Repot No. 6680-CH China External Trade and Capital Reform Issues and Options (In Two Volumes) Volume 1: Main Report October20, 1987 Asia Regional Office FOR OFFICIAL USE ONLY * oc' m ,e of the ,'brid Bank This report has a restrkcted dOstnibution and may be uised by recipients only in the perormnce of their official duties. its contents may not otherwise ibe disclosed without Vibrd Bank authorization. CURRENCY EQUIVALENTS The Chinese currency is called Renminbi (RMB). It is denominated in Yuan (Y). Each Yuan is 1 Yuan = 10 jiao = 100 fen Calendar 1985 December 1986 US$1.00 = Y 2.94 US$1.00 = Y 3.70 Y 1.00 = US$0.34 Y 1.00 = US$0.27 FISCAL YEAR January 1 - December 31 WEIGHTS AND MEASURES Metric System LIST OF ACRONYMS DL/C - Domestic letter of credit FTC - Foreign Trade Corporation GATT - General Agreement on Tariffs and Trade GDP - Gross domestic product GTC - General Trading Corporation ICOR - Incremental capital-output ratio LDC - Less developed country L/C - Letter of credit MFA - Multi-Fiber Arrangement MOFERT - Ministry of Foreign Economic Relations and Trade NIC - Newly industrialized country PBC - People's Bank of China SAEC - State Administration of Exchange Control FOR OMCIAL USE ONLY This report and iLs annexes are based on the findings of a mission that visited China during October 1986. In addition to interviews and discussions in Beijing, the mission visited Zhengzhou (Henan Provinc3), Shanghai, Fuzhou and Xiamen Special Economic Zone (Fujian Province). The mission consisted of Gerhard Pohl (mission leader), Peter Harrold (deputy mission leader, trade policy) and Joel Bergsman (foreign direct investment), Basil Kavalsky (foreign exchange allocation), Margaret Kelly (IMF, foreign exchange policy and debt management) Hyung-Ki Kim (technology transfer), Wouter Tims (primary commodity trade), Mieko Nishimizu (import policies), Marko Voljc (export policies), Christine Wallich (external capital flows and debt management) and Candy Wong (mission secretary). Research assistance in Washington was provided by Nicola Favia, Yasheng Huang, Lili Liu, Min Zhu and Umnay Sae-Hau. The mission was accompanied by a Chinese counterpart team from a number of government agencies, led by Zhu Ang, Deputy Director, International Trade Research Institute, Ministry of Foreign Economic Relations and Trade. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CHINA EYTERNAL TRADE AND CAPITAL REFORM ISSUES AND OPTIONS MAIN REPORT Table of Contents Page No. SUMMARY AND CONCLUSIONS ......................... ... i-xiii CHAPTER I: THE "OPEt DOOR" IN INTERNATIONAL PERSPECITVE ............ 1 A. Recent Progress and A New Juncture............................. 1 B. The Role of Foreign Trade: International Experience0.0....6.64 2 C. Successful Trade Policies........... ...,e** ................ 9 D. Foreign Trade Prospects ........... ......... ......... ..... . ..... 13 CHAPTER II: CHINA'S FOREIGN ECONOMIC RELATIONS: ISSUES AND POLICY 18 A. Key Features of the Chinese Trade System....................... 18 B. The Impact of the Current Trade System ........................ 24 C. Reforming The Trade Institutions ............................... 33 D. Reforming The Trade Environment................................ 42 E. Technology and Foreign Investment.............................. 45 CHAPTER III: TRADE POLICY AND MACROECONOMIC MANAGEMENT.............. 60 A. The Role of Macroeconomic Policies............................. 60 B. Macroeconomic Policies in Practice--Recent Developments in China's Balance of Payments............................... 68 C. Macro-Policies and Trade Reform................................ 71 Exchange Rate Policies and Foreign Exchange Allocation ....... 71 External Debt Management..... ..................... 73 D. Macroeconomic.Outlook With and Without Reforms................. 76 E. Conclusions for Macroeconomic Strategy......................... 79 CHAPTER IV: INTERRELATIONS AND SEQUENCING OF REFORMS ................ 81 A. Fundamental Preconditions for Efficient Decentralization..4.... 81 B. Priorities and Sequencing of Trade Policy Reforms*.............. 84 C. Interrelations with Other Reform Measures...................... 89 - ii - Page No. TABLES IN TEXT 1.1 Trade Strategy, Export Growth and Economic Performance, Selected Countries6............................ * * ........... ** 6 1.2 Trends in Manufactured Exports7................................ 7 1.3 Trade Policy and Adjustment to External Shocks .................* 9 1.4 Growth of World Production and Trade Volume .....,............. 15 1.5 Market Penetration of Manufactured Imports in Selected Industrialized Countries, 1975-837....... 16 3.1 China - Selected Economic Indicatorsdc..oto.r.s......... ..... 69 3.2 Macroeconomic Scenarios... . e n a r i os... . ............ .. ..... 78 BOXES 1.1 Classification of Trade Strategies.............................* 4 2.1 The Pre-1978 Trading Environment in China..................... 19 2.2 MOFERT's Report on the Reform of the Trade System of September 1984 .......21 2.3 The Effect of Direct Trading Rights on a Chinese Enterprise.... 28 2.4 Export Promoting Subsidies and GATT........................... 32 2.5 What Foreign Buyers Do 36 2.6 Preshipment Export Financinga n c in.g..... ... ....... ......... 46 2.7 Organizational Technology - The Case of the Automobile Industry.... 48 2.8 Sources of ................................... ................ 50 2.9 Different Strategies Towards Foreign Investment#****#*********# 55 3.1 Indicators for Assessing the Appropriateness of an Exchange Rat. 62 3.2 Recent Experience of Developing Countries with Floating Exchange Rates 64 3.3 The Transition to a Market Clearing Exchange Rite - Experiences of Two East Asian Economies...................... 65 3.4 Inflation and Floating Exchange Rst e s 67 ANNEXES 1. China's Trade System and Institutions and Recent Trade Trends 2. The Emerging Trade Policy Environment 3. China's Primary Commodity Trade and The Impact of Distorted World Markets 4. Foreign Direct Investment: China's Experience and Options for Policy Reform 5. External Borrowing and Debt Management CHINA EXTERNAL TRADE AND CAPITAL Summary and Conclusions 1. As a central element of its overall strategy to improve allocative efficiency and promote the modernization and growth of the economy, the Chinese government introduced at the end of 1978 a process of reform aimed at decentralizing production and investment decisions and subjecting them increasingly to market influences. So far, the most far-reaching reforms have occurred in the rural sector with the introduction of the household responsi- bility system, under which househoLds are the principal production and decision-making units in agriculture and related rural activities. 2. While reforms in the urban industrial and commercial sectors pro- ceeded at a slower pace, the introduction of the "open door" policy in early 1979 was one of the earliest and most wide-ranging elements of the reform package, and a major departure from the preceding two decades of self-reliance and autarkic economic and technological development. Overall, the "open door" policy has been very successful, yielding dramatic increases in international trade and exchanges and contributing to higher efficiency and higher growth of the Chinese economy. 3. In October 1986, a World Bank mission visited China to analyze, in collaboration with a Chinese :ounterpart team from a number of government agencies, the present status of China's foreign trade and capital system and, on the basis of this analysis, make suggestions for further improvements. This summary provides a short overview of the missions findings and recommendations that are further detailed in the Main Report and a companion volume of Annexes that provides further analysis and detail. The "Open Door" Policy in International Respective 4. Since the "open door" policy was adopted at the end of 1978, rapidly increasing exports of raw materials and manufactured goods, increased earnings from tourism and other services, and the use of foreign direct investment and other capital inflows have permitted a dramatic increase of imports of capital goods, machinery, technology and materials in short supply. This was clearly a major factor in the dramatic improvement in the efficiency of the Chinese economy, although other factors--such as the rural refor,ls and partial enter- prise and price reforms so far introduced--were also important, and perhaps more important than the "open door" policy alone. 5. While initial results of the "open door" policy have thus been impressive, both in terms of the immediate objective to increase the capacity to import and the ultimate objective of raising economic efficiency through technology imports, present policies wich respect to foreign trade and capital flows are far from optimal. Further policy changes could contribute to sustain recent rapid growth of external trade and thus to increase technclogy - ii - transfer and economic efficiency. Without further changes it is quite possible that foreign trade growth will slow down, as the high rates of recent years reflect primarily the transitional effect in moving from almost complete insulation to a partly open, but by international standards still relatively inefficient foreign trade system. 6. Different economies have adopted different trade policies, ranging from the almost complete autarky adopted by China during the past two decades, to the extremely open "laissez faire" approach adopted, for example, by Hong Kong or (since 1967) by Singapore. In between these extremes are a great number of different policy regimes, characterized by different degrees (and different instruments) of government intervention designed mainly to protect certain sectors from foreign competition. In developing countries protection is usually extended to industry, while industrialized countries often protect agriculture or declining industrial activities. 7. International experience suggest that developing countries that have adopted more outward-oriented trade strategies have had better economic performances than countries that have adopted an inward-oriented economic strategy (paras. 1.12-1.19). Outward oriented trade policies involve no, or very little, use of administrative restrictions on foreign trade, promote exports through flexible exchange rate policies which maintain competitiveness, apply relatively low and uniform tariffs on imports, and provide import duty and indirect tax rebates for export production. 8. Under outward-oriented (or "neutral") trade policies, export and import-substituting activities are equally (or almost equally) favored. By contrast, a trade system that relies extensively on administrative import restrictions and high (and often highly variable) tariff rates inevitably creates a bias against exports, because the exchange rate tends to become overvalued (lower in terms of domestic currency units per unit of foreign exchange) as import demand is artificially lowered. This overvaluation of the exchange rate inevitably makes import-substituting activities more profitable than export activities. The effect of an administratively fixed (and often even more highly overvalued) exchange rate usually implies an even larger anti-export bias. 9. Developing countries that have moved to an outward-oriented (or "neutral") trade system have not only experienced far more rapid growth of exports and imports, they have also experienced faster growth of domestic economic activity, often considerably exceeding the direct impact of increased foreign trade on domestic economic activity (para. 1.14 and Table 1.1). These indirect benefits of an outward-oriented trade strategy are related to productivity increases made possible by more rapid technology transfer and diffusion into the lomesti: economy. Direct contacts with buyers and suppliers and transfer of technical personnel between enterprises are among the most important channels for enhanced technology transfer and diffusion. 10. Some costs as well as benefits may be associated with increased foreign trade and external economic relations. One potential disadvantage is that foreign competition could stifle infant industries--activities that could become internationally competitive, but only after some time. Other - iii - disadvantages include the possible transmission of primary product price volatility into the domestic economy, and the possible emergence of large trade deficits or surpluses. But the skillful use of appropriate policy instruments can soften this tradeoff--minimizing the costs while securing the benefits. International experience also suggests that developing countries that have adopted an outward-oriented economic strategy have been able to adjust to external shocks more easily than inward-oriented countries (paras. 1.18-1.19 and Table 1.3). 11. By comparison to outward-oriented developing countries, China's "open door" policy today would have to be characterized as "inward-looking." It continues to place heavy reliance on administratiye import and export decisions, including mandatory planning; it restricts imports and some exports through licensing and other quantitative restrictions; it imposes high (and highly variable) tariffs on imported goods to protect domestic industry from foreign competition; it limits direct contacts between domestic enterprises and foreign buyers and suppliers through the intermediation of monopolistic Foreign Trade Corporations; and it uses a cumbersome administrative foreign exchange allocation system, and is still somewhat inflexible in its use of the exchange rate as an economic lever. 12. The impressive record of countries applying outward-oriented trade and development strategies deserves China's consideration. 8ut, is a switch to an outward-oriented development strategy possible and feasible for a large country like China in today's world of slow economic growth and protec- tionism? What would happen in international markets if other large countries such ss India were to pursue the same strategy at the same time? 13. The most important reason for an outward-oriented trade policy is to reap the indirect benefits of a more open trading system, including the myriad of efficiency gains that enterprises can make when they are in direct contact with buyers, suppliers and competitors. China should not adopt an outward- oriented strategy merely, or even primarily, to maximize export revenues. Foreign competition provides a spur to the efficiency of domestic enterprises and provides strong incentives to keep up with progress in technology, organ- ization and management practices made elsewhere. The present administratively intermediated foreign trade system, in contrast, creates an "air-lock", in which most of the indirect benefits of external trade are blocked from reaching the domestic economy. Key Features of the Present Trade System 14. During the 1949-78 period, China's economic strategy was essentially one of self-reliance. Domestic needs were to be satisfied as much as possible by domestic production, and foreign trade was used to supplement shortfalls in such production--primarily in food, essential raw materials, and capital goods. 15. Trade flows were determined through a centralized planning system under the authority of the Ministry of Foreign Trade (MFT). The annual foreign trade planning process wouLd identify the required imports, and exports would be selected to finance the desired level of imports, so as to - iv - avoid any significant trade deficits that would require foreign borrowing. Once the plan was determined, it would be entrusted to the twelve main centralized foreign trade corporations (FTCs) for implementation. To ensure the insulation of the domestic market, the FTCs had monopoly powers within specified scopes of business, and they would procure all goods for export and sell all procured imports at the domestic prices for these goods. 16. From 1979 to 1985 the trade system evolved rapidly. The centralized FTCs lost some of their monopoly powers, and their branch offices began to operate as separate units. In addition, provincial authorities created their own FTCs to fulfill provincial export aspirations, and line ministries found it convenient to establish corporations to engage in external tradt,in their products directly. A formal system of foreign exchange retention - at the provincial and enterprise level was also introduced, and, when combined with the introduction of import licensing and an initial decentralization of licerse-issuing authority, this period saw a rapid change in the locus of trade activity. From being a residual activity carried out in a highly centralized manner, trade became, during this period, a central focus of effort. These various reforms in trade practices over the period were summarized in the September 1984 document on the reform of the trade system adopted by the State Council (para. 2.15 and Box 2.2). Nevertheless, many trade practices, such as pricing and administrative allocation remained essentially unchanged. The Impact of the Present Trade System 17. The "Air-Lock". The FTCs are the main point of contact between China and the world market, and Chinese enterprises are to a large extent insulated from world price developments and have only limited contact with foreign buyers and suppliers. This isolation was an explicit aim of the trade system as originally constructed, as world market forces were viewed as irrational and potentially harmful to national interests. Although much has changed since 1979, the air-lock still exists to a considerable extent. The air-lock creates four problems, which may not have been critical in the early years of export development, but which will become increasingly important as China seeks to expand exports of new, less homogeneous products: (a) Inefficiency of Exporting. Under the current system, exports are not determined efficiently according to China's comparative cost advantage. The economic .osts of earning foreign exchange vary widely among different export products with profits made on some, and losses on others. (b) Inappropriate Export Patterns. Because world prices are not passed on to producers, the production of profitable exports is not encouraged nor is production of unprofitable exports discouraged, because the producers are not made aware of, and thus do not respond l/ A foreign exchange retention quota is a right to purchase foreign exchange based on past export sales. - v - to, the profits and losses incurred. (Even if passed on, enterpri- ses might not respond in a fully appropriate way, because they are not yet fully responsible for their profits and losses.) (c) Lack of Information. The air-lock of the FTC between the producer and the market prevents enterprises from reacting in a timely fashion to market opportunities. At the same time, the absence of contacts between foreign buyers and sellers and Chinese enterprises means that these enterprises are foregoing much "free" technical assistance in such areas as product specifications and design, new product developments, capital equipment, factory design, quality control, etc. Experience elsewhere has shown such direct contacts between foreign buyers and suppliers and domestic enterprises to be one of the most important, most effective and cheapest sources of technology transfer and marketing know-how. (d) Lack of Competition from Imports. Import controls and high tariffs prevent import competition from stimulating efficiencv in domestic production. 18. Foreign Exchange Allocation. Foreign exchange allocation mechanisms have two essential, overlapping functions: they allocate foreign exchange among the various potential users, and they determine the price at which foreign exchange is traded. The importance of this latter function cannot be overemphasized, and has two major effects. Firstly, the exchange rate alters the relative attractiveness of producing a specific good or service for export markets vis-a-vis the domestic market and the ability of the domestic product to compete with imports. In addition, if changes in the exchange rate are perm.itted to have an impact on domestic prices, the relative prices of tradeable goods and nontradeables (such as power, domestic services, restaurants) will be altered, thus influencing the production and investment decisions of enterprises. It is partly because China does not permit the exchange rate to move in such a way as to equalize demand and supply of foreign exchange that it is necessary to allocate foreign exchange administra- tively, since at the present price of foreign exchange there is inadequate supply to meet demand. 19. Protection of Domestic Production. Protection of domestic production from imports acts automatically as a disincentive to exports, and limits the many indirect benefits of engaging in foreign trade (paras. 1.7 and 2.30). Protection in the Chinese trade system is provided through four mechanisms: trade planning, import pricing, licensing and tariffs. As trade decision-making is decentralized, and international prices are passed on to final users through wider application of the "agency system", import licensing and tariffs will become the predominant sources of protection. Reforming Trade Institutions 20. With its enormous resource base and population, China could even- tually produce and export a wide variety of goods and services. The question is what to produce and what to import and export, and how these decisions should be made. Such decisions become increasingly complex as an economy - vi - develops, and international experience suggests that they can be made most efficiently by enterprise managers guided by market prices (modified, if necessary, by "economic levers"). In foreign trade, it will be important to establish a policy framework that promotes efficient decisions in trade and production. Such a trade reform will necessarily inter-relate with other areas of reform, notably enterprise reforms, pricing policies and macro- economic policies (para. 2.36 and Chapter IV). 21. International experience suggests that China could benefit substantially by removing the tight compartmentalization or "air-lock" between domestic and international markets that results from the present operating rights and procedures of the FTCs. However, if decentralized decision-making is to lead to efficient operations, competition needs to be introduced, and prices and other economic levers (in particular the exchange rate) have to reflect the true costs of resource use. Some of the specific measures that could be considered-include: (a) Eliminating, over time, the product-specific trading rights of FTCs, permitting them to engage in the trade of any commodity, thus becoming de facto general trading corporations (GTCs). (b) Making FTCs fully responsible for profits and tosses, and elimin- ating any plan exports that imply substantial losses to the FTCs. At the same time, subsidies for command plan imports that remain necessary because of large differentials between domestic and inter- national prices which continue for domeitic policy reasons (e.g. fertilizer) should be provided by specific product subsidies in the state budget, rather than from FTC revenues, or general subsidies to @TCs. (c) Eliminating the exclusive right of Foreign Trade Corporations to engage in foreign trade and leaving the choice of whether or not to use an FTC, and which FTC to use, to enterprises. FTCs would then become service organizations that would have to justify their role by being more efficient than other foreign trade channels. The only exception would be "special" exports, for which China faces inelastic demand or has a very large market share (see paras. 2.53- 2.56). (d) Eliminating the strict separation of domestic and foreign trade, giving FTCs the right to engage in all types of dumestic wholesale and retail trade and domestic commercial organizations the right to engage in foreign trade. (e) The role of foreign buyers and foreign trade corporations in China's external trade could also be increased. Joint ventures between Chinese FTCs and foreign GTCs could, for example, bring benefits from the latter's extensive marketing networks, and facilitate the transfer of marketing know-how to China's FTCs. 22. These recommendations are quite consistent with the September 1984 report of the Ministry of Foreign Economic Relations and Trade on the future - vii - reform of the trade system. However, in many key respects the reforms out- lined in that report have not been implemented, and remain the key institu- tional reform issues. Once the decision to decentralize to enterprises has been taken, it will be necessary not only to make the institutional reforms described above, but also to adjust some of the economic levers affecting trade decisions. Reforming the Trade Environment 23. Exchange Rate Policies and Foreign Exchange Allocation. If trade decision-making is decentralized to enterprises, and international prices are passed on to both exporters and importers, the exchange rate will become a more important factor in determining the volume and pattern of exports and imports. The exchange rate can only play this role efficiently if exchange rate policy is implemented flexibly and actively to ensure that Chinese exports remain competitive in world markets and, in combination with other policies, ensures a viable medium-term balance of payments. 24. At the present time, China maintains a managed floating exchange rate system, under which the exchange rate is based on developments in the balance of payments and movements in costs and exchange rates of China's major trading partners. However, the present exchange rate has only limited impact on trade decisions. Foreign trade and other external transactions are still largely determined by administrative interventions, including the trade plan, mandatory imports and exports, administrative restrictions on a substantial number of import and export transactions, and so on. These decisions, how- ever, are being increasingly decentralized to local governments, and enter- prises can use their foreign exchange retention quotas to import machinery and other unrestricted goods for modernization investments ("technical transforma- tion"). 25. Trade reforms of the type proposed in this report would make the exchange rate a more effective policy instrument by increasing links betwee.n foreign and domestic prices. While trade reforms will make the exchange rate a more effective policy instrument, it is also true that a more appropriate and flexible exchange rate policy would make trade reforms more effective. This objective could be achieved by a more active use of exchange rate policy within the context of the present exchange system. Alternatively, China may decide to allow market' forces to play a more direct role in exchange rate determination and move to a system in which the exchange rate would be determined in the market by the supply and demand for foreign exchange. This could be particularly useful during the transition towards a more market- oriented foreign trade system. It would be appropriate and useful to discuss changes in foreign exchange allocation and exchange rate policy with the IMF. 26. If the present managed floating exchange rate system is maintained, exchange rate adjustments would be needed to compensate for the removal of administrative import restrictions or subsidies. Balance-of-payments develop- ments, desired reserve levels and indicators of international competitiveness (see para 3.6 and Box 3.1) could be used to estimate the required adjust- ments. An advantage of this option is that it would not represent a major change from the present exchange rate system, but would introduce more - viii - flexibility than at present. A disadvantage of this option is that it would take quite some time (several w1-nths or quarters) to ascertain whether any exchange rate adjustments made have been sufficient to compensate for the removal of trade restrictions or other changes in the domestic or inter- national economy. This might make it difficult to phase out administrative allocation of foreign exchange rapidly. 27. An alternative approach that would permit a more rapid transition towards market determined allocation of foreign exchange would be to permit trading of foreign exchange retention rights among enterprises and other units at a freely negotiated price. This would be similar to the approach used in the domestic market where transactions at flexible prices have been introduced while maintaining fixed prices for the same goods allocated by the Plan. 28. While not desirable as a permanent solution, such a system would improve on current practice. Since the exchange rate for foreign exchange retention rights would be freely determined, it would permit rapid removal of administrative restrictions without running the danger of incurring large balance of payments deficits. The exchange rate established in the market for foreign exchange retention rights would give an indication of the direction and magnitude of the requirod adjustment of the official exchange rate to compensate for the removal of administrative restrictions. 29. The problems with such a system are similar to the use of dual (plan-market) prices in the domestic economy. Enterprises or individuals that have access to foreign exchange at the official rate will always be tempted to sell it at the market price for foreign exchange established in the market for foreign exchange retention rights. It would thus be desirable to move the official exchange rate as rapidly as possible towards the rate established in the market for foreign exchange retention rights, eliminating any gap between the two rates after a relatively brief transition period. 30. To ensure that the rate established in the market for foreign exchange retention rights is realistic, it would be important for the market in foreign exchange retention rights to cover a large proportion of foreign exchange transaction and to have a large number of enterprises participate, particularly those enterprises that are already responsible for profits and losses. It will also be desirable to increase the foreign exchange retention ratio for enterprises, ultimately to 100%. 31. The State will, of course, remain a major user of foreign exchange for key projects. The creation of such a market for foreign exchange does not mean that the State would have lower access to foreign exchange, but that it may be obtained in different ways. Indeed, creation of a market in foreign exchange can be expected to increase the supply of foreign exchange available to the economy as a whole--and thus potentially to the State--both by reducing the hoarding of foreign exchange, and by stimulating exports. If the increased retention ratios reduce the foreign exchange available directly to the State below the desired level, the State can always enter the foreign exchange market directly to purchase additional foreign exchange to fund key projects. In addition, hovever, the State could save itself considerable levels of foreign exchange by changing the ways it funds the inputs for key - ix - projects. Instead of maintaining low prices for the command plan imports and subsidizing all such imports, it would be preferable to provide specific subsidies direct to the key projects, but adjust the prices of the strategic products, so that the subsidy is provided for the priority uses, but not for non-priority uses, as is the case at present. 32. It is also important to consider the relationship between the exchange rate and domestic prices. The exchange rate links international and domestic prices and determines the relative price between tradeable and non- tradeable goods and services. In general, changes in the exchange rate will affect the relative prices of different goods and services, and not the general level of prices. Inflation is primarily caused by excess money creation, and this will eventually necessitate an exchange rate change. Of course, a depreciation of the exchange rate will raise the prices of imports, but international experience has shown that such price effects have frequently been anticipated prior to such a depreciation. It is thus usually the case that rather than an exchange rate change causing inflation, it is inflation which causes the exchange rate change. Import Licensing and Tariff Reform 33. Trade reforms will be more effective if a more flexible exchange rate policy is implemented. Appropriate fiscal and monetary policies would also need to be implemented. Additional enterprise and price reforms would also complement exchange reforms. At the same time, trade reforms would also make the exchange rate a more effective policy instrument. A flexible exchange rate policy would allow quantitative restrictions to be gradually phased out. However, there may remain valid industrial 2qlicy reasons to restrict certain imports to protect domestic enterprisesz/ for a limited period of time until they have become internationally competitive. Such protection should be provided mostly by tariffs rather than import licensing on other administrative import restrictions. Only in very exceptional cases are quantitative restrictions equivalent, or perhaps even superior, to price (tariff) interventions (para. 2.64). It would thus be desirable to replace licensing by tariffs as soon as possible. 34. The present tariff schedule, with rates ranging from 0-200%, results in a highly variable 'effective' protection of different industrial subsec- tors. International experience has shown that such highly variable protection can lead to substantial inefficiencies. Tariff rates should therefore be made more uniform, perhaps with a range of 10-50%. On the other hand, the wide- spread exemptions from import duties for machinery and capital goods imported for capital construction and technical transformation result in a penalty on domestic producers of such goods, which could, in many cases, achieve effi- cient production in China. Tariff exemptions for capital construction and technical renovation should thus be eliminated. Tariff exemptions or rebates would then be resLricted to imported inputs for export production (see 21 Ideally, production subsidies would be preferable to trade measures, but budgetary resources might not be sufficient. - x - below). Tariff reform would probably be best phased over several years. In the first stages it will be desirable to pre-announce planned tariff adjust- ments planned to occur over a period of several years so that domestic enterprises can adjust accordingly. Export Incentives 35. The most important export incentive at present is the foreign exchange retention system for enterprises, which compensates to some extent for the anti-export bias of the present trade system. The retention system could be expanded by making access to foreign exchange more automatic, both for exports and for import-substituting enterprises, since a dollar saved through efficient import substitution is as good as a dollar earned through increased exports. However, as long as tariffs or other import restrictions are applied for reasons of industrial policy (e.g. infant industry protec- tion), the system will continue to have an anti-export bias. One way to reduce this anti-export bias is to rebate to exporting enterprises the tariff duties collected on inputs used in the production of export goods. This is administratively challenging, but there is a large international experience which could help avoid some of the difficulties. In this way, any enterprise anywhere in China would reap the advantages that the Special Economic Zones now provide, and this would at the same time, provide the indirect advantages of greater export orientation (para. 2.67) throughout the economy. 36. A second measure that could be taken to reduce anti-export bias is to rebate domestic taxes on export production. An administratively efficient and internationally acceptable way of doing so would be to replace the present industrial and commercial tax with a value-added tax system. This would, of course, be a major undertaking and should await a more general reform of China's tax system. It may also be possible to link such rebate systems with a system for permitting access by indirect exporters to foreign exchange retention rights. 37. A third measure, and one that could perhaps be taken earlier, would be to provide improved access to domestic working capital financing to all exporters and, through a system of "domestic letters of credit", to all enterprises that provide inputs into production for exports. Such a credit facility would, of course, have to be consistent with the overall credit plan of the People's Bank of China. Such a system would also make it possible to extend rebates of tariffs on imported goods used in export production to indirect exporters (suppliers to export enterprises). 38. These export incentives would all require advance planning and administrative sophistication and should be considered for implementation only in the medium term--once substantial progress has been made in the reform of the foreign exchange allocation mechanism, the decentralization of foreign trade rights, the reduction of detailed administrative foreign trade planning and licensing, and a reduction and harmonization of tariff rates. An implementation of, say, five years would appear feasible. - xi - Foreign Direct Investment 39. Foreign direct investment that brings in desired advanced technology and capital can be stimulated by a number of policy actions. These include, most importantly, measures that would make it less necessary to allocate foreign exchange. Recently, measures have been announced that would make it easier for foreign joint ventures to gain access to foreign exchange by becoming de facto foreign traders on behalf of domestic enterprises. While this is progress in the right direction, access to foreign exchange could be made even more automatic through implementing a more flexible exhange rate policy. Transport costs, slight quality differences, brand-name recognition and other similar factors often make import substitution easier than export promotion--particularly in new products using more advanced technology. Thus, there is a very useful role to be played by foreign joint ventures in pro- ducing for the domestic market, and in a number of sectors it may take ten years or more until production for exports will become competitive. 40. More important than preventing 'exploitation' by foreign investors (that might, in the worst case, involve a small fraction of production value in excessive royalties, overpricing of supplies or services) is to prevent economically inefficient foreign direct investment. Apart from domestic economic inefficiencies, the potentially most worrisome source of inefficiency is excessive protection from import competition. Fortunately, China has a very large domestic market that will justify numerous enterprises (and joint ventures) in nearly any line of business, thus permitting vigorous competition among domestic enterprises as well as joint ventures. But in some cases this will not be enough, and high levels of import protection could seriously reduce the incentives for efficiency. Interrelations with Other Reform Measures 41. While the trade policy measures summarized in the preceding section are, as stated, closely related to other economic reforms, progress can be made in implementing these trade reforms independently of progress in other economic reform areas. The most crucial precondition is the maintenance of domestic macroeconomic balance and stability. All too many developing countries have in the past attempted to reform their traue systems in the face of large government budget deficits, permissive monetary and credit policies to finance these deficits, and resulting rapid inflation. Their task has often been rendered even more difficult by heavy external indebtedness that left little room to maneuver, as short-term reductions in exports or surges in imports could simply not be financed. But very often also, trade reform measures have failed because initial steps were too timid, particularly with respect to exchange rate system reforms. 42. Maintaining Macro-Economic Balance. Efficient decentralization of decision-making and introduction of further reforms would be much facilitated by stable aggregate demand and stable prices. Of particular importance in this regard are further steps to restrain government spending and to give the People's Bank full authority to limit the expansion of money and credit. Increased interest rates, particularly on enterprise deposits and capital construction loans, could help to dampen excess demand pressures and to - xii - channel resources to the investments with the highest yield. Stable macro- economic management is particularly important in managing the balance of payments and in limiting foreign borrowing to a viable level (Chapter III). Apart from prudent budgetary and credit policies, appropriate exchange rate adjustments as discussed above are essential to maintain macroeconomic balance. 43. China does not, at this moment, face extreme macroeconomic imbalances. However, the experience of the past two or three years shows that China's economic and political system is not immune to such dangers either. There are pressures for continued high government expenditures and easy bank credit, at the very time when the role of the central government is being redefined and revenues decentralized to enterprises and local governments. When combined with a tow political tolerance for inflation, this leads to calls for renewed or continued administrative controls over prices, imports and so on. Macroeconomic balance is therefore crucial during a transition towards a more efficient trade system based on decentralization of decision- making and competition. 44. Domestic Economic Reforms. In other respects, trade reforms could begin to be implemented relatively independently of the progress of domestic reforms. This is so, because in many respects the decentralization to enter- prises of foreign trade decisions is currently trailing behind the decentrali- zation of domestic production and marketing decisions. Similarly, competition among enterprises in the domestic market is already reasonably advanced in those commodities where China's export prospects in the near term are brightest: relatively labor-intensive, diversified manufactures, including not only traditional exports such as handicraft, processed food, textiles and garments, but, more importantly, new products such as consumer electronics, standard electrical and mechanical machinery, fabricated metal products, and so on. The most important missing element to become internationally competi- tive in these products is better design and quaLity that can be acquired efficiently only under a more open system of foreign trade and international economic relations (para. 2.45 and Box 2.5). 45. This is not to say that trade reform can be carried out without regard to further progress on price reform. Further price reform is desirable and necessary both to increase the efficiency of resource allocation in the domestic economy--particularly for capital-intensive basic raw materials (steel, chemicals, power, etc.)--and also to ensure an efficient set of aignals to govern trading decisions. A wide-ranging trade reform without price reform would lead to irrational exports of some goods, such as steel- products and energy intensive goods, and excessive imports of consumer goods. However, for many of the goods mentioned in the previous paragraph there has already been considerable progress in price reform, and it is our judgement that the proposed initial steps in trade reform do not have to wait for further progress in other areas. 46. These considerations suggest a phasing for the recommended trade reforms. By identifying those sectors and subsectors where price and enter- prise reform have already made good progress, candidates for early trade reform could be selected. This could involve liberalization of trading rights - xiii - in those sectors, together with increased foreign exchange retention, elimination of plan allocation for imported inputs, and removal of import licensing. Once trade reforms had 'caught up' with price and enterprise reforms in these sectors, progress in extending the trade reforms to other sectors could be linked with further progress in price and enterprise reforms (paras. 4.32-4.37). 47. The creation of a more sophisticated 'neutral' environment for export enterprises (including duty and indirect tax rebating, automatic access to working capital finance, etc.) will of course require simultaneous movement in the areas of tax and financial systems reform. The primary constraint there is the administrative capacity required to prepare and implement these reforms. It will thus take at least 3-5 years to prepare and implement these measures. 48. Similarly, in the area of foreign direct investment there are not too many constraints arising from the domestic reform process. The most important constraint is the requirement to balance foreign exchange receipts and payments. Apart from this, progress will have to be made in eliminating dicferences in the legal and administrative environment between domestic enterprises and foreign joint ventures. This would in a first phase mostly involve giving domestic enterprises the same managerial autonomy and responsi- bility that joint venture enterprises already enjoy, except for the over- generous corporate income tax provisions that joint ventures now enjoy. How- ever, enterprise reform should not stop there. Further steps are required, particularly to enhance factor mobility by creating markets for labor services, land use, and capital, or improving the efficiency of these markets. 49. Thus, foreign trade and other foreign economic relations play an essential role in the domestic economy. Foreign trade not only provides access to machinery, technology and goods in short supply, but also the multitude of contacts with buyers, suppliers and competitors that have proved to be an essential and low-cost source of technology transfers and information about market opportunities. To maximize these indirect benefits of foreign trade, it is essential that enterprises have direct conLacts with buyers and suppliers and can make foreign trade decisions autonomously. The larger the number and types of enterprises that can engage in such contacts, the mote rapid technology transfer and diffusion are likely to be. Foreign trade should thus not be regarded as a separate sector or activity, but as an integral part of enterprise decision-making on production and investment. However, if such decentralized decision-making is to lead to efficient resource allocation, prices will have to reflect economic costs, competition will have to be vigorous, and macro-economic stability has to prevail. Reforms will thus have to move simultaneously on these several fronts. I. THE "OPEN DOOR" IN INTERNATIONAL PERSPECTIVE 1.1 As a central element of its overall strategy to improve allocative efficiency and promote the modernization and growth of the economy, the Chinese government introduced at the end of 1978 a process of reform aimed at decentralizing production and investment decisions and subjecting them increasingly to market influences. So far, the most far-reaching reforms have occurred in the rural sector with the introduction of the household responsi- bility system, under which households are the principal productiun and decision-making units in agriculture and related rural activities. 1.2 While reforms in the urban industrial and commercial sectors have proceeded at a slower pace, the introduction of the "open door" policy in early 1979 was one of the earliest and most wide-ranging elements of the reform package, and a major departure from the preceding two decades of self- reliance and autarkic economic and technological development. Overall, the "open door" policy has been very successful, yielding dramatic increases in international trade and contacts and contributing to greater efficiency and higher rates of growth of the Chinese economy. 1.3 In October 1986, a World Bank mission visited China to analyze, in collaboration with a Chinese counterpart team from a number of government agencies, the present status of China's foreign trade and capital system and, on the basis of this analysis, make suggestions for further improvements. The mission also benefitted from an earlier study of domestic enterprise and financial sector reform ipues jointly undertaken by another Bank mission and Chinese counterpart team.- This report summarizes the mission's findings and recommendations, and a companion volume of annexes provides more in-depth description and analysis. Chapter I first looks at the benefits and costs of increasing foreign trade and other external economic relations in the light of international experience and then briefly discusses trade policy measures adopted by successful exporters and how these compare to China's "open door" policy. Chapter II then discusses the present trade system and options for reform. Chapter III looks at macro-economic management issues relating to external trade and the balance of payments, and Chapter IV discusses the sequencing of trade reform and its interrelation with other elements of economic reform. A. Recent Progress and a New Juncture 1.4 Since the "open door" policy was adopted at the end of 1978, increased exports of raw materials and manufactured goods (rising at 15X p.a. in terms of US$ during 1979-85), increased earnings from tourism and other services, and the use of foreign direct investment and other capital inflows 1/ World Bank: China--Finance and Investment, Report No. 6445-CHA, June 1987. - 2 - have permitted a dramatic increase of imports of capital goods, machinery, technology and materials in short supply. Imports in 1985 were $38 billion, five times the level of 1975. This massive inflow of technology, capital goods, and intermediate goods was a major,factor in the dramatic improvement in the efficiency of the Chinese economy*,_ although other factors--such as the rural reforms and partial enterprise and price reforms so far intro- duced--were also important, and perhaps more important than the "open door" policy alone. 1.5 While initial results of the "open door" policy have thus been impressive, both in terms of the immediate objective to increase the capacity to import and the ultimate objective of raising economic efficiency through technology imports, many Chinese as well as foreign observers believe that present policies with respect to foreign trade and capital flows are far from optimal. The impressive gains of the past eight years reflect, to a large extent, the comparatively easy initial gains of opening up to external trade. Further improvements could not only raise the efficiency of foreign trade and therefore lower the domestic resource costs of importing foreign technology, they are essential for China to increase its exports (and capacity to import) at the rapid rate that is required if the high targets for economic growth and modernization are to be chieved. 1.6 A related issue arises from the fact that China is quickly becoming a major factor in world trade, at least in some product categories. While China's share in total world trade is still less than 2%, its share of world exports of textiles and clothing has doubled to 8% since the introduction of the "open door" policy, and it is now among the top five suppliers in both categories. Rapid growth in these and other products, where China has comparative advantage resulting from its low wages and skilled labor force, will require structural change--and possibly economic and social disloca- tions--in importing countries that might lead to increased protectionist pressures. China can, to some extent, reduce these pressures by adopting a trade system that is more transparent and lees likely to attract countervailing actions by major trade partners. B. The Role of Foreign Trade: International Experience 1.7 There are at least three reasons why countries engage in foreign trade. First, differences in natural resource endowments lead to differences in relative cost of production, and trade permits countries to specialize in the production of those commodities of which they are lower-cost producers. Second, trade permits less developed countries to overcome differences in human resources (technology, management skills, etc.) by acquiring goods and services embodying more advanced technologies, and thus avoiding "reinventing 2/ During 1981-85, the investment rate of 30% of GDP resulted in a rate of growth of 10% p.a. while an almost identical investment rate during the preceding 10 years resulted in a growth rate of only 6% p.a. Put differ- ently, the incremental capital-output ratio declined from an average of 5 during the 1970s to 3 during 1981-85. the wheel" and thereby speeding up the process of economic development. Third, external trade permits countries to expose their domestic enterprises to international competition and thus ensures that domestic industry will live up to the highest atandards of efficiency. 1.8 While all three factors play a role, the second (technology transfer and development) is of particular importance for a large developing country, such as China. By contrast, the first is predominant for some resource abun- dant countries with small populations such as, for example, capital-surplus, oil-exporting countries. The third factor is particularly important for small countries where, without external trade, even modest economies of scale would lead either to sub-optimal and inefficient scale of plants, and/or to mono- polistic industry structures and inefficiencies. This is not to say that the competitive impact of foreign trade is not relevant and beneficial for larger countries--quite to the contrary. Among the advanced industrialized countries, the competitive stimulus is often regarded as one of the most important benefits of external trade. 1.9 Increased external trade could also have disadvantages. One poten- tial disadvantage is that foreign competition could stifle infant industries--activities which could become internationally competitive, but only after a period of time in which necessary experience and skills are developed. Other possible disadvantages include the transmission into the domestic economy of primary product price volatility, fluctuations in world demand for industrial products, and the possible emergence of undesirable large trade deficits or surpluses or unwanted inflows and outflows of capital. (These concerns are discussed in Chapter III and Annex V.) International experience suggests that these costs are usually outweighed by the benefits, as discussed in the following paragraphs. 1.10 While external trade is the most important component of international economic relations, there are other complementary means to increase external economic contacts, technology transfer and economic development. Tourism and exports of other labor services serve to augment the earnings from commodity exports and thus to increase the capacity to import; foreign borrowing permits increased imports today to be paid for by future export earnings, and can thus accelerate technology transfer; and foreign direct investment permits the import of capital goods, technology, management skills, foreign financing and risk-sharing in one bundle. Last, but not least, overseas education and training can be one of the most effective forms of technology transfer. Each of these special forms of foreign economic relations has its own advantages and disadvantages, but they are dwarfed by the far larger volume of commodity trade. Trade thus takes a special and dominant place in international economic policy-making. 1.11 Different economies have adopted different trade policies, ranging from the almost complete autarky adopted by China from 1960 to 1978, to the extremely open "laissez faire" approach adopted, for example, by Hong Kong or (since 1967) by Singapore. In between these extremes are a great number of different policy regimes, characterized by different degrees (and different instruments) of government intervention designed mainly to protect certain sectors from foreign competition. In uaveloping countries, protection is - 4 - usually extended to industry, while industrialized countries often protect agriculture or declining industrial activities, such as textiles, ship building or steel. These trade policy interventions are either of an administrative and quantitative nature (import restrictions, licensing, foreign exchange rationing, "voluntary" export restraints) or are effected through "economic levers" using the price mechanism (tariffs and other border levies). Box 1.1 classifies trade policy regimes into four different categories. Box 1.1: CLASSIFICATION OF TRADE STRATEGIES Strategy Exmles I. Closing of economy with tight Burma from 1962 administrative restrictions on trade Sri Lanka to 1976 and foreign investments. USSR to 1986 China to 1978 II. High protection, using administative South Korea to 1962 restrictions on imports and/or high Brazil to 1967 and tariffs (resulting in overvalued after 1979 exchange rates). India China from 1979 III. Relative openness, emphasizing export South Korea from 1962 promotion through realistic exchange Brazil from 1967 to 1973 rates and export incentives, but using selective import protection through tariffs and limited administrative import restrictions. IV. Substantial openness with little pro- Singapore from 1967 tection against manufactured imports Chile from 1976 (but sometimes high protection for South Korea from 1988 /a agriculture) and few restraints on Industrialized countries investment and capital flows. /a After completion of the five-year import liberalization program pre- announced in 1983. 1.12 International experience suggests that developing countries that have adopted more outward-oriented trade strategies have had better economic performances than countries that have adopted an inward-oriented strategy. Outward-oriented trade policies (Categories III and IV in Box 1.1) involve no,' or very little, use of administrative restrictions on foreign trade, promote exports largely through a market-related exchange rate, apply relatively low - 5 - and uniform tariffs on imports and provide import duty and indirect tax rebates for export production.- In terms of the categories in Box 1.1, most developing eountries since World War II have adopted polici-s that can be classified as belonging to either category II or III. Larg, improvements in export and national income growth can be observed between categories II and III (Table 1.1). The main difference between these two strategies is that category II relies extensively on administrative restrictions on trade combined with high tariffs, while category III places predominant reliance on a market-based exchange rate system, combined with lower and more uniform tariffs. A market-based exchange rate system implies the absence of administrative foreign exchange allocation, with the market itself ensuring adequate export revenues to meet foreign exchange needs, through automatic adjustment of the exchange rate to balance demand for and supply of foreign exchange.- In terms of economic incentives, China's "open door" policy compares most closely to category II rather than III (export promotion), modified by a political commitment and some economic incentives towards export promotion (see Chapter lI, paras. 2.16-2.18). 1.13 Under outward-oriented (or "neutral") trade policies, export and import-substituting activities are equally (or almost equally) favored. They should not be confused with "export-promoting" strategies under which some countries have adopted positive bias in favor of exports. Outward-oriented strategies involve the absence or near-absence of bias against exports, creating reasonable opportunities for exports, but not implying necessarily a reliance on exports as the major source of growth in the economy. By contrast, a trade system that relies extensively on administrative import restrictions and high (and often highly variable) tariff rates inevitably creates a bias against exports, because the exchange rate will become overvalued (lower in terms of domestic currency units per unit of foreign exchange), as import demand is artificially lowered. This overvaluation of the exchange rate, combined with high import protection, makes import- substituting activities more profitable than export activities. The over- valuation of the exchange rate is usually more extensive and serious if import protection is provided by administrative import restrictions (which could be seen as equivalent to a tariff with a rate of infinity). 1.14 The superior performance associated with an outward-oriented strategy not only holds across countries (where differences in culture, work ethic, savings pattern etc. could account for some of the difference), but has been equally true for a country that has changed policies over time. The countries shown in Table 1.1 all changed their trade strategies at some time from an import substitution to an outward-oriented strategy: South Korea around 1960, Brazil around 1967, Colombia and Tunisia between 1965 and 1970. 3/ Countries that have adopted outward-oriented trade policies also generally rely on decentralized production, marketing and investment decision-making by autonomous enterprises, influenced by market prices, for both domestic and foreign sales and purchases. 4/ See Chapter III for a fuller discussion. All showed dramatic increases in exports and domestic economic growth after switching to a more outward-oriented economic policy. The effect of an outward-oriented strategy on economic growth appears to be largest for small countries, which tend to have a higher ratio of external trade to national income. But even in large developing countries, such as Brazil, the effects of an outward-oriented strategy on domestic growth and efficiency have been large. More recent evidence also suggests that the benefits of an import- substitution strategy tend to diminish over time, as the opportunities for efficient substitution are exhausted, and successive import-substituting investments become more costly. Table 1.1: TRADE STRATEGY, EXPORT GROWTH AND ECONOMIC PERFORMANCE, SELECTED COUNTRIES Trade strategy Export earnings/a Real GDP (see Box 1.1) growth (X p.a.F3 growth (X p.a.) Brazil 1955-65 II (import substitution) 1.1 5.5 1965-76 III (neutral) 28.2 7.6 Colombia 1955-65 II (import substitution) -1.4 1.9 1970-76 III (neutral) 16.9 6.5 South Korea 1953-60 II (import substitution) -6.1 5.2 1960-76 III (neutral/export promotion) 41.6 9.2 Tunisia 1960-70 II (import substitution) 6.8 4.8 1970-76 II/III (partial export prom.) 23.4 9.4 China 1960-78 I (autarchy) 7.8 4.6 1978-85 II (modified import subst.) 15.7 8.8 /a At current prices in US$. Source: A.0. Krueger: The Effects of Trade Strategies on Growth, Finance and Development, June 1983, World Bank staff estimates. 1.15 The four newly industrializing economies (or NICs ) of East Asia-- South Korea, Singapore, Hong Kong and Taiwan, China--have been particularly successful in their outward-looking strategy. Despite their relatively small size, their combined merchandise exports now amount to some $110 billion-- equivalent to about one-half of Japanese or US exports and accounting for fully one-half of all exports of manufactured goods of developing countries. Their export performance has been largely based on manufactured goods, and they have tripled their combined share of world trade in manufactured goods over the past fifteen years (Table 1.2). - 7 - 1.16 While China has made good progress in promoting exports of manufac- tured goods, its growth rate performance still remains slightly below that of the other developing economies of South-East Asia, although somewhat above that of Latin American countries (Table 1.2). Other developing countries have barely held their combined market share in world trade, with the gains in some countries offset by substantial losses in others. Table 1.2: TRENDS IN MANUFACTURED EXPORTS Share of world trade (Z) Growth (% p.a.) 1970 1984 1970-80 1980-84 Industrial countries 88.1 80.1 18.2 -1.9 Developing countries 6.7 14.8 25.4 5.6 East Asian NICs 2.4 7.5 28.3 8.5 South East Asia 0.1 1.2 40.2 10.9 China 0.5 1.3 26.7 9.3 Latin America 1.0 2.4 25.4 7.4 Other LDCs 2.6 2.3 19.7 -5.7 Source: World Bank Trade Data Base. 1.17 Some costs as well as benefits may be associated with increased foreign trade and external economic relations. Two categories of potential economic costs of an outward-oriented economic strategy were identified above: the stifling of infant industry by foreign competition and greater vulnerability to fluctuations in international prices and world demand. The international evidence suggests that these costs have been outweighed by the benefits. Countries that have adopted outward-oriented policies have exper- ienced far more rapid growth of manufactured exports, and have also exper- ienced more rapid and more efficient industrial development generally. For example, some of the most successful outward-oriented NICs have recently shifted from earlier export successes in textiles, clothing and consumer elec- tronics to machinery, automobiles, shipbuilding, and personal computers. They are not only gradually diversifying their industrial sectors but have also become internationally competitive in these new sectors that have received relatively little protection. Moreover, skillful use of appropriate policy instruments can soften the trade-off between the costs and benefits, such as selective impo~rt industry protection, and phased liberalization policies (see Chapter II). 1.18 International experience also suggests that outward-oriented developing countries have weathered external shocks much better than inward- oriented countries, even though external shocks have been larger for outward- oriented countries, since their external trade is larger relative to the size - 8 - of the domestic economy. For example, a World Bank study if of the adjustment of 28 developing countries to the oil price increases of the mid-1970s showed that the outward-oriented economies were able to maintain higher growth rate of national income than the inward-oriented economies, even though the increased cost of oil imports (or "shock") was far larger relative to the size of their economies. 1.19 A recent study undertaken for the World Development Report 1987 provides additional evidence from a larger sample of 41 developing countries and the recent years of slower growth and adjustment to renewed external shocks. While growth rates of national income have slowed down in 1973-85 compared to 1963-73, the differences between outward- and inward-oriented countries have been maintained. Table 1.3 also shows that the efficiency of investment, as measured by the incremental capital-output ratio (i.e. the investment required for a unit increase in net output) has remained far higher (i.e. a lower ICOR) in outward-oriented countries than in inward-oriented countries. Table 1.3 also provides some evidence about the long-term impact of different trade strategies: in outward-oriented countries the share of the labor force employed in manufacturing has nearly doubled over the past two decades, while it has increased only marginally in inward-oriented countries. 51 B. Belassa: "Adjustment to External Shocks in Developing Economies", World Bank Staff Working Paper No.472, 1981. - 9 - Table 1.3: TRADE POLICY AND ADJUSTMENT TO EXTERNAL SHOCKS Outward-oriented Inward-oriented countries /a countries /b Real CDP Growth (Z p.a.) 1963-73 7.9 5.2 1973-85 5.0 3.7 Real GNP per capita Growth (1 p.a.) 1963-73 5.2 2.7 1973-85 2.5 1.0 Incremental Capital-Output katio 1963-73 2.5 4.1 1973-85 4.9 7.0 Manufacturing Employment (1 of total) 1963 13.2 12.7 1980 23.0 14.1 /a 13 countries in 1963-73, 11 countries in 1973-85. /b 28 countries in 1963-73, 30 countries in 1973.85. Source: World Bank: World Development Report 1987, Chapter 5. 1.20 While the evidence provided above does not prove a causal link between different types of trade strategies and economic performance, the data are nevertheless quite significant. If the countries are further subdivided by different degrees of outward and inward orientation, the differences in economic performance are even more striking (see World Development Report 1987, Chapter 5). C. Successful T-ade Policies 1.21 By comparison to outward-oriented developing countries, China's "open door" policy today would have to be characterized as an "inward-looking" strategy. This might appear paradoxical, given China's success in export markets since the introduction of the "open door" policy. China has moved from an extreme position of nearly complete autarky during the past two decades to a policy regime that is partially open. China's impressive export performance over the past eight years may thus be largely a transitory phenomenon. If this supposition is correct-and more detailed corroborative analysis is provided in the following chapters and the Annexes--China will probably have to undertake substantial additional reforms of its trade system to maintain these gains. - 10 - 1.22 What constitutes a successful. "outward-oriented" strategy? Different countries have pursued somewhat different policies, but a number of common elements falling into three different categories can be cited: (a) decentralization of decision-making to enterprises; (b) a favorable incen- tive environment for exports; and (c) stable macro-economic management. Decentralization of Decision-Making 1.23 International experience strongly suggests that decentralization of trade decisions to the enterprise level can be very important in helping to realize the potential benefits of increased external contacts. Direct exposure of exporting firms to foreign buyers and competitors has proved to be an extremely effective way not just of learning in the abstract about new and better products and production processes, but also of learning how to intro- duce them in practice. Decentralization of export and import decision-making to enterprises is consistent with the central thrust of China's economic reforms in the productive sectors of decentralizing production and investment decisions to enterprises and subjecting them increasingly to competition and market forces. 1.24 Decentralization of decision-making implies that enterprise managers would determine the details of external transactions. To ensure that their decisions were in accordance with China's overall economic interests and broadly consonant with government strategy, internal reforms to make enter- prises more sensitive to costs, to customer requirements, and to profits and losses would be essential, as would be an appropriate system of prices. Of particular importance for the economic rationality of decentralized export and import decisions is the relationship between domestic and world prices. In many respects, the more direct the linkage the better, since world prices can provide appropriate signals to Chinese producers and consumers about the value or cost of particular goods to China in world markets. To a considerable degree, trade reform is thus an integral and essential part of enterprise reform, and there are close linkages between domestic price reforms and trade reform. (This subject is discussed further in Chapter IV.) Export Incentives 1.25 A realistic exchange rate has probably been the most important element of the successful trade strategies of other countries. With the decentralization of export and import decisions to enterprises, the exchange rate will also become an increasingly important element of trade policy and export promotion in China. Indeed, it should become the most important instrument. In general, the exchange rate should tend toward that lgyel for which the country's earnings from the exports of goods and services - are sufficient to cover all its import needs, assuming that most import and export decisions are made by independent enterprises on the basis of profitability considerations. 6/ Plus policy-determined capital in flows (see Chapter III). - 11 - 1.26 A market-based exchange rate can be modified by other policy instruments. For example, if a tariff is imposed to protect domestic enter- prises from import competition, some enterprises or consumers will switch to domestically produced goods, because imports are now more expensive. The demand for imports will thus be less, and the market-clearing exchange rate (in terms of dollars per yuan) will be higher. Quantitative import restric- tions have a similar affect. If, for example, imports of large numbers of commodities are restricted, there will be less total demand for imports, and the exchange rate will be market-clearing at a higher rate (in terms of dollars per yuan) than if there were no such restrictions. But this higher exchange rate also creates an anti-export bias. In many countries extensive quantitative import restrictions are the main reason for an overvalued exchange rate. Instead of letting their exchange rates depreciate to a level which would reestablish balance between demand for and supply of foreign exchange, many developing countries have adopted extensive licensing and other quantitative restrictions, resulting in import-substitution bias and poor export and economic performance, as discussed in the preceding section. 1.27 Of course, China does not have a market-determined exchange rate at present. As discussed in more detail in Chapter II, a significant share of exports and imports is still determined administratively. As in other developing countries, administratiys import restrictions and high tariffs are bound to lead to an overvaluation - of the exchange rate. Furthermore, the uncertainties inherent in administrative allocation lead enterprises to hoard imported goods, to hoard foreign exchange allocations and to pad import requests. This leads to apparent excess demand for imports, or "import hunger", quite similar to the more familiar phenomenon of "investment hunger" in China due to administrative allocation of low-cost investment funds. Thus, it becomes very difficult to determine at what level the exchange rate would settle once administrative restrictions on trade were eliminated or signifi- cantly reduced. 1.28 Some tentative steps in the direction of more decentralized foreign exchange allocation and partial market determination have been undertaken with the introduction and expansion of foreign exchange retention quotas for export enterprises (see Chapter II). But more systematic and wide-ranging measures are required. Some of the options for the transition toward a more efficient exchange rate regime are discussed in Chapter III. While decentralized import and export decisions by independent enterprises responsible for profit and losses, made on the basis of a unified exchange rate, are likely to lead to the most efficient allocation of resources, domestic prices also have to be rational to make decentralized foreign trade decisions efficient. The connec- tion between reforms of foreign exchange allocation and domestic price reform is further pursued in Chapter IV. 7/ Thus, an exchange rate can be judged to be overvalued either when there is an excessive balance-cf-payments deficit, or when restrictions on imports or exports subsidies are required to achieve balance in the market for foreign exchange (see para. 3.5). - 12 - 1.29 Most countries impose tariffs on imports, either to protect domestic industry from foreign competition or to generate government revenues. In many ways, tariffs are preferable to quantitative import restrictions as a tool of protection. Successful exporting countries have tended to rely more on tariffs and less on quantitative import restrictions than countries that have adopted inward-looking (and generally less successful) policies. The reasons for preferring tariffs to quantitative restrictions as instruments of infant industry protection are quite similar to those that lead to preferring price determined rather than administrative allocation of foreign exchange. Price interventions--such as tariffs--are more transparent, create less uncertain- ties for enterprise managers and eliminate the pursuit of favors and windfall gains (such as, for example, obtaining restricted imports at the official exchange rate for resale at a higher price). Such pursuit of windfall gains or "economic rents" is a major source of inefficiency in many developing countries. With a tariff, such windfall gains are avoided, and, moreover, the revenues generated accrue to the government. 1.30 However, the imposition of tariffs or other import restrictions is likely to lead to an overvalued exchange rate, even if the exchange rate is market-determined. Protecting domestic industry--or, more precisely, import- substituting activities--imposes a burden on export industries. The higher the protection extended to import-substituting activities, the higher the penalty on exporters. One solution, adopted by a number of successful exporting countries, is to rebate tariffs on all inputs used in export produc- tion. While this reduces the anti-export bias of tariff protection, it does not fully eliminate it. Some successful exporting countries have provided additional subsidies to exporters to make up for the remaining anti-export bias. However, importing countries have become increasingly sensitive to export subsidies that put their own industries at a competitive disadvantage and have started to impose countervailing taxes. 1.31 The domestic tax system can also lead to an anti-export bias, particularly if turnover taxes are levied on several successive production stages. This occurs, for example, in the case of China's industrial and commercial tax. Many export-oriented countries have therefore reformed their tax systems to minimize anti-export bias. Tariff drawback and value-added tax systems are internationally accepted ways to minimize anti-export bias for exporters, by making imported inputs available at internationally competitive prices, and avoiding undue taxation of export production. 1.32 Some of the most successful outward-oriented economies have completely dispensed with infant industry protection by imposing no or very low import tariffs. Hong Kong and Singapore have followed this str:tegy very successfully. Others have used protective measures in the past and have recently phased out quantitative restrictions and substantially reduced tariff protection. South Korea has recently followed this path and will soon complete a ten-year program of trade reforms that includes a complete phasing- out of import restrictions for manufactured goods and a drastic reduction in tariffs. Only agricultural activities will remain protected by import restrictions, ina;UMtrial activities only by tariffs. Korea's recent export succees suggests that this is at least as good a policy as the mixed export promotion/import substitution policy pursued during much of the 1970s. The - 13 - excellent performance of East Asian economies that have adopted a less restrictive import policy much earlier also attests to that. Sl Maro-conomicManaement 1.33 Another important element in the success of outward-oriented trade strategies has been prudent and stable macro-economic management. This is particularly important during the transition from an inward-looking to an outward-looking strategy. Many developing countries have failed in their efforts to shift towards a more outward-oriented strategy, because macro- economic imbalances led to unsustainable balance-of-payments deficits during the early stages of the transition. These deficits have often been followed by quantitative restrictions on imports and foreign exchange rationing and a return to an overvalued exchange rate, bias towards import substitution and so on. 1.34 The most frequent sources of macro-economic imbalances have been inappropriate fiscal and monetary policies. Governments have been unable or unwilling to finance ambitious expenditure programs with adequate taxation, or have promoted large "strategic" industrial development programs outside the government budget through directed credit and permissive monetary policies. If a major move towards a more effective outward-oriented strategy is to be successful in the future, appropriate fiscal and monetary restraint will be essential. D. Foreign Trade Prospects 1.35 China's "open door" policy still entails a rather substantial import-substitution bias and many other elements usually associated with an inward-looking development strategy. The impressive record achieved by outward-oriented trade and development strategies deserves China's considera- tion. But, is a switch to an outward-oriented development strategy possible and feasible for a large country like China in today's world of slow economic growth and protectionism? What, for example, would happen in international markets if other large countries such as India were to pursue the same strategy at the same time? 1.36 The first point that can be made is that China should not adopt an outward-oriented strategy merely, or even primarily, to maximize export revenues. Internationally traded goods will necessarily represent a small portion of China's total output. The most important reason for an outward- oriented trade policy, therefore, is to reap the indirect benefits of a more open trading system, including the myriad of efficiency gains that enterprises can make when they are in direct contact with buyers, suppliers and compe- titors. Foreign competition provides a spur to the efficiency of domestic enterprises and provides strong incentives t.. keep up with the progress in technology, organization and management practices made elsewhere. The rate of growth of export earnings is far less important than the extent of contacts that production enterprises and commercial units have with the outside world and the effectiveness of the diffusion of the information thus gained to other domestic enterprises. In contrast, the present administratively intermediated foreign economic relations create an "air-lock", in which most of the indirect benefits of external trade are blocked from reaching the domestic economy. - 14 - 1.37 A higher degree of diversification of exports and imports is also likely to carry with it more technology diffusion and higher productivity gains. In all successful outward-oriented economies, labor mobility (within and between enterprises), especially among technical personnel, has been an essential ingredient in this transmission. Overseas education and training have also been important. 1.38 The growth of world output and world trade has slowed down considerably in recent years, and it is reasonable to assume that the high growth rates of the 19509 and 1960s will not return in this or the next decade. The reduction in the responsiveness (or elasticity) of world trade with respect to world economic growth can be largely attributed to the slow growth in consumption and trade of primary goods (Table 1.4). This trend is likely to be lasting, as economic growth in advanced industrialized countries is becoming less and less material-intensive. 1.39 Exports of manufactured goods have held up much better, however, (Table 1.4) and East Asian exporters of manufacturers have continued hy enjoy export growth rates of close to 10% p.a. in volume terms (Table 1.2)., Worldwide, the responsiveness (or "elasticity") of growth of world trade in manufactures with respect to growth of production (and national income) has not markedly changed. During 1980-85 this elasticity 9 was actually somewhat higher than in the 1960s (but slightly lower than in the 1970s). The relatively good export performance of developing country manufactures holds even for those manufactured goods for which producers in industrialized countries have been strongly affected by the increased competition, such as textiles, clothing, footwear, automobiles, steel and so on. While the required structural adjustment in industrial countries-displacement of workers, plant closures etc.--is undeniably resisted and has resulted in a mushrooming of import barriers, market penetration has continued to increase and has reached already very high levels in some product categories, such as garments (Table 1.5). 8/ Prices in dollar terms fell slightly during 1980-85. 9/ i.e., the rate of growth of world trade volume divided by the rate of growth of world production. - 15 - Table 1,4: GROWTH OF WORLD PRODUCTION AND TRADE VOLUME Production growth (1 p.a.) Export growth (X p.a.) 1960- 1970- 1980- 1960- 1970- 1980- 70 80 85 70 80 85 Agricultural products 2.5 2.2 2.6 3.9 3.5 1.2 Minerals and fuels 5.4 2.7 -2.6 7.2 1.7 -3.6 Manufactured goods 7.4 4.3 3.1 10.2 7.1 4.8 Source: GATT: International Trade 1985-86, Table A.1. 1.40 The share of primary commodities in China's exports has changed little in recent years, remaining at about 50%. But, within this 501, there have occurred significant changes in composition. The share of non-oil primary commodity exports has falles.. from 50% of total exports in the early 1970s to 26X in 1985, and the share of agricultural commodities fell somewhat more rapidly (to 14%) than the share of other primary commodities. The good performance of primary commodity exports in total can be entirely attributed to the rapid growth of petroleum exports, a trend already changing as a consequence of falling petroleum prices. In the future, it is quite possible that China's exportable surplus of domestic oil production will stagnate or even decline, given rapidly growing domestic demand and only moderate exploratiQn results. 1.41 While China's export performance in manufactured goods was respectable during the past eight years (Table 1.2 and Annex 1), it was highly variable, with manufactured exports increasing rapidly between 1978 and 1981, then growing only modestly between 1981 and 1985. After the exchange rate adjustments vis-a-vis the US dollar and, even more importantly, the decline of the US dollar vis-a-vis other currencies, exports of manufactures improved again, increasing by about 30% (in terms of US$) in 1986 (see Annex 1). Textiles and clothing account for about 40% of manufactured exports, followed by miscellaneous exports (handicrafts, toys, etc.) with 25%. Machinery, electrical and electronics products are only 6% of manufactured exports (or less than 31 of total exports). - 16 - Table 1.5: MARKET PENETRATION OF MANUFACTURED IMPoRTS IN SELECTED INDUSTRIALIZED COUNTRIES, 1975-83 (t of apparent consumption) Total Nonelectrical Manufacturing Clothing Machinery 1975 1983 1975 1983 1975 1983 All Imports Us 7.0 10.3 9.8 20.3 7.3 11.8 Japan 4.9 5.3 8.3 13.0 5.5 4.3 West Germany 24.2 35.1 44.9 73.2 30.4 39.6 Imports From Developing Countries Us 2.0 3.4 7.7 17.8 0.5 2.3 Japan 1.7 1.9 6.2 10.1 0.4 0.3 West Germany 2.4 4.0 14.7 28.7 0.9 1.2 Source: OECD Compatible Trade and Production Data Base. 1.42 One explanation for the continued market penetration in manufactured goods, despite nontariff barriers in industrialized countries, is that these trade barriers are far more porous than they appear. Inventive exporters often find ways around these restrictions, for example by shipping the goods in a form or with components not covered by the restrictions. Coats with removable sleeves that are imported as vests (which are not covered by the testrictions) come to mind, or woolen sweaters that are now made in silk/alpaca/angora blends that are not subject to restrictions. Decentraliza- tion of marketing decisions to enterprises appears to be essential, if such market opportunities are to be pursued effectively. At the same time, China will have to make strong efforts to shift to new and different export products, especially machinery. But to become successful in these more sophisticated products, quality and design are essential, and it is unlikely that China can become a successful exporter in these product categories, unless export decision-making is decentralized to enterprises, and the trading system becomes more open to permit more rapid technology acquisition and diffusion. 1.43 An important trend that could help to stave off increased protec- tionism in industrialized countries is the greater interest-and more active involvement of developing countries in the General Agreement on Tariffs and Trade (GATT). While developing country interest was very limited during the 19609 and even 19709, and largely focussed on obtaining special treatment and on limiting reciprocity under GATT, many developing countries now realize that they have a stake in the preservation of an open trading system and ars very actively participating in the new Uruguay Round of multilateral trade negotia- tions. They also realize that they will have to accept more reciprocity if the trading system is to be kept open. - 17 - 1.44 Recently, China formally indicated her intention to resume member- ship in GATT. Given the changed trading environment, it will probably become necessary for China to accept more reciprocity in GATT than other developing countries had to accept in the past, if China wants to have the better assured access to major export markets that participation in GATT provides. External trade reforms that would enable China to reap the many benefits of an outward- oriented economic strategy would also serve to make China's trade system more transparent and acceptable to other GATT members. - 18 - II. CHINA'S FOREIGN ECONOMIC RELATIONS: ISSUES AND POLICY OPTIONS 2.1 In light of the lessons of international experience outlined in Chapter I, this chapter presents a description and analysis of China's system of foreign economic relations. Part A outlines the key features of the trade regime, both historically and as of October 1986. Part B provides an analysis of the impact of the current regime. Based on this analysis, Part C recom- mends a set of institutional reforms and Part D a set of policy reforms that would have to be implemented concurrently if the institutional reforms are to succeed. Finally, in Part E we analyze other issues in China's foreign econ- omic relations, notably the management of technology imports and foreign investment. A. Key Features of the Chinese Trade Regime The Historical Context 2.2 During the 1949-78 period, China's economic strategy was essentially one of self-reliance. This was reflected in the approach to trade and the trading system adopted. Domestic needs were to be satisfied as much as possible by domestic production, and foreign trade was to supplement short- falls in such production--primarily in food, essential raw materials, and capital goods. Nevertheless, there was always a recognition that China could not be entirely self-sufficient, especially in capital goods, and that progress would be served by obtaining foreign technology either by purchase or through assistance from friendly countries. However, the enthusiasm for technology imports varied over the period, and the trends in such import activity are generally referred to as the "four waves" of imports (see Annex 1, paras. 1.4-1.7). 2.3 Although these "four waves" resulted in considerable levels of tech- nology imports, they were generally characterized by the purchase of embodied technolggy: i.e., whole plants or production lines acquired on a turnkey basis." This was consistent with a general approach of isolating the domestic market from the "destabilizing" influence of world markets, which characterized trade policy during this period. The level and composition of trade were determined through a centralized planni system under the authority of the Ministry of Foreign Trade (MFT).A The annual foreign trade 1/ The only significant exception to this was during the 1956-60 period, when significant numbers of plants were imported from the USSR. These imports were generally accompanied by blueprints and technical assistance and were replicable and more easily maintained. Many of these plants are still in operation. 2/ After a government reorganization in March 1982, the Ministry of Foreign Trade was combined with the Ministry of Foreign Relations, and was renamed the Ministry of Foreign Economic Relations and Trade (MOFERT). - 19 - planning process would identify the required imports, and exports would be selected to finance the desired level of imports, so as to avoid any signi- ficant trade deficits that would require foreign borrowing. Once the plan was determined, it would be entrusted to the twelve main centralized foreign trade corporations (FTCs) for implementation. To ensure the insulation of the domestic market, the FTCs were given monopoly powers within specified scopes of business, and they would procure all goods for exports and sell all procured imports at the domestic prices for these goods (Box 2.1). Trade plans were expressed in quantitative terms, and any resulting financial losses for the FTCs were covered by budgetary grants (see Annex 1, paras. 1.8-1.10). Box 2.1: THE PRE-1978 TRADING ENVIRONMENT IN CHINA Since 1978, China's trade volume has more than doubled and some US$8 billion in foreign investment has been made. These statistics are indicative of the remarkable change in attitude to trade and the role of the outside world since 1978. Before 1978, contacts were extremely limited, with virtually no trade business possible between the twice- yearly Canton Trade Fair. No officials carried business cards and would only identify themselves by surname and corporation. At the Trade Fairs, foreign businessmen would meet with the FTCs to do business. "Between the Spring and Autumn Canton Fair, foreign trade corporation managers were loath to answer letters or telexes, so the Fair assumed enormous importance. On the opening day crowds of visit- ing business people from all over the globe gathered at the steps of the trade fair complex, and once the opening ribbon was cut, a frenzied stampede took place." Another key feature was the extreme 'airlock' imposed by the FTCs between the foreign supplier and the Chinese end-user. "Ultimate end-users were not only not present for commercial and technical negotiations, but were generally not even identified. Apart from the time factor, the potential for misunderstanding was enormous. Many companies that sold technical products during the early or mid-1970s found out in later years, when they were allowed access to their earlier sales' end-users, that huge waste had resulted, which sometimes reflected unfairly on their products' reliability and their reputa- tions. From: Thomas D. Gorman's "China's Changing Foreign Trade System, 1975-85," in Robert Delfs and Thomas Gorman and Owen Nee: "China", Euromoney Publications, London 1986. - 20 - 2.4 The "open-door" policy was officially launched in December 1978 at the Third Plenary Session of the 11th Central Committee of the Chinese Communist Party, as one component of the overall economic reform program launched at that time. However, the first clear indication of the government's intent to "open the door" came in July 1979, with the adoption of the "Law of the People's Republic of Chins on Joint Ventures Using Chinese and Foreign Investment", which thus permitted foreign investment in China for the first time since 1949. 2.5 From 1979 to 1985, the trade system evolved rapidly. The central- ized FTCs lost their monopoly powers, and their branch offices began to oper- ate as separate units. In addition, provincial authorities created their own FTCs to fulfill provincial export aspirations, and line ministries found it convenient to establish corporations to engage in trade in their products directly. A system of foreign exchang, retention at the provincial and enterprise level was also introduced,- and, when combined with the intro- duction of import licensing and an initial decentralization of license-issuing authority, this period saw a rapid change in the locus of trade activity. From being a residual activity carried out in a highly centralized manner, trade became, during this period, a central focus of effort, with provincial authorities and institutions in the vanguard. These various reforms in trade practices over the period were summarized in the September 1984 document on the reform of the trade system adopted by the State Council (see Box 2.2). 2.6 Perhaps the most important changes during this period were in the scale of activities, as trade doubled, and in the general attitude towards trade. From being a slightly suspect activity (see Box 2.1), foreign trade became one of the most desired and respected occupations. Nevertheless, many trade practices, such as pricing and the role of planning, remained essen- tially unchanged. The Trade Regime in October 1986 (a) Foreign Trade Planning 2.7 The foreign trade plan remains central to the trade regime in China, albeit with three important changes from the pre-1979 system: it is now driven more by export possibilities than by import requirements; it was a top- down process until 1979, but has become more of a bottom-up process, with 3/ Foreign exchange retention was introduced in 1979, but was formalized in January 1984 with standard rates, and was subsequently modified in January 1985 such that the retention rights were shared equally between the enterprise exporting and provincial authorities. 4/ For readers not familiar with Chinese trade institutions and policies, a more detailed description is provided in Annex 1. It should also be noted that all figures quoted in this section are World bank estimates based on interviews at the enterprise level, and the foreign trade plan remains unpublished. - 21 - Box 2.2: MOFERT'S REPORT ON THE REFORM OF THE TRADE SYSTEM OF SEPTEMBER 1984 On August 14, 1984, MOFERT submitted a report to the State Council on Reform of the Foreign Trade System, which was approved by the State Council on September 15, 1984. The State Council urged all responsible units to "carry out these reforms vigorously ... step by step in a planned way." There were five key elements of reform identi- fied by MOFERT: (a) Separate Government Functions from Enterprise Management and Strengthen Administrative Management of Foreign Trade. Administrative departments were to leave all day-to-day functions related to trade to the foreign trade enterprises. This would leave MOFERT free to concen- trate on such areas as drawing up regulations, formulating long term plans, advising on the appropriateness of economic levers related to trade, conducting negotiations with foreign countries, and conducting research and training. (b) Simplify Administration and Transfer Power to Lower Administrative Levels and Bring into Full Play the Managing Initiative of Various Foreign Trade Enterprises. "Foreign trade enterprises of all types should become independent of their original administrative depart- ments, whilst keeping their own accounts, assuming responsibility for their own profit and losses and developing in the direction of special- ization and socialization. There must be a large number of small and medium sized companies and enterprises to participate in foreign trade. Big production enterprises may handle foreign trade business subject to approval." (c) Adopt Import and Export Agency System and Improve Operations and Management of Foreign Trade. "By import and export agency system, we mean that foreign trade enterprises should provide diversified services, handle import and export business as entrusted by production units and order-placing departments and collect service charges, while profits and losses are to be borne by the entrusting unit." This would "facilitate the combination of foreign trade enterprises with production units" and "raise economic efficiency." (d) Reform the Foreign Trade Planning System and Simplify the Contents. This was aimed at "transferring more decision-making power of management to the foreign trade enterprises and production units," so that they could be more flexible and responsive to changing market needs, and "fulfilling the state plan with the maximum economic return." (e) Reform the Foreign Trade Financial System and Strengthen the Economic Means of Regulation. This required the elimination of the state's responsibility for profits and losses in trade. The foreign trade enterprise should pay taxes instead of turning over profits to the state. Source: Almanac of China's Foreign Economic Relations and Trade, 1985, pp. 388-398. - 22 - estimates of export availability and import needs made at the local level; and we estimate that it now covers only about 70% of trade. The planning system is described in detail in Annex I (paras 1.32-1.57). One key feature of the planning process that has remained during this period of change is that it is essentially a negotiating process between the local and national levels of administration. 2.8 The export plan has two components: the command plan and the guidance plan. The command plan covers some 120 commodities (compared with 3,000 in 19'') and we estimate that it accounts for about 70% of the plan in value terms (or 50-60% of all exports). These include such commodities as oil, coal, agricultural products, handicrafts, textiles and garments. The guidance plan covers the remaining 30% of the plan, or about 20% of all exports. The command plans are mandatory, are stated in physical quantities and are very specific, and producers receive inputs for export production under the materials allocation system. In contrast, the guidance plans contain value targets assigned to provincial authorities and FTCs, which have considerable flexibility in determining how to achieve them. 2.9 The import plan has undergone much more reform than the export plan. The import command plan in general now covers only the seven key raw materials 5 which are under "unified management" (albeit comprising an estimated 40% of all imports) and is closely related to the traditional role of filling shortfalls in domestic production for the materials allocation system. The guidance plan (covering about 30% of all imports) is essentially a foreign exchange allocation mechanism covering essential imports for invest- ment projects and of raw materials, spare parts and capital goods. The balance of imports (30%) are financed either out of foreign exchange retention by provinces and enterprises or by foreign borrowing by noncentral authori- ties, and are controlled by the import licensing system (see paras. 2.13- 2.15). (b) Foreign Trade Corporations and Trade Pricing 2.10 It has already been noted that, prior to 1978, all trade was con- ducted through twelve centralized FTCs. In 1986, it remained the case that nearly all trade was conducted by FTCs, but by a much larger number of them. Only Sino-foreign joint ventures and a few Chinese corporations have been granted direct trading rights. In 1981, FTCs (and their branches) under MOFERT controlled 89% of trade, but by 1984 this share had fallen to 72%, as a result of the proliferation of FTCs created under ministries other than MOFERT and uy provinces. The total number of FTCs is unknown but is at least 1,200. Moreover, provincial branches of national FTCs became separate financial and operational entities. Each FTC has a prescribed "scope of business," outside which it may not operate. Most FTCs continue to be specialized in particular commodities or commodity groups, except for new FTCs established by provincial authorities, whose scope of business is basically 51 The "unified management" commodities are steel, chemical fertilizers, rubber, timber, tobacco, grain, and polyester and other synthetic fibers. - 23 - defined as "trading in anything for which there is no FTC with monopoly power. 2.11 The FTCs conduct all trading activities, from the identification of markets and negotiation of contracts to the procurement of goods and their shipment. For exports, the large part of procurement takes place at domestic prices, which can be either fixed, floating, or free-market. Because FTCs' primary concern is to achieve overall export targets, they are not concerned to ensure that each individual export activity is profitable, or that each branch is profitable, provided that export targets overall can be achieved with given budgets. 2.12 The import system is very different from that for exports. Command plan imports and imports for major investment projects are assigned to specific FTCs, but for all other imports end-users can select an appropriate FTC. However, the more significant distinction relates to pricing. Whereas exports continue to be procured in general at domestic prices, the agency system is used for all noncommand plan imports; i.e., the price to the end- user is the import price plus the FTCs' costs, unless this would be lower than the fixed domestic price, in which case the domestic price is charged. Of course, losses are incurred in importing certain "unified management" commodities--such as steel and fertilizers, where the domestic prices are lower than world prices--but FTCs are compensated directly for such losses. (c) Import Licensing and Tariffs 2.13 As the trade reforms have progressed, the import licensing regime and the customs tariff have emerged as important economic levers, used both to control "excessive" import growth and to protect domestic industry. If a good is defined as restricted under the January 1984 regulations (see Annex 2, paras. 2.20-2.30), it requires an import license. As of October 1986, there were 45 goods in this category, which accounted for about 30% of' all imports. The list has grown over time, in part as a reaction by the Government to actual import trends. 2.14 Major or sensitive imports (command plan goods, assembly lines, major consumer goods) are handled centrally by MOFERT, and other goods locally (see Annex 2, Table A2.1). For restricted imports not included in the Plan, an application must go through several layers of administration, and can be rejected or cancelled after approval for a variety of reasons. Even if approved, the State Administration for Exchange Control (SAEC) may not permit access to foreign exchange rights, as happened for several months in 1986 for computer imports. 2.15 Until the adoption of the open-door policy, the practical purpose of the customs tariff was to raise revenue, as the planning mechanism was the primary determinant of the level and structure of imports. Tariffs are now becoming important economic levers in influencing import decisions. The principal purpose of the tariff, in addition to its revenue-raising function, is to protect and promote domestic industry, as well as to discourage "waste- ful" imports. Tariffs (including temporary surcharges called the "regulatory tariff") range from 0-200%, with the highest rate applying to motor vehicles, - 24 - and the zero tariff to a limited number of essentials such as grain and live animals and plants. We estimate that the average rate is 38X, cascading in the usual way--27X on intermediate products/raw materials, 311 on capital goods and 63Z on consumer goods. The actual average rate of collection is, however, only 12-16X of import value, because of various exemptions, mostly on capital goods for technical transformation, joint ventures, and for the special economic zones. The potential impact of the tariff structure in terms of effective protection is discussed in Annex 2, paras 2.69-2.81. (d) Incentives for Exporters 2.16 The trade planning and import regime described above is one that offers considerable protection for domestic industry and, consequently, creates a high degree of anti-export bias within the economy. To counteract this bias, several forms of export incentives have been created. By far the most significant of these is the foreign exchange retention mechanism. Under this mechanism, 25Z of all foreign exchange earned is retained at the provincial level, with equal shares going to the enterprise producing the export and to the provincial authorities. It is also reported that these retention rights can rise to as high as 70X for above-plan exports, and are also hiiher for certain provinces, such as Fujian, Guangdong and Inner Mongolia. 2.17 Other export incentives are offered through the FTC system. Firstly, there is a more or less formal system of performance bonuses in direct proportion to performance against plan targets. For each US$ earned, if the plan target is achieved, the bonus is 3 fen, and for each US$ earned above the target, 10 fen. (Although such bonuses are targeted for producing enterprises, they are paid through FTCs and it is not clear to what extent they are passed on.) In addition to these bonuses, FTCs--rather than enterprises--carry any losses on individual exports, balancing such losses with profits made on other exports. Thirdly, it is reported that some provincial authorities also provide subsidies to thei'? provincial FTCs in order to generate higher local foreign exchange retention rights. 2.18 A third type of incentive is provided to so-called export production bases; i.e., production facilities geared exclusively to exports under the control of FTCs or ministries. Various incentives may be offered to such facilities, such as low interest or grant financing, tax rebates, and preferential access to foreign exchange and required inputs. FTCs often arrange "compensation trade" financing of capital equipment for such facili- ties (see Annex 4, para. 4.28). The Government is considering expanding the production base system. This would offer to Chinese enterprises similar incentives to those offered to foreign investors in joint ventures (see Annex 4, Table 4.8). B. The Impact of the Current Trade System 2.19 The trade system has undergone considerable reform in recent years, which is reflected in the volume and structure of trade (see Annex 1) as well as in the dramatic increase in the number of participants in foreign trade. However, further expansion of trade may now be more difficult, unless steps - 25 - are taken to improve the efficiency of the present trade system. This section assesses the main problems that remain to be addressed in the current trade regime. The four main issues identified are: (a) the "air-lock" that has been created between producers and the world market; (b) the limitations of the foreign exchange allocation mechanism; (c) the degree of protection given to domestic industry; and (d) the absence of automatic and effective export incentives. (a) The "Air-Lock" 2.20 We use the term "air-lock" to describe the buffer that has been created between Chinese enterprises and world market forces. The FTCs are the main point of contact between China and the world market, and Chinese enter- prises are to a large extent insulated from world price developments and have very limited contact with foreign buyers and suppliers. Indeed, this isolation was an explicit aim of the trade system as originally constructed, as world market forces were viewed as irrational and potenti&lly harmful to national interests. Although much has changed since 1979, the air-lock still exists to a considerable extent. 2.21 Its existence derives from the rules for the operations of FTCs and from limited trade rights. Each FTC has a fixed scope of business, outside of which it may not trade. This severely limits the extent of competition in tradigg, although the degree of monopoly varies a great deal between subsec- tors.-I At the same time, producers have no choice but to export through one of the FTCs, as only a small number of enterprises (except Sino-foreign joint ventures) have been given direct trading rights. Thus, for most producers there is no choice about whether to export directly or through an agent, and very limited choice about which agent to use. 2.22 The FTC system also permits a procurement pricing system to exist which isjffected by world prices to only a limited extent. Whilst noncommand plan imports do now reflect world prices, most export procurement prices offered by FTCs are the same as the prevailing domestic prices. 2.23 The air-lock creates four problems, which may not have been critical in the early years of export development, during which China rapidly expanded traditional exports such as oil and standard textiles in which it has an obvious comparative advantage, but which will become increasingly important as China seeks to expand exports of new, less homogeneous products. (i) Inefficiency of Exporting. Under ths current system, exports are made in an economically inefficient way. The domestic costs of earning foreign exchange through exports varies widely, either because FTCs are able to subsidize many exports, 6/ For example, only the China National Textiles Import and Export Company (Chinatex) is permitted to trade in textiles, and it rigidly allocates quotas to its provincial branches. In contrast, there can be considerable competition for procurement of processed foods. - 26 - by cross-subsidy from profitable exports or imports, or because of distorted domestic input prices. Thus, some exports may be financially profitable but economically inefficient. For example, low electricity and steel prices would tend to encourage steel products exports, which may not be viable if input prices reflected world market values. tii) Inappropriate Export Patterns. Because world prices are not passed on to producers, or used as the primary determinant of which products will be exported, the pattern of exports will probably not correspond to China's comparative advantage. The production of profitable exports will not rise, nor will production of unprofitable exports fall. Similarly, because changes in world prices are also not passed on, the Chinese economy is prevented from reacting to changing external market conditions. The essential problem is that the signals from the world market are generally blocked at the level of the FTC. (iii) Lack of Information. The air-lock of the FTC between the producer and the market also has serious consequences for the flow of information. Firstly, there is the problem of reacting in a timely fashion to market opportunities. If all export opportunities must be processed by the FTC, it may take so long to pass the necessary information on to a producer, that, by the time there is a response, the opportunity has gone. There are examples of this in China (see Box 2.3). Secondly, the absence of contacts between foreign buyers and sellers and Chinese enterprises means that these enterprises are foregoing much "free" technical assistance concerning such areas as product specifications and design, new product development, capital equipment, factory design, quality control, etc. Experience elsewhere has shown such direct contacts bUt-_-l.r foreign buyers and suppliers and all kinds of domestic enter- prises to be one of the most important, most effective, and cheapest sources of technology transfer and marketing know-how (See Box 2.5 and 2.8 below). Moreover, as China develops exports of more sophisticated manufactures, including capital goods, direct contacts with foreign purchasers will also be important for the development of such skills as after-sales services, and installation. While FTCs attempt to obtain such information and to pass it on to enterprises, this becomes increasingly difficult, as exporters become more numerous, and exports become more diversified. (iv) Lack of Competition from Imports. The final problem is that the air-lock prevents imports from performing one of their most important functions (para. 1.7), that of stimulating efficiency in domestic production by ensuring the maintenance of competitiveness. - 27 - (b) The Foreign Exchange Allocation Mechanism 2.24 Foreign exchange allocation mechanisms have two essential, over- lapping functions: they allocate foreign exchange among the various potential users, and they determine a price at which foreign exchiange is traded. The importance of this latter function cannot be overemphasized and has two major effects. Firstly, the exchange rate alters the relative attractiveness of producing a particular product for export markets or for the domestic market as well as the ability of domestic production to compete with imports. In addition, if changes in the exchange rate are permitted to have an impact on domestic prices, the relative prices of tradeable goods and nontradeables (i.e. nontradeable services such as power, domestic services, restaurants) will be altered, thus influencing the production and investment decisions of enterprises. Thus, the exchange rate is of key importance in determining a country's balance-of-payments situation. 2.25 The pricing and allocation functions of the exchange rate are overlapping in those economies in which its determination is left to market forces of supply and demand, because then it is the price of foreign exchange that determines its allocation; i.e., potential users of foreign exchange will only choose to acquire 't for uses whose value justifies the price paid for it. Many countries, including China, do not permit market forces to determine foreign exchange allocation, but instead allocate it administratively. As already noted, 252 of foreign exchange earnings are allocated in equal shares to the provinces and enterprises which generate them, with most of the balance being allocated directly by the state plan for imports, debt services, reserves accumulation, etc. Moreover the exchange rate is determined adminis- tratively (options for alternative ways of doing this are discussed in Chapter III). It is partly because China does not permit the exchange rate to move in such a way as to equalize demand and supply of foreign exchange that it is necessary to allocate foreign exchange administratively, since, at the present price of foreign exchange, there is inadequate supply to meet demand. In China, this is compounded by the absence of strict budget constraints on enterprises, which dampens further the impact of any rise in the exchange rate on the demand for imports. 2.26 The foreign exchange retention system for provinces and enterprises has been established as a means of offsetting, at least partially the anti- export bias in the trade system, and is used as an export incentive. In a situation in which the market clearing function of the exchange rate is suppressed, resulting in excess demand for foreign exchange, the ownership of and permission to use foreign exchange is highly valued. Moreover, the scheme provides exporters relatively more assured access to imported inputs and enables them to reduce, to some extent, the heavy administrative costs of applying for foreign exchange through the regular channels. - 28 - Box 2.3: THE EFFECT OF DIRECT TRADING RIGHTS ON A CHINESE ENTERPRISE The No. 2 Abrasive Wheel Factory in Henan Province is a long- established producer in China, accounting for a large share of domestic market production. some 301 of its output is exported to 69 different countries. Until March 1986, all its exports were arranged by the China National Machinery Import and Export Corporation (CMIEC). In March 1986, it obtained direct trading rights. In our discussions, factory managers gave three reasons for seeking these rights: (a) they wanted direct contact with customers; (b) they wanted to avoid unnecessary intermediaries; and (c) they wanted more direct access to world market information. When asked about the effects of having direct foreign trading rights, factory managers gave a good example showing how important direct foreign trade rights and contacts are to export success in diversified manufactures. A Japanese customer recently telexed directly an inquiry regarding 100,000 tons of abrasive materials, involving 12 different specifications and different delivery dates. The factory was able to telex back a price quote the same day and was awarded the contracts within 24 hours. When asked what would have happened if exports had still to be arranged by CMIEC, the factory manager said that CMIEC would not have had enough information about sourcing the order; it would have taken two weeks to one month to reply to the inquiry, and, by that time, the order would have probably been lost to a competitor in another country. It should also be noted that the factory finds several of the contracts it inherited from CMIEC to be unprofitable. As soon as these are fulfilled, they will be terminated, and only profitable exports will be made. (The company's calculation of profit includes a premium that it attaches to retained foreign exchange.) 2.27 For provincial authorities, this incentive is very powerful, as the possession of foreign exchange increases provincial independence and enables them to carry out some projects without central approval and funding. This independence is so valued by provincial authorities that some have found it worthwhile to supplement national export incentives (see paras. 2.32-2.35 below) with incentives of their own. For example, as already noted, it is reported that some provincial aud municipal authorities provide subsidies to their FTCs so as to encourage increased exports and augment provincially retained foreign exchange. 2.28 The retention scheme has undoubtedly encouraged the development of industries which can compete internationally, and it has permitted exporters to undertake technical transformation projects. It is also a welcome recogni- tion on the part of the authorities that foreign exchange is important and merits policy consideration. However, there are severe limitations to the - 29 - present foreign exchange allocation mechanism, and the retention scheme in particular, both in terms of promoting efficiency, and in encouraging exports and avoiding balance-of-payments deficits: (i) Its most serious shortcoming is the same as that of the FTC system (para. 2.23(i) and (ii)) in that it does not provide signals as to where China's comparative advantages in foreign trade lie, because the "air-lock" insulates the effects of changes in exchange rates from enterprises. FTCs do find exporting more profitable when the exchange rate is devalued, but, even so, the effect of an exchange rate change is severely muted. Enterprises which can earn foreign exchange without difficulty, such as textile producers, but have relatively small foreign exchange needs, have little incentive to export more or to upgrade the quality of their export production, because they derive little benefit from additional foreign exchange earnings. On the other hand, producers who can make substantial profits on the protected domestic market are induced to export, even at a financial loss, in order to be able to finance import requirements. (ii) The power of the incentive implied in foreign exchange retention quotas is seriously diminished by limitations on their tradeability. If retention rights were tradeable, this would create stronger export incentives for producers with surplus retention rights. (iii) Another major drawback is that, whilst the retention system offers a major incentive to export under current conditions, its value cannot be measured and varies widely for different types of exporters, as noted above. . ta., - (iv) Given the scarcities of foreign exchange, the current retention system encourages provincial authorities to establish produc- tion facilities geared for export production, but, as the system is not giving signals about which facilities to estab- lish, it may be encouraging the development of exports which are not justified from an economic point of view. (v) The system as currently operated makes the control of foreign exchange usage more difficult, as was witnessed in 1985, when at least some part of the rapid import expansion can be attributed to the unforeseen usage by provincial auth'^-ities of retention rights accumulated in previous years, despite the import planning and licensing system. (vi) The retention system only provides foreign exchange to the final exporter, and not directly to the indirect exporter who provides inputs to the final exporter. While some informal mechanisms exist to overcome this, there is no established foreign exchange incentive sharing procedure. - 30 - (vii) Finally, the incentive is also diminished by uncertainty over the availability of foreign exchange, and by the possibility of losses resulting from exchange rate changes between the date of the acquisition of the right and the date of usage. (c) Protection of Domestic Production 2.29 Protection of domestic production from foreign competition (e.g., through a tax on imports) acts automatically as a disincentive to exports (i.e., like a tax on exports). Excessive protection usually leads to very substantial losses in economic efficiency (see Chapter I, para. 1.7). China has tended not to regard imports as a source of beneficial competition for domestic industry, but rather as a complement to domestic production, filling domestic shortfalls and supplying technical expertise (usually embodied in machines) not available domestically. Protection in the Chinese trade system is provided through four mechanisms: trade planning, import pricing, licens- ing and tariffs. For command plan commodities, the plan determines the level of imports and ensures that import prices will not undercut domestic prices. This is supplemented by a licensing system which provides highly variable levels of protection by quantitatively limiting specific imports. In addition to offering protection, the increasingly important licensing system also arbitrarily awards economic rents to license holders, carries high administra- tive costs, and offers inducements for corruption. The tariff system itself also offers a high (and variable) degree of protection, as was discussed earlier (para. 2.15). 2.30 All these elements of protection ensure that production of goods for the domestic market in China is more profitable than export production and considerably more so for some products than for others. Even a uniform level of protection (provided, for example by a uniform tariff rate on all imports) reduces the level of exports and imports and thus the rate of technology transfer from abroad compared with a zero tariff situation. The quantitative restrictions and highly variable tariff rates presently found in China imply, in addition, significant distortions in domestic resource allocation as a result of the variable (and almost unpredictable) protection provided to domestic industries. In many countries, such a variable pattern of protection has led to highly inefficient industries and monopolistic practices. 2.31 In China, distortions arising from protection may not have been a major problem in the past, since pricing and investment decisions have been made administratively. However, China's economic reform will make these issues more important, as enterprises are granted greater freedom in pricing, production and investment decisions. If the present trade regime is maintained as these reforms are introduced, resources will be drawn towards those more heavily protected sectors--consumer goods, light machinery--and away from less protected sectors--agriculture, mining, intermediate products, capital goods. These problems may not be as serious in China as in smaller countries, because China's large market for nearly every product could make competition among domestic enterprises a potent force in reducing the worst effects of variable import protection. Nevertheless, economic efficiency could be significantly reduced. In particular, resources would be attracted way from export-related production. - 31 - (d) The Encouragement of Exports 2.32 For many of today's exports from China, the decision to export is either administrative--in the command plan or by an FTC--or else it is taken by a producer or province in order to gain access to foreign exchange. Exporting is regarded essentially as the means of financing a desired level of imports without recourse to excessive external borrowing. This is, of course, the fundamental purpose of exports, but China has created a system in which the imports to be made and the individual exports to finance them are deter- mined to a significant degree by administrative decision, rather than permitting the volume and composition of imports and exports to be determined by economic levers. 2.33 As explained in para. 2.23, this approach may lead to "inefficient" exporting. Moreover, as enterprises (including FTCs) become fully responsible for profits and losses, it may become increasingly difficult to achieve the desired level of exports, as unprofitable exports are redirected to the more profitable domestic markets. The current system of export inducements (including performance bonuses) help to compensate for the anti-export bias in the system, but carry a fiscal burden (see also Box 2.4). 2.34 The strongest export incentive offered in China is the foreign exchange retention system as discussed in paras. 2.24-2.28. Although problems associated with the retention system have been noted above, it is nevertheless a powerful incentive. This power, however, derives from the weakness of the foreign exchange allocation system. If that system were able to satisfy China's foreign exchange needs, there would be no need for an incentive, whose power lies in the fact that it avoids the pitfalls of the allocation mechan- ism. Under alternative foreign exchange allocation systems (see Chapter III), the power of the incentive provided by the retention scheme would disappear, especially if that system included a market-based exchange rate. 2.35 The third set of questions about the present incentive regime concerns the "export production bases." This is a more general issue in connection with export planning. Experience in many countries has proved the difficulty of "picking winners"; i.e., it is very hard to guess in advance what the sectors of most rapid export growth will be. Of course, it is easy simply to plan for further growth of existing exports, but if this is where all the incentives go--which seems to be happening at present in China--many potential exports will be missed. If there is no generalized mechanism available, whereby exporters--not just the pre-selected ones--can get the incentives, then the potential new exporters will be discriminated against and will not be able to fulfill this potential. - 32 - Box 2.4: EXPORT PROMOTING SUBSIDIES AND GATT Ideally, shifting to an outward-oriented trade strategy can best be accomplished by removing existing trade barriers, accompanied by appropriate exchange rate adjustments. In practice, many developing countries have attempted to achieve a similar goal by introducing export incentives--including subsidies--without dismantling import barriers. In justifying this second approach, it has been argued that import liberalization can take place only slowly because of vested interest groups; exchange rate adjustment may cause domestic inflationary pressures; import taxes constitute a major source of government revenues and cannot be reduced; and the use of export incentives without dismantling import barriers encourages both import substitution and export expansion. Export subsidies, however, have in recent years increasingly come under the threat of countervailing actions by some industrialized countries. Even domestic subsidies not directly related to exports have become subject to scrutiny and retaliation. In the US alone, 117 subsidy-related countervailing duty actions were initiated during 1980- 85, compared to 45 during 1975-79 and only 2 during 1970-74. Since subsidy-related countervailing actions could possibly be abused and used as a protectionist device, a number of developing countries have dismantled export subsidies that are not acceptable under the GATT rules and have joined the GATT Subsidies Code and/or entered into bilateral agreements to obtain better protection against sucn actions. (Under the Subsidies Code and national legislation in the case of the US, material injury has to be proved before such actions can be taken). Promoting exports by an export subsidies-cum-import barrier scheme--a policy option successfully implemented by some developing countries during their transition from inward to outward-oriented trade strategies is thus no longer a politically feasible second-best alternative to an import-liberalization cum currency devaluation scheme. From: Nam, Chong-Hyun: "Export Promoting Policies Under Countervailing Threats: GATT Rules and Practice", World Bank, Discussion Paper. VPERS 9, 1986. - 33 - C. Reforming Trade Institutions Introduction 2.36 With its enormous resource base and population, China could eventually produce and export a wide variety of goods and services. The question is what to produce, and what to import and export, and how these decisions should be made. Such decisions become increasingly complex as an economy develops, and international experience suggests that they can be made most efficiently by enterprise managers guided by market prices (modified, if necessary, by "economic levers"). In foreign trade, it will be important to establish a policy framework that will promote efficient decisions on trade and production. (Inevitably, such a trade regime will inter-relate with other areas of reform, notably enterprise reforms, pricing policies and credit policies, and these relationships and sequencing issues are discussed in Chapter IV.) 2.37 The recommendations that follow would move China towards a regime that has worked successfully for many of the world's most rapidly developing countries, although the recommendations have been adapted to the particular circumstances of China. Such a system could be compatible with a planned commodity economy and could be implemented without serious dislocation to domestic production activities, although careful phasing and monitoring would be needed. In this chapter, the discussion focusses on medium-term reform goals in the area of foreign trade, and discussion of phasing, immediate steps and links with other reform areas is left to Chapter IV. The recommendations are essentially an elaboration of MOFERT's recommendations on trade reform, adopted by the State Council in October 1984, and the principles for the reform of the Customs Tariff, as adopted in March 1985. They would also be consistent with other economic reforms being adopted in China that stress the role of economic levers in place of administrative allocation. 2.38 A reformed foreign trade system would operate essentially through economic levers, and these levers would in turn depend on appropriate macro- economic policies. In particular, exchange rate policies, domestic monetary policies, and foreign borrowing policies would need to work in a complementary way. The trade policy recommendations which follow fall into two types: (a) institutional reforms designed to bring about a true decentralization of decision-making in trade; and (b) economic policy reforms designed to change the economic environment within which the new decision-makers would operate. Institutional reforms are discussed in this section, and economic policy reforms in Section D below. Decentralization of Trade Decision-Making 2.39 At the present time, there are many different persons and institutions involved in making decisions concerning foreign economic relations: central planning authorities, MOFERT, national FTCs, local FTCs, provincial authorities, enterprise managers. It is undoubtedly true that considerable decentralization has taken place already from the center to the provinces. However, the decision to trade has, in most cases, not passed down - 34 - to the level of the enterprise. The analysis of this chapter (and Annexes) has focussed on the key issues of the problems created by the separation of the producing unit from world markets, and the absence of an appropriate environment to induce efficient exports. 2.40 Decentralization in this sense does not mean that provinces should take over decision-making for trading, but rather that enterprises should take such decisions in reaction to the prices and economic levers they face. To achieve this will ultimately require major reforms across the spectrum of economic policies, notably including enterprise reform and domestic price reform. However, major progress can be made in the meantime, although for the reasons stated earlier, it is likely that simultaneous action would be required on exchange rate management. Several steps would be required to achieve efficient decentralization to the enterprise level and, as has been done in other areas of economic reform, it would cause least disruption if changes were introduced over time. 2.41 International experience has shown that no single organizational form of foreign trade is "best" for every product and enterprise. Product characteristics and the organization of production and distribution for a particular product strongly influence the choice of international trade and distribution channels. For some products, particularly raw materials and other standard products, relatively little knowledge of product characteris- tics and user requirements is needed. For other products, for example high quality garments, detailed knowledge of (often rapidly changing) consumer tastes is required. For still other products, for example, specialized machinery, detailed knowledge of product characteristics and user requirements may be needed and some technical after-sales services might also be required. It is thus very unlikely that the same organizational structure would be appropriate for these rather different situations. 2.42 The appropriate marketing channel will also depend on the charac- teristics dPt the buyer. Consumer goods will almost always require the inter- mediation of a distributor (wholesaler, retail chain), while sales of producer goods might not require any intermediary and are often carried out through direct contacts between producer and user. The size of the market is also an important factor. For sales to large markets, direct marketing or the use of specialized (domestic or foreign) traders might be the most appropriate, while for small markets a (domestic or foreign) general trading corporation may be the most appropriate. 2.43 Finally, the stage of development is an important factor. Having a comparative advantage in the production of a particular product does not necessarily imply having a comparative advantage in marketing that product. Most of China's East Asian neighbors have concentrated in their development of export capabilities first on attaining proficiency in production and have relied primarily on foreign buyers to market their products, only gradually moving into export marketing themselves. 2.44 The role of foreign buyers is particularly important in those manufactured goods where familiarity with consumer requirements (e.g. product design, styling, packaging) is of utmost importance (Box 2.5). It is exactly - 35 - in these types of products that China's export capabilities and competitive- ness is most constrained, because of the still too limited contacts between domestic producers and foreign buyers and suppliers. (The importance of these contacts for technology transfer and acquisition is discussed in Box 2.8). 2.45 The international experience suggests that China could benefit very substantially by removing the tight compartmentalization or "air-lock" between domestic and international markets that results from the present operating rights and procedures of the FTCs. As for domestic enterprises, substantial efficiency gains could be made by decentralizing decision-making. However, if decentralized decision-making is to lead to efficient operations, competition needs to be introduced, and prices and other economic levers (in particular the exchange rate) have to reflect the true costs of re_qurce use. Some of the specific measures that could be considered include:_ (i) Eliminating, over time, the product-specific trading rights of PTCs, permitting them to engage in the trade of any commodi A, thus becoming de facto General Trading Corporations (GTCs)._ (ii) Making FTCs fully responsible for profits and losses, and elim- inating any plan exports and imports that imply substantial losses to the FTCs. At the same time, subsidies for command plan imports that remain necessary because of large differentials between domestic and international prices which continue for domestic policy reasons (e.g. fertilizer) should be provided by specific product subsidies in the state budget, rather than from FTC revenues, or general subsidies to FTCs. (iii) Eliminating the exclusive right of Foreign Trade Corporations to engage in foreign trade. If appropriate indirect levers (prices, exchange rate, tariffs etc.) are used, there is no reason to exclude any enterprise from engaging in foreign trade, unless that enterprise produces one of the "special" exports discussed in paras. 2.53-2.56 below. Alternatively, direct foreign trade rights might be given first to all production enterprises whose expo-ts or imports exceed a certain amount, say, US$l million. 7/ These recommendations are quite consistent with the September 1984 report of the Ministry of Foreign Economic Relations and Trade on the future reform of the trade system (see Box 2.2). However, in many key respects the reforms outlihied in that report have not yet been implemented and remain the key institutional reform issues. 8/ Careful phasing of such a policy for different types of products would, of course, be necessary given different rates of progress of price reform in different areas (see paras. 4.31 - 4.36). - 36 - Box 2.5: WHAT FOREIGN BUYERS DO The following excerpt from a study of 113 successful exporting enterprises in South Korea, undertaken by World Bank staff in 1976, highlights the importance of direct external trade contacts between production enterprises and foreign buyers: "The relations between Korean firms and the foreign buyers went far beyond the negotiation and fulfillment of contracts. Almost half the firms said they had directly benefitted from the technical informa- tion foreign buyers provided: through visits to their plants by engi- neers or other technical staff of the foreign buyers, through the provision of blueprints and specifications, through information on production techniques and on the technical specifications of competing products; and through feedback on the design, quality, and technical performance of their products by letter and telex. Three-quarters of the firms said that the requests and recom- mendations of foreign buyers influenced the design, style, packaging, or technical specifications of products exported. Most of the firms con- firmed that some of their exported products were made directly in accord with designs, patterns, or other specifications supplied by foreign buyers. Did the foreign buyers mainly affect the packaging? Did they affect the design or styling (as would be expected for garments, for I example)? Or did they affect the bAsic technical specifications? I Almost three-quarters of the firms mentioned product design and styling. Packaging was also frequently mentioned, but few firms mentioned packag- ing alone. About half the firms indicated that requests by foreign butyers "affected the technical specifications. In a more detailed evaluation of the advantages to firms of direct contacts with foreign buyers, more than half the firms surveyed considered those contacts to be important for the adaptation of product design and styling to market requirements--and for the development of new products or new product varieties. About two firms in five said that contacts with foreign buyers were important for improved techniques of quality control and for improved techniques of production; one in five for improved cost accounting and control. Because of the activi- ties of foreign buyers in supervising and checking export shipments, the exporting firms had a strong motive to implement effective methods of quality control. Three-quarters of the firms confirmed that foreign buyers attempted some supervision over shipments, usually by preshipment inspections. Other methods included periodic visits to the factories and constant surveillance by inspectors that buyers had stationed in the factories. In addition, foreign buyers made suggestions about techniques of cost accounting and control." From: Rhee, Ross-Larson and Pursell: "Korea's Competitive Edge - Managing the Entry into World Markets"; World Bank/Johns Hopkins University Press, 1984, p. 61-62. - 37 - (iv) Eliminating the strict separation of domestic and foreign trade. FTCs could be given the right to engage in all types of domestic wholesale and retail trade (and manufacturing too). This would help to bring increased competition and efficiency to domestic commerce. At the same time, domestic commercial organizations could be given foreign trade rights, perhaps with some minimum requirements first that might be eliminated later. (v) Consideration could be given to permitting a larger role for foreign buyers and foreign trade corporations in China's external trade. Joint ventures between FTCs and foreign GTCs could bring benefits from the latter's extensive marketing networks, and facilitate the transfer of marketing know-how to China's FTCs. 2.46 It is not possible to predict with any certainty or precision what the results of such a trade reform would be on the level of exports. Since exports that were not profitable would decline or be eliminated, total exports would probably fall in the absence of other reform measures. It was thus argued earlier that exchange rate reform would also be necessary to induce the appropriate level of exports needed to finance imports without excessive borrowing. If this were done, there would be no net change in the value of trade, but the composition of trade would change, as unprofitable or ineffi- cient exports were replaced by profitable ones. The Foreign Trade Planning System 2.47 If trade decisions are to be decentralized to enterprises, it will not only be necessary to change the way in which FTCs operate but also the way trade planning is carried out. We estimate that command planning continues to cover about 50% of total trade. Thus, half of all trading decisions are taken administratively and not necessarily according to economic criteria. Of course, it is likely that many of these imports and exports would take place within a non-command framework, but the absence of choice leads to distortions within the trade system. MOFERT has clearly stated the intention of moving from comaand to guidance planning, and some progress has been made in this respect. It has not been possible to reduce the scope of command planning faster because of the absence of appropriate economic levers--tariffs, exchange rate, export rebates, etc. As these are developed, it will be both possible and necessary to reduce command planning; otherwise, the economic levers would not be fully effective. 2.48 This is not to say that there should be no trade planning. The nation's trade balance is crucial to macroeconomic stability and the nation's creditworthiness, ane the government must be able to adjust its economic policy levers, as necessary. Trade planning should continue, therefore, in the form of projections of likely exports and imports, and the careful monitoring of developments to ensure that the projected pattern is being, at least in the macro sense, achieved. This will be particularly important during the period of adjustment from the present system to the recommended future system, since it is very difficult to project adjustment problems and - 38 - to time the adjustment path correctly. This also requires that continued attention be given to the preparation and improvement of trade statistics, so that the situation can be monitored with the most timely information. The Agency System 2.49 In its October 1984 policy statement, MOFERT identified the introduction of the agency system as a key step in trade reform. We would strongly endorse this proposal and urge its implementation at the earliest opportunity. The changes recommended in para 2.45 (i-iii) above would be equivalent to a full implementation of the agency system. Although MOFERT recommended the introduction of the agency system for all trade except in agricultural and handicraft products, it appears to have been introduced to only a very limited extent, probably covering less than 25% of total trade thus far. 2.50 One of the main problems of the present system is that some exports take place at a loss, and some imports have to be quantitatively controlled. To eliminate these two problems, producer and importer must be able to feel the full effect of world prices. The main reasons that the agency system has not yet been put in place appear to be threefold: (a) it is feared that exports would decline; (b) it is feared that increased imports of consumer goods would occur because of high domestic prices, resulting both in balance of payments problems and industrial dislocation; and (c) conversely, it is feared that there would be shortages of key raw materials, since these could not be imported profitably at low fixed domestic prices. These are legitimate fears, if no accompanying measures are taken. However, a package of reforms, such as those recommended in Part D below and in Chapter III, could help to avoid such results. Successful introduction of the agency system requires the simultaneous introduction or management of other economic levers and, at the same time, these economic levers require the agency system to be in place to function well. 2.51 As one example of this, it is interesting to note the effect that the absence of the agency system has on the impact of one of the main economic levers, the exchange rate. The exchange rate was devalued by 17% in July 1986. One of the purposes of the devaluation was to improve the profita- bility of exports, by raising the Renminbi equivalent of world market prices. Amongst all the enterprises (except joint ventures) that the mission visited, there was only one enterprise whose export price had changed since the deval- uation, and this enterprise had direct trading rights. Not one of the enter- prises which exported through an FTC had received any benefit from the deval- uation. This is not to say that a devaluation would have no impact on exports at all, since it would increase FTCs' profits (or reduce their losses) and permit them to finance a higher level of exports with any given budget constraint. Nevertheless, the absence of any direct benefit to enterprises from the devaluation--and indeed some enterprises did suffer increased import costs--seriously reduces the impact of a devaluation on exports. An increase in exports was an explicit objective of the devaluation, but the absence of the agency system is undermining the impact of that action. - 39 - Competition and Entry Barriers 2.52 If the trade system is to become efficient, the present monopolies must be removed. The October 1984 statement foresaw this, and some consider- able progress has been made, but there do remain some artificial barriers. These are in two forms: (a) monopoly controls over imports and exports of certain commodities are vested in particular FTCs; and (b) MOFERT continues to prescribe the 'scope of business' of all FTCs and has to approve the creation of new FTCs, and the granting of trading rights to producers. If producers decide to export through an FTC (which many will), they should be free to select the FTC that best meets their needs. This means not only that FTCs should not have monopolies over particular commodities, but also that provin- cial boundaries should be no hindrance either. If an FTC in Shanghai or Guangdong can export Sichuan's products more efficiently, they should be permitted to do so. Clearly, to overcome the barriers to inter-provincial trade in exportables would require reforms of the foreign exchange retention scheme, given current export incentives. The "Special" Exports and Strategic Imports 2.53 There are special market considerations affecting exports of oil, textiles and garments, and fresh and live food to Hong Kong, and exports of certain metals and minerals in which China is a dominant world supplier. China may wish at times to constrain oil exports out of political solidarity with other developing countries; China is required to limit exports of textiles and garments to comply with its obligations under the Multi-Fibre Arrangement (MFA); and China supplies 95% of Hong Kong's fresh food market, and, in the past, excess competition for this lucrative market led to price declines and an overall reduction in export revenues. At the present time, these exports are controlled quantitatively through strict monopoly powers vested in national FTCs. These FTCs then allocate quotas to their provincial branches, which in turn allocate quotas between factories. In the case of food exports, these quotas are allocated as frequently as every month by the National Cereals, Oils and Foodstuffs Import and Export Corporation in Beijing. 2.54 The present arrangements for these special exports work quite well for China, and, although we would suggest that they eventually be dismantled, this should be done with care and in a well-planned way. There are two ways in which these exports can be limited in the future, which would eliminate the essential problem of the present system that it locks in the present export pattern and suppresses new, potentially more efficient exporters. (i) Export Taxes. Especially for oil and food exports, the imposition of an export tax could both achieve the desired level of export restriction and would generate revenue for government, instead of excess profits for quota-holders. It would also permit the export shares of different producers to change over time in line with their efficiency. Economic research could determine the appropriate tax levels. - 40 - (ii) Quota Auctions. In a restricted environment, export quotas have a price. The more efficient the exporter, the higher the price the producer can offer to buy a quota. For textiles and garments, auctions could be a much better way of allocating available export quotas and would have the same merits as an export tax. Quota auctions would provide strong incentives for increased exports of higher quality varieties and thus increase export revenues. The same could be tried for food exports, even in conjunction with an export tax. 2.55 Similar considerations also apply to the imports of certain essentialy raw materials, which are generally the same as those of the import command plan. The fixed domestic prices of steel, fertilizer and grain, for example, are well below import parity levels. Consequently, large losses are incurred in carrying out these imports. Clearly, if all the reforms recom- mended were carried out, there could be conflicts in arranging for imports of these loss-making, command plan commodities. Conversely, there could be attempt to export some of these same items procured at the low domestic price. It must be recognized that pricing and allocation reforms may take longer in these areas than in, say, light manufactures, and the timing of reforms of trading arrangements in such products would have to be linked very closely to progress in price reform (see paras. 4.32-4.37). 2.56 However, this is not to say that current arrangements could not be improved upon, especially in the context of trade sector reforms. In particu- lar, the current level of subsidies on these products are not accounted for separately or specifically, but rather are included within overall subsidy accounts. Whilst trading arrangements involving monopolies/monopsonies are likely to be required for these commodities for some time, it would be appro- priate to develop specific subsidy accounts related to the imports of these commodities required for state purposes. Alternatively, instead of main- taining low prices and subsidies for all such imports, it would be preferable to provide strategic subsidies direct to the key projects, but adjust the prices of the strategic products, so that the subsidy is provided for the priority uses, but not for non-priority uses, as is the case at present. Not only would this avoid cross-subsidy practices, it would also reveal explicitly to policymakers the budgetary costs of current pricing policies for these products. The New Role for the FTCs 2.57 The proposal that the exclusive foreign trading rights of FTCs be substantially or fully eliminated (para. 2.45 (iii)) does not mean that FTCs would lose their role and and go out of business. Most enterprises would still prefer to use FTCs for all or part of their exports or imports. How- ever, they would have considerably more choice and would use the trading corporation that provided them with the most efficient services. Since there are some economies of scale involved in trading (e.g. maintenance of overseas offices, telecommunications facilities etc.), one could expect the most effi- cient FTCs to grow into very substantial general trade corporations (GTCs). But unless their managers have considerably more decision-making freedom on such questions as where to open foreign offices, how to staff them, how much - 41 - to invest in telecommunications facilities, or how many trips salesmen make, they can hardly expect to provide services equivalent to and competitive with foreign trade corporations from other countries. The more China's comparative advantage shifts towards diversified manufactured goods, the more important will effective overseas marketing become in determining success or failure. 2.58 For independent and profit-oriented FTCs to operate efficiently, it is essential that they be exposed to competWion from other FTCs. To promote competition, two measures appear essential:_ (i) Independent FTCs responsible for profits and losses should not be unified under one owner. Otherwise, they are likely to operate in collusion rather than in competition, and their efficiency will suffer. (ii) Government responsibility for ownership or control should be separated institutionally from its responsibility for regulat- ing an enterprise's (FTC's) external environment. Otherwise, there will always be a tendency by the regulatory agenc, to protect the enterprises under its control from competition, e.g., by restricting the entry of other enterprises into that line of business. 2.59 Ownership Options. The above principles suggest that the FTCs should be separated from MOFERT. One possibility would be to make each prov- incial branch an independent profit-oriented enterprise that would compete with other FTCs from other provinces for export and import business. The danger of this proposal, especially under the present foreign exchange reten- tion system, is that provincial authorities would be tempted to monopolize all foreign trade of their province through the provincial FTC in order to maxi- mize foreign exchange retention within the province. An essential complemen- tary measure would thus be to pass on all foreign exchange retention rights to production enterprises. Competition could be strengthened either by permit- ting free entry into foreign trade activities or, at least, by substantially lowering entry barriers. At the same time, these new exporters should be permitted to carry out foreign trade for any other enterprise on an agency basis. The recent decision to allow foreign joint ventures to engage in foreign trade is a very sensible step in this direction. However, the poten- tial benefits are reduced by the present quantitative restriction on their export activities (i.e., limited to their foreign exchange requirements). 2.60 In order to limit provincial attempts to monopolize foreign trade under the provincial FTCs, it would also be desirable to permit FTCs to engage in domestic trade. With their considerable technical and commercial skills, they could prove a potent force toward promoting interprovincial and inter- regional trade in China. Joint ventures by several provinces to create such 9/ For a discussion of enterprise ownership and management issues see: World Bank: China: Finance and Investment, Report No. 6445, March 1987, Chapter III. - 42 - general trading corporations could be useful. Similarly, international trust and investment corporations at the national and provincial levels are well suited to extend their activities further into international and domestic trade and could provide trade and financial services to exports and importers at the same time. D. Reforming the Trade Environment 2.61 Once the decision to decentralize has been taken, it will be necessary not only to make the institutional reforms described above, but also to adjust some of the economic levers directly affecting trading decisions. In particular, attention would need to be given to the three areas of foreign exchange allocation, the import regime, and export incentives, as these will critically affect enterprise trading decisions. Foreign Exchange Allocation 2.62 As has already been stated several times, the key accompanying reform will be to ensure that there is a market-clearing exchange rate. options for ways to ensure this are discussed in Chapter III. However, there are two questions in foreign exchange issues: the rate and allocation. Ideally, there will one day be a full-fledged market for foreign exchange, and anyone who needs foreign exchange would be able to enter that market to buy it. In line with the general approach of reforming in stages, it is important to consider how the current allocation system can be modified so as to improve its economic functioning and move the system towards greater reliance on the market. 2.63 The existing allocation mechanism, and in particular, the retention scheme, can be refor TS in two directions, before a more comprehensive solution is adopted.- (i) Tradeability of exchange retention rights. At the present time, enterprises are permitted to transfer their foreign exchange retention rights to another enterprise, but only at the current official exchange rate. Consideration could be given to developing a market in which such rights could be traded at a negotiated exchange rate. Such an option would act as an inducement to efficient exporters and assist with the allocation of foreign exchange to the highest value uses. (ii) Increased retention rates. If the first approach were adopted, it would then also make sense to raise gradually the proportion of foreign exchange retained by enterprises. As this occurred, two things would happen: the central government would have less foreign exchange available to allocate, and enterprises would have greater surpluses to trade. For the first problem, the government could either enter the market itself as a buyer, or it could cease providing foreign exchange to some users and send them instead to the market. 10/ China will also want to discuss the advantages and disadvantages of the approaches outlined with the International Monetray Fund. - 43 - A New Environment for Imports 2.64 The environment for imports will be governed to a large extent by progress in two other areas: reform of the FTCs as described above, and reform of the exchange rate regime. Many of the problems of the present import regime derive from the use of import licensing and other forms of quantitative restrictions. The key element in the reform of the import regime is, therefore, to introduce an exchange rate regime that eliminates the need for such restrictions. As progress is made on this front, it is strongly recommended that all quantitative restrictions be phased out, so that control of the level of imports is achieved solely by the exchange rate and the customs tariff. 2.65 One instrument is already at hand to assist in the transition from quantitative restrictions to tariffs: the regulatory tariff, which is a surcharge to the tariff schedule. This could be used to dampen overall demand for imports, so that the initial changes could be more gradual than otherwise, thus avoiding the need for "shock treatment" to the system. A similar mecha- nism in its impact could be the adoption of a system for auctioning import licenses. (This is discussed further in Annex 2, paras. 2.53 and 2.204). 2.66 The effective protection rates generated by the present tariff structure vary considerably from one product to another with results that are perhaps not really intended by policymakers. Thus, once it becomes the principal instrument of protection, the tariff structure would need to be revised, with the primary intention of introducing more equality of protection amongst different products. This is a time-consuming process, particularly in terms of planning a transition. Preparation should therefore begin at an early time, under a reconstituted Customs Tariff Committee. In the meantime, one of the most important initial steps that can be taken is to streamline the current exemptions to the tariff. A New Environment for Exports 2.67 The major problem of the current export regime is that there is an inefficient system of compensation for the inherent anti-export bias of the present trade regime. In addition to the simple export incentive scheme suggested below, the most important export incentive will come from reform of the foreign exchange regime, both in terms of rate determination and alloca- tion, as already discussed. 2.68 The success of many East Asian countries in promoting exports has rested on a trade and industrial strategy that stresses specialization based on comparative advantage. The export policy complementing such a strategy should focus on removing the disadvantages domestic firms may face vis-a-vis foreign competitors. As China move towards decentralization, it will be important to ensure that Chinese enterprises are also freed of any such dis- advantages. This is generally referred to as creating a 'neutral status' for exporters, whereby exporters are permitted to compete on an equal footing with their foreign competitors. In addition to exchange rate issues, there are three key areas where reforms could be undertaken to achieve this "neutral status." - 44 - (i) Duty-free Inputs for Exports. In principle, Chinese exporters can obtain duty-free imported inputs for exports, but the system is not automatic, or even well-known, and its applica- tion seems to depend on an enterprise's relations with its FTC. Moreover, there seems to be no provision for the draw- back of duties actually paid on such imports for exports (See Annex 2, para 2.146-2.162). Assuring exporters free access to imported raw materials and intermediate inputs at world prices is an important element of providing neutral status. The simplest way to do this is to permit exporters to import duty- free the imported components of exports, with licenses for such imports being granted automatically. Again, such a regime seems to apply already for sino-foreign joint ventures. The current arrangements which apply to larger exporters (duty-free entry plus export verification) seems reasonable, but should be made automatic and universal, and in particular, it should operate as well for small, less-well connected enterprises as for laege ones. Provisions are also desirable for drawing-back duties actually paid or for exemption or drawback status for domestic suppliers of inputs for exporters, and for exempting domestic suppliers of their obligations for the consolidated industrial and commercial tax. Such provisions would both enable the Chinese exporter to compete on an equal footing, and eliminate any resulting discrimination against domestic suppliers. An automatic system for such exemptions or draw- backs should be introduced as soon as possible (see Box 2.6). (ii) Financing of Exports. As with access to foreign exchange and duty-free imports, some ad hoc arrangements exist for pre- and post-shipment financing of exports from China, both in local currency and foreign exchange, through the Industrial and Com- mercial Bank and through the Bank of China. Experience from other countries has shown that export financing at uniform interest rates for all economic activities generating value added for exports has proved to be critical if a country is to exploit fully its export potential (see Box 2.6). In many countries, such facilities are frequently accompanied by guar- antee and export credit insurance schemes. Much of the reason for the absence of such facilities in China is probably the present role of FTCs in foreign trade. As enterprises become increasingly responsible for their own trade, it will be most important for such facilities to be established by the banks with redi sllcunting at an institution such as the People's Bank of China. Alternatively, these exporters' credit facilities could be combined with export credit facilities, which China may wish to offer to overseas buyers, into an export-import bank. Although, such facilities would guarantee exporters 11/ For a fuller discussion of how all these export incentives operate, see Annex 2, and in particular, paras. 2.163 to 2.171 and Box 2.3. - 45 - access to credit, it should be stressed that there is no economic justification for subsidizing the interest rates charged on such credit. The guaranteed access to credit itself should help to compensate for any residual anti-export bias in the system, and should itself only be extended so long as the anti-export bias remains. (iii) Access to Foreign Exchange. The limitations of the current foreign exchange retention scheme were discussed earlier. For exporters, it is critical to ensure that foreign exchange is readily available to finance the imported component of exports. Failure to do so can cause exporters to miss delivery dates or be unable to respond to order inquiries, which can seriously damage an exporter's reputption. There are many ways to arrange such access for exporters, including the current provisions in place for sino-foreign joint ventures. The final form would depend on any reforms that China decides to intro- duce in the foreign exchange allocation mechanism. E. Technology and Foreign Investment 2.69 In the past, China has preferred to develop known technologies by copying imported equipment or through completely independent research and development, rather than by importing technological know-how. This strategy of creating, and often recreating, technology contrasts sharply with the absorptive strategies of Japan and, more recently, the newly industrializing countries of East Asia, which aimed to assimilate and adapt foreign tech- nologies as fast as possible. International experience has shown that the absorptive strategy has many advantages for a technological latecomer, because selective importing of technologies generally involves lower risks, a shorter time lag, and a lower cost of acquisition. Indeed, even the most advanced industrialized countries rely extensively on the absorptive strategy--not to do so would soon mean falling behind in technology and productivity. - 46 - BOX 2.6: PRESHIPMENT EXPORT FINANCING There are three types of financing needed by exporters: preshipment, postshipment and investment. Of these, experience has shown that it is the first which is most critical. It is also the one which is most difficult to organize. In particular, it has proved difficult to arrange such schemes to include indirect exporters (suppliers of inputs to final exporters) and those direct exporters which utilize general trading corporations. One innovation in several East Asian countries that seems to address this latter problem well is the domestic letter of credit (DL/C). The principle of the DL/C is to provide credit to the indirect exporter on the basis of the direct exporter's creditworthiness. When the exporter receives an irrevocable letter of credit for his exports-- or other firm evidence of an export order--his bank is able, on the basis of this to open a second similar credit account on his behalf, with the indirect exporter as the beneficiary. The DL/C states that the bank will pay the indirect exporter when the latter submits evidence of delivery of goods to the final exporter. Armed with this DL/C, the indirect exporter can obtain production financing for his output. The second benefit of this scheme--as well as providing credit directly to the exporter and indirect exporter--is that it provides a dirm basis on which to grant export incentives to the indirect exporter. The documentation provided to the exporter's bank as verification for payments against the DL/C is the very same verification needed to assure authorities about the amount of export incentives--usually in the form of indirect tax rebates--that the indirect exporter has earned. In some countries, the government has also taken the next logical step and has assigned responsibility for administration of the export incentives to the banking system. Thus, the DTJ/C solves one of the major problems of adminis- tering export incentive schemes: involving and ensuring full benefits for the indirect and small exporters, who would otherwise find themselves discriminated against. - 47 - 2.70 The issue is not whether China should upgrade its technological capability through its own research or through imports. Studies in other countries have shown that these two routes are complementary. Enterprises with strong research and development capabilities can make more productive use of imported technologies and can be selective in importing. Selectivity is the key to using imports in upgrading technology. In the past, China has tended to alternate between the extremes of importing turnkey plants or relying wholly on domestic research and development. But there are many intermediate options. Elements of unpackaged technologies that can be imported include licenses, designs, key equipment, and consultants to help solve management, marketing, or engineering problems. Recent Chinese policy has stressed the need to import "software" (know-how) in addition to, or instead of, "hardware" (equipment). This is an important advance in tech- nology policy, because it recognizes that individual elements of technology can be traded, and that equipment is not necessarily the most important element. While these are steps in the right direction, they do not yet go far enough. In order to maximize the benefits from technology transfers, further changes are needed. 2.71 As the subject of technology transfer and development is beyond the scope of the present study, the present discussion is limited to those aspects of technology transfer and development thaV ire most closely related to trade and external economic policies in general, focussing first on different types and channels of technology transfer, and then more extensively on one vehicle, foreign direct investment. External Economic Policies and Technology Transfer 2.72 The Nature of Technology. The term 'technology' refers to a very broad concept that has many different aspects. Technology can be "embodied" in machinery or human beings, or ;t can be "disembodied"; e.g., in the form of manufacturing instructions, product specifications or computer codes. An important aspect of technological knowledge is that much of it is not explicit, but tacit; that is, it resides within human beings or organizations and cannot easily be transferred except by transferring people or a whole organization. Very often, technology is "organizational" technology; that is, the ability to produce something with greater efficiency or at higher quality without necessarily using different machinery or other 'hardware" (Box. 2.7). 12/ A more comprehensive treatment of technology policy issues is contained in: World Bank: "China, Long-term Development Issues and Options" (1985), Chapter 7: "Managing Industrial Technologies" and in Background Papers 5-8 to that report. - 48 - Box 2.7: ORGANIZATIONAL TECHNOLOGY - THE CASE OF THE AUTOMOBILE INDUSTRY - The success of Japanese automobile companies in world markets in recent years provides an interesting example of the importance of organization as an element of manufacturing technology and productivity. The success of Japanese automobile companies has been based on their much higher labor productivity compared to US or European producers. This higher labor productivity has been brought about not so much by more or better machinery (such as assembly robots) but rather by organizational innovations, in particular the just-in-time (kan ban) inventory system that greatly reduces inventories and material handling, and a quality control system that emphasizes the role of assembly workers and largely eliminates the need for quality inspectors and supervisors. These two organizational innovations have permitted Japanese automobile companies to produce the same volume of cars in technically nearly identical factories with far fewer employees. Most of the savings are in inventory handling, quality inspectionq, supervisors and other overhead personnel. Today, Japanese automobiles plants are generally also more automated than most other plants. Together, these organizational innovations and higher automatization have increased labor productivity to twice the level achieved by competitors in other industrialized countries. 2.73 What the example in Box 2.7 shows is that technical hardware is only one element in the acquisition and development of technology, and has to be complemented by other elements. While the importance of 'software' has recently been stressed by Chinese policymakers, particularly in the context of foreign direct investment, in practice most technology transfer is still hardware-focussed. The transfer of 'soft' elements of technology is impeded by the "airlock" of the FTCs and the limited authority of enterprise managers over the use of foreign exchange (except in the case of hardware imports from foreign exchange retention quotas). 2.74 Experience in other countries has shown that channels of technology transfer other than machinery or foreign direct investment can be very important and less expensive and more effective. Examples include not only overseas education and training of technical personnel, but also hiring of foreign technical specialists, retired managers and so on. The case of a foreign manager being hired to run an engine manufacturing plant in Wuhan has become well-known and celebrated. However, what is unusual about this is how little use has been made by Chinese enterprises of this possibility in comparison to other countries. Japan and many of China's neighbors have made extensive use of foreign managers and technical advisors, especially during - 49 - the start-up phase of a new enterprise or industry. Many Chinese enterprises would probably use this channel much more, if they had greater authority over the use of their foreign exchange. While the salary of a foreign manager may be high by comparison to Chinese salaries, they are often cheaper than unnecessary imported equipment or inefficient use of imported equipment. Studies of successful exporting enterprises in other countries have shown the importance of these and other 'informal' channels of technology, such as the advice provided by buyers of products and suppliers of equipment (Box 2.8). 2.75 Equally important as more, and more efficient, technology transfer is better domestic diffusion of the technology acquired. Very often, tech- nical and managerial competence varies widely among domestic enterprises. Allowing greater labor mobility, especially among technical and managerial personnel would perhaps be the most important measure. Technology diffusion can also be enhanced by the development of a domestic technical and management consulting industry. 2.76 The examples above show that there is no single best way of trans- ferring technology. All possible avenues--direct foreign investment, licens- ing, turnkey projects, purchases of material and equipment, consulting ser- vices, employment of foreign managers and technical specialists etc.--should be used, and each has its own advantages. The question is not which channel to use, but how to make best use of each channel, and this will be different in every case, depending on capabilities within each enterprise, the avail- ability of local sources of technology, the availability of alternative foreign sources of technology, and so on. Only the enterprise managers directly involved can really judge which channels will be most appropriate. One of the most effective ways to absorb foreign technology is the 'appren- ticeship' pattern, in which foreign assistance is relatively extensive at the beginning, such as for a plant using new technology, but is gradually reduced ai further capacity is added and engineering services are localized. 2.77 Some Chinese observers argue that the liberalization of technology transfer has already gone too far. They point to the negative side-effects of decentralized technology import decisions, such as the emergence of excess capacity in a number of fields, brought about by the indiscriminate duplica- tion of productive capacity in many parts of the country, either through ignorance of each other's planned activities or because of unreasonably high demand estimates. The case of some 120 television assembly plants being established in the last several years under the aegis of provincial and local governments is an example. Another negative side-effect has been the preemp- tion of potentially invaluable opportunities for local suppliers to fill some of the machinery and plant equipment requirements. Given the packaged nature of the transfer of complete production lines, the scope for local suppliers has tended to be limited. - 50- Box 2.8: SOURCES OF TECHNOLOGY A survey of 113 South Korean export enterprises undertaken by World Bank staff in 1976 (see also Box 2.5) provides an interesting account of how successful exporting enterprises have acquired technolo- gies (see Table). Of the various sources of technology, contacts with buyers and suppliers was the most important source, followed by employ- ment of technical staff with work experience in other domestic enter- prises or abroad. By contrast, licenses and technical agreements were cited as important sources of technology in only 8% of all cases. Overall, local and foreign sources were about equally important, but many domestic sources of technology were initially foreign, having already been assimilated by other domestic enterprises. This study shows the overwhelming importance of informal channels of technology transfer and the crucial role of efficient domestic diffusion of technology acquired from abroad. IMPORTANCE OF DIFFERENT SOURCES OF PRODUCTION TECHNOLOGY (Z) Local Foreign Total Buyers and suppliers 9 20 29 Transfer of labor 11 13 24 Local technical know-how 19 - 19 Technical assistance 3 8 11 Licenses and technical agreements - 8 8 Government R&D and information support 9 - 9 Total 51 49 100 Source: Rhee, Ross-Larson and Pursell: "Korea's Competitive Edge - Managing the Entry into World Markets"; World Bank/Johns Hopkins University Press, 1984, Chapter 4. - 51 - 2.78 However, the source of these difficulties does not lie in the recently reduced control over technology imports, but rather in the still only partly reformed enterprise environment, in particular the still limited decentralization of decision-making to enterprises, the limited responsibility and accountability of enterprise managers, and the limited extent of competi- tion, profit orientation and performance incentives under which enterprises and their managers operate. One of the main advantages of foreign joint venture enterprises in China is that they can overcome these restrictions far more easily than domestic enterprises, and this is one reason for the popularity of joint ventures in China. While these joint ventures provide an important experiment and role model for domestic enterprises, changes need to be made to give similar rights to domestic enterprises. 2.79 Quality Control. China also needs to give greater emphasis to quality control and quality assurance in the introduction of foreign tech- nology, since prospects for finding windows of opporcunities for Chinese manu- factured goods in foreign markets hinge critically on meeting international standards. Allowing foreign quality control and assurance entities to come into China, either on a wholly-owned basis or through joint ventures with Chinese counterparts, could lead to more rapid international acceptance of Chinese products. For example, the absence of internationally accepted certi- fications of quality, (such as United Laboratories, UL), removes many of China's electronics components factories (i.e. printed circuit boards) from their own local market, as end-users do not wish to procure components that lack such certification. 2.80 Technology Transfer Legislation. Some government agencies have pressed for legislation with respect to technology transfer in order to pro- tect domestic enterprises against unfair and restrictive practices by foreign licensors. A more effective way to address 'unfair restrictive clauses' in technology transfer agreements might be through enacting an anti-monopoly and fair trade law that would cover both domestic and foreign contracts. What is needed is a "promotional" rather than a 'regulatory' approach to technology transfer, but with an ultimate safeguard mechanism against abuses by foreign suppliers of technology. As technology transfer becomes more an enterprise- initiated rather than a state-initiated matter, it is also crucially important to streamline administrative procedures for effecting technology transfer. Foreign Direct Investment 2.81 Until 1979, China was essentially closed to foreign direct invest- ment. Technology was imported, first through turnkey plants and technical assistance from the Soviet Union, and later through turnkey projects and mach- inery purchased from capitalist industrialized countries. Since 1979, as part of its policy of opening to the outside world, China has permitted and tried to attract foreign direct investment from private foreign companies, mostly in the form of joint venture enterprises. 2.82 Progress in this area since 1979 has been, in many ways, remarka- ble. Indeed, it can be said that the growth of foreign direct investment is the symbol of the policy of "opening to the outside world." China has rapidly - 52 - adapted to the needs and methods of the foreign investor, and the accumulated level of actual foreign investment in place--about US$6.5 billion is the end of 1986--is a testament to the seriousness of this policy. Moreover, China has showed a clear and sensible willingness to experiment and to learn from experience. 2.83 Nevertheless, the Government is rightly concerned about the amount and composition of the ventures that have been established, which have fallen short of the goals set. Many have been in (a) exploitation of coal and offshore oil; (b) labor-intensive manufacturing, mostly by overseas Chinese; and (c) hotels and ancillary services for tourists and other foreigners. These investments are useful but have not brought as much advanced technology as desired. Furthermore, the total amount has been less than China had hoped to, and could, attract; and, recently, the level of new commitments has been falling. Among the many reasons behind the less than fully satisfactory record of foreign direct investment so far is an inconsistency between (i) China's foreign investment regulations, which have put a premium on joint ventures being able to export, right from the start of their existence; and (ii) China's natural attraction to foreign investors, which is mostly its domestic market. 2.84 Among the things that China wants from foreign investment, the most frequently emphasized is "advanced technology." In earlier periods, great emphasis was put on "hardware"--machinery and equipment, or the plans to build it. More recently, the importance of management skills, labor training, and ability to organize, administer, and continue to develop an enterprise has been recognized, as well as the ability to continue to develop and create new hardware. Having the supplier of the technology committed to the success of the enterprise that is to absorb it has come to be seen as something of value--an important reason why the managers of many Chinese enterprises want to enter into joint ventures with foreign partners instead of merely purchas- ing machinery. Production and marketing of exports has also been desired, while the capital that comes with foreign direct investment has been a less important goal. This is consistent with experience elsewhere, which shows that, while the capital flows associated with foreign direct investment are useful, the other aspects are its most important benefits. 2.85 Most countries permit several different kinds of cooperation between foreign and local enterprises. In China, six different forms have been included in statistics and discussions of foreign investment; these are described in Annex 4. Of these six kinds of activities, the ones that would normally be thought of elsewhere as foreign investment are: wholly foreign- owned ventures, equity joint ventures, some contractual joint ventures and some co-production agreements for development of coal and oil. Little or no actual investment is involved in the two other forms: compensation trade, and processing and assembling. The availability of all these varied forms of "absorbing foreign capital" is, in general, good for China, because it gives enterprises great flexibility in choosing the best way to associate with foreign companies. While equity joint ventures potentially offer a wider range of benefits, and for a longer time than do the other forms, they are also more difficult to negotiate and may be more complex to operate. The contractual joint venture has been very popular, in large part because of the flexibility it permits. - 53 - 2.86 Excluding compensation trade, and processing and assembling (which in most countries would not be counted as investment), and also the small number (but significant amount) of joint oil production agreements, foreign investment in China, as of the end of 1985, in wholly foreign owned ventures, equity joining ventures and contractual joint ventures consisted of: 4,742 project agreements signed and approved, for $8.5 billion, of which only $2.4 billion were actually invested. To put these numbers in perspective, $8.5 billion would have been about 3% of China's GDP, while $2.4 billion was about 0.8 percent. A comparison with Brazil, whose economy is about two thirds as large as China's, but which has welcomed foreign investment for most of the twentieth century, might give a very approximate idea of how much foreign investment China could ultimately attract; Brazil has a stock of foreign investment estimated at something above $30 billion, or about 12% of its CDP. The Role of Foreign Direct Investment 2.87 As one of the many forms of technology transfer, foreign direct investment in joint ventures or wholly foreign-owned subsidiaries provides a whole bundle of different things. What distinguishes it from other technologi "bundles", such as turnkey plants, technology licensing, or management contracts, is that it provides for some degree of risk sharing, since the provision of equity funds entails a residual ownership claim on the earnings of an enterprise, and usually a direct participation in, or control over, management. This direct involvement and the claim on profits provides a strong incentive for the foreign partner to perform well. A foreign partner with a significant share in the ownership of a joint venture will be interested in continuing to upgrade the technology as the joint venture develops. He will also be interested in providing organizational, management and marketing skills, rather than just selling some machinery at a profit. 2.88 Multinational corporations possessing technology and management skills that give them an edge over competitors are often highly reluctant to sell this knowledge, but may transfer it to companies in which they have an ownership interest in order to gain access to other markets or to production sites where costs are lower. Another characteristic that makes some types of knowledge more profitable to exploit within a corporate group (as opposed to selling or licensing the knowledge to companies not associated with the parent company) is that the knowledge is difficult or impossible to embody in a set of machines, plans, etc. Examples of such knowledge include the organization skills needed for complex production processes, such as automobile production (see Box 2.7); or skills and organizational capabilities needed for marketing a product or service abroad. When these problems are most difficult, complex and long-lasting, foreign direct investment is most useful. 2.89 Foreign direct investment also has limitations. It tends to concentrate in certain sectors and in certain types of countries. For very large countries such as China, much of it, especially the high-tech varieties, tends to be motivated primarily by the domestic market. Like any capitalist corporation, foreign investors are motivated by profits and increased markets. However, foreign investment can be attracted if laws and practices are made consistent with both the attractions of the host country and the motivations of the foreign investors. Moreover, recent history in several developing - 54 - countries, for example Brazil, has shown that joint venture enterprises can generate export surpluses, even if their original primary motivation was the domestic market. Box 2.9 describes some of the different approaches that different kinds of countries have taken towards foreign investment. Policy Issues 2.90 A major policy weakness underlying China's problems in attracting foreign direct investment is that Chinese policy-makers tend to pursue too many objectives at the same time. The natural advantage of foreign direct investment and joint ventures is to introduce a complex bundle of technology, management and marketing skills that cannot be easily embodied in machinery or turnkey plants, and these activities are most likely to be found in import- substituting activities, where China's large market provides the major attrac- tion, and its skilled and low-cost labor force provides the possibility to become internationally competitive in due course. Often, however, this will take a number of years. For example, it took 10-15 years until the automobile industries of Japan and South Korea became internationally competitive, and exports became a significant share of total production. 2.91 This time is needed for the enterprise to get established, train its work force, achieve an efficient scale of production, find or help create local sources of high quality and competitively priced inputs and components, and get its costs of production down to where it can compete in international markets. The Chinese pursuit of motivating exports from foreign investment ventures, which is admirable in itself, is applied too zealously to all ventures regardless of their particular needs and abilities, and the difficulties this causes deter other potential investors from coming to China. Both the foreign investors and the Chinese side need to be more realistic in their expectations of how long it takes some ventures to export, and the Chinese authorities need to provide ways for the enterprise to have access to the foreign exchange it needs in the interim. 2.92 Another problem facing China in dealing with foreign investment, common to many other countries as well, is that the implementation of policy directions often falls short of the laws, regulations, and intentions of the Government. For example, many high officials recognize the situation with respect to import substitution and the foreign exchange balance requirement. But, in practice, adequate solutions to the problem are not always implemented. Additional attention is needed to ensure that actual practice is consistent with the policy directions laid down. - 55 - Box 2.9: DIFFERENT STRATEGIES TOWARDS FOREIGN INVESTMENT Different developing countries have adopted different approaches to foreign investment. There have been three different approaches: (a) some of the middle-income countries have been rela- tively open to foreign direct investment, such as Brazil, Mexico and the Philippines; (b) others have been more defensive, limiting foreign investment either moderately (India, Turkey until recently) or severely (Japan in the 1950s and 1960s, South Korea before 1980); (c) finally, a few countries have tried to attract foreign investment, but have failed to create an appropriate environment (Egypt, Yugoslavia). Brazil, particularly in the late 1960s and early 1970s, has been more open to foreign investment than almost any other developing country--except a few small Asian economies such as Hong Kong and Singa- pore. There were no restrictions on profit remittances, no maximum or minimum restrictions on equity shares held by the foreign partner, and foreign investment was welcomed in most manufacturing subsectors. No special tax incentives were given to foreign investment per se, although substantial inducements were offered to export-oriented investments. Foreign investment first was largely attracted into import substitution activities. However, with a shift towards outward-oriented trade policies in the late 1960s, many joint ventures reoriented their produc- tion to profitable export sales. South Korea, like India, has taken a less welcoming, more defensive approach. Until the 1980s, joint ventures were limited to a narrow list of sectors. Foreign direct investment was excluded from activities that competed with Korean firms in export markets or for local inputs. The foreign share of ownership was limited to 50%, except for firms that exported all their output or had particularly valued technology. Neither Korea nor India attracted much foreign direct investment, but have recently relaxed their restrictions on foreign investment. South Korea, of course, accomplished a remarkable record of growth and development without much help from foreign direct investment. Yugoslavia and Egypt decided to open their economies to foreign investment in the 1960s and 1970s, respectively. Both were countries with substantial markets and therefore attractive to foreign investors. However, both involved themselves in a contradictory situation not unlike China's: their overall trade policy framework discriminated against exports and they also imposed an enterprise-level foreign exchange balance requirement on joint ventures. Both attracted very little foreign investment. Source: Annex 4 - 56 - 2.93 The Foreign Exchange Balance Requirement. From a balance-of- payments point of view it really does not matter whether foreign exchange is earned by exporting or by substituting domestic production for what would otherwise have been imported. What does matter, however, is the efficiency of export or import-substituting activities. For activities that require advanced technology, management and marketing skills, it will often be more efficient to engage first, and perhaps for a considerable length of time, in import substitution. The present policies that require, with a few negotiated exceptions, foreign exchange balance for joint venture enterprises is the single-most important obstacle to attracting foreign direct investment into those activities where they could potentially make the greatest contribu- tion. The fact that such activities would indirectly save foreign exchange (by reducing import requirements) is not given adequate recognition by the present policy environment. 2.94 The most important policy change needed to attract foreign direct investment into activities where it can make the best contribution is thus the elimination of the foreign exchange balance requirement. The most desirable way to do this would be by introducing a market for foreign exchange for all enterprises (not only foreign joint ventures), where foreign exchange could be freely bought and sold at a market clearing exchange rate. Alternative options for introducing such a foreign exchange market are discussed in Chapter III. The recent announcement that foreign joint ventures will be permitted to export Chinese commodities up to an amount equivalent to their foreign exchange requirements is a step in the right direction although the limitations placed on which products are eligible has severely restricted the usefulness of this arrangement. A better arrangement, introduced in the October 1986 "Regulations on Encouraging Foreign Investment," is allowing the ventures to buy and sell foreign exchange for RMB amongst each other, at rates they themselves are free to negotiate. This practice, so far restricted to intra-municipal or intra-provincial transactions, could be expanded to the national level. However, this is likely to be a far less efficient vehicle than a foreign exchange market that would also open these possibilities to domestic enterprises. 2.95 Screening, Approval Procedures and Incentives. Potential foreign investors have encountered serious difficulties in China in negotiating not only with potential partners but also with the government. One pervasive problem is the complexity and the many changes over time in the rules, procedures, incentive structure, etc. As Box A and Table 8 in Annex 4 show, joint ventures are subject to approval by many different authorities at the municipal, provincial or central level. Different ministries (or their municipal or provincial equivalents) enter the picture, depending on the sector of the proposed venture. Each Special Economic Zone has its own set of procedures and incentives. Municipal and provincial governments frequently issue new rules and permit new incentives, which of course differ from place to place. Moreover, these are often not collected in any one published docu- ment, and in some cases are not published at all. Steps need to be taken to simplify the process, to make all tte relevant information more readily avail- able, to aim for more stability in the rules of the game, and to assure that provincial and local authorities follow both the letter and the spirit of the basic laws and regulations. - 57 - 2.96 Tax and Other Incentives. Investors usually ask for high protec- tion, which in general should not be granted. Similarly, China should take care in accepting foreign investment in those sectors which currently benefit from high rates of effective protection (such as consumer goods industries), unless it can be shown that such investment can survive without such protec- tion. Income tax exemptions and other incentives that may increase after-tax profits have little or no influence on investment decisions of domestic market-oriented ventures, and thus in most cases are needless giveaways from the government to the investors. Brazil, for example, has attracted a great amount of foreign direct investment without income tax exemptions. Export- oriented investment, on the other hand, requires conditions for competing with other countries, such as highly cost-effective labor, easy access to inputs at world market prices, excellent infrastructure, etc. For competitive reasons, income tax exemptions may be useful in attracting such investment, although they do not make up for deficiencies in other areas. In a country such as China, it may make sense to concentrate a package of fiscal and other incen- tives for export-oriented firms on certain geographic areas that have the transportation, labor, and infrastructure needed by such firms--as is the in- tention in the Special Economic Zones. However, a "neutral" trade environ- ment, available throughout the country for any enterprise, is likely to be more effective, as discussed above (para. 2.66-2.67). 2.97 Independence of Management. Chinese regulations and other actions have restricted the ability of joint venture managers to implement and achieve the good management practices and efficient operation that are among the major benefits that foreign investment can bring. Among the major problem areas have been limitations on the ability to select workers, to reward them for good performance or penalize them for bad performance, and to fire them if necessary. These restrictions have included limitations on the organization of productive processes and of the size and composition of the work force. Other, more subtle modes of interference occur when Chinese managers or workers in joint ventures respond to the dictates of political officials (both inside and outside the enterprise), rather than to the appropriate official of the enterprise, on operational matters. The government has recognized the necessity of increasing the independence of enterprise management, and Articles 12 and 15 of the October 1986 regulations have guaranteed considerable freedom to managers. 2.98 Exploitative or Collaborative Behavior. Another pervasive problem is the habit of some Chinese partners and many officials of continually trying to extract the most short-run benefit in dealing with foreign investors, to the detriment of China's long-term gains from foreign investment. 2.99 In examining the experience of other countries, it is clear that the successful countries have treated the process as a positive sum game--i.e., they approached foreign investment as an activity that should leave all parties better off than they were before. They have treated the foreign investors as potential long-term allies in the development of the country. To insure a positive contribution, they have regulated foreign investment; they have directed it to where they wanted it; they have prohibited it from some activities and have given it incentives to engage in others. But they have done this in ways that have helped joint venture enterprises to prosper, thus attracting more reinvestment of profits from existing enterprises and more new investment from others. Some officials agree that China should adopt such an approach to aim at maximizing long-term benefits from all potential foreign - 58 - investors, not just immediate benefits from those already there. Adopting this more far-sighted approach, as well as taking steps to assure its imple- mentation throughout the country, are worth considerable effort. 2.100 Local Sources of Finance. It is normal practice in most countries to finance a certain part of a company's needs with debt. The appropriate debt/equity ratio varies with the riskiness of the venture, the expected variability in its cash flows, and other factors, but in most cases investors find it unnecessarily expensive and risky to finance companies completely with equity. In China, some progress has been made in providing institutions and channels through which joint ventures can borrow or obtain loan guarantees-- either in RMB or in foreign currency. But more could be done. One problem is the legal ceilings placed on the debt/equity ratio of each foreign joint venture, which depend only on the size of the company. (Small companies must be financed 100X with equity). These restrictions make no sense from the point of view of the enterprise. There are two other problems--different but related-with the current state of affairs regarding sources of loans and guarantees. One is a problem for the enterprises: it may be difficult for them to borrow what they need. The other is a problem for China: almost every existing source oi loans or loan guarantees in effect involves the Chinese State, ultimately, as the lender or the guarantor. This obligation may be in part responsible for the government's mixed feelings about foreign investment as a source of capital. What is needed, therefore, is the creation and strengthekling of institutions and channels that can lend to joint venture enterprises and are not subsidized by the State, but responsible for their own profits and losses. 2.101 Policies for Export-Oriented FDI. To attract FDI that is focussed on producing manufactured goods primarily for exports, any country must compete with other countries that are also trying to attract the same invest- ments. This competition is very keen. The characteristics that tend to attract such investment are: (a) disciplined, hardworking and skilled labor; (b) good transportation and communication facilities; and (c) freedom from government interference in the management of the enterprise. Other characteristics of some importance are low prices for other inputs such as land, utilities, etc.; political stability; and low taxes. 2.102 Geographically, China's location in the Far East places it in direct competition with the world's best performers in the competition for this kind of investment. Potential foreign investors in export-oriented ventures have found that labor in China is neither as cheap as they had expected nor as highly productive. Regulations or practices have often required some sort of parity of wages for local workers with workers in similar occupations in more advanced countries. Foreign investors have also encountered high rents for factory sites, high charges for electricity and water, and deficiencies in communication and transport systems. They have encountered bureaucratic difficulties in doing business. 2.103 Recognizing these problems, the Government has been taking steps to solve them. There has been progress on every one of the problems mentioned. The 22 Articles of October 11, 1986 promise better treatment in regard to surcharges on wages, site rentals, utility supply and charges, taxes, and the - 59 - freedom to manage enterprises without interference, including the freedom to hire, fire, and set wages independently. Additionally, regulations issued on December 1, 1986, provide for importation of inputs without the need for licenses or the payment of duties or other taxes. Thus, although the results will depend on how well these regulations are implemented and how well other problems facing export-oriented enterprises are dealt with, policies for such enterprises are being improved significantly. 2.104 Technology Transfer. The authorities rightly stress that foreign investment is especially welcome when it brings advanced technology that China needs. The regulations of October 11 specifically provide for better treat- ment of such enterprises. However, past experience suggests that the criteria of what constitutes useful "advanced technology" may sometimes be too narrow. At times, there is excessive focus on "hardware"--machinery and equipment, or the plans, patents, or process description for making such equipment. As noted elsewhere in this report, the so-called "soft" aspects that increase efficiency--good management, labor training, efficient organization of produc- tion and of all other functions of the enterprise--are just as important. 2.105 The potential for technology transfer from foreign investment is by no means limited to the joint venture itself. Joint ventures can help to modernize the rest of a developing economy as well. Contacts with purely local enterprises can diffuse technology and good management practices throughout the economy. But such contacts must be permitted and indeed encouraged; joint venture enterprises should be encouraged (but not forced, without regard to price and quality) to purchase inputs from local sources, to sell products to local users, and to compete with Chinese enterprises. Of course, for the local institutions to benefit fully from this contact, they must themselves be motivated to improve and be free to improve. By moving in this direction, China's ongoing enterprise reform will enhance the usefulness of foreign investment in the modernization of the entire economy. However, to attract such foreign investment in significant amounts, China will have to make further progress on the problems discussed above, especially those associated with the foreign exchange balance requirement. - 60 - III. TRADE POLICY AND MACROECONOMIC MANAGEMENT 3.1 The second chapter of this report dealt with various issues relating to the efficiency of China's foreign economic relations: how to influence decisions on trade and exploit China's comparative advantage; how best to arrange for the utilization of technology and manage technology transfer; and how to promote appropriate foreign direct investment. The present chapter is concerned with the macroeconomic environment within which these decisions will be made. Whilst efficiency questions are critical, they can be overwhelmed if there is a failure to maintain macro-balance. 3.2 If macroeconomic policies are not appropriate, micro reforms can cause macroeconomic imbalances, which can result in a reversal of the micro- economic reforms, with adverse consequences for improvements in efficiency. Moreover, macroeconomic imbalances are likely to lead to large and unsustain- able balance-of-payments deficits and an excessive amount of external debt, as illustrated by the experience of many countries, particularly the large Latin American states in the 1980s. Therefore, if China decides to decentralize foreign trade decisions to enterprises and independent foreign trade corpora- tions, it will be vital to ensure that these are carried out in an appropriate macroeconomic environment. Of particular importance are the exchange rate, monetary and fiscal policies, and external debt management. A. The Role of Macroeconomic Policies The Exchange Rate 3.3 The exchange rate is one of the most important prices in an economy. It links international and domestic prices and determines the relat-'ve price between tradeable and nontradeable goods and services. If the exchange rate depreciates or is devalued, the price of tradeables--both exports and import substitutes--rises, and, in response, the supply of these products increases, and the balance of trade moves to the direction of a smaller deficit (or greater surplus). The size and speed of the supply response depends on the extent to which it results from (a) putting to use previously idle resources; (b) increased productivity through more intensive use of resources; (c) movement of resources from nontradeables, such as housing construction and services, to tradeables; or (d) movements of resources within the tradeables sector to more productive employment (e.g., from previously protected import substitutes to exports). The supply response will be quicker where the first two conditions prevail than in cases where a transfer of resources is required. If the exchange rate appreciates, the reverse of what happens in the case of a depreciation occurs, and the trade balance moves in the direction of a greater deficit (or smaller surplus). 3.4 Exchange rates can be determined by market forces or administra- tively. Where the rate is determined administratively, indicators are needed to signal whether the exchange rate is appropriate. These indicators may also be useful when the exchange rate is freely determined to assess the likely future path of the rate and to determine whether or not it is consistent with other policies and with short-term and medium-term economic policy objectives. - 61 - 3.5 In assessing the appropriateness of an exchange rate, it is impor- tant to take into account the stance of other policy instruments. In particu- lar, excessive administrative restrictions on external transactions, high import tariffs, and/or export subsidies result in an exchange rate (whether it is administered or freely floating) which is higher than would occur in their absence. While such an exchange rate may be consistent with external balance, it is not fully appropriate, in that it relies on restrictions that result in an inefficient pattern of foreign trade and a misallocation of domestic resources. If it becomes necessary to increase such restrictions to keep the balance of payments within sustainable limits, this is evidence that the exchange rate has become more overvalued. Alternatively, if restrictions are not increased, and an excessive balance-of-payments deficit emerges, this is also evidence that the exchange rate has become more overvalued. 3.6 Within the above context, there are a number of indicators for assessing the extent to which an exchange rate is inappropriate. The most often used indicators are elasticity analyses, various indicators of competi- tiveness, commodity-specific analysis, and exchange rates in legal or illegal parallel markets (see Box 3.1). While no one indicator is wholly reliable, if a number of indicators are used, and if proper account is taken of their statistical and methodological limitations and of the stance of other policies, an informed judgment can be made. Such a judgment must also take into account the objectives with regard to exports, debt and international reserves of the country concerned. 3.7 In cases where the exchange rate is administered, exchange rate policy must be implemented flexibly to ensure that the rate remains appropri- ate. For that reason, it is necessary to ugdate exchange rate indicators fre- quently and to have the necessary political commitment to make exchange rate adjustments as required. Of course, an exchange rate adjustment can be aVoided by an increase in administrative restrictions and tariffs. In terms of the aggregate demand for imports, administrative restrictions, tariffs, and an exchange rate devaluation can all have the same effect. However, economic theory and international experience strongly suggest that administrative re- strictions tend to be less efficient than tariffs, and these are in turn less efficient than exchange rate adjustments. This is because an exchange rate devaluation discourages imports in a neutral fashion (i.e., does not distort relative prices among them as a selective tariff or administrative import restriction would do) and encourages exports at the same time. Consequently, the use of exchange rate adjustment is likely to lead to a more efficient pattern of foreign trade. - 62 - Bos 3.1: INDICATORS FOR ASSESSING TIE APPROPRIATENESS OF AN EXCUAMSE RAT The elasticities approach estimates the effect of exchange rate adjustments on export supply and import demand and uses such estimates in a simulation to arrive at an exchange rate adjustment that is consistent with medium-term balance of payments objectives. Unlike indicators of competitiveness (discussed below), this method has the advantage that it is not limited to comparison with a particu- lar base period, and it allows other factors that influence the balance of payments to be taken into account. Like indicators of com- petitiveness, however, it is based on historical experience or econo- metric estimates and must be modified to take account of structural change. Data limitations make use of this method in its explicit form difficult. The most frequently used index of changes in a country's overall competitiveness is the Real Effective Exchanpe Rate (REER). This is the nominal effective exchange rate index adjusted for rela- tive price movements in the country concerned and abroad. The main difficulty with this index is that it requires choosing a base period when the country's external position was sustainable. If domestic and external policies and conditiors are more adverse than in the base period, competitiveness would need to be improved beyond what it was in the base period. In addition, it needs to be modified to take into account structural or policy changes (e.g., a reduction in severe restrictions that have been used to keep the balance of payments deficit to a financeable level). Care is also required in interpret- ing the price indices used in this calculation, if extensive price and wage controls exist. commodity-specific analysis usually involves caLculating the domestic cost of producing exports and import substitutes per unit of foreign exchange thereby earned, after correcting for all price dis- tortions and netting out taxes and subsidies. This calculation pro- vides an implicit exchange rate for each commodity and allows a rank- ing of activities according to comparative advantage and an assessment -of the prevailing exchange rate. This method focuses directly on profitability and can be useful in countries where production of tradeables is concentrated in a few key commodities. Because of its data requirements, however, it is seldom used. The existence of a parallel market where foreign exchange trades at a price different from the official rate is prima facie evidence that the official exchange rate is failing to reflect demand and supply fully. A legal parallel exchange rate is sometimes intro- duced as a transitional measure as a means of determining the appro- priate level at which to reunify the exchange rate. The exchange rate in an illegal parallel market is a less reliable indicator of the equilibrium exchange rate, because the size of the market is limited by penalties for dealing in the market. Sources Based on: "Formulation of Exchange Rate Policies in Adjustment Programs," IMF, Occasional Paper No. 36. - 63 - 3.8 Rather than fixing the exchange rate, a number of developing countries have adopted floating rate systems in recent years. In some of these countries, dual or multiple exchange rates were maintained for short periods of time as transitional devices before unification of the exchange system. Most countries that adopted floating systems did so because of severe balance-of-payments difficulties, which previously had been addressed by extensive controls on foreign exchange transactions. This had led to disin- termediation in official exchange systems, with large volumes of transactions going through unofficial exchange markets, making it difficult to predict the equilibrium exchange rate. At the same time, the lack of official foreign ex- change resources had made it difficult to support the administered exchange rate. The floating of the exchange rate was intended to bring foreign exchange transactions back into legal channels and to permit a liberalization of exchange and trade restrictions. In other cases, it was believed that a floating exchange rate would lead to a more efficient allocation of resources than the previous administered system, while in still other cases there was a desire on the part of the authorities to reduce political responsibility for changes in the exchange rate. (Box 3.2). 3.9 As indicated above, some developing countries have maintained dual or multiple exchange rates for short periods as transitional devices to uni- fied floating exchange systems (Box 3.3). Generally, multiple exchange rate systems are not encouraged, because they are likely to result in a misalloca- tion of resources, are difficult tc administer, provide incentives to seek illegal windfall gains, and are harmful to other countries which retaliate aSainst such systems. Apart from these considerations, such systems often do not have their intended economic effects (e.g., to limit inflation or to influence the pattern of trade through their implicit taxes and subsidies). The lesson from experience is that if dual or multiple exchange rate systems are adopted as a means of determining the appropriate level for the exchange rate, they should be as part of a well-conceived plan that includes measures that till help eliminate the need for such a system, and brings about the unification of the exchange system during a specific and, preferably, brief period. - 64 - Box 3.2: RECENT EXPERIENCE OF DEVELOPINC COUNTRIES WITH FLOATING EXCHANGE RATES Until the early 1980s, the conventional wisdom was that the volume of currency transactions and the relatively underdeveloped state of financial markets in most developing countries ruled out the use of freely floating exchange rates for these countries. However, in the last few years, the number of developing countries with float- ing arrangements has risen to 15, from only one prior to 1983 (this was Lebanon which had little choice given its situation). There are two main forms of arrangement that have been developed: foreign exchange auctions (6 countries), and domestic interbank markets (9 countries). In a foreign exchange auction (Bolivia, Ghana, Guinea, Jamaica, Uganda, Zambia), the general method of operation is for the central bank to put up a certain amount of foreign exchange for auc- tion. There are rules to qualify for entry to the auction, and bidders submit written bids for the amount of domestic currency they are willing to pay for a certain amount of foreign exchange. The exchange rate is then determined as the price bid for the last unit of foreign exchange available at the auction. Auctions vary in frequency from daily to weekly, and the percentage of foreign exchange available to the central bank that is auctioned varies considerably from country to country. Interbank r- kets permit commercial banks to determine the exchange rate, the rate is determined on an ongoing basis by negotia- tion between banks and clients, with the group of banks fixing the price so as to match available supply with their clients' demands. It is clear from the list of countries that have adopted these schemes that most of them were in very serious economic diffi- culties, including low foreign exchange reserves, external arrears, and growing illegal transactions in foreign exchange. The experience to date has been generally positive, although in general the interbank experience is 'happier' than that of auc- tions, three of which (Jamaica, Uganda, Zambia) are currently in difficulty. Not only have the arrangements permitted a correction of exchange rate disequilibriat they have also permitted countries to liberalize trade and capital arrangements. This, together with exchange rate changes, has frequently served to stem capital flight and has even encouraged reflow in some casei. Moreover, it has also been the case that fears that such arrangements would lead to a 'free fall' with unstable rates encouraging cost-push inflation, have been largely unfounded. Source: Peter J. Quirk et alt "Floating Exchange Rates in Developing Countries:" IMF Occasional Paper #53, May 1987. - 65 - Box 3.3: THE TRANSITION TO A MARKET CLEARINC EXCHANGE RATE - EXPEIKMICES OF IWD ESCONOIEs of South Korea had extensive administrative restrictions on foreign trade in the late 1950S. Not only was the exchange rate over- valued and thus discriminated against exports, but the economy also had an inefficiert multiple exchange rate system that applied different exchange rates for different types of imports and exports (e.g. government imports, other imports, different types Of export goods). In its economic impact, the multiple exchange rate systeal was not very different from the present trade system in China, where export enterprises receive domestic, rather than international prices for their goods. The transition to a market clearing exchange rate in South Korea in 1964 was the initial and single most important element of trade reform. During 1961-62, the government of South Korea pursued very expansionary fiscal and monetary policies. As a result, the official exchange rate became highly overvalued, and foreign exchange certifi- cates (similar to China s foreign exchange retention rights, but with 100X retention) traded at an increasing premium (501 by April 1964) over the official exchange rate. In Hay 1964, the official exchange rate was devalued by 501. However, the foreign exchange certificate continued to trade at a--albeit lower--premium. A year later, in March 1965, the official and the m&rket-determined foreign exchange certificate rate were unified by setting the official rate on the basis of the certificate price quoted in the free exchan*e market. As fiscal and monetary policies remained expansionary, and inflation rates were consequently higher than those of trade partners, the nominal exchange rate continued to decline (in terms of dollars per won), but the real (inflation adjusted) exchange rate was very stable in subsequent years. In 1982, the Ugandan economy was in a serious condition after a period of civil strife, and the exchange rate was seriously overvalued, but by an unknown amount. In July 1982, Lhe Government established a two-tier exchange rate regime, with an official rate set by the Covernment, and a parallel rate determined by a foreign exchange auction. As one of the key purposes of using an auction was to ascertain the equilibrium exchange rate, this rate was used as the guide, and the official rate was gradually devalued so as to achieve convergence. This was done in stages over a period of 22 months, until convergence was achieved in early 1984. Initially, the parallel auction rate had been three times as high as the official rate, but, interestingly, the parallel rate did not depreciate any further over the dual rate period, and when unification of the rate was achieved, it was at a rate very similar to the opening parallel rate. It should, however, be noted that this successful transition was followed several months late.by a surge of inflation as fiscal policies were relaxed, to;ether with accommodating monetary policies to finance the rising public sector deficit. For a similar experience in another East Asian economy, see Liang and Liang in Hong: "Trade and Growth of Advanced Developing Countries in the Pacific Basin," KDJ Press, 1981. Sources: For South Korea: Frank, Kim and Westphal: Foreign Trade Regimest South Korea, NBER/Columbia University Press, 1975, Chapter 41 For Uganda: as for Box 3.2. - 66 - Monetary and Fiscal Policies 3.10 A discussion of monetary and fiscal policies is beyond the scope of this report and has been dealt wtjh in more depth in the Bank's recent 'Finance and Investment' report,4 However, it should be noted that monetary and fiscal policies impact on the external sector, and can either contribute to or help to alleviate balance of payments problems. Monetary policy should be formulated taking into account the objectives for the rate of natio37l income growth and inflation and likely developments in money velocity._ 3.11 What happens if money creation is higher than would be justified on the above grounds? Credit creation increases the buying power of individuals and enterprises, who use the credit to attempt to purchase goods and services. Rapid credit creation can lead to demand which cannot, in the short run, be satisfied domestically because of supply constraints, especially in an economy such as China's, which has been relatively slow to respond to changing market conditions. This excess demand results in domestic price pressures, or a deterioration in the balance of payments, or some combination of the two. in a situation of limited availability of foreign exchange, with fixed exchange rates and strict import controls, the increment in demand will be channelled entirely to the domestic market, and domestic prices will rise in line with the excess credit creation; The price increases will, in turn, bring about an overvaluation of the exchange rate, and exports will be less competitive (and fall), causing a deterioration in the current account of the balance of payments. 3.12 If there is foreign exchange available, and imports are less strictly controlled, part of the excess credit will be channelled into imports, limiting the impact on domestic price inflation. If a loss of international reserves is acceptable, this situation may not have harmful 'long-term effects, provided that the excess credit creation is quickly curbed. If foreign exchange reserves are scarce, a balance-of-payments deficit can only be avoided following excess credit creation if the exchange rate is permitted to depreciate. In this case, there will be domestic price changes, but compensated by exchange rate changes. This has been the typical situati2-n of several Latin American countries in recent years, notably Brazi'. This circular process of inflation and devaluation can only be stopped if the domestic sources of inflationary pressures are brought under control (e.g., budgetary deficits, excessive credit growth). 3.13 However, it is important to stress that a devaluation will not necessarily cause an increase in the general level of prices. The primary effect of an exchange rate cha:nge is to alter the relative prices of tradeable 2/ See: World Banks 'China - Finance and Investment' (Report No. 6445- CHA), March 1987, Chapter VI. 3/ 'Velocity' refers to the ratio between national income and the stock of money. Velocity could decline as a result of increased complexity and monetization of the economy. - 67 - and nontradeable goods. Of course, prices of imported goods will rise, but experience has shown that such rises have frequently been anticipated prior to any devaluation, because the scarcity of foreign exchange to purchase such imports in a pre-devaluation period permits high prices to be charged for such imports on the domestic market. More frequently, it is the case that it is inflation due to excess money creation which leads to an eventual need for a devaluation. Indeed, in several of those countries which have recently adopted floating exchange rates, there has been a decline in inflation during the period following the floating of the exchange rate (see Box 3.4). 3.14 External borrowing may also facilitate an increase in imports. A rapid accumulation of external debt--particularly commercial debt and debt of short-term maturity--can quickly result in debt servicing problems. Firm control over the level and composition of external borrowing is necessary to , oid such problems (see paras. 3.30-3.35). Box 3.4: INFLATION AND FLOATING EXCHANGE RATES The impact on the rate of inflation of a switch to a float- ing exchange rate regime depends crucially on the stance of monetary and fiscal policies, as well as other policies adopted at the same time as exchange rate reform. In the case of many of the countries that have recently moved to floating rates (see Box 3.2) domestic price liberalization has been pursued simultaneously, not least because price repression had been necessary in order to maintain previously overvalued rates, with accompanying distortions in resource allocation. Therefore, in several of these countries--notably Bolivia, Philippines, the Gambia, Ugandat and Zaire--most price controls were removed at the same time as the change in exchange rate regime, and controlled prices were adjusted upwards in several other countries. For four of the countries for which sufficient post-float data is available, there was a decline in the rate of inflation after an initial rise reflecting the exchange rate change (Dominican Republic, Jamaica, Philippines and Zaire), and indeed the rates fell to below the pre-float level. Most notably, in Zaire the pre-float rate of inflation had risen to 123% by end-1983, but had fallen to 10- 15% p.a. by end-1985 after the float. Source: As for Box. 3.2. - 68 - B. Macroeconomic Policies in Practice--Recent Developments in China's Balance of Payments 3.15 From 1980 to 1984, China ran a current account surplus. 8oth the trade and services balances were positive. The trade surplus peaked in 1982 at $4.6 billion (1.8% of CDP), before declining to a position of near balance in 1984 (see Table 3.1). Because of high levels of reserves and low outstand- ing debt, interest receipts exceeded payments. The services balance remained at about $2 billion in 1983 and 1984. In 1984, therefore, the current account still showed a surplus of $2.5 billion, or nearly 1% of GDP. The buoyant external situation over the period had resulted in a near tripling of reserves to $17.8 billion (equivalent to over 9 months of imports) in 1984. There t'as little external borrowing, and the recorded stock of medium and long-term debt was only 4.5% of GDP. A debt service ratio of only 3.9% (including substan- tial prepayment of debt) was extremely low in comparison to other developing countries. 3.16 Imports surged from $24 billion in 1984 to $38 billion in 1985--a 60% increase. Much of the increase occured in imports of raw materials in short supply domestically (steel, chemicals, fertilizer, timber) and capital goods with new technologies. As a result, the ratio of capital and inter- mediate goods imports to GDP was more than doubled in 1985. The current account moved to a deficit of $12 billion in 1985, with about 601 of this deficit financed by borrowing, and 40% by reserves drawdown. 3.17 Although some progress was made in reducing the deficit in 1986, reserves fell by about US$1 billion and there was external borrowing at almost the same rate as the previous year. The modest improvement in 1986 reflected improved export performance and a containment of imports, which nevertheless remained at a very high level by historical standards. Earnings from petro- leum exports fell to about half the previous year's level, and the improved export performance was the result of an increase of about 30% in non-oil exports, mainly manufactures. The continued reserve drawdown and the cost of servicing the growing volume of foreign debt meant that the sizeable surplus on the services account disappeared in 1986. In future years, this can be expected to be negative factor of increasing significance, as debt continues to accumulate. - 69 - Table 3.1: CHINA - SELECTED ECONOMIC INDICATORS 1982 1983 1984 1985p 1986e ---------------US billion---- -- Exports, fob 21.1 20.7 23.9 25.1 27.9 petroleum 4.9 4.3 5.6 6.8 3.4 manufacture 11.5 12.4 13.9 13.4 17.0 Imports, cif 18.4 20.4 26.0 41.7 39.9 Current account balance 5.8 4.9 2.5 -11.9 -9.5 Reserve changes (- = increase) -6.3 -3.6 -2.4 4.6 1.3 … -------------… change-------------- Real GDP 8.3 9.8 13.5 12.3 8.0 Money and quasi money 14.9 19.2 41.2 18.2 16.7 …--…----I----- of GDP…------------- Overall fiscal balance -1.5 -1.8 -1.6 -0.5 -0.4 Current account balance 2.2 1.6 0.9 -4.5 -3.7 Investment 30.3 30.9 29.4 38.4 38.6 Source: State Statistical Bureau, World Bank an IMF staff estimates. 3.18 Four main factors appear to be associated with the unusually rapid growth in imports: (a) Credit expansion: during 1981-83, total credit, as measured by the net domestic assets of the specialized banks, grew relatively modestly. During 1984 and the first nine months of 1985, however, credit increased very sharply, by 45% in 1984 and by a further 25% during 1985. By June 1986, the rate of expansion had eased, although by this date net domestic assets were at double the level of December 1983. Figure 3.1 shows how credit expansion and the growth of imports and the balance-of-payments deficit compare. Much of the problem of excess credit creation during 198h-85 relates to the decentralization of credit decisions to the specialized banks, in advance of administrative or market instruments to control the overall quantity of such credit by those banks. Improved controls were established in April 1985, and the situation has since improved, although strains on the system seem to be emerging again, with a 30% credit expansion in 1986. - 70 - (b) Import liberalization: The easier access to imports described earlier (and in Annex 1) contributed to the problem. This was a two-part phenomenon. Firstly, the liberalization facilitated the spilling over of excess credit into imports, rather than domestic prices. Secondly, the liberalization may have sparked some of the demand for credit, since the availability of capital goods imports probably caused many enterprises to seek credit to finance such imports. (c) Reduction of excessive foreign exchange reserves: The increase in domestic credit was translated into imports, partly because the Government decided that foreign reserves had risen to rather high levels (nine months of imports at end-84), and that some of these should be used to finance the modernization of the economy. Exchange and import controls were therefore liberalized, and enterprises and provinces were encouraged to use domestic credit to purchase foreign exchange up to the limit of foreign exchange retention rights built up in previous years. (d) External debt. External borrowing increased sharply during 1985 and 1986. In 1985, this mainly reflected increased borrowing from foreign commercial banks--much of it at short-term maturity--by the Bank of China and other entities. The granting of foreign borrowing rights to provincial governments and enterprises without appropriate procedures for controlling debt also was instrumental in the rapid accumulation of external debt. External borrowing provided finance for imports in the context of more liberal import restrictions and, in some cases, boosted foreign exchange reserves to a level that permitted enterprises to use accumulated foreign exchange retention rights without running down foreign reserves to an inadequate level. The further significant accumulation of external debt in 1986 was partly responsible for the continued high level of imports in 1986, although, unlike 1985, a smaller proportion of such borrowing appears to have been of short-term maturity. 3.19 At the same time, exports did not grow much in 1985. There is no doubt that a major explanation of this is that FTCs had difficulties in procuring supplies for above-plan exports. This is because the partial liberalization of domestic prices for consumer goods, combined with overly rapid expansion of money and credit, led to price increases, which caused domestic markets to be much more attractive than export markets for above-plan production. A more depreciated exchange rate would have enabled FTCs to secure a greater proportion of above-plan production for export. At the same time, it would thave limited the impact on import demand of import liberaliza- tion, increased external borrowing, and excess credit creation. 3.20 The deterioration in China's balance of payments reflects to a large extent a lack of coordination between macroeconomic and microeconomic policies and inappropriate use of macroeconomic policies, partly because of a lack of familiarity in using macroeconomic controls. In particular, the authorities were not able to control access to borrowed foreign resources, credit expansion was excessive, and exchange rate policy was not operated with - 71 - sufficient flexibility. A more appropriate and coordinated use of macro- economic policies would have helped to prevent import and exchange liberalization and economic reform measures from having adverse effects on the balance of payments. Two areas of policy are discussed in detail in the following section. FIGURE 3.1: MONEY SUPPLY AND IMPORTS, 1983-86 (% change) 13 114 7"s C. Macro-Policies and Trade Reform tExcag«e Rat Policies and Foreign Exchange Allocation 3.21 In China, as explained in para. 2.23, the effect of the exchange rate is muted because of the "air lock" between world prices and domestic prices--especially producer prices of exported commodities. Administrative controls on domestic prices, production, income distribution, and factor mobility also dampen the potential effects of exchange rate policy. In particular, most enterprises are not fully responsible for their profits and losses and do not, at present, face the prospect of bankruptcy if they make substantial losses. Trade reforms of the type proposed in this report would make the exchange rate a more effective policy instrument by increasing 1links between foreign and domestic prices* However, the effectiveness of these reforms depend on their being complementea by further enterprise and price reforms. 3.22 While trade reforms will make the exchange rate a more effective policy instrument, it is also true that a more appropriate and flexible exchange rate policy would make trade reforms more effective. This objective - 72 - could be achieved by a more active use of exchange rate policy within the context of the present exchange system. Alternatively, China may decide to allow market forces to play a more direct role in exchange rate determination and move to a system in which the exchange rate would be determined in the market by the supply and demand for foreign exchange. This could be particulary useful during the transition towards a more market-oriented foreign trade system. It would be appropriate and useful to discuss changes in the foreign exchange allocation and exchange rate policy with the IMF., 3.23 Currently, China maintains a managed floating exchange system, whereby the exchange rate for the renminbi is based on developments in the balance of payments and movements in costs and exchange rates in China's major trading partners. The U.S. dollar is the intervention currency. In practice, adjustments in the rate for the renminbi vis-a-vis the U.S. dollar tend to be made infrequently, with the result that over time the rate tends to move away from (or further away from) its equilibrium value. Infrequent adjustments provide some nominal stability in the cost of foreign goods. However, this advantage is outweighed by the resulting decline in competitiveness, which can occur, and the instability of the real (inflation-adjusted) exchange rate, and of expectations regarding the sustainability of the exchange rate. 3.24 If the present managed floating exchange rate system is maintained, exchange rate adjustments would be needed to compensate for the removal of administrative import restrictions or subsidies. Balance-of-payments develop- ments, desired reserve levels and indicators of international competitiveness (see para. 3.6 and Box 3.1) could be used to estimate the required adjust- ments. An advantage of this option is that it would not represent a major change from the present exchange rate system, but would introduce more flexi- bility than at present. A disadvanage of this option is that it would take quite some time (several months or quarters) to ascertain whether any exchange rate adjustments made have been sufficient to compensate for the removal of trade restrictions or other changes in the domestic or international economy. This might make it difficult to phase out administrative allocation of foreign exchange rapidly. 3.25 An alternative approach that would permit a more rapid transition towards market determined allocation of foreign exchange would be to permit trading of foreign exchange retention rights among enterprises and other units at a freely negotiated price. This would be similar to the approach used in the domestic market where transactions at flexible prices have been introduced while maintaining fixed prices for the same goods allocated by the Plan. 3.26 While not desirable as a permanent solution, such a system would improve on current practice, not only because of the limitations of the current foreign exchange allocation system (see para. 2.28), but also because 4/ In formulating exchange rate policies, China has certain commitments as a member of the IMF under Articles VIII and XIV of its Articles of Agreement and its general role in the surveillance of members' exchange rate policies. - 73 - it could lead to an elimination of some of the present multiplicity of de facto exchange rates resulting from the current, highly variable levels of export incentives. Since the exchange rate for foreign exchange retention rights would be freely determined, it would permit rapid removal of administrative restrictions without running the danger of incurring large balance-of-payments deficits. The exchange rate established in the market for foreign exchange retention rights would give an indication of the direction and magnitude of the required adjustment of the official exchange rate to compensate for the removal of administrative restrictions. 3.27 The problems with such a system are similar to the use of dual (plan-market) prices in the domestic economy. Enterprises or individuals that have access to foreign exchange at the official rate will always be tempted to sell it at the price established in the market for foreign exchange retention rights. It would thus be desirable to move the official exchange rate as rapidly as possible towards the rate established in the market for foreign exchange retention rights, eliminating any gap between the two rates after a relatively brief transition period, measured in months, rather than years. 3.28 To ensure that the rate established in the market for foreign exchange retention rights is realistic, it would be important for the market in foreign exchange retention rights to cover a large proportion of foreign exchange transactions and to have a large number of enterprises participate, particularly those enterprises that are already responsible for profits and losses. It would thus be necessary to increase gradually the foreign exchange retention ratio for enterprises. The State would always be able to obtain foreign exchange for Plan requirements by entering the foreign exchange market. 3.29 Once the transition to a more market-oriented foreign exchange rate determination system has been completed by closing the gap between the offi- cial exchange rate and the market-determined rate for foreign exchange reten- tion rights, it would be possible to opt either for a floating exchange rate SyStem (no intervention by the central bank), or a managed floating exchange rate system where the central bank would intervene to stabilize the exchange rate to avoid undesirable short-term fluctuations in the real (inflation- adjusted) exchange rate. External Debt Management 3.30 As already noted, the external imbalances of 1985 and 1986 resulted in a rapid accumulation of external debt, which rose from US$12 billion at end-1984 to an estimated US$27 billion at end-1986, with the ratio of debt to export earnings rising from 51% in 1984 to 97% in 1986. While the level of debt is by no means excessive in international creditworthiness terms, the trend needs to be monitored. If China were to continue to contract debt at the present rate, debt would be a serious economic issue by the 1990s. There are also questions related to the composition of debt with, at present, a high component of short-term debt. (See Annex V for a full description of China's external debt situation). - 74 - 3.31 China's foreign borrowing in 1985 and 1986 was not simply a conse- quence of the large trade deficit, but a cause of it as well. The access of a number of actors to foreign borrowing and the inadequate controls over their actions, enabled a large volume of imports to be brought in. Insofar as creditors normally require a guarantee by the Bank of China or an other guarantor approved by the State Administration for Foreign Exchange Control (SAEC), there is a mechanism for controlling such borrowing through the approval limits of these guarantors. Insofar as approvals are not required (or not obtained), and guarantees are not asked by foreign banks, debt can be contracted by these enterprises and other units with relative ease. Indeed, in 1985 and 1986, foreign commercial banks were highly liquid and had become convinced that China was a profitable and secure market, and began to lend without guarantees. The understanding appears to have been that, since they were lending to provincial authorities or public bodies, these loans would in fact be guaranteed. Indeed, in the Latin American debtor countries, even private borrowing has effectively become sovereign risk with a blurring of the distinction between guaranteed and non-guaranteed private borrowing (see Annex 5, Box 5.3). Indeed, in Chile, private sector debt was taken over by the government as a condition of a debt rescheduling agreement. In the absence of effective control over foreign borrowing, there is serious risk that it could continue to increase inconsistent with a viable medium-term balance of payments. 3.32 There are essentially three steps involved in managing the external debt position: (a) assessing the size of external debt; (b) formulating a debt strategy; and (c) monitoring and supervising borrowing to ensure imple- mentation of the strategy. Significant progress has been made with the appointment, in April 1986, of the State Administration for Exchange Control, (SAEC) within the People's Bank, as the sole agency responsible for the monitoring of external debt. SAEC conducted a debt survey in 1986, the first comprehensive debt survey in China. However, there continues to be a major discrepancy between the tocal public debt estimated by SAEC and that derived from creditor-reported statistics (see Annex V). This discrepancy appears to reflect short-term debt, particularly trade credits and borrowing by joint ventures. Efforts are underway to eliminate these discrepancies. Progress on this needs to be continued, as up-to-date information is essential for proper debt management. 3.33 Equally important is the need to extend the reporting system to include contingent liabilities and other debt-like flows. Contingent liabili- ties are guarantees of foreign obligations issued by government-affiliated institutions or enterprises in China, which are not presently reported centrally (although, in theory, the underlying borrowings are). If debt planning is going to be meaningful, it must be on the basis of as complete a coverage as possible. This means including debt-like flows such as obliga- tions under leases, compensation trade or similar arrangements which are very similar to debt financing, but in China are often included in "foreign invest- ment." Finally, there is a need to clarify the distinction between sovereign and other debt. Consider the example of a joint venture between a state enterprise and a foreign partner. If this joint venture borrows abroad--which it can do without any central permission--and is subsequently unable to repay, it may well be that the central government will feel compelled to assume the joint venture's foreign liabilities. - 75 - 3.34 China's debt strategy has not yet been clearly formulated. A one- year plan for official borrowing is formulated at the same time as the trade and foreign exchange allocation plans, and there is also a long-term planning horizon with global targets for import and export growth and, consequently, an implicit debt strategy. However, there is no explicit medium-term debt strategy on which to base the one-year plans. Such a strategy needs to be concerned with both the volume and composition of debt. 3.35 The formulation of a strategy for the volume of debt should be based on the development of a medium-term (5-10 year) outlook for the balance of payments. The key issue is affordability. Contracting debt today permits a higher level of imports, investment and consequently--if investments are made wisely--higher economic growth and increased debt-servicing capacity. Chinese authorities have indicated a desire not to see debt-servicing (interest plus principal) exceed 15% of export earnings. As discussed below, these are con- servative but appropriate targets. However, unless placed in a medium-term context, China could either end up exceeding such targets because of inade- quate planning, or have to undergo sharp contractions in imports and borrowing at some future date to avoid exceeding them. 3.36 Having decided upon an appropriate volume of debt, it will be equally important to plan a debt composition strategy. This relates to maturity, interest rate risks, and exchange rate risks. Debt servicing could rapidly rise and become much higher than desired, if the planned volume of debt is contracted at variable rates in appreciating currencies, as was much of the 1985 commercial debt. Thus, these aspects of the borrowing plan should also be identified in advance, so that indivitual debt transactions can be judged in a framework. Of course, the appropriate volume of debt is determined, in part, by the expected composition of the debt, including cost (concessional vs. financial market debt) and maturity (short- vs. long-term debt). (See Annex 5 for a fuller discussion of these issues.) 3.37 The third and most difficult aspect of debt management is the control of debt. The conflicts here are several: balancing the need to decentralize micro-economic decision-making while maintaining overall macro- economic control; balancing the need to control non-sovereign debt with the desire to avoid accepting explicit or implicit sovereignty of the debt; and balancing provinces' and enterprises' desires to promote their own economic development with national economic management priorities. However, two issues are paramount: most foreign lenders do regard all lending to China as sovereign debt, given the public ownership status in China; and the sum of all potential borrowers' desires to contract external debt undoubtedly exceeds the level of de'bt which central authorities would consider prudent. Therefore, we would strongly recommend the maintenance, and strengthening of central control and approval procedures for external debt, but that subject to these proce- dures the conduct of debt operations should continue to be carried out by a variety of institutions. Detailed recommendations in this area are given in Annex V, but can be summarized as follows: (a) Prior approval by a central government agency (such as the State Administration of Exchange Control, which we understand the Government has selected for this purpose) should continue to be - 76 - required for all debt operations of the eight financial institutions with foreign borrowing rights. Extension of such rights to other institutions should be very gradual. (b) Prior approval should be required of borrowing by joint ventures, direct overseas borrowing by subsidiaries of Chinese corporations, and the assumption of quasi-debt such as compensation trade. It is not suggested that detailed and difficult bureaucratic procedures be established for these forms of borrowing. Rather, they should be placed on a prior notification basis, with the right of objection by the central government debt agency. Certain minimum conditions (e.g., maturity) could be set for such borrowing. (c) Contingent liabilities should come under the control of central authorities. These relate to guarantees issued by the 68 institu- tions so empowered. D. Macro-Economic Outlook With and Without Reforms 3.38 The purpose of the reforms recommended in this report are to encourage efficient trade, which will permit technical transformation to take place and thus encourage economic growth, whilst avoiding debt crises. In the following paragraphs, these aims will be assessed in a simple quantitative framework. However, prior to presenting these analyses, it is important to stress that these are not projections for the Chinese economy, but rather indications of the direction and implications of changes of major macro- variables under different assumptions. 3.39 Given the focus of the report, the scenarios have been constructed using variations in two main variables: the rate of growth of exports and the level of external debt. The case was made in Chapters I and II that with the external constraints China is facing, future export growth will rely heavily on the growth of manufactured exports, and that the latter will be determined to a significant degree by the pace of trade policy reforms. Therefore, scenarios are worked out using two different assumptions for export growth rates of manufactures (5% and 101). Similarly, the implications of different rates of growth of debt are examined. The assumptions made reflect the actual experience of developin, economies pursuing different trade policies (Chapter I, Table 1.1).- 5/ These scenarios are not constructed to be projections of the Chinese economy over the next eight years, but rather they are intended to illustrate the magnitude of the implications of alternative economic policy strategies. - 77 - 3.40 The other two parameters which differ between the alternative scenarios are GDP growth and the level of consumption and domestic savings. The higher level of imports in Scenarios 3 and 4 pgrits a higher level of investment and thus GDP than in Scenarios 1 and 2.- GDP grows at 8% p.a. in Scenarios 3 and 4, and at 6% p.a. in Scenarios 1 and 2. Similarly, the two high debt scenarios permit a higher level of consumption and lower level of domestic savings--at least in the first 6-7 years--than in the low debt scenarios. These scenarios have all been constructed using a macro-economic consistency model. Although there are limits on the extent that such models can simulate long-term structural changes, the assumptions have been chosen so as to reflect the actual experience of developing economies pursuing such different strategies. 3.41 Putting together these two sets of assumptions gives four scenarios, whi:h can be typified by reference to particular country examples: Export Growth Debt Growth Example Scenario 1 Low Low China pre-79, India Scenario 2 Low High Poland Scenario 3 High Low Taiwan, China Scenario 4 High High South Korea 3.42 The results of these scenarios are shown in Table 3.2. It is instructive to compare the results in two sets of pairs. Consider first Scenarios 3 and 4 in comparison with Scenarios I and 2. By 1995, import levels are 38-46% higher in the first pair than in the second. The difference is less marked in 1990, showing the impact of sustained export promotion efforts. It is also interesting to note, however, that the early (1990) difference between Scenarios 3 and 4 in terms of import levels tends to narrow over time. In 1990, imports are 7% higher in Scenario 4 than in Scenario 3, but by 1995 the difference is down to 2.4%. This is because the higher debt service burden in Scenario 4 begins to absorb an increasing share of foreign exchange earnings. 6/ It is assumed that concessional funds available to China--at 0-2% interest rates--are the same under all scenarios, at about US$1.3 billion per annum. Other funds carry an average interest rate of 8% per annum, and consequently average rates of interest are somewhat higher in the high debt scenarios. - 78 - Table 3.2: MACROECONOMIC SCENARIOS (Current US$ billion, unless stated) 1986 1990 1995 Scenario 1 (low exports, low debt) Exports 27.9 36.4 55.4 Imports 37.6 40.1 59.4 Current Account Balance -9.6 -4.1 -4.z Debt Service/Exports of Goods and Service (%) 8.5 12.9 9.6 _--_____--____________________________________________________________________ Scenario 2 (low exports, high debt) Exports 27.9 36.4 55.4 Imports 37.6 43.2 60.8 Current Account Balance -9.6 -8.0 -8.0 Debt Service/Exports of Goods and Service (X) 8.5 16.5 17.8 …------------------------------------------------------------------__-------- Scenario 3 (high exports, low debt) Exports 27.9 42.9 80.8 Imports 37.6 46.4 84.0 Current Account Balance -9.6 -4.0 -3.8 Debt Service/Exports of Goods and Service (%) 8.5 11.5 7.8 -----------------------------------------------------------------------__----_ Scenario 4 (high exports, high debt) Exports 27.9 42.9 80.8 Imports 37.6 49.7 86.0 Current Account Balance -9.6 -8.2 -8.3 Debt Service/Exports of Goods and service (Z) 8.5 14.6 13.9 Source: Mission estimates. - 19 - 3.43 The other two pairs to examine are Scenarios 1 and 3 in comparison with Scenarios 2 and 4, thAt is, lower or hiiglher debt. The debt service ratios of each country within a pair are similar, but in each pair Lhe country with the higher export level can afford 45% more imports. 3.44 These are admittedly simple scenarios, but they bear strong resemblance to the recent experience of many countries. They demonstrate very clearly that over the medium-term, trade strategy is far more important than debt in increasing import levels. A debt accumulating policy is only sustainable for a short period, and should only be undertaken in conjunction with and not as a substitute for export promoting policies. Scenario 4 includes debt accumulation of US$9 billion per annum, but because of high export growth rates, debt servicing remain tolerable. Scenario 2, however, shows the results of using debt instead of export promotion, and demonstrates how a countrv can end up with high debt servicing but without high exports and domestic economic growth. The scenarios therefore emphasize the urgency for China to formulate appropriate trade policies and to develop both a debt strategy and the means to implement it. E. Conclusions for Macroeconomic Strategy 3.45 The foregoing analys's suggests the need for a reorientation of economic planning, but with fundamentally similar objectives as past planning activities. China has already moved in such a direction. While trade planning and annual foreign exchange and debt plans continue to have a central function, there is already an 'above-plan' level of trade and debt, being influenced by economic levers. 3.46 The preceding analysis has focussed on the importance of defining a debt strategy and putting mechanisms in place to ensure that it can be achieved, as well as the critical role of the exchange rate in ensuring macro- balance. These two areas come together in macroeconomic planning. In carrying out such planning--to put it very simplistically--there are four key steps: (a) based on a medium-term framework for the balance of payments, the government should identify a target for the current account of the balance of payments in the current planning year; (b) the government should decide how this deficit (if any) is to be financed, thus establishing the debt ceiling, as well as targets for concessional and official lending, foreign direct investment, and private capital; - 80 - (c) the government should then adjust economic levers and other macro- economic policies to achieve the export and import targets. This notably includes exchange rate policy but also fiscal and monetary policies; and (d) since events are unlikely to be fully predictable and controllable, the situation needs to be monitored during the course of the year-- especially the level of foreign reserves and the accumulation of debt, which give early signals--so that economic levers can be adjusted as needed. It is in this context that a flexible exchange rate policy can be critical, for as noted it can act as the safety valve, so that if, say, the debt ceiling is enforced, the exchange rate can adjust for any inappropriateness in fiscal or monetary policy. - 81 - IV. INTERRELATIONS AND SEQUENCING OF REFORMS 4.1 As outlined in the preceding chapters, substantial measures would be required to make China's external trade system more efficient and maximize the benefits to the domestic economy from further opening to foreign trade and in- vestment. Such changes would necessarily take a number of years. At the same time, China is also implementing other economic reforms leading, in particu- lar, to greater decentralization of production and investment decisions to enterprises. Not only are the various trade policy measures discussed here closely interrelated, there are also important interrelations between trade policy reforms and other areas of economic reform. Indeed, trade policy reforms can be seen as an integral part of enterprise reforms. The question then arises what the appropriate sequencing and pace of trade and other policy measures should be. 4.2 Economic theory and international experience offer little help in deciding upon the sequencing and pace of trade reforms. In China the question is made even more difficult by the complex interrelations between trade and domestic economic reforms. There appears to be no alternative, therefore, to the cautious, experimental approach that the Government has adopted, acknow- ledging that some mistakes will be made, and that mid-term corrections will often be required in order to limit the costs of mistakes. Nevertheless, some observations about priorities and sequencing of reform measures can be made, and these are outlined below.-1 Section A will discuss fundamental precon- ditions of efficient decentralization of economic decision-making to enter- prises, followed by an application of these principles to external economic policies in Section B. Finally, Section C will analyze some of the inter- relations between domestic and external economic reforms. A. Fundamental Preconditions fot Efficient Decentralization 4.3 As already noted, reforms of the trade system can be seen as one aspect of the ongoing enterprise reforms. By the late 1970s, China's leaders had concluded that improved economic efficiency and sustained growth required the decentralization of production decisions to the line managers of produc- tion enterprises. But decentralization alone does not assure efficient pro- duction. Enterprise managers may have the technical know-how and hands-on familiarity with their respective enterprise's operations and markets that central planners lack, but what is to lead them to produce those goods in greatest demand and to do so in the most efficient manner? Two further con- ditions appear essential: first, enterprises must also be subjected to competition, so if one fails to satisfy the needs of buyers at the lowest possible costs, another more efficient enterprise will capture the business. Second, there must be a direct link between the results of an enterprise and the rewards to owners, managers and workers. Most important in forging this 11 The recent companion volume to this report: "China - Finance and Investment," Report No. 6445-CHA, World Eank, June 11, 1987 provides a more detailed discussion of domestic economic reform meazures and their phasing and sequencing. - 82 - link is to make enterprises separate accounting units responsible for profits and losses. 4.4 These principles are well understood in China, and although the reform measures implemented over the past few years have brought enterprises and other economic units closer in line with these principles, progress so far has been variable, depending on the type of enterprise (state, collective, private), the sector in which it operates (heavy industry, light industry, commerce), and the type of decision involved (production, investment, domestic and foreign trade, etc). In general, competition is strongest among the smaller private and collective enterprises, but state enterprises in some sectors (e.g. textiles) now also face substantial competition. Competition is still virtually absent in many heavy industry sectors, where most or all output continues to be allocated by the Plan, Also, while production and domestic marketing decision are now mostly in the hands of enterprise mana- gers, they have far less autonomy in investment and foreign trade decisions. The former are still mostly subject to review by supervisory agencies and planning and economic commissions at the relevant levels, while the latter are controlled through administrative decisions on foreign trade and foreign exchange allocation. As discussed further below, there are good reasons to treat foreign trade decisions exactly like domestic production ar.d marketing decisions. 4.5 ?rice Reform. Decisions made by enterorise managers in the pursuit of profit will only be consistent with efficient production if the prices of goods and the resources used in their production reflect the scarcities of, and the demand for, these resources. Substantial progress has, in fact, been made over the past several years in deregulating the prices of a large number of consumer goods. Some progress has also been made in adjusting prices of those commodities (largely heavy industry products) whose prices continue to be administratively set. Equally important has been the marketing of above- quota or outside plan production of these commodities at market-related prices, which has helped to ease shortages appearing under administrative allocation. The underpricing of financial capital (interest rates) is one of the most important sources of remaining price distortions, leading to excess demand for investment in the same way that an overvalted exchange rate (i.e. too low a price of foreign goods in terms of renminbi) leads to excess demand for imports. The interest rate and the foreign exchange rate are the two most important prices in a decentralized economy open to foreign trade, given the impact they have on resource allocation across the entire economy. 4.6 Other price distortions also influence foreign trade decisions. In particular, the underpricing of raw materials and other capital-intensive goods could lead to excessive exports of these goods and too little export of labor-intensive goods (such as light industry products), in which China's comparative advantage likely to be the is largest. This is because of the high explicit and implicit tax burden on labor-intensive manufactures that is used to provide the financi&I resources required to subsidize heavy industry production. Consequently, China's export pattern remains suboptimal, with exports using more domestic resources than necessary. Further progress in domestic price reform is thus desirable. Chinese enterprises have proven quite resilient to the relative price adjustments already introduced, and - 83 - price reform (adjustment and decontrol of prices) could probably be accele- rated, if the authority of the People's bank in restraining the growth if money and credit is strengthened, and inflationary pressures are thus mini- mized. 4.7 Competition. Markets only function efficiently if enterprises are exposed to vigorous competition. The gradual decontrol of prices is the most important element in promoting competition and compelling enterprises to seek progressive improvements in quality and reductions in production costs. However, competition is not always a naturally self-sustaining state of affairs. Enterprise owners and managers will always have the incentive to try to monopolize markets and to reduce the threat of competition from other enterprises. The situation is particularly vulnerable where ownership is vested in government agencies that also have the power to regulate important elements of the enterprise environment and can, for example, restrict entry of other enterprises into a particular product line. As discussed in paras. 2.10-2.11, this does not only apply to domestic production and commerce, but also applies to foreign trade. In China, the benefits of restricting compe- tition amonS domestic enterprises in export activities have often been over- estimated, while the costs of such restrictions have not been sufficiently recognized (see paras. 2.20-2.23 and Box 2.3). 4.8 Enterprise Ownership and Responsibility. The assumption that enter- prise owners will be motivated by profits in evaluating managerial performance in production and investment decisions requires that ownership rights, including those of state enterprises, are clearly defined, and that owners' interest are linked to profits and the long-term prospects of the enter- prise. In China, state enterprises often are led to pursue noneconomic objectives which conflict with economic objectives, and government agencies have used their regulatory power to protect enterprises under their control from competitive pressures. Efficient decentralization will require some clarification of the nature and locus qg enterprise ownership rights and res- ponsibilities. As discussed elsewhere!' diversification of ownership and competition can help to reduce the temptation to promote noneconomic objectives. Further tax reform and the gradual phasing-out of adjustment taxes and other negotiated subsidies or levies would also be desirable; otherwise, prices will remain distorted. Finally, the legal rights and responsibilities of enterprises and their owners, creditors and workers need to be clarified (e.g., in an enterprise law and by implgmenLing the bankruptcy code), and the rights and obligations of contracting parties need to be elucidated (e.g., in a commercial code). 4.9 Factor Mobility. As discussed in Chapter II (paras. 2.73-2.77), technology transfer and diffusion could be considerably accelerated if labor mobility, particularly among technical personnel, were to be increased. A number of changes could be made to facilitate greater labor mobility, in particular, freeing enterprises from the direct responsibility for social 2/ World Bank: "China - Finance and Investment," Report No. 6445, March 198k, Chapter III. - 84 - services such as retirement, unemployment, health care and housing, by establishing appropriate social insurance schemes and housing finance institutions. 4.10 Aggregate Demand Management. Efficient decentralization of decision-making and introduction of further reforms would be much facilitated by stable aggregate demand and prices. Of particular importance in this re- gard are further steps to restrain government spcading and to give the People's Bank full authority to limit the expansion of money and credit. Increased interest rates, particularly on enterprise deposits and capital construction loans, could help to dampen excess demand pressures and to channel resources to the investments with the highest yield. Stable macro- economic management is particularly important in managing the balance of payments and in limiting foreign borrowing to a viable level (Chapter IJI). B. Priorities and Sequencing of Trade Policy Reforms 4.11 Among the many policy changes discussed in Chapters II and III, not all are equally important and urgent. Also, some measures could be implement- ed relatively quickly, while others should be phased-in over a number of years. Foreign Exchange Allocation 4.12 One of the most important and urgent measures is the introduction of greater decentralization and market determination in foreign exchange alloca- tion. As discussed in Chapter III, several options are available to manage this transition. While China has adopted a managed, floating exchange rate system, it is not market determined and consists, .n effect, of a large number of different exchange rates, as Foreign Trade Corporations continue to cross- subsidize different exports by purchasing goods at domestic prices and selling at (often significantly different) international prices. This also applies to some extent to imports (see Annex 1). Even more important than the ineffici- encies implied in these cross-subsidies are efficiency losses resulting from the administrative allocation of foreign exchange. International experience suggests that decentralized, market-intermediated allocation is more efficient, especially as the complexity of an economy increases. 4.13 Two main options are available for moving towards a foreign exchange allocation system in which the exchange rate, rather than administrative import restrictions or export subsidies, becomes the principal instrument in maintaining a sustainable balance of payments. In the first option, admini- strative exchange rate adjustments would be made to compensate for the removal of administrative import restrictions or subsidies. Balance of payments developments, desired reserve levels and indicators of international competi- tiveness (see para. 3.6 and Box 3.1) would be used to estimate the required adjustments. The advantage of this option is that it would be essentially a continuation of the present exchange rate system, but introducing more flexibility and more frequent adjustments than at present. The disadvantage of this option is that it might take several months or quarters to ascertain whether exchange rate adjustments have been sufficient to compensate for the removal of trade restrictions, and this uncertainty might make it difficult to phase out administrative allocation of foreign exchange rapidly. - 85 - 4*14 The alternative approach would be to permit trading of foreign excbage or foreign exchange retention rights among enterprises and other units at freely negotiated prices. During a, preferably, short trhnsition period, this approach would result in a dual exchange rate system, but the approach would call for the gradual adjustment of the official rate towards the parallel rate so as to eliminate the differential over the course of the transition period. However, there are certain dangers in this option which would need to be guarded against: (a) If the market is too small, it could be very volatile, or have distorted results. Therefore, care should be taken to ensure a reasonable level'of supply to such a market. The best way to do this is to increase the foreign exchange retention ratio for enterprises, whilst simultaneously drawing up rules governing who would enter such a market to buy foreign exchange and for what purpose. (b) There is also a risk that differentials in exchange rates could emerge between provinces, because provincial authorities may attempt to prevent inter-provincial transfers of foreign exchange. Two measures could be taken in this regard. Firstly, care should be taken to ensure that any increased retention ratios are assigned to enterprises--including indirect exporters (as noted in para. 2.28 (vi))--and not to provincial authorities, and the People's Bankt through its provincial branches, should oversee and guard for any attempts by provincial authorities 'o prevent inter-provincial transfers of foreign exchange. Secondly, there is the organizational question. Clearly, in a country such as China, a large number of market locations would be necessary, but there are no established interbank mechanisms to ensure arbitrage between these locations. In this situation, it would make sense for the People's Bank itself, to operate such markets directly, possibly in the form of a variant on the auction mechanism, moving supply from one location to another to equilibrate marginal bids. (c) The third danger is that such a market may overshoot because of excess demand for foreign exchange by enterprises which are not yet fully responsible for profits and losses. To ease this problem, initial access to the market could be limited to enterprises and other units that are responsible for their own profits and losses. Such access could be linked to other areas of reform, such as trade reform itself, reduction of central planning and allocation, and price reform. (See paras. 4.32-4.37 for further suggestions in this regard). Access to the market could be expanded over time in line with increased retention and further progress in trade and price reform. 4.15 The introduction of a foreign exchange market would be compatible wnth a requirement that all foreign exchange be turned over to the People's Dank, the Bank of China, or other licensed foreign exchange depositories. In this case, a foreign exchange market becomes a market in foreign exchange rights, quite similar to the existing foreign exchange retention rights. - 86 - However, the development of such a market could be facilitated by the issuance of transferable financial instruments ensuring access of the holder to the retention rights. Such instruments would thus replace the present practice of retention accounts. 4.16 The State will, of course, remain a major user of foreign exchange for key projects. The creation of such a market for foreign exchange does not mean that the State would have lower access to foreign exchange, but that it may be obtained in different ways. Indeed, creation of a market in foreign exchange can be expected to increase the supply of foreign exchange available to the economy as a whole--and thus potentially to the State--both by reducing the hoarding of foreign exchange, and by stimulating exports. If the increased retention ratios reduce the foreign exchange available directly to the StatA below the desired level, the State can always enteri$he foreign exchange market directly to purchase additional foreign exchange to fund key projects. In addition, however, the State could save itself considerable levels of foreign exchange by changing the ways it funds the inputs for key projects. Instead of maintaining low prices for the command plan imports and subsidizing all such imports, it would be preferable to provide specific subsidies direct to the key projects, but adjust the prices of the strategic products, so that the subsidy is provided for the priority uses, but not for non-priority uses, as is the case at present. Decentralization of Trade Decisions 4.17 The decentralization of foreign trade decision-making and the simul- taneous introduction of more competition among foreign trade corporations would be complementary to changes in the foreign exchange allocation system and is of equal priority. Among the many measures discussed in Chapter II (para. 2.45), eliminating product-specific trading rights (monopolies) of FTCs and making FTCs fully responsible fo profits and losses (i.e., full applica- tion of the "agency system") are the most important measures. At the same time, consideration should be given to further removal of entry barriers into foreign trading. Chinese enterprises are currently losing many export oppor- tunities because of the time-consuming and sometimes cumbersome intermediation of FTCs that have trade monopolies. But further progress in this direction depends on progress made in reforming the foreign exchange allocation system. The introduction of a full-fledged "agency system" for imports and exports would, in any case, require compensatory foreign exchange rate adjustments. While the commodity pattern of exports and imports would remain broadly the same under an "agency system," some unprofitable exports would cease, and incentives would therefore be required to stimulate additional exports of other commodities through exchange rate adjustments and a re'orm of the foreign exchange retention system, as outlined above. 4.18 However, the timing and phasing of these reforms would need to be linked to other areas of reform, and could be done on the basis of sectors. As in the case of reforms in foreign exchange allocation and rate determina- tion, there are dangers in granting trading rights to enterprises that are not fully responsible for profits and losses, and it was noted earlier (para. 2.55) that trade liberalization in the case of products with heavily distorted prices could be harmful, both in terms of ensuring supplies of - 87 - essential imports, and in er'ouraging non-economic exports, which could have the secondary danger of having an adverse impact on world commodity markets. (See paras. 4.32-4.37 for further suggestions in this area.) 4.19 Introducing competition among FTC's and giving foreign trade rights directly to production enterprises does not mean that the government will lose control over foreign trade. But that control would be primarily affected through macro-economic levers: chiefly, the exchange rate, control over the expansion of domestic money and credit, and control over the size of the budget deficit. Trade planning and policymaking would primarily operate through such iadirect levers. Import Licensing and Tariff Reform 4.20 The key element in the reform of the import regime is the introduc- tion of an exchange rate system that eliminates the need for administrative restrictions to maintain the desired macro-balance of exports and imports. Once this were achieved, most quantitative restrictions would become redundant and could be phased out. Indeed, it would again be possible and desirable to link such re.orm to other reform areas. As the foreign exchange market is developed, the administrative restrictions on imports of certain commodities could be lifted in line with the shift of the financing of such commodities from the foreign exchange plan to the foreign exchange markets. For example, the retention ratio for garment exporters could be increased, and, simul- taneously, they could be required to obtain all imported imports via foreign exchange either retained or purchased on the market. This could permit restrictions to be lifted on inputs for this sector. However, there may remain valid industri¶j policy reasons to restrict certain imports to protect domestic enterprises - for a limited period of time, until they have become internationally competitive. Such protection should be provided mostly by tariffs rather than import licensing or other administrative import restric- tions. 4.21 The present tariff schedule, ranging from 0-200%, results in highly variable 'effective' protection of different industrial subsectors. Inter- national experience has shown that such highly variable protection can lead to substantial inefficiencies. Tariff rates should therefore be made more uni- form, perhaps with a range of 10-50%. On the other hand, widespread exemp- tions from import duties for machinery and capital goods imported for capital construction and technical transformation result in penalty on domestic producers of such goods, which could, in many cases, achieve efficient production in China. Tariff exemptions for imports for capital construction and technical renovation should thus be eliminated. Tariff exemptions or rebates would then be restricted to inputs into export production (see below). Tariff reform would probably be best phased over several years. In 3/ Ideally, production subsidies would be preferable to trade measures, but budgetary resources might not be sufficient. - 88 - the first stages, it would be desirable to pre-announce the tariff adjustments planned to occur over a period of several years, so that domestic enterprises can adjust accordingly. Export Incentives 4.22 The most important export incentive at present is the foreign ex- change retention system for enterprises, which compensates to a substantial extent for the anti-export bias of the present trade system. The retention system could be expanded by making access to foreign exchange more automatic, both for exports and for import-substituting enterprises, since a dollar saved through efficient import substitution is as good as a dollar earned through increased exports. However, as long as tariffs or other import restrictions are applied for reasons of industrial policy (e.g. infant industry protec- tion), the system will continue to have an anti-export bias. One way to re- duce this anti-export bias is to rebate to exporting enterprises the duties collected on inputs used in the production of export goods, or arrange for exemption for exporters. This is administratively challenging, but there is a large international experience which could help avoid some of the difficul- ties. In this way, any enterprise anywhere in China would reap the advantages that the Special Economic Zones now provide, and this would at the same timet provide the indirect advantages of greater export orientation throughout the economy. 4.23 A second step that could be taken to reduce anti-export bias is to rebate domestic taxes on export production. One internationally acceptable way of doing so would be to replace the present industrial and commercial tax with a value-added tax system. This would, of course, be a major undertaking and should await a more general reform of China's tax system. It may also be possible to link such rebate systems with a system for permitting access by indirect exporters to foreign exchange retention rights. 4.24 A third step, and one that could perhaps be taken earlier, would be to provide automatic access to domestic working capital financing to all exporters and, through a system of "domestic letters of credit", to all enterprises that provide inputs into production Lor exports. Such a credit facility would of course have to consistent with the overall credit plan of the People's Bank of China. Such a system would also make it possible to extend rebates of tariffs on imported goods used in export production to indirect exporters (suppliers to export enterprises). 4.25 These export incentives would all require advance planning and administrative sophistication and should be considered for implementation only in the medium term--once substantial progress has been made in the reform of the foreign exchange allocation system, the decentralization of foreign trade rights, the reduction of detailed administrative foreign trade planning and licensing, and a reduction and harmonization of tariff rates. An implementa- tion period of, say, five years would appear feasible. - 89 - Voreign Direct Investment 4.26 Reform of the foreign exchange allocation mechanism would also be the most important policy measure to stimulate foreign direct investment that would bring in the desired advanced technology and investment. Recently, measures have been announced that would make it easier for foreign joint ventures to gain access to foreign exchange by becoming de facto foreign traders on behalf of domestic enterprises. While this is progress in the right direction, access to foreign exchange could be made even more automatic through the reform of the foreign exchange allocation system described earlier. In this context, it is again isportant to remember that a dollar saved through import substitution is as good as a dollar earned through export promotion. Transport costs, slight quality differences, brand-name recogni- tion and other similar factors often make import substitution easier than export promotion--particularly in new products using more advanced techno- logy. Thus, there is a very useful role to be played by foreign joint ven- tures in producing for the domestic market, and in a number of sectors it may take ten years or more until production for exports will become international- ly competitive. 4.27 More important than preventing 'exploitation' by foreign investors (that might, in the worst case, involve a small fraction of production value in excessive royalties or overpricing of supplies and services) is to prevent inefficient foreign direct investment. Apart from domestic economic ineffi- ciencies, the potentially most worrisome source of inefficiency is excessive protection from import competition. Fortunately, China has a very large domestic market that will justify many enterprises (and joint ventures) in nearly any line of business, thus permitting vigorous competition among domestic enterprises or joint ventures. But in some cases this will not be enough, and high levels of import protection could seriously reduce the incentives for efficiency. C. Interrelations With Other Reform Measures 4.28 While the trade policy measures summarized in the preceding section are, as stated, closely related to other economic reforms outlined in Section A, progress can be made in implementing these trade reforms independently of progress in other economic areas. The most crucial precondition is the maintenance of domestic macroeconomic balance and stability. A number of developing countries have attempted to reform their trade systems in the face of very large domestic imbalances, chdracterized by large government budget deficits and permissive monetary and credit policies to finance these deficits, resulting in rapid inflation. Their tasks have often been rendered even more difficult by heavy external indebtedness that left little room to maneuver, as short-term reductions in exports or surges in imports could simply not be financed. But very often also, trade reform measures have failed because initial steps were too timid, particularly with respect to exchange rate system reforms. 4.29 China does not, at this moment, face extreme macroeconomic im- balances. However, the experience of the past two-and-a-half years shows that China's economic and political system is not immune to such dangers. There - 90 - are pressures for continued high government expenditures and easy bank credits at the very time when the role of the central government is being redefined and revenues decentralized to enterprises and local governments. When combined with a low political tolerance for inflation, this leads to calls for renewed or continued administrative controls over prices, imports and so on. Macroeconomic balance is therefore crucial during a transition towards a more efficient trade system based on decentralization of decision-making and competition. 4.30 In other respects, trade reforms could begin to be implemented relatively independently of the progress of domestic reforms. This is so, because in many respects the decentralization to enterprises of foreign trade decisions is currently trailing behind the decentralization of domestic production and marketing decisions. Similarly, competition among enterprises in domestic markets is already reasonably advanced in those commodities where China's export prospects in the near term are brightest: relatively labor- intensive, diversified manufactures, including not only traditional exports such as handicrafts, processed food, textiles and garments, but more impor- tantly new products such as consumer electronics, standard electrical and mechanical machinery, fabricated metal products and so on. The most important missing element to become internationally competitive in these products is better design and quality that can be acquired more efficiently under a more open system of foreign trade and international economic relations (para. 2.45 and Box 2.5). 4.31 This is not to say that trade reform can be carried out without regard to further progress on price reform. Further price reform is desirable and necessary both to increase the efficiency of resource allocation in the domest.c economy--particularly for capital-intensive basic raw materials (steel, chemicals, power, etc.)--and also to ensure an efficient set of signals to govern trading decisions. A wide-ranging trade reform without price reform would lead to irrational exports of some goods, such as steel products and energy intensive goods, and excessive imports of consumer goods. However, for many of the goods mentioned in the previous paragraph there has already been considerable progress in price reform, and it is our judgement that the proposed initial steps in trade reform do not have to wait for further progress in other areas. 4.32 These considerations suggest a way to proceed with the recommended trade policy reforms. Firstly, based on a examination of those sectors where price reform has been substantive, and where enterprises are generally respon- sible for their profits and losses, a list of candidate sectors/subsectors for early trade reform could be drawn up. For these sectors, there could be an immediate opening up of competition in trade, through the abolition of busi- ness scopes for FTCs and the effective monopoly powers of the relevant FTC for that sector/subsector. Any FTC or other trading organization would then have the right to trade in these products. Similarly, for these sectors/subsectors there could be a widespread granting of direct trading rights to enterprises in these sectors/subsectors, perhaps with the initial safeguard of only granting direct trading rights to enterprises that achieved a specified level of exports in the previous year, say US$1 million. At the same time, it would be necessary to ensure that the FTCs concerned become fully responsible for profits and losses. - 91 - 4.33 This same classification could then be used to make progress on foreign exchange allocation and import licensing restrictions. For those sectors/subsectors where trade is to be liberalized, the foreign exchange retention ratio for the enterprises could be raised, perhaps substantially. Simultaneously, there would be a cessation of any foreign exchange allocations from the plan for these sectors/subsectors, with foreign exchange-deficient enterprises having to purchase foreign exchange from the *oreign exchange market, which could be launched formally at the same time. Of course, such a market would not be limited to these enterprises, but would be open to any enterprise with surplus foreign exchange, or in need of foreign exchange. The higher retention ratios would provide the surplus foreign exchange to seed the market. 4.34 Again, a link could be made with import licensing. For these same sectors/subsectors, there couid be a simultaneous removal of most quantitative restrictions on trade, including not only competing imports, but also freer access by enterprises in these sectors to imporLed inputs for their own use. In some cases, such as certain consumer goods, it may be necessary to facilitate the adjustment process by raising customs tariffs for a period of time, and the regulatory tariff could be used for this purpose. In general, however, the foreign exchange market should be relied upon to control the level of liberalized imports. 4.35 It is beyond the scope of this study to identify those sectors or subsectors where advances could be made rapidly. However, it would seem that textiles, most light industry sectors, and some cEpital goods sectors would be candidates. The concept could also apply to above-plan production of some heavy industry sectors, perhaps including coal. 4.36 Such an approach would be in line with the Chinese practice of experimenting with reforms in certain areas prior to full implementation. While such experimentation has usually been limited by geographic area, the proposal in this case is simply to use the same app-oach, but with limits in a sector/subsector basis. To avoid the problems of regional differentials in exchange rates and inter-provincial conflicts on trade flows, it is strongly recommended that trade reform experiments are not conducted in geographically limited areas. 4.37 Once a period of initial trial has been conducted with trade reforms, and the unavoidable errors that will occur have been corrected, the reforms could then be gradually expanded. It is recognized that progress in other reforms will act as a constraint to the speed with which such an expansion could take place. For example, trade in the seven command plan commodities may have to continue on present lines until progress on price reform has made much more progress. However, the point to be stressed is the coordination of these policy reforms in the future. Institutional trade reforms, trade planning, import licensing, foreign exchange allocation, price reform, and enterprise responsibility should in future be seen to make progress simultaneously. The trade reforms suggested for immediate implemen- tation can thus be seen as 'catching up' with the progress already made in price reform and enterprise reform. Thereafter, they can move forward simultaneously. - 92 - 4.38 The creation of a more sophisticated 'neutral# environment for export enterprises (including duty and indirect tax rebating, automatic access to working capital finance, etc.) will of course require simultaneous movement in the areas of tax and financial system reform. The primary constraint is the administrative capacity required to prepare and implement these reforms, which will consequently take at least 3-5 years to prepare and implemert. 4.39 Similarly, in the area of foreign direct investment there are not too many constraints arising from the domestic reform process. The most important constraint is the reform of the foreign exchange allocation system, and this could be carried out relatively quickly. Otherwise, progress will mainly involve the elimination of differences in the legal and administrative environment between domestic enterprises and foreign joint ventures. This would, in a first phase, mostly involve giving domestic enterprises the same managerial autonomy and res;onsibility that joint venture enterprises already enjoy, except for the over-generous corporate income tax provisions that joint ventures now enjoy. However, enterprise reform should not stop there. Further steps are required, particularly to enhance factor mobility by creating markets for labor services, land use and capital, or improving the efficiency in these markets. 4.40 Thus, foreign trade and other foreign aconomic relations play an essential role in the dowestic economy. Foreign trade not only provides access to machinery, technology and goods in short supply, but also the multitude of contacts with buyers, suppliers and competitors that have proved to be an essential and low-cost source of technology transfer and informAtion about market opportunities. To maximize these indirect benefits of foreign trade, it is essential that enterprises have direct contacts with buyers and suppliers, and can make foreign trade decisions autonomously. The larger the number and types of enterprises that can engage in such contacts, the more rapid technology transfer and diffusion are likely to be. Foreign trade should thus be regarded not as a separate sector or activity, but as an integral part of enterprise decision-making on production and investment. However, if such decentralized decision-making is to lead to efficient resource allocation, prices will have to reflect economic costs, competition will have to be vigorous, and macro-economic stability has to prevail. Reforms will thus have to move simultaneously on these several fronts.