China's exports to the United Sttes. For example, the increase of roughly three and one-half times in the clothing duty (from an MFN rate of 15.3 percent to a general race of 55 percent) would significantly reduce, if not eliminate, exports of this key product (its projected decline is between 50 and 100 percent, depending on the assumptions, from its present level of $2.2 billion). Overall, China's annual export losses to the United States are likely to be between 42 and 96 percent, i.e., between $7.0 and $15.2 billion. Chinese exporters are, however, not the only ones that would lose. According to one estimate, United States consumers could end up paying as much as $14 billion per year in higher prices, resulting from a combination of costlier substitutes from alternative supply sources, and higher tariffs on the products that would continue to be imported from China. On the whole, it appears that the dislocation of trade flows likely to result from withdrawal of China's MFN status by the United States would range from the dramatic to the disastrous, with the associated costs being high for both parties. Other aspects of China's export prospects remain entirely within its own hands. Revealed comparative advantage calculations suggest that China's exports have been moving in line with its comparative advantage, which lies in labor-intensive (and especially in skiled labor- intensive) exports as well as in higher technology exports that can be assembled locally. Over the medium term (three to five years), China's present comparative advantage is unlikely to change significantly. The report's analysis indicates that China does not yet have a broad-based comparative advantage in machinery and electronics, despite all the programs of support to this sector, and it is unlikely to develop, for a few years yet, any significant advantage in the exports of heavy industrial equipment or in high technology exports that cannot be assembled locally. On the other hand, the report finds that there exist both underexploited geographical markets and new products that China could pursue without trying to alter the nature of its comparative advantage, while reducing the risk of market access problems. In conclusion, therefore, priorities over the next few years should lie in upgrading quality, diversifying into other skilled labor-intensive products and assembly-type exports, diversifying into underexploited geographical markets, and participating as fully as possible in initiatives to promote multilateral trade liberalization.  I. CI]NA'S MERCHANDISE TRADE-TRENDS AND PERSPECTIVES A. INTRODUClON Since the initiation of its "open-door" policy in 1979, the contribution of foreign trade to China's economy has grown at an extraordinary pace. For the last fourteen years, China's average annualized rate of export growth has been about 17 percent and its imports have grown at over 15 percent per year. Over the period, China's total exports (on a customs basis) increased almost ninefold and in 1992 were estimated at $85 billion, while imports grew more than sevenfold, and stood at $80.6 billion. In 1991, China was the thirteenth largest exporting nation in the world and it ranked as the sixteenth largest importer; its trade accounted for 1.8 percent of world merchandise trade.1 The World Bank analyzed China's trade regime in 1987. The objective was to assess the status of China's foreign trade and capital system and make suggestions for improvements. Since then, much has happened. On the one hand, there has been an increase in trade frictions the world over. China itself has been engaged in bilateral trade negotiations with the United States, even as it has been pressing its case for resuming its membership of the General Agreement on Tariffs and Trade (GATI). On the other hand, after seeing a temporary slowdown in 1988, China's trade has picked up momentum again. The fast paced growth in China's merchandise trade since 1988 is in part explained by the economy's recovery that has followed the austerity program of 1988189, but two major spurts of foreign trade system reform, in 1988 and 1991, respectively, have no doubt also contributed to it. These reforms have been far reaching in scope and substance, and their implications are still unfolding. China's 'Hong Kong connection" has also developed very rapidly over this period, and is another contributing factor to the country's impressive trade performance. Given these developments, China's foreign trade regime and prospects merit a second in-depth look. This report takes stock of the rapid changes that have occurred in China's trade performance and evaluates the country's emerging trade policy and the prospects for its exports. It is also the report's objective to propose directions for future policy. In order to do so, it seeks to address a number of questions that have not as yet been tackled. First, China's continuing shift from a planned to a market economy poses important challenges for the country's foreign trade policy. Of particular relevance is the question of how China's trade and foreign exchange planning systems should evolve so as to allow the use of indirect as against direct instruments of trade policy. Second, China's approach to trade policy so far as been "mercantHist," i.e., motivated by achieving export growth for the sake of generating foreign exchange without sufficient regard to its costs and linked with attempts to contain import growth. China has not used its trade regime as a well defined instrument of industrial policy and, as a result, reform of the import regime has hitherto been neglected. Much is still unclear about the functioning of China's import regime. The issue is  -2- not merely to examine the extent of market access that the regime provides, but also to define an approach for (i) maling the import regime more rational from the point of view of incentives to domestic industry; and (ii) liberalizing the regime over time, and thereby exposing a greater cross-section of domestic economic activity to international competition. Third, so far the Chinese government has played a very active role in providing support for exports. Aside from regulating trade activity and providing financial assistance for exports, the government has also directly provided a variety of other support services including marketing and quality control. As China continues its drive towards a market economy, an important question that needs to be addressed is the future role for public intervention in export development. Finally, growing uncertainties in world markets and a marked trend towards regionalism in mternational trade relations make the question of the prospects for China's exports a pertinent one. Whether world markets can absorb China's exports, and how vulnerable are Chinese exports to trade barriers in parter country markets are issues that need to be reexamined as China becomes a more important presence in world trade. With these issues in mind, the report is structured as follows: this chapter reviews the trends in China's merchandise trade, and provides an appraisal of the evolving composition and character of China's trade with the rest of the world. Chapter 2 traces the recent evolution of China's trade and foreign exchange planning system and analyzes the remaining problems. Chapter 3 provides a quantitative evaluation of China's current system of foreign trade controls, particularly from the point of view of protection of the domestic market. Chapter 4 explores options and strategies for future reform of this system of trade controls. Based on the analysis of previous chapters, Chapter 5 seeks to elaborate a program of trade liberalization for China. Chapter 6 focusses on export support measures. Lessons are drawn from the experience of other East Asian countries for defining the future role of public intervention for export development in China. Finally, Chapter 7 examines future prospects for China's exports and issues relating to access to international markets. B. TRENDS IN THE TRADE BALANCE Although trade grew over the 1979-89 period as a whole at very high rates, it is since 1984 that the most impressive growth has been observed. In 1983, total trade was only $43.6 billion. It then exploded over the next two years, to reach almost $70 billion in 1985 (Figure 1.1). Since then, trade has more than doubled, reaching $149.6 billion in 1992. Throughout this period, the trade balance has followed other macro economic variables with a lag of about 6 months to a year. In general, there has been a tendency for imports to rise faster than exports, except when the government has intervened. Thus, over the last decade, a trade deficit has been recorded in every year except 1982183 and 1990/91, the years immediately following the two retrenchment programs of 1981 and 1989, respectively. In 1988, the trade deficit was almost down to zero, following a third period of retrenchment which lasted from 1985 to 1987. In the early 1980s, exports were driven by the trade plan. Export performance was, therefore, largely explained on the basis of the availability of an exportable surplus. Excess aggregate demand translated into a poorer export performance and a larger trade deficit. This provoked the government's first retrenchment program, which involved scaling down import requirements through the import plan and other administrative controls. The results of this program manifested themselves in the form of a temporary trade surplus of 1982/83.  -3- Figure 1.1: CHNA: TREND iN TRAE, 1978-91 Exporls/Imports in $ billions Trade Balance in $ billions 100 10 --- - - - -- - ----- - - - --5 80 t 0 60 40--- ________________ -5 40- ________________- --------------____ - 10 20 _- 0 l l l l l l -20 1978 1980 1982 1984 1986 1988 1990 1992 Year -Merchandise exports +Mercha.ndise imports IlTrade balance Note: In current prices and on customs basis (exports are .ob. and imports are c.i.t). Since then, China's macroeconomic management has remained plagued by sharp cycles. While the reform of the country's trade regime has progressed considerably, the govenument has continued to resort to all manner of administrative controls in order to address macro imbalances. In 1984185, major trade reform initiatives were implemented.1I These had the effect of liberalizing imports and the foreign exchange allocation system, and reducing the importance of the trade plan. As a result, a greater proportion of China's imports and exports were left to be determined by market forces. The very rapid growth rates that China experienced in 1984/85 were accompanied by a deteriorating trade balance. The trade balance reached a record deficit of $15 billion in 1985, as exports remained stagnant, while imports surged by over SO percent. In response, the government resorted to stricter controls. Credit was tightened, and heavy use made of administrative controls such as import bans, quotas and licenses, the importance of such instruments having gone up as that of the import plan had diminished. As a result of these various measures, import growth was stabilized. Meanwhile, exports finally 'took off,' growing at the dizzying pace of 20 percent per annum, and by mid- 1988, the trade balance was virtually restored. At that time, controls were relaxed and imports were allowed to pick up again. However, the underlying macroeconomic trends were such that export growth, though rapid, could not keep up with imports. A major austerity program was I/ The Mnistry of Foreign Economic Relations and Trade's (MOFERT's) report on the reform of the tade system appmrved by the State Council in September, 1984.  -4 - launched in 1988. By mid-1989, the trade deficit was back up at $12.3 billion, not far from the record of 1985, prompting the major austerity program of 1989. As on previous occasions, the trade balance responded, but with a lag. Thus, although in mid-1989 the trade deficit was still $12.3 billion, close to the record of 1985, 1990 saw a surplus of $13.1 billion, reflecting an 18.1 percent growth in export value and a 9.8 percent decline in merchandise imports. As the recovery continued in 1991, there was yet again a sharp reversal in pattern, with exports continuing to grow at 15.8 percent to $71.8 billion, but with imports rising sharply by 19.5 percent to $64 billion. However, unlike in previous years, China was able to maintain a surplus on the trade balance in 1991. All the evidence suggests that over the decade of the 1980s, export performance has become increasingly independent of the level of within-plan domestic economic activity. The secular growth of exports can no longer be explained merely on the basis of the availability of an exportable surplus, resulting from an excess of planned output over domestic demand. Since the mid-1980s the overall export trend has been much more stable than it was in the early 1980s. In fact the growth rate of the few key sectors such as clothing and footwear that have underpinned China's recent export performance was hardly affected by the excess domestic demand of 1988/89. This trend is likely to continue, and it is safe to assume that export perfbrmance will depend much more on such factors as the exchange rate, policies for export development, and the prospects for world trade.21 On the other hand, inports appear to be explained by a combination of three factors. Recent analyses (Figure 1.2)1/ have confirmed a strong correlation between broad money and currency, industrial production and imports The level of domestic industrial activity appears to be the most important determinant of import demand in China. In parallel, the expansion of the system of foreign exchange retention has undoubtedly also made imports more sensitive to the exchange rate.4l Finally, the trend in the trade balance discussed above demonstrates how important administrative controls still are in managing China's import demand. The one important lesson that emerges is that, as long as China is unable to better manage macroeconomic balances, it will be very difficult to dismantle the array of discretionary controls that still distort the country's import regime.5/ C. PERSrECVS ON CHINA'S EXPORT PERFORMANCE Over the decade of the 1980s, China's exports outperformed those of most other economies, including such Newly Industrialized Economies (NIEs) as Malaysia. The only exporters that registered average annualized export growth rates higher than China were Thailand 2I Chapter 2 examines the evolving link between the exdcange rate and export performance, Chapter 6 explores China's policies of export development, and Chapter 7 examines the prospects for China's exports. 3/ Intemational Monetary Fund (IMF) (1991). U The reminbi recorded an effective devaluation in the swap market rate of over 13 percent between the third quarter of 1989 and end 1990, while access to swap centers reached new heights (the volume of transactions on the foreign exchange adjustment centers (FEACs) rose 53 percent to reach $13 billion in 1990). Imports fell by around 10 percent in 1990. 51 See Chapters 3 and 4.  Table 1.1: SRUCruRAL CHNGES IN CHINA'S MAJOR EXPORTS: SELECrED YEAS FROM 1965 TO 1990 La Commodity group L 1965 1975 1980 1985 1990 (value of trade in terms of $ million) Total exports 1,718 6,303 18,237 27,764 80,541 All foods 642 2,088 3,272 4,073 6,862 Agricultural raw materials 189 450 1,170 1,810 2,198 Mineral fuels 32 897 3,974 7,158 5,290 Crude petroleum 13 778 2,572 5,347 3,654 Refined petroleum 7 92 1,246 1,556 1,070 All manufactures 783 2,632 8,521 13,657 64,220 Chemicals 75 322 1,176 1,460 3,420 Textiles and clothing 364 1,140 4,089 7,304 23,204 Nonelectric machinery 20 107 248 282 1,930 Electrical machinery 14 78 240 536 8,666 Transport equipment 15 74 71 88 574 Ores, minerals and metals 68 212 560 760 1,486 Miscellaneous goods 4 24 740 303 485 (as a percentage of total exports - Total exports 100 100 100 LIA 100 All foods 37 33 18 15 9 Agicultural raw materials 11 7 6 7 3 Mineral fiels 2 14 22 26 7 Crude petroleum 1 12 14 19 5 Refined petroleum - 1 7 6 1 All manufactures 46 42 47 49 80 Chemicals 4 5 6 5 4 Textiles and cothing /c 21 18 22 26 29 Nonelectric machinery 1 2 1 1 2 ElectiLA machiery 1 1 1 2 11 Transport equipment 1 1 - - 1 Ores, minras and metals 4 3 3 3 2 Miscellaneous goods Id - - 4 1 1 LI Data used here are based on imports of partner countries from China. This procedure was necessitated by the fact that China did not begin reporting exports to the United Nations until the mid-1980s. I Total exports consist of all products classified in Standard International Trade Classification (S]TC) 0 to 9; foods are groups (0+1 +22+4); agricultural raw materials (2 less 22, 27, 28); mineral fuels (3); crude petroleum (331); refined petroleum (332); manufactures (5 to 8 less 68); chemicals (5), textiles and clothing (65 + 84); nonelectrical machinery (71); electrical machinery (72); transport equipment (73); ores, minerals and metals (27+28+268); miscellaneous goods (9). Jc The share of clothing alone went from 13.1 percent in 1985 to 19.6 in 1990. Td The share of footwear went from 1.2 percent in 1985 to 4.5 percent in 1990. Source: Statistics are compiled from the United Nations Commodity Trade System (COMTRADE) database.  -6 - Figure 12: BROAD MONMY AND IMPORTS Z Oily. Chunge In Imports % Qily. Change Tn Broad Money 60 16 40 1 4 4 12 20 10 0 -20 b 6 4 -40 2 -60 0 12341 23412341234123412341234 1986 | 1987 1 1988 1 1989 1 1990 1 1991 1 1992 Quarter/Year -% Gr. Imports +% Gr. of Broad Money Source: IMF and Bank Economic and Social Database (BESO). (13.2 percent), Korea (12.8 percent), Taiwan (China) (12.1 percent) and Portugal (11.7 percent).2 Institutional decentralization, depreciation of the real effective exchange rate, foreign investment (especially from Hong Kong), and duty free access to imported inputs for export assembly, all seem to have contributed to this strong performance. The Changing Compositior, of China's Exports 61 Table 1.2 shows the value and share (in total trade) of China's major export products for selected years from 1965 to 1990 (on the basis of import data from partner 61 Unless stated otheiwis, all data pertaining to sectoral composition of merchandise trade are presented throughout the reDort per classifications of Standard Intemnational Trade Classification (SITC) Revision 1. This was done for the purpose of historical continuity.  Table 1.2: T1I VA ANo SimARE OF MJoR FlURDtaiT S1TC PROWCIS IN 'Cums EwORms-Swzcm YR FROM 196S To I9 Product IStTCI 1965 1975 1980 1955 S199 1965-90 Value Shari of Value Share of Value Share of Value Share of Value Share of Share $ mill.) total (23 S *mil1.) total (2) (4 *sll.) total (2) (4 mill.) total (2) (S mill.) total (l) change ToUtl trade (0 to 9) 1,718 100.0 6,305 100.0 18,237 100.0 27,764 100.0 80,541 100.0 - Clothing (641) 54 3.1 26S 4.3 1,592 6.7 3,629 13.1 15,760 19.6 12.3 Toy. and sporting goods (694) 14 0.6 38 0.6 108 0.6 653 2.4 6,049 7.3 6.7 Tolec inicatione equipment (17243 1 - 11 0.2 51 0.3 256 0.9 4,715 5.9 5.9 Crude petroleum (331) 13 0.8 778 12.3 2,572 14.1 5,347 19.2 3,654 4.5 3.7 Footweer (351) 16 1.0 60 1.0 201 1.1 330 1.2 1,632 4.5 3.5 Travel goode (831) 1 - 23 0.4 92 0.5 456 1.6 2,987 3.7 3.7 Other miecallanecue manufacture. (8992 22 1.3 115 2.2 403 2.2 534 1.9 2,060 2.6 1.' Woiwn teatile artielee (653) 53 3.1 148 2.3 500 2.7 923 3.3 2,443 3.0 -0.1 Doseetic electrical equipment (725) 1 - a 0.1 32 0.2 74 0.3 1,725 2.1 2.1 Cotton fa-brics (652) 174 10.1 327 5.2 697 3.8 973 3.5 1,652 2.1 -8.0 Presb fimb (031) 44 2.6 150 2.4 347 1.9 346 1.3 1,583 2.0 -0.6 Sound recording equipment (8913 2 0.1 8 0.1 25 0.1 62 0.2 1,472 1.8 1.7 Articlee of plestlc (8931 1 - 6 - 27 0.1 87 0.3 1,367 1.7 1.7 Wade-up textile articles (656) 43 2.5 181 2.9 554 3.0 586 2.1 1,290 1.6 -0.9 Vatchee and clocka (164) 3 0.2 18 0.3 76 0.4 186 0.7 1,270 1.8 1.4 Electric power machinery (722) 2 0.1 24 0.4 54 0.3 102 0.4 1,233 1.5 1.4 Textile yarn (6512 24 1.4 117 1.9 278 1.5 753 2.7 1,101 1.4 - Refined petroleum products (332) 7 0.4 92 1.5 1,246 6.8 1.556 5.6 1,070 1.3 0.9 Organic chemicele (512) 8 0.5 52 0.8 265 1.6 334 1.2 803 1.0 0.5 Other electrical mbchinery (729) 6 0.5 30 0.5 79 0.4 64 0.3 766 1.0 0.5 Prepared or preserved vgetablea (035) 14 0.8 93 1.5 279 1.5 338 1.2 713 0.9 0.1 Hachinery and appliane.s, nec (719) 4 0.2 22 0.3 66 0.5 111 0.4 703 0.9 0.7 office achinery (714) - - 2 * 8 - 51 0.2 694 0.9 0.9 Purniture (621) 3 0.2 29 0a5 127 0.7 157 0.6 633 0.8 O.6 Vegetables fresh or frozen (054) 45 2.6 97 1.5 311 1.7 293 1.1 631 0.6 -1.8 Scientific instnumentu (861) 12 0,7 10 0.2 27 0.1 43 0.2 583 0.7 floor coveringe (657) 6 0.5 54 0.9 270 1.5 277 1.0 565 0.7 0.2 Oil ceeds and nutC (2213 92 5.4 135 2.1 165 0.9 413 1.5 556 0.7 -4.7 Coal and coke (3211 13 0.8 26 0.4 135 0.8 243 o.s 547 0.7 -0.1 Pottsry and coramice (6d6) 15 0.9 71 1.1 140 0.6 147 0.5 546 0.7 -0.2 Total of above Items 699 40.7 3,011 47,6 10,767 59.1 19,354 69.7 62,609 75.0 37.3 Other tbree-dSgit exports 1,019 59.3 3,290 52.2 7,450 40.9 8,410 30.3 17,736 22.0 -37.3 Source: Sahtideatu camcmpUd ffom the UniteA Naiona COMTRADE data bue uslng the reporS bprtb of parb= countui frm China.  -8- countries).21 The top half of the table indicates that China's total exports roughly tripled three times over this period; between 1965 and 1975; from 1975 to 1980, and thirdly between 1980 and 1990. In the early years, China's exports were guided by the philosophy of exporting only surplus commodities. Thus1 initially, China's export growth was dominated by crude petroleum and nonstaple foods. Raw materials that could have been exported very profitably in view of domestic price distortions-such as coal-were restrained by tightly administered export quotas. As the reforms have progressed, the structure of China's exports has changed dramatically. Most importantly, the share of China's traditional exports, foods, agricultural raw materials and petroleum, has declined progressively since 1975 from 52 percent down to 18 percent in 1990. On the other hand, the share of manufactures has grown by 34 percentage points from 46 percent of total exports in 1965 to 80 percent in 1990.8/ What is notewordty is that China's manufacturing exports did not "take-off" until after 1985. Just within the five year period 1985-90, the share of manufactures in total exports rocketed by 31 percentage points. The explanation for this sudden upturn in the performance of manufacturing exports lies undoubtedly in two factors: (i) the specific foreign trade reforms that were initiated at around that time including in particular, the decentralization of FTCs, the creation of swap markets for foreign exchange and the introduction of foreign exchange retention for exporters3; and (ii) the quantum jump in the utilization of foreign direct investment! The steady depreciation of the swap market exchange rate cerainly provided a strong boost to exports in general and to exports of sectors entitled to higher retention rates in particular.2/ The sector whose share in China's exports grew the fastest over the 1985-90 period was electrical machinery (SITC 72), which recorded a 10 percentage point expansion of its share, going from 1 percent of China's exports in 1985 to 11 percent in 1990. However, clothing and footwear also enlarged their share by 9 points over the same period, to reach 23 percent of China's exports, more than double that of the share of electrical equipment. 71 China did not officially bea reporting trade data to the United Nations until 1984. Historical statistics on China's exports were, therefore, reconstructed from reported imnpons from China by other countries. Several points need to be bome in mind when interpreting the data. First, parter country data on China's exports are systematically higher than the value of exports reported by China, because the former are in most cases based on the c.i.f value of partner country imports from China, whereas the latter are based on the fo.b value of China's exports. Thus, while Table 1.1 shows China's total exports in 1990 as $80.5 billion, the value of exports reported by China for the same year was $62 billion (see Table A1.2)-the difference between the two being in large part the value of tanwsport and insurance payments. Second, for some product categories, the difference between partner country and Chinese data is difficult to explain. In the case of transport equipment (SITC 73), for example, China's exports per partner country records were $0.6 billion in 1990, whereas Chinese data indicate a sotal export value of $4.1 billion. Third, some of China's trading partners-particularly the (former) socialist countries of Eastem Europe-did not report to the United Nations (UN) during the full 1965-80 period so some of China's tade (perhaps about 10 percent) is not accounted for in those years. In general though, statistics on imports are thought to be more accurate than those for exports (particularly since the former are used for applying import tariffs) so the partner country (import) information should provide a more accurate profile of China's exports. 8I Based on Chinese reported data, the share of manufactues reached 71 percent in 1990 and 80 pflpct in 1992. 9f Between 1985 and 1990 the real effective exchange rate (REER) applying for exports depreciated by over 120 percent (mneasured in yuan per dollar, see Chapter 2).  - 9 - Table 1.2 provides a more detailed look into the factors undedying China's changing export structure over 1965 to 1990 by ranking the 30 largest three-digit SITC products (in terms of 1990 values). Crude petroleum ranked as China's most important export as recendy as 1985, when it was replaced by clothing. Clothing more than doubled its share in ten years to account for 19.6 percent of China's total exports in 1990. Toys and sporting goods were the second most important export item for China in 1990, followed by footwear, accounting for 7.5 and 3.7 percent, respectively, of China's exports. Both these subsectors saw their shares more than treble between 1985 and 1990. The rapidly expanding share of electrical equipment in China's exports seems to be accounted far in large part by telecommunications equipment (comprised essentially of black-and-white televisions, radio receivers, and telephone equipment) and by domestic electrical equipment (which includes washing machines, air conditioners and refrigerators). These two subsectors saw their share of China's exports rise from zero in 1975 to 7.3 percent in 1990, with almost all the growth coming after 1985. Table 13: TmE VALUE AND SHARE OF VARIoUS TYPES OF PioDucTs IN CmNA'S MANUAcuRD Exroars: SELECtED YARS 1965 Xn 1990 Growt rate ( Product category 1965 1975 1980 1985 1990 1965-90 1980-90 (Value of China's exports in tenrs of S millions) Total exports 1,718 6,303 18,237 27,764 80,541 16.6 16.0 Labor-ienive manuficturs 570 2,253 7,168 12,319 59,787 20.5 23.6 Unskilcd labor-intensive -oods 454 1,557 5,254 9,742 41,222 19.8 22.9 Capital-itcensive manufictes 1,113 3,128 6,353 7,984 14,978 10.9 8.9 Human capital-intensive goods 148 473 1.292 1,708 12,325 19.3 25.3 Naturl resource-based products 961 3,665 9,116 13,339 16,585 12.1 6.2 Coal, petroleum and gas 32 897 3,974 7,157 5,290 22.7 2.9 (as a shae of total exports - f) Total cpots 100 100 100 100 100 - - Labor-intensive manufactures 33 36 39 44 74 - - Unskiled labor-intensive goods 26 25 29 35 51 - - Capital-intensive manufactures 55 50 35 29 19 - - Human capital-intensive goods 9 8 7 6 IS - - Natural rcsource-based products 56 58 50 48 21 - - Coal, petroleum and gas 2 14 22 26 7 - - /a The procedures used for idnifying labor and capital-intensive goods (as well as a broad list of the former) can be found in Annex 1.1. Note that although the two classes of goods are mutually cxclusive, they do not encompass aL product categories, some of which are excluded from both classifications. See Annex 1.1 for details. Unskilled labor intensive, human capital intensive, and natural resource based products arc defined in Lawrence B. Krause, United Swates Economic Policy Toward the Associaion of Southeast Asian Nations (Washington: Brooldngs Instiution, 1984). Annex 7.2 to this report provides a complete tabulation of all the SITC codes that are included in the category of labor-intcnsive manufactures' product groups.  - 10- The sectors that have emerged as China's most dynamnic exports appear to conform well to the country's natural comparative advantage. The analysis of Table 1.3 confirms this impression. The important point that emerges is that the composition of China's exports has become increasingly labor incensive over the years. Over 1965 to 1990 the share of labor-intensive products in China's exports has grown from roughly one third to about 75 percent.101 Over the same period, the share of capital-intensive products was cut down to less than 20 percent, less than a third of what it was in 1965, and that of natural resource- based exports was more than halved, declining from 56 percent in 1965 to 21 percent in 1990. A second point to note is that, within the category of labor-intensive manufactured exports, the share of products using unskilled labor has been declining since 1985. That China has begun to turn to products requiring higher skill levels for its exports is evident from the rapid rise in its exports of relatively simple telecommnunications equipment and domestic electricals since 1985. Table 1.4 compares changes in the concentration of Chinese exports (i.e., the range of products involved) over the years with that of selected East Asian economies. Some Table 1.4: CONCEaNRAmON oF Cr&s AN COMPARATOR COUNTRY ExPORTS No. of three-dinit uroducts exported (Hirschmann index in parentheses) 1962 (Rev. 1) 1972 (Rcv. 1) 1980 (Rev. 2) 1988 (Rcv. 2) EVoner Singapore 144 (0.324) 176 (0.234) 232 (0.235) 231 (0.143) Rep. of Korea 30 (0.245) 162 (0.262) 207 (0.085) 215 (0.098) Hong Kong 91 (0.337) 128 (0.374) 164 (0.164) 173 (0.141) Taiwan, China 44 (0.429) 57 (0.401) 210 (0.117) 217 (0.093) China 47 (0.411) 58 (0.390) 190 (0.156) 211 (0.120) memo item Japan 174 (0.103) 176 (0.158) 224 (0.118) 220 (0.142) United States 175 (0.084) 180 (0.107) 236 (0.064) 235 (0.086) Source: United Nations Coafcrencr on Trade and Development (UNCTAD) Handbook of International Trade Stadstics, various issues. Indices for China esimated by the World Bandk To be counted a dtr-digit product must have exports of more than $50,000 in 1980 or $100,000 in 1988 or be more dtan 0.3 pement of total exports. The 1962 and 1972 statistics are based on SfTC Revision 1 data while 19W and 1988 employ Revision 2 data. About 180 three-digit products are dermed in the former and 240 in the lazier. indication of concentration is provided by the number of tiree-digit SITC products exported by each country. Another indicator used is the Hirschmann concentration index.5 This index lO See Annex l. I to this report for an explanation of how these products were identified and defined and for a complete tabulation of all SITC codes that comprise each product category. It should be noted that the groups are not necessarily mutually exclusive and some double counting is involved in the tabulations shown in Table 1.3.  - 11 - ranges between 0 and 1.0, with higher values representing more concentrated expoits. Two points are evident from these tabulations. First, in all countries a progressive deconcentration of their exports has occurred, although the rate of deconcentration appears to slow down appreciably (or in the case of Korea was even reversed) over time as the export drives of these countries reach maturity. China appears to be following the same trend. After its Hirschman concentration index more than halved over the 1970s, the Index registered a decline of only 20 percent over the 1980s. Second, China's exports have shifted from being far more highly concentrated than those of the East Asian NIEs to a point where they are no longer significantly more or less concentrated than other East Asian economies. In 1988, China's Hirschinan concentration index was smaller than that of Hong Kong or Singapore, but somewhat higher than that for Taiwan (China) or Korea. The Role of Assembly Operations In 1984, the State Council approved two schemes designed to facilitate exports based on processing or assembly type activity. The schemes are known as "processing with supplied materials" and 'processing with imported materials," respectively. The former entitles a local enterprise to import free of duty all raw materials provided to it by an overseas supplier to help the firm meet an export contract. The latter exempts a local firm from import duties on 9all raw materials that it uses to manufacture its exports.j/ These schemes have proven to be enormously successful. They have contributed in a major way to the marked rise in manufactured exports from China since 1985, and have helped China move into product categories such as telecommunications equipment, whose exports appear to have grown in large part based on assembly type operations using cheap yet reasonably skilled labor. Exports based on processing of imported or supplied materials accounted for 23 percent of total exports in 1988 (Table 1.5). Since then they have almost doubled their share, and now account for 45 percent of all exports or 64 percent of all of China's manufactured exports. 12/ Although a substantial proportion of exports based on processing activity was generated in Special Economic Zones (SEZs), over the years such activity has spread to firms outside the zones. Thus, while SEZs accounted in 1986 for over 27 percent of exports based on processing of supplied or imported materials, their share of such exports declined to 19 percent in 1991. Complete data on the sectoral composition of exports based on processing activity were not available. It is estimated, however, that the share of machinery and electronics exports is the largest, at about a third of all exports based on processing activity, followed by clothing (24 percent) and then toys (11 percent). It appears, therefore, that processing-based exports of machinery and electronics were in the order of $10 billion or almost 80 percent of all Chinese exports from these sectors in 1991. Much of this activity is clearly low in domestic value added. This is evident from estimates of the import content of exports based on processing activity. In 11/ The details of these schemes and how they operate are discussed in Chapter 3. It should be noted that these schemes are distinct from compensation tade. Compensation trade is similar to processing with supplied raw materials, except that the local firm gets paid in kind and does not receiv any foreign exchange. The value of exports under compensation trade schemes in 1991 was $221 million (Source: CustDms Directorate). 12/ Based on export data reported by China. The share of manufcctured exports on this basis was 70.9 percent in 1991, as opposed to an estimated 80 percent based on partner country data.  - 12 - Table 1.5: EXPORTS FROM ASSEMBLY OPERATIONS (X billion) 1988 1991 1. Exports processed with supplied materials 6.5 12.9 2. Exports processed with imported materials 6.4 19.5 3. Total value of processed exports 12.9 32.4 of which, from SEZs 2.6 6.2 4. Imports of materials for export processing 13.7 25.0 of which, into SEZs 2.6 5.1 Memo Item Total merchandise exports 47.5 71.8 Total merchandise imports 55.3 63.8 Source: Customs Directorate. 1991, imports for processing activity represented 77 percent J3t of the value of processed exports. Markets for China's Exports China would appear to have experienced a marked increase in the geographical concentration of its export markets (Table A1.3 414), due to the more than doubling of Hong Kong's share and the disappearance of Eastern European markets.L5/ However, this is misleading because Hong Kong has served increasingly as a conduit for Chinese exports rather than as a consumer of such products. China reports having exported 53 percent of its exports to Hong Kong alone in 1990, this share having risen from 26.5 percent in 1984. In fact, almost all of the growth of China's exports to Hong Kong is due to the fast growth in reexports of Chinese goods to other countries-only a small and rapidly declining proportion of China's 131 This is likely to be somewhat of an overestimate of the import content of processed exports. Time lags in the utilization of imports mean that not all inputs imported in any one year are likely to be used in the export production for that year. 14/ The statistics in Table 1.7 are based on data reported to the United Nations by China. A longer time pewspective was not possible since China did not report trade to the UN prior ti 1984. 15, While, in 1984, five Eastern European countries, including the then Union of Socialist Soviet Republics (USSR), were amongst China's top thirty markets, absorbing a little under 5 percent of its exports, by 1990, and not surprisingly, their share of China's exports had become negligible.  - 13 - exports tO Hong Kong is actually consumed in Hong Kong. Of the $6.9 billion that China exported to Hong Kong in 1984, $3.3 billion was reexported. In 1990, of the $32.9 billion exported from China to Hong Kong, $29 billion was reexported to countries other than China. The major markets for these reexports were the United States, which in 1990 absti 36 percent, followed by the European Community (EC) (17 percent), and Japan (7 percent). Taking into account the final destination of Hong Kong's reexports of Chinese products (Table 1.6), it turns out that Hong Kong's share of Chinese exports has in fict declined from about 14 percent in 1984 to only 6 percent in 1990. The fastest growing market for China has been th EC, whose share more than doubled to reach an estimated 19 percent in 1990, followed closely by the United States (US) whose share went from 13.6 to 25.6 percent between 1984 and 1990. As the share of the US and EC markets has grown, that of Japan has shrunk. Japan went from being the largest importer of Chinese products in 1984, to being the third largest in 1990, after the US and the EC. All these changes have made the market structure for China's exports very similar to that of other East Asian economies. For each of these countries the US is the largest export market, absorbing anywhere from 28 to 34 percent of their exports, followed by the EC and then Japan Cfable A1 6). Table 1.6: MARKFI5 FOR CHINA ExPRTS, 1990 Share of China exports (%) Export market Adjusted for Unadjusted for reexport thru reexport thim Hong Kong Hong Kong Japan 11.5 8.2 USA 25.6 8.7 EC 19.2 9.2 Other of which: Hong Kong 6.2 53.9 Memo Items Value of China's exports to Hong Kong ($ billion as reported by China) 32.9 Value of Hong Kong reexports of Chinese products to the rest of the World ($ billion as reported by Hong Kong) 29.0 Value of China's total merchandise exports 62.1 Source: United Nations COMTRADE data base, China's Customs Statistics and Hong Kong Review of Overseas Trade, various years. The Role of the Nonstate Sector Definitional issues pertaining to ownership make it difficult to pin down the contribution of China's burgeoning nonstate sector to exports. However, data are available on  - 14 - certain types of nonstate enterprises.j.i/ The share of township and village enterprises (TVEs) in total exports, for example, increased fivefold between 1985 and 1990 (Table 1.7), and their share was estimated to have exceeded 25 percent by 1992. An estimated 90 percent of IVE exports are manufactured products, of which a little less than half are spread evenly between textiles, clothing and arts and crafts. Also, not surprisingly, almost all TVE export earnings (88 percent in 1990) are generated in Eastern China, with the coastal provinces of Guangdong, Jiang-su, Shanghai, Zhejiang, Shandong, Tianjin and Hebei accounting for the lion's share. Table 1.7: TOWNSIUP AND VILLAGE ENTERPRISE ExPORTS Export-oriented Exports As share of total enterprise ($ billion) (%) (number) 1985 1.20 4.4 8,500 1986 2.67 8.6 9,000-11,000 1987 4.35 11.0 18,000-20,000 1988 8.03 16-9 La. 1989 10.00 19.1 n.a. 1990 12.501a 20.2 56,000 /a Beiing Review, Vol. 24, No. 4, January 24, 1992, p. 29. Source: For exports 1985-88: A. Ody (1992). Data for foreign-invested enterprises (FIEs) indicate that the contribution ofthese to China's exports has also been growing. The share of FlEs in total exports went from less than half of 1 percent in 1984 to 5 percent in 1988, and then quadrupled to reach $17.4 billion or 20 percent in 1992. All FIE exports have been in manufacturing, and 94 percent were generated in the coastal provinces in 1990.' It can be concluded from the above that the contribution of the nonstate sector to China's exports has become very important. The share of this sector, taking account only of the exports of TVEs and FIEs, stood at a minimum of 40 percent of China's total exports or L61 Bromadly speaking, the nonstate sector is considered to comprise urban collectives, nual collectives, private enterprises, individual businesses and foreign-invested firms. See Yusuf (1992). Township and village enterprises (IVEs) are distinct from foreign-invested enterprises or urban collectives, but constitute a subset of the other categories. Foreign-invested enterprises inchde equity joint ventures, wholly foreign-owned ventures, cooperative operations and cooperative development ventures. See Khan (1991).  - 15- 50 percent of its manufacturing exports,17/ with the bulk of the contribution of nonstate enterprises originating in the coastal provinces. The Role of Hong Kong China's 'Hong Kong connection" has been vital to the success of its export drive. It was noted earlier that more than half of China's exports to the rest of the world are now handled by Hong Kong. Although the role of Hong Kong as trade intermediary is critical, its contribution to the development of China's exports goes much further! Most importantly, 70 percent of the cumulative value of $58.1 billion in foreign direct investment (FDiI) commitments to China has come from Hong Kong and been mostly directed to export-oriented joint-ventures in the coastal provinces, Guangdong in particular. Guangdong has attracted an estimated 50 percent of all the country's foreign investment commitments and the province accounted for almost 40 percent of China's total exports (and certainly a much higher proportion of manufacturing exports) in 1992. Hong Kong's involvement in export-oriented production in China is not just limited to joint ventures. A lot of Guangdong's export production is supervised under contract by partners in Hong Kong. Much of Gluangdong's success with processing using supplied or imported materials is based on partnerships with counterpnrts in Hong Kong. It is no wonder that Guangdong recorded $16.6 billion in exports from processing activity or about 70 percent of the country wide total for such exports in 1991.' All evidence suggests that the economies of Hong Kong and Guangdong are becoming increasingly integrated with one another. As exports of certain products such as toys and clothing, from Guangdong have increased, those of similar products from Hong Kong have declined.' This suggests that production of a variety of low value products has been displaced from Hong Kong to the hinterland, where wages are lower and the labor force is reasonably skilled. This has enabled resources in the very tight labor market in Hong Kong to be reallocated to higher value products such as office machines, for example, the share of which in its domestic exports has been rising steadily. Clearly, the growing interdependence between Guangdong and Hong Kong is working to their mutual benefit, and to the benefit, of course, of China's overall export performance. D. TRENDS IN CxA'S MERCHANDISE IMPORTS The Structure of Merchandise hIports The story of China's merchandise imports since the "open-door" policy is much simpler than that of its merchandise exports. Three tendencies are noticeable (Table 1.8). First, the share of foodstuffs in China's imports has seen a marked decline-compared to shares of between 16 and 23 percent in the early 1980s, foodstuffs now account for around 6 percent of total imports. Second, the share of intermediates and raw materials, although smaller now than in the early 1980s, appears to have stabilized at a little over a third of total imports. And third, capital goods (machinery and transport equipment) have seen a steady increase in their shares. -Capital goods now account for roughly 43 percent of total imports, compared to 19 percent at the start of the reforms. The share of mineral fuels has ink-eased somewhat, while that of consumer goods did not exceed 5 percent at any time during the decade of the 1980s. 17/ Prybala (1992), reported in Yusuf (1992) estimates that about half of China's total exports originate in the nonstate sector.  - 16- Table 1.8: CmNA: STRUCTURAL CHANGE IN CIUNA'S IMORTS (CIJ CUSrOMS BASIS (% share to Total) Commodity 1984 1987 1988 1989 1990 Food 9.8 7.2 7.7 9.1 8.6 Mineral fuels 0.5 1.2 1.4 2.8 2.4 Intermediate 53.8 43.2 46.1 43.1 39.4 of which: Chemicals 16.6 11.8 16.7 13.1 12.9 Crude materials 10.0 7.8 9.6 8.4 7.6 Iron and steel 17.1 11.1 8.4 9.8 5.3 Consumer goods 2.3 4.5 4.1 3.7 4.7 Capital goods 33.5 43.5 40.1 41.0 44.3 of which: Transport equipment 7.7 9.4 8.7 9.0 11.2 Miscellaneous 0.0 0.4 0.6 0.4 0.5 Source: Chinese Trade Data (SlTC Revision 1), Customs Directorate. In the earlier years of reform, the import plan explained much of China's import struture. The plan has been used, by and large, to ensure minimum imports of key foodstuffs, intermediate goods and raw materials.l8/ The products most subject to plinning have been typically ones with the largest domestic price distortions. Quantities to be imported have been determined through a gap-filling" exercise, rather than on the basis of relative prices or quality. The importance of the plan has been declining, however. Less than a third of imports are now subject to mandatory planning or to canalization, i.e., the practice of restricting the imports of such commodities to a few designated foreign trade corporations (see Chapter 2). Whereas, in 1984, over 80 percent of the imports of food and intermediate goods came under the plan, by 1991, only half of food imports and only two thirds of intermediate good imports, respectively, were comprised of planned commodities. Within the category of foodstuffs, cereals remain by far the most important planned commodity. Chemical fertilizers, plastic materials (mainly plastic sheeting), textile yarn, iron and steel and wood are the largest intermediate goods imports that remain subject to import planning. Meat and dairy products, along with animal feed, account for the bulk of the growth in nonplanned commodities in the food group. Within the category of intermediate goods, chemical elements and paper products have underpinned the rising share of ncnplanned imports. This changing composition of imports suggests that an increasing proportion of food and raw material imports has become market driven. Although the share of import planning has declined progressively, China's import structure still bears the mark of management. Thus, although capital goods are not part of the import plan, the steady rise in their import share is due in large part to a systematic IS! Almost none of the plannedlcanalized commodities fa1 into the categoq of consumer or capital goods. Tlle only exceptions are televisions, cathode ray tubes and diesel engines.  - 17 - government strategy of using imports as a way of importing embodied technology for modemization. Over the last decade, China has imported the equivalent of about 3 percent of GDP in capital goods, accounting for about 7.5 of all investment, and about 15 percent of all investment in equipment.'0 The growth in such imports reflects, more than anything else, trends in directed domestic investment, and has been supported by controlled access to foreign exchange. The central government still directly controls over 50 percent of the country's foreign exchange earnings, which are allocated to planned imports and imports of materials and equipment needed for priority projects (Chapter 2). The inconsequential share of consumer goods in China's total imports also suggests the use of administrative controls and foreign exchange allocation. Consumer goods have to a significant degree been the "residual" component of imports, their share falling in years of retrenchment and rising in years of relaxed controls. The managed nature of China's import structure is reflected quite clearly in recent shifts in import shares. Over the 1985187 period, which was a period of retrenchment following the build up of domestic excess demand, the share of consumer goods fell from 5.5 to 4 percent of imports, while that of capital goods and critical commodities subject to import planning, such as cereals, petroleum and fertlizers, and textile yarn was allowed to rise. Again, the immediate effect of the retrenchment of 1989 was to restrain the growth in consumer goods imports, while the share of cereals, textile yarn, iron and steel and machinery and transport equipment went up. China's import structure thus clearly reflects a conscious import strategy, which has been to ensure the supply of key raw materials and to acquire embodied technology through the import of capital goods, while imports of consumer goods have been regarded as a residual. Overall. however, China's import structure resembles that of its East Asian neighbors, except that its share of food imports is somewhat larger. What is significant is the relative openmess of China, like Korea, Taiwan (China) and Malaysia to the imports of capital goods. In fact, the share of capital goods in China's imports is not excessive by the standards of these other economies. On the other hand, this is what sets China and the East Asia NIEs apart from other large developing countries such as India and Brazil, where the share of capital goods imports is significantly smaller. Fmally, a word about the origin of China's imports. As in the case of exports, Hong Kong plays an important role of intermediary in the procurement of China's imports. In 1990, 27 percent of China's imports were recorded as having come from Hong Kong, when in fact almost all these were reexports from other places of origin. A large proportion (45 percent) of Chinese imports coming in as reexports through Hong Kong are from Taiwan (China) and Japan. Until the normalization of diplomatic relations earlier this year, China's imports f.om South Korea were also handled in this manner. Taking account of the country of origin fbr products reexported from Hong Kong, Japan emerges as the single largest supplier of China's imports accounting for one fifth of total imports (Table 1.11). In this regard China is hardly different from other East Asian economies although overall, China's dependence on Japanese imports is markedly smaller than for the other countries of the region and appears to be declining fast. The second largest supplier of Chinese imports is the United States which has had a more or less stable 15-percent share since 1984. The US share of China's imports is higher than in other East Asian countries, such as Thailand, and Indonesia, but it remains substantially lower than in Taiwan (China) and Korea. The EC is the only bloc of member countries of the Organization of Economic Cooperation and Development (OECD) countries that has posted even modest gains in its share of China's imports. Between 1984 and 1990, the EC's share of  - 18- Table 1.9: STRUCTURE OF CIUNA'S IMPORTS (CIF) CUSTOMS BASIS (A Comparison with Other Importers for 1990) (%) Commodity China Korea Taiwan Brazil India Malaysia (China) Food 8.6 5.2 5.5 9.3 3.2 6.8 Food 6.5 4.7 4.7 8.7 2.4 6.1 Beverages 0.3 0.3 0.6 0.2 0.0 0.4 Animal fat 1.8 0.3 0.2 0.3 0.8 0.3 Petroleum (mineral fuels) 2.4 15.8 11.0 26.8 27.3 5.3 Intermediate 39.4 36.4 34.6 27.3 40.4 24.3 Chemicals and related products 12.9 10.7 13.0 15.9 12.9 8.8 Crude materials (nonfood) 7.6 12.4 8.2 5.3 9.4 3.4 Leather 0.7 1.1 0.5 0.9 0.3 0.1 Cork 1.1 0.6 0.6 0.1 0.0 0.1 Textile yarn (yarn, fabrics, etc.) 9.9 2.8 1.9 1.1 1.0 3.4 Nonmetallic minerals 0.8 1.3 1.2 0.7 9.1 1.1 liron and steel 5.3 4.7 5.4 1[3 4.9 5.3 Nonferrous metals 1.1 2.8 3.8 1.9 2.6 2.1 Consumer Goods 4.7 3.4 5.3 4.7 2.3 6.3 Paper and related products 1.4 0.6 1.2 1.2 1.0 2.0 Rubber 0.1 0.3 0.3 0.5 0.2 0.3 Furniture 0.1 0.2 0.2 0.0 0.0 0.2 Travel goods 0.0 0.0 0.1 0.0 0.0 0.1 Clothing 0.1 0.2 0.5 0.3 0.0 0.3 Footwear 0.0 0.0 0.1 0.1 0.0 0.1 Photo supplies 0.0 0.0 0.0 0.0 0.0 0.0 Miscellaneous 2.9 2.1 2.7 2.5 L10 3.5 Capital goods 44.3 38.8 41.2 31.9 20.5 57.1 of which: Transport equipment 11.2 - 3.9 7.2 3.5 3.9 10.0 Miscellaneous 0.5 0.4 2.4 0.0 6.2 0.2 Total 100.0 100.0 100.0 100.0 100.0 100.0 Source: United Nations COMTRADE Data (SlTC Revision 1).  - 19 - Table 1.10: ORICIN OF CmNA's IMPORTS: A CoMPARISoN wrmi O0n EcoNomms oF EAs AsA SELECED YEARS (Tn Year Thailand Indonesia Taiwan Korea China In Origin of Share of Share of Share of Share of Shar of Sham of irports imports imports imnports imports imports impots (adjusted for (unadjusted ror rcaxports thu rexpoits thru Hong Kong) Hong Kong) 1934 Japan 27 24 29 25 35 31 USA 13 18 23 22 16 15 EC 13 15 9 9 13 13 Other 47 42 39 44 29 41 of which: Hong Kong - - - - - 12 1990 Japan 31 24 30 27 21 14 USA 11 12 24 24 15 12 EC 15 19 13 12 15 15 Other 44 46 33 37 49 59 of wbich: Hong Kong - - - - 4 27 /a Discrpancies between thesc and figures rcported by China possible because of transport margins and period differcut information reporting lags. Source China's Customs Statistcs, Hong Kong Rcview of Overseas Tsde, various years, and United Nations COMTRADB data base. 1984 1990 Memo Iems Value of China's imports from Hong Kong (S billion as reported by China) 3.1 14.4 Value of Hong Kong's reexports to China from other places (S billion as rcported 3.3 12.4 by Hong Kong, of which: Share of reexports originating in Taiwan (China) (%) 13 23 Sharc of reweorts originating in Japan (%) 36 22 Share of rexpots originating in USA (L)e 11 9 China's imports went from 13 to 15 percent. It appears that the more advanced East Asian economies are the ones that have made the largest inroads into China's market by exporting through Hong Kong. The joint share of Taiwan (China) and South Korea in China's imports went from an estimated 4 percent in 1984 to over 9 percent in 1990.  - 20 - Table 1.11: SIARE OF MERCHANDISE TRADE IN GDP: SELECTED COUNTRIES (%$) China 31 USA 16 Japan 18 India 14 Brazil 12 Mexico 36 Turkey 34 Source: Data for China are for 1991 (see World Bank, 1992c). Data for other countries are from the General Agreement on Tariffs and Trade (GAIT) (1992b), Volume 1, Appendix Table 5. E. TRADE IN CHNA'S EcONOMy: SOME CONCLuDING OBSERVAnONS 'The trends in exports and imports discussed above illustrate how far China's approach to trade has evolved since the open door policy was first announced. Traditionally, China viewed exports as a surplus over domestic production, to be sold in order to generate financing for the purchase of commodities, such as grain, in short supply at home, and of imported technology for purposes of modernization. This philosophy of 'gap filling led to an autarkic approach to trade, with the handful of foreign trade corporations (FTCs) granted permission to carry out external trade operating an 'airlockl between the outside world and the domestic economy."1 Since then, the number of FTCs has gone from 12 to a peak of 6,500 in 1989, and 3,600 at the present time. From no fbreign investment in 1978, the total has risen to over 29,000 joint ventures with a contracted value of $45 billion. And most remarkably, the share of merchandise in China's Gross Domestic Product (GDP) went from 10 percent in 1978 to 31 percent in 1991.19/ This is a significant and permanent change in the face of the Chinese economy. As compared to other large countries, China has become a very open economy (Table 1.l)-by this measure, China appears to be more than twice as open as India and Brazil, and significantly more open than USA or Japan. China now exports about 17 percent of the gross value of industrial output (GVIO) of its overall manufacturing sector, and imports an estimated 28 percent of the GVIO of its machinery and transport equipment sector (Table 1.12). 19! This figure needs to be treated with caution. First, China's Gross National Product (GNP) is widely regarded as underestimated. Second, customs statistics on exports include the funl value of exports based on processing of imported inputs, which tends to exaggerate the role of trade in the Chinese economy. Excluding the latter would reduce the value of China's exports in 1990 by about $11.? billion, and the trade to GDP share to about 28 percent. Depending upon what estimate of China's GDP is taken, the trade to GDP share could be anywhere betwee 18 and 26 percenL  - 21 - Table 1.12: CIuNA: THE STRUCTURE OF PRODUCTION, IMPORTS AND ExpoRrs, 1985 AND 1990 Share in total Imports/GVIO Exports/GVIO Sector GVIO(%) L(% . L%) 1985 1990 1985 1990 1985 1990 Food, beverages 11.6 11.6 5.0 8.0 10.6 14.7 & tobacco Raw materials 20.8 20.9 8.2 8.4 15.4 10.8 Manufactures 67.5 67.5 16.6 16.2 5.0 16.9 of which: Intermediates 39.4 40.2 12.6 11.3 4.9 10.2 Machinery & transport 20.0 19.2 28.0 28.5 1.3 14.4 Consumer goods 8.1 7.9 8.0 10.8 14.6 56.9 Memo Items Light Industry 49.1 47.0 Heavy Industry 50.9 53.0 Clothing 2.0 2.1 0.2 0.6 34.9 118.0/a Travel goods 0.1 0.1 0.9 1.5 27.0 90.5 Footwear 1.2 1.2 0.2 0.2 7.3 41.3 /a An export share of greater than 100 percent indicates large export volumes from outside the independent accounting sector of the Chinese economy. Source: Based on Table A3.3. GVIO shares pertain to output of firms with independent accounting only. These firms accounted for 77.8 percent of total GVIO in 1990. Data on light and heavy industry are from Chin's Statistical Yearbook, various years. On the one hand, the growing openness of the Chinese economy has had a perceptible impact on the quality of a range of Chinese products that have benefited from the use of updated technologies and exposure to the demands of consumers in external markets. On the other hand, it is also noteworthy how little the structure of the China's industry has changed, despite a non-negligible degree of import penetration and high export ratios. Table 1.12 compares the sectoral share of GVIO between 1985 and 1990 and the import penetration and export ratios of each board sector over the same period. It indicates that, while import penetration ratios have remained more or less stable across sectors, the share of exports in (;iVIO has grown more than threefold in the case of manufacturing in general, and almost fourfold in the case of consumer manufactures. Despite these very substantial increases in export ratios, however, the overall sectoral composition of output, including the share of consumer manufactures (with such heavily export-oriented sectors as clothing, footwear and travel goods) has remained virtually unchanged. While it is probable that capital-output ratios in the more export-oriented sectors have fallen during this period, it appears that the allocaion of investnent  - 22 - has remained unaffected by the growing openness of the economy. Investment is evidently still heavily directed, resulting in a "planing' phenomenon, such that investable rescurces are divided more or less equally amongst all sectors instead of being channelled to the most efficient ones.2Q/ As open as the Chinese economy is today, trade still does not yet play a sufficient role in domestic resource allocation. An important challenge for the future will be to allow a stronger link to be established between trade and investment. 20/ See I. J. Singh (1992).  - 23 - Endnotes 1. GATT (1992b). The figures for shares in world trade include exports, imports and reexports and the trade of Eastern European countries. 2. World Bank (1992e). 3. MOFERT's report on the reform of the trade system was approve& by the State Council on September 15, 1984. The system of foreign exchange retention, though first introduced on an experimental basis in 1979 was formalized in Januay 1984, and was subsequently modified in January 1985 such that te retention rights were shared eqially between the exporting enterprise and the province. See World Bank (1987). 4. FBI annual utilization, sluggish until then, doubled from $0.6 billion in 1982 to $1.3 billion in 1983, and grew at an average rate of over 34 percent per amum thereafter till 1989. See Khan (1991). 5. Hirschmann index normalized to make values ranging from 0 to 1 (maximum concenration), according to the following fonnula: where: j = country index; 239 xj = valueofexportsofcommodityi; x x, and 239 = number of products the three-digit SITC, Revision 2 level. 6. Almanac of China's Foreign Economic Relations 1991/92, page 88. 7. See Yun-Wing Sung (1991). S. Almanac of China's Foreign Economic Relations 1991/92, page 88. 9. While Hong Kong's domestic exports of toys declined from $1.5 to $1.3 billion between 1985 and 1988, reexport of toys, almost all originating in China, rose from $0.6 billion to over $2.1 billion over the same period. Similarly, from 1986 to 1988, while the share of clothing in Hong Kong's domestic exports declined from over 40 perceat to 31 percent, reexports of garments from China almost doubled, rising from $1.6 billion to $3.1 billion over the same period. 10. World Bank (1992c). 11. See World Bank (1987a).  - 24 - II. THE TRADE PLANNING AND FOREIGN EXCHANGE SYSTEMS-REFORMS THROUGH A PERIOD OF TRANSITION Until recendy, the fundamental feature of the Chinese foreign trade system was the planning mechanism used to control exports and imports and to allocate foreign exchange to priority uses. Since the launching of the open-door policies, this mechanism has undergone significant change. In general, the thrust of successive waves of system reform has been to reduce the importance of the plan. Although remarkable in its impact on China's foreign trade, the process has not been an entirely smooth one. Numerous difficulties have arisen along the way, most stemming from the unique circumstances of China's transition from a rigidly planned economy to a more market-oriented one. Since the mid-1980s the Chinese authorities have launched two major reforms of their foreign trade and exchange planning system, one in 1988 and the second in 1991. The basic objectives of the two major reforms were similar. They were designed to increase the role of the market in determining the pattern of imports and exports and the use of foreign exchange, and to reduce the burden of financial subsidies imposed by the external sector on the state budget. This chapter takes stock of the reforms of the trade planning and foreign exchange management system that have occurred since 1986 and makes recommendations for the future. A. FOREIGN TRADE PLANNING, FOREIGN TRADE CoRPoRTnoNs, AND SuBMEs Foreign trade planning has entailed not only formulating the export and import plans, but also defuiing the role of China's foreign trade corporations that have been used as an institutional vehicle for implementing the plans. The original objective of trade planning was to identify the key raw materials and commwdities that were is short supply and had therefore to be imported, and then ensure that sufficient foreign exchange was generated through selected exports-the plan was thus driven by the country's import requirements. In 1978, all trade was handled by a dozen FTCs and their branches, which had the responsibility of executing the plan. Following approval of the reform plan of the Ministry o: Foreign Economic Relations and Trade (MOFERT), now called the Ministry of Foreign Trade and Economic Cooperation (MOFTEC) in 1984, the foreign trade system was decentralized very considerably, with the provincial branches of national FTCs allowed to become independlent financial and operating bodies, and each province allowed to create its own FTCs. By 1986 there were already 1,200 FTCs in operation. The foreign trade plan also became more export driven. A target for exports would be fixed, and the objective of the planning exercise became one of managing imports within the foreign exchange constraints implied by this target. The export plan was split into two components: the command plan and the guidance plan. The command plan was mandatory, fixed in quantitative terms, applied to specific products, and was accompanied by an assured supply of necessary inputs to the producing enterprises. In contrast,  - 25 - the guidance plan contained value targets assigned to provincial authorities, which were accorded considerable flexibility in determining how to achieve them. Products subject to the command export plan were in turn split into two lists: the list of so-called Category I exports, comprising products that could be handled only by a few designated national FTCs, and Category II exports that could be handled by a wider range of FTCs, including local and provincial ones. An estimated 60 percent of exports were subject to the mandatory export plan and 20 percent to the guidance plan in 1986, with the balance being "above plan" exports. The import plan was comprised of three components: a mandatory plan for key raw materials, the steady supply of which was considered essential-these imports were to be handled only by designated national and/or provincial FTCs (see also Chapter 3); a system of foreign exchange allocation for imported raw materials and spare parts for key established national projects, and for the import component of priority new investment projects .1; and an import licensing system (see also Chapter 3). As in the case of exports, commodities subject to the mandatory import plan were divided into two lists: Category I imports that could only be handled by a few designated national FTCs; and Category I[ imports that could be handled by several FTCs, including local and provincial ones. In 1986, an estimated 40 percent of imports fell under the mandatory import plan, and another 30 percent were procured under the foreign exchange allocation mechanism for priority projects-the remaining 30 percent, financed either out of retained foreign exchange or foreign borrowing by noncentral authorities, were subject to selective import licensing.? All procurement for the mandatory export and import plans took place at fixed prices. For the export guidance plan, the price could be fixed, floating or free market, depending on the how each province decided to meet the value targets of the guidance plan,21 and on the negotiating position of producing firms vis-a-vis the procuring FTCs. Given that all export transactions still had to be routed through FTCs, which in many cases were designated, the link between international prices and the procurement price offered to exporters was still not systematic. On theother hand, such a link was much better established for nonmandatory import transactions. The widespread use of the so called agency system fcr irnports meant that importers were free to choose any FTC for procuring nonmandatory imports and would pay the import price plus the FTC's costs. Under such a system, with pervasive domestic price distortions on the one hand, and FITCs bound by obligations of the mandatory export and import plans on the other, domestic currency losses on some international transactions were inevitable. FTCs incurred financial losses measured in domestic currency each time they were required, under the import plan, to procure imports of such products as food grains and chemical fertilizers, which were purchased at international market prices, but then had to be sold on the domestic market at the state fixed price applicable to domestic producers of the same products. Likewise, they incurred losses when the export plan required them to purchase relatively high priced domestic goods, such as certain types of machinery and electronic products, and then to sell them on the international market.2 In 1986 direct fiscal subsidies to foreign trade companies to cover their losses were 1/ This included imports of complete plant for industrial expansion, and centrally funded general investment projects. 2J Some provinces essentially converted the guidance plan into a provincial command plan with fixed quantities and prices.  - 26 - more than Y 24 billion, more than two percent of China's gross national product and larger than the size of the official budget deficit. By 19B8, total fiscal subsidies to FTCs had reached an annual level of over Y 26 billion C(able 2.1). Table 2.1: LOSSES OF FOREIGN TRADE CORPORATIONS FINANCED BY CENTRAL GOVERNMENT BuDGEr (Y 100 million) Year 1986 1987 1988 1989 1990 1991 Losses 249.6 282.1 268.5 336.4 224.4 176.1 Memo Item Total losses of SOEs within budget 417.1 481.7 520.6 749.6 932.6 931.1 Source: Industrial and Commercial Deparanent, Ministry of Finance. The 1988 and 1991 Reforms The Foreign Trade Contract. The key institutional feature of the 1988 refonn of the foreign trade system was the contract system. Every provincial level administrative unit and all specialized national FTCs signed contracts with MOFERT. These specified three targets: the amount of foreign exchange earnings; the amount of foreign exchange to be remitted to the central government; and a fixed amount of domestic currency that the center would provide to subsidize losses on export sales. MOFERT, the Ministry of Finance, and the State Planning Commission jointly determined the values of each of these targets and the State Council approved these numbers before they were incorporated in the signed contracts? The contract was the key policy instrument the central government used to control indirectly the increasing amount of decentralized trade that was to be undertaken by local govermment while, at the same time, controlling the magnitude of export losses for which the central government was responsible! Starting in 1991, the foreign trade contracting system was modified so that targets for the value of exports, foreign exchange earnings, and foreign exchange remitted to the central monetary authorities are now set annually, rather than every three years.af Also, the process of setting targets, though still involving some negotiation, has beccme 'bottom up." Typically targets are proposed by enterprises and local authorities bas'd conservatively on growth trends from previous years. 3/ In the case of conuacts between the cenzer and national FTCs spealized in Category I exports, quantitative targets are still specified.  - 27 - Fiscal Subsidies. One of the important features of the foreign trade contracts introduced in 1988 was its attempted limitation of the size of the fiscal subsidies that the central government would provide to offset domestic currency losses incurred on the sale of exports. Nationally the aggregate amount of export subsidies was fixed at an amount equal to four percent of the value of exports in 1988 or around Y 7 billion. This total was divided among all of the contracts signed so that each provincial level government received a fixed amount of export subsidies. Not only did the government try to put a cap on export-related subsidies, but in 1991, it went further and eliminated such subsidies altogether. The key provision of the 1991 reform made all specialized national foreign trade companies and all provincial-level administrative units responsible for their own domestic currency profits and losses, at least on exports. Central government fiscal subsidies for money-losing exports were said to be cut to zero beginning in 1991. In parallel, the central govermnent has also moved to reduce the burden of fiscal subsidies to offset money-losing import transactions. If the cap of Y 7 billion on export subsidies was in fact respected in 1988, Table 2.1 suggests that central government subsidies for import transactions amounted to Y 20 billion. Efforts to reduce this burden focussed on price increases for domestic products that historically had been sold at low prices. During the course of 1989 the state raised the domestic prices of steel, nonferrous metals and several other products to near world market prices. And the procedure for fixing the domestic prices of imports of these goods was changed. Beginning in 1989, they began to be priced like noncommand impoas, i.e., on an agency basis, meaning that the world price converted to domestic currency at the official exchange rate formed the basis of the domnestic price. To this the authorities added import duties, port fees and other costs. The reform meant that the domestic prices of these goods would change in response to changes in either world market prices or the exchange rate. Previously domestic users of these imported products were fully insulated from these changes. These changes reduced the value of import subsidies paid by the central government by Y 2.5 billion in 1989. Overall, however, subsidies on import transactions went up substantially in 1989 owing to the devaluation of the renminbi which aggravated rTCs losses on the imports of important items such as fooc grains for which no price adjustments had been made at that tirne. This presumably explains why total subsidies to FTCs reached a record Y 33 billion in that year. In April 1990, the domestic prices of six other imported goods, soda ash, caustic soda, aniline, cattle hides, tallow, and coconut oil were raised and imports of these goods were priced to fully reflect international market prices. That action reduced the value of import subsidies paid by the central government by Y 270 million. . Action was also taken on China's single largest money-losing import product-food grains.5 In the spring of 1990, the internal distribution price of imported grain increased by Y 30 per ton. The state followed this up in the spring of 1991 by raising the retail prices of edible vegetable oils, flour, and rice for the first time in 25 years. The increases were large. The average price paid by urban residents for wheat flour, rice, and corn rose by 0.2 yuan per kilogram or 71 percent The average price for edible vegetable oil (peanut oil, sesame oil, rapeseed oil, refined cottonseed oil, and soybean oil) rose by even more-Y 2.70 per kilogram or 160 percent. The main effect was to reduce the losses incurred on the retail sale of domestically produced output. But the reduction of subsidies on imported food grains and vegetable oils in 1991 was also significant-Y 1.27 billion. Even  -28 - so, financial losses on import transactions are far from eliminated. The internal distribution price of food grains still remains below the import cost plus handling and transportation costs, and although overall subsidies to FTCs have declined substantially from their peak in 1989, they still amounted to Y 17.6 billion in 1991. If all export subsidies were indeed eliminated as claimed starting January 1991, all this Y 17.6 billion can be assumed to be on account of subsidies on import transactions. The Export and Import Plans. Perhaps the most important feature of the reform introduced in 1988 was a reduction in the importance of the foreign trade plan. Ihe number of so-called Category I export products that were subject to mandatory planning was reduced to 21, roughly half the number previously falling in that category.' Category II exports, comprised an additional 91 commodities, down from 120 in 1986. The reform initiatives of 1991 went further in this area and it seems that all mandatory export planning has been abolished. However, the state still retains some control through the continued use of canalization and licensing. Exports that used to be subject to mandatory planning are still channelled through a few designated FTCs if they are classified as Category I or II exports. In the first quarter of 1992 about 15 percent of China's exports were either Category I or Category II. The import plan has also been scaled down since 1988, although not to the same extent as the export plan. The number of import products in Category I, largely under the control of specialized national FICs, has been cut almost in half, and was down to 14 in 1992. Only an additional six imports now fall in Category I. In the first quarter of 1992, mandatory planned imports constituted only 18.5 percent of all imports compared to 40 percent in 1986, and covered 11 broad product categories.7 The process of scaling down the plan has been accompanied by the decentralization of responsibility for implementing the plan and an expanding number of foreign trade companies, mostly at the local level. The number of FTCs allowed to handle Category II imports and exports has grown over the years as has the total number of FTCs authorized to undertake trading activity. From about 800 in 1986, the number of FTCs has increased to over 3,600 FTCs in operation, more than 300 times the number existing at the start of the reforms. B. THE EXCHANGE RATE REGIME From 1981 until 1984, China had dual exchange rates: official and secondary. The official rate depreciated gradually under a system of managed floating while the secondary rate was fixed at a more depreciated rate. The secondary rate, termed the internal settlement rate, was used for settlement of payments between FTCs and the supplying enterprises. In January 1985, the official exchange rate was set at the internal settlement rate, and the latter was abolished. However, dual exchange rates reappeared with the establishment of foreign exchange adjustment or "swap" centers (FEACs) in late 1986. At present, the administered official rate is used for foreign trade and other extemal transactions included in the annual areign exchange plan, including probably, debt transactions. A second, more depreciated rate is determined in FEACs, where enterprises are pennitted -t buy and sell foreign exchange as well as retention quotas which can be used to acquire foreign exchange at the official rate to finance primarily trade transactions not included in the plan. Between July 5, 1986 and December 15, 1989, the official exchange rate was pegged de facto at Y 3.72 per US dollar, leading to a real appreciation of the yuan in the face  . 29 - of rising inflation. At the same time, the exchange rate in the FEACs depreciated, allowing exporting enterprises to maintain their profitability. The devaluation of the official rate by 21.2 percent (to Y 4.72 per US dollar) in December 1989 and by a further 9.6 percent in November 1990 returned the real exchange rate to the level prevailing in 1986; the nominal bilateral rate was maintained at Y a.22 per US dollar from November 1990 till April 1991. On April 9, 1991, a new system of a managed float was adopted, under which the administered official rate is adjusted frequently through small periodic changes based upon several factors including: (a) developments in the balance of payments, (b) developments in foreign currency markets, (c) developments in FEACs, and (d) changes in the domestic resource cost of earning foreign exchange. Although the rate has been adjusted in both directions, it has tended to depreciate and, as of end-March 1993, had reached to Y 5.73 per US dollar. In the FEACs, the average exchange rate depreciated by about 21 percent to a peak of Y 6.7 per US dollar in early 1989. It appreciated to about Y 5.9 per US dollar in December 1989 in the wake of the devaluation of the official rate. Since then it appreciated slightly to Y 5.7 per US dollar in December 1990, before depreciating to about Y 6.9 per US dollar towards end-September 1992, and further to Y 8.4 per US dollar, by the end of the first quarter of 1993. Thus, as of end-March 1993, the spread between the two exchange rates had once again become substantial and stood at over 45 percent. Foreign Exchange Plan The annual plan for foreign exchange receipts and expenditures has been, and continues to be, formulated by the State Planning Commission (SPC) and approved by the State Council. The State Administration of Exchange Commission (SAEC) is responsible for supervising the implementation of the plan, which has traditionally taken explicit account of the foreign exchange implications of the mandatory export and import plans for trade in key products. With the elimination of mandatory export planning in 1991, the foreign exchange plan now takes account of the foreign exchange targets negotiated with each province in the context of the foreign trade contracting system. Importers of goods included in the import plan are allotted quota accounts by the SAEC and a quota notice form that is required by MOFERT before import licenses are issued. When payment for the import is due, foreign exchange is made available to the importer upon surrender of local currency against the quota account in the SAEC. In addition, the foreign exchange plan is used to allocate foreign exchange to finance part of the import requirement of priority investment projects. Ihis would appear to be redundant given that funding for key projects and their associated imports is already allocated as part of the government's investment program. Retention Scheme China traditionally combined an inconvertible currency with a rigid system of exchange control requiring all exporters to turn over all of their foreign exchange receipts to a specialized bank, the Bank of China, in exchange for domestic currency. Exporters thus were left with no foreign exchange to finance their imports. Like any other would-be importer they had to depend on the State Planning Commission, which allocated all foreign exchange earnings via an annual import plan. The government began to decentralize the administration of foreign exchange earnings in 1979 by allowing local authorities, departments, and enterprises to retain  - 30- the rights to buy back a certain proportion of their foreign exchange earnings.4/ Initially, retention quotas,' which are transferable between enterprises, were transacted at the administered exchange rate, but by 1988, all MIEs and domestic enterprises with retention quotas were permitted to operate in the FEACs. Retained foreign exchange can be sold at the FEACs for renminbi or used to purchase imports. Retention quotas can be traded in the foreign exchange centers or used for acquiring foreign exchange at the official rate to purchase approved imports. Because of the premia that swap market rates fetch over the official exchange rate, the right to retain foreign exchange has constituted an important financial inducement for beneficiary enterprises. The original retention rates, or the proportion of its foreign exchange earnings that enterprises can retain as their quota, were relatively low. In 1984/85, local authorities and enterprises retained rights to only 25 percent of their planned export earnings while the remaining 75 percent went to the central authorities. Over time, the government has adapted the foreign exchange retention system to fiuther its industrial policy objectives. Beginning in early 1988, corporations trading in priority sectors-light industries, arts and crafts, clothing, machinery, and electrical products 5/-were permitted to retain oDnsiderably higher proportions (between 70 and 100 percent) of foreign exchange retention quotas in order to give greater incentives to those sectors.9 A higher rate of retention was also allowed on foreign exchange eanings above planned targets.6/ Lilewise, special higher retention rates applied to cerin provinces such as Guangdong and Fujian, several of the autonomous regions populated by minodty peoples, and the SEZs, the latter enjoying retention rates of up to 100 percent. Although the specific afrangements varied from locality to locality and depended on the type of good involved, retained foreign exchange was in general split evenly between the FTC handling an export transaction and the firm producing the good. In the case of general commodities exported from ordinary provinces, for example, of the 25 percent of foreign exchange earnings that could be retained, 12.5 percent would normally go to the FTCs and the remaining 12.5 percent to the producing enterprise. Several important modifications were made to the retention scheme in 1991. First, the retention system, as it had evolved, had given rise to considerable distortions and provided an unfair competitive advantage to some coastal provinces. Thus, in February 1991, a uniform retention rate was set throughout the counry.7/ For general commoditiesj/ this rate 4t The fiul amotmt of the freign exchange eceipts was required to be surredered to the state at the official rte by exportrs. 5I In the case of light industry, arts and cafts, and clothing this provision seems to have been a quid pro quo for them having taken on responsibility for their own profits and losses. 6/ These disappeared in 1991, with the dimination of the mandatory export plan. 7, Tibet, the only exception, was allowed to retain a 100 percent retention quota for its modest expors- 81 For commodities such as crude oil and petroleum deNivatives which are in effect monopolies of the State, retention rates are set very low (less than 5 percent).  - 31 - was set at 80 percent. Of the 80 percent, 10 percent accrues to local government, 10 percent to the producing enterprises and the remaining 60 percent to the foreign trade corporation.2/ Second, special rates for certain sectors were retained and adjusted upwards. For machinery and electronics products, the retention rate is now 100 percent of which 10 percent accrues to the producing enterprises and 90 percent to the foreig,' trade corporation. FIEs and joint ventures can still retain 100 percent of their export earnings in foreign currency accounts in resident banks. The retained share of new foreign exchange earned in processing activities was pegged at 90 percent, up significantly from the 30 percent rate prevailing in 1985. Third, although retention ratios were raised across the board, the central govenment ensured access to foreign exchange sufficient to meet its own requirements by reserving the right to purchase, albeit at the prevailing FEAC rate, an additional 30 percent of the foreign exchange initially retained-20 additional percent points from the FTCs and 10 from export producers. All indications are that the center fully exercised this option in 1991. In effect, therefore, 50 percent of the overall foreign exchange earnings of the local authorities and enterprises was still appropriated by the central government, and producers of export goods were left with no retained foreign exchange (T.'le 2.2) with which to directly buy nonplan imports. If a firm producing export goods want xd to purchase imports outside the plan it had to seek access to the FEACs in order to do so. 10I Finally, the authorities have started experimenting with a cash retention system on a limited scale in several centers around the country. In such a system, the enterprise would only be required to surrender a portion of their foreign exchange earnings to the SAEC and would bc allowed to keep a portion of their foreign exchange earnings in resident bank accounts. Foreign Exchange Markets As early as 1980, in some localities, companies with excess retained fbreign exchange were allowed to sell it to other firms who sought access to foreign exchange to purchase imports outside the plan. Initially the volume of transactions in these local secondary foreign exchange markets was limited, perhaps in part because the state sought to constrain the price at which these swaps occurred to the official exchange rate which highly overvalued the domestic currency. More regularized markets formally opened in several cities in 1985. However, following a rndown in its foreign exchange reserves in 1984, the central govermnent effectively froze the cumulative foreign exchange retention rights of exporters and used the underlying foreign exchange to finance its own imports. As a result, the volume of transactions on the swap market was limited, and the major participants were FlEs which from the outset had been allowed to retain one hundred percent of their foreign exchange earnings from exporting. 9J In the case of those selected producing enterprises which have been granted direct trading rights (i.e., not required to deal through FTCs), the share of the foreign trade corporations accrues to the producing enterprises. This also applies to producing enterprises which export under the agency system. 10/ All of the foreign exchange sold to the State is eventually credited to the state reserves at the People's Bank of China (PBC), although large amounts are in transit on the books of the Bank of China (BOC) (the amounts in transit on the books of the specialized banks are aggregated in the PBC data).  - 32 - Table 2.2: FoREicN EXCHANGE RETEnoN RATES, 1991 Pcrcent to cental Of which Percent Calegory governmcnt market rate retained /a Of which 1. Gcenral 50 30 50 8 provincial govt commoditics 2 municipal/local 40 foreign trade co II. Machinery & 30 30 70 5 provincial govt electronic products 65 foreign trade co and science and technology products m. Petroleum (including crude oil and rtemed petroleum products) Unified plan 96 0 4.0 2.7 local govt amounts 1.3 Sinochem Over plan amounts 30 30 70 WV. Fees from 10 0 90 10 provincial govt processing 45 foreign trade co contracts 35 cnterprisc is The distribution of retentions is that prevailing aftcr the central government exercises the right to purchase, at the swap market rate, some of the initil distnbutions of retine foreign ecchange. The 1988 reform linked expanded foreign exchange retention rights with a liberalized foreign exchange market. Liberalization of the FEACs was assured by several measures. First, quota controls on the utilization of retained foreign exchange, which had been imposed by the central government in 1985, were abolished beginning in 1988. Second, the number of authorized local foreign exchange markets increased. Each province, autonomous region and centrally administered city was authorized to establish at least one foreign exchange swap center within its territory. By end-December 1992, there were over 100 FEACs in operation. Third, a national foreign currency swap center was also to be established in Beijing to facilitate currency transactions between central government agencies and to facilitate transactions between local swap markets in different administrative jurisdictions. Finally, the swap markets were opened formally not only to foreign-invested enterprises but to sta-owned and collectively owned enterprises as well. These reforms led to significant increases in the volume of foreign currency transactions on secondary markets. Volume rose from $4.2 billion in 1987 to $6.3 billion in 1988, $8.6 billion in 1989, and then $25 billion in 1992. However, inter-market transactions appear still to be restricted, leading to nonnegligible differentials in rates across markets in different parts of the country.  - 33 - Criteria for Access to FEACs. Sales of foreign exchange at the swap rate have been virtually unrestricted since December 1991, when all domestic residents were allowed to start selling foreign exchange at the swap rate at designated branches of banks.l1/ Sales of foreign exchange consist of sales of actual foreign exchange mainly by PIEs and of foreign exchange retention quotas by Chinese enterprises (the actual foreign exchange having been surrendered to the state). By contrast, access to FEACs to purchase foreign exchange is subject to approval and is restricted mainly to enterprises which need foreign exchange either to service their foreign currency debt or to import goods which are not inconsistent with the industrial policy of the state. Purchases of actual foreign exchange are usually limited to sums needed by FIEs which, according to the termns of their contracts as approved by MOFERT, can be used for designated purposes including their own operating needs, debt repayment, and remittances. 121 Domestic enterprises, which are approved by MOFERT to import, can purchase retention quotas in the FEACs; the purchased quotas must be used within a six-month period to acquire foreign exchange from the state reserves at the prevailing official rate. SAEC authorization is dependent on conformity with the priority uses of foreign exchange as set out in current government regulations. The priority list (Annex 2. 1 reflects plan objectives regarding key commodities, and it provides for favorable treatment for exporting activity as well as for the acquisition of advanced technology. On the other hand, requests for foreign exchange to finance imports of consumer durables, luxury goods or goods judged to be speculative in nature are not permitted. Two points are essential to understanding the manner in which enterprises can have access to foreign exchange to conduct import operations outside the plan.lIN First, applications for foreign exchange are checked by the local MOFERT office, the Commission of Foreign Economic Relations and Trade (COFER1), for conformity with the purposes of the enterprise, as specified in its business license.j14/ Enterprises are allowed to import products that are not subject to import licensing, provided that COFERT verifies that they are within the scope of the enterprise's business license. Second, every enterprise authorized in this manner has a foreign exchange budget (or quota) assigned to it by the local SAEC based upon its export targets and expected demand for imports, debt service, and remittances. Each application for foreign exchange is also checked against the enterprise's foreign exchange quota before access to the FEAC is approved. Only after the documentation of the applicant is checked and an import license is obtained, where required, are buy orders processed or the applicant allowed to trade in the swap market. For applicants with retention quotas, the procedures are simpler, since only verification of the enterprise's foreign exchange retention quota is required. Il/ [ at least one local branch of the Bank of China, the rate at which domestic residents can sell their foreign exchange was fixed by the bank at about 3