SWP759 Altemative International Economic Strategies and Their Relevance for China John B. Sheahan WORLD BANK STAFF WORKING PAPERS Number 759 A Background Study for China: Long-Term Development Issues and Options 1pY. Long-Term Development Issues and Options Tbis report is supplemented by six annex volumes publisbed by tbe World Bank: I China. Issues and Prospects in Education 2 China: Agriculture to the Year 2000 3 China: The Energy Sector 4 China: Economic Model and Projections 5 China. Economic Structure in International Perspective 6 China: The Transport Sector In addition, nine background papers, numbered bere as they are referred to in the text, bave been prepared in connection witb the report. They are available as World Bank staff working papers. I The Asian Experience in Rural Nonagricultural Development and Its Relevance for China (Staff Working Paper 757) 2 International Experience in Urbanization and Its Relevance for China (Staff Working Paper 7 5 8) 3 Alternative International Economic Strategies and T1-heir Relevance for China (Staff Working Paper 759) 4 International Experience in Budgetary Trends during Economic Development and Their Relevance f'or China (Staff Working Paper 760) 5 Productivity Growth and Technological Change in Chinese Industry (Staff Working Paper 76 1) 6 Issues in the Technological Development of China's Electronics Sector (Staff Working Paper 762) 7 The Environment for Technological Change in Centrally Planned Economies (Staff Working Paper 718) 8 Managing Technological Development. Lessons from the Newlv Industrializing Countries (Staff Working Paper 717) 9 Growth and Structural Change in Large Low-Income Countries (Staff Working Paper 763) WORLD BANK STAFF WORKING PAPERS Number 759 A Background Study 4947 for China: Long-Term Development Issues and Options "a /DY Alternative International Economic Strategies and Their Relevance for China John B. Sheahan The World Bank Washington, D.C., U.S.A. Copyright (C 1986 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing February 1986 This is a working document published informally by the World Bank. To present the results of research with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The publication is supplied at a token charge to defray part of the cost of manufacture and distribution. The World Bank does not accept responsibility for the views expressed herein, which are those of the authors and should not be attributed to the World Bank or to its affiliated organizations. The findings, interpretations, and conclusions are the results of research supported by the Bank; they do not necessarily represent official policy of the Bank. The designations employed, the presentation of material, and any maps used in this document are solely for the convenience of the reader and do not imply the expression of any opinion whatsoever on the part of the World Bank or its affiliates concerning the legal status of any country, territory, city, area, or of its authorities, or concerning the delimitation of its boundaries, or national affiliation. The most recent World Bank publications are described in the annual spring and fall lists; the continuing research program is described in the annual Abstracts of Current Studies. The latest edition of each is available free of charge from the Publications Sales Unit, Department T, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from the European Office of the Bank, 66 avenue d'lIna, 75116 Paris, France. John B. Sheahan is professor of economics at Williams College and a consultant to the World Bank. Library of Congress Cataloging-in-Publication Data Sheahan, John, 1923- Alternative international economic strategies and their relevance for China. (World Bank staff working papers ; no. 759) "A background study for China, long-term development issues and options." Bibliography: p. 1. China--Foreign economic relations. 2. Developing countries--Foreegn economic relations. 3. China-- Econimic policy--1976- . 4. Developing countries-- Economic policy. I. China, long-term development issues and options. II. Title. III. Series. HF1604.S54 1986 337.51 86-1540 ISBN 0-8213-0709-6 Abstract When a country moves away from tight restrictions on trade and investment toward a more open system, it can gain significantly from increased efficiency but may also encounter intensified problems in such domains as unemployment, setbacks to particular industries, increased inequality, issues of foreign influence, and even absolutely reduced income levels in poorer regions. Many countries changed from relatively closed to more open develop- ment strategies in the last 30 years; their experiences can be enlightening both for the problems they reveal and for the diversity of the methods used to deal with them. The issues do not come down to a choice. between free trade or a tightly closed system: there are an infinite variety of ways to handle relations with the external world, some much more productive than others. China has both more to gain and more to lose from a change toward an open system than most other countries. The scope for gain is especially great because of the pervasive character of prior departures from efficiency in relative prices and in the structure of production, because the country became so far out of touch with modern technology, and because it has an industrial base and the necessary skills to move rapidly to higher levels of productiv-' ity. But the scope for loss is also great because the society has so many distinctive characteristics worth a great deal of effort to preserve. Some of the main issues concern inequality within and between regions, highly differ- ential capacity among regions in their near-term capacity to respond to increased trade and investment, and possible dislocations of a unique system of social participation and protection. In the process of adapting to new structures of production and to external relative values in trade, it would be particularly important to distinguish carefully between: (a) unambiguously desirable changes away from arbitrary relationships which clearly impede efficiency and (b) changes which upset socially determined relative prices or location decisions oriented toward welfare objectives. This is always a treacherous matter for any country; the way it is handled by China will have a great deal to do with shaping the character of the society in the decades ahead. Table of Contents Page No. 1. CONTRASTING GROWTH PATTERNS AND INTERNATIONAL ECONOMIC STRATEGIES ........................................... 3 A Spectrum of Strategies .. 4 Exports and Growth . . . 7 From Import Substitution to Export Promotion . . 12 Varieties of Export Orientation: Type 1 Open Economies ....... 14 Varieties of Export Orientation: Type 2 Economies ............ 18 Export Orientation in East Europe: Hungary and Yugoslavia .... 26 European Industry and The European Economic Community . . 30 Two Economies Closed for the Sake of Autonomy and Equality: Burma and Sri Lanka . . 33 Effects of Trade on Regional Inequalities ..................... 37 2. ASPECTS OF THE CURRENT CHINESE SITUATION . .................. 42 Direction, Balance and Composition of External Trade . . 42 Internal Trade and the Location of Industry . . 45 Balance of Payments, Capital Flows, Foreign Investment ........ 52 Export Performance, Exchange Rates, and Prices .............. . 55 3. CONFLICTS AND OPTIONS .............................. 59 External Trade and Finance: Incentives and Controls ......... 60 Foreign Investment . ........................................... 65 Low-Income Regions ....... ............... ..................... 68 Relative Prices: Efficiency and Social Goals . . 72 BIBLIOGRAPHY ........... 79 (i) International economic policies involve an inescapable tension between the wish to guide and protect the character of internal development and the wish to take advantage of the gains possible through contact with the external world. China has more achievements of its own to safeguard than many other countries do: a greater degree of equality, more diversified industry and more widespread local initiative than most low-income countries, a pro- foundly rooted national identity, and an exceptional degree of social cohe- sion. But at the same time, living standards remain low, population pressures are intense, some regions continue in serious poverty, the capacity to produce enough food is recurrently in question, and both the efficiency of the produc- tive system and the quality of technology in use are much below levels which should be possible. Efficiency, income, and the level of technology could be raised through greater internal and international trade and through acquisi- tion of technical knowledge from other countries, but opening up flows of trade and investment has to be done carefully; it could do harm as well as good. (ii) So many changes are going on in the Chinese economic system that it escapes classification in any familiar category. It is definitely socialist with strong institutions of planning and control, but it also allows decen- tralized initiative and an increasing role for market forces. It is in the process of becoming a new kind of economic system whose essential features are not yet fully defined. Both international and internal trade are becoming more open but still operate within a great many constraints. This process of constrained opening raises issues similar to those faced by many other - 2 - countries, both capitalist and socialist, in moving from protected to more open systems during the postwar period. The results of such movement have sometimes been highly positive and sometimes unexpectedly costly. There are countless ways to be "open." China will find new ways of its own, but it may find better ways more quickly through study of the experiences of others. (iii) Chapter 1, following, reviews a wide range of policies toward inter- national trade and finance used by different countries in the period since 1950. The issues have been selected to direct attention to aspects which might be of particular interest for choices of policy by China, but in this chapter the concern is primarily with analysis of results in the rest of the world. Chapter 2 then considers some of the main characteristics of China's situation, partly in the form of questions about aspects which are uncer- tain. Chapter 3 summarizes the discussion in terms of the likely gains and costs of some of the major options of national economic strategy. - 3 - 1. CONTRASTING GROWTH PATTERNS AND INTERNATIONAL ECONOMIC STRATEGIES 1.01 The world economy as a whole has probably never before had such a sustained period of rapid growth as in the three decades from the end of the Second World War, or such marked changes in relative incomes among countries. Some countries failed almost completely to participate in this prosperity, others achieved striking individual successes, and many others just drifted along with the general tide. These differences in performance became further accentuated when the growth of the world economy slowed from 1974 and turned downward from 1980 to 1983. In those last three years, per capita income fell sharply in some countries but kept growing in ot-hers. Such differences in capacity to control events and keep growing cannot be explained adequately by any single set of factors. They involve such fundamental questions as the cohesion and character of the nation itself, the nature of the economic system, and the quality of national leadership. But they also involve factors which are, more narrowly, matters of the options taken with respect to parti- cular economic policies and, of all such options, those concerned with exter- nal economic relationships can be among the most decisive.-1 1/ Among the many empirical studies of international economic policies in relationship to problems of development, three large-scale group projects comparing individual country trade and investment policies within a common analytical framework are especially useful: (1) those of the OECD carried out in the late 1960s and summarized by Ian Little, Tibor Scitovsky and Maurice Scott (1970); (2) those of the National Bureau of Economic Research made in the mid-1970s summarized in Bhagwati (1978) and Krueger (1978); and (3) further studies sponsored by the NBER in the late 1970s, brought together in Krueger (1982). Apart from these group projects, perhaps the most comprehensive analyses of trade policies and growth have been those of Bela Balassa (1981 and 1982). - 4 - 1.02 Perhaps the strongest single influence differentiating country growth rates in the postwar period has been the degree of success in raising industrial exports. Those countries which emphasized export promotion relativel-y' early did exceptionally well. Those which paid less attention to exports found it much harder to sustain high rates of growth and often found that this neglect aggravated many other strains of development. The negative consequences were shared both by capitalist countries trying to grow through protected import substitution and by most socialist countries concentrated on internal markets. Both capitalist and socialist countries then turned, in many cases, toward greater participation in the world economy to escape the constraints of their prior orientation.. As they did, so the results proved to be mixed: some achieved significant improvements but many encountered new tangles of contradictory methods and objectives. These more complex problems of policy reversal are not simply matters of export success but also involve troublesome consequences in the domains of poverty, equality, regional imbalance, and the general character of the societies concerned. A Spectrum of Strategies 1.03 As a framework for consideration of the advantages and disadvantages of alternative strategies, Table 1 summarizes five broad types of policy orientation, with examples of countries associated with particular orienta- tions in specified periods. These categories are meant to serve as points of departure for analysis; they are not narrowly defined strategies with common internal details. Each of the five includes a multitude of detailed differ- ences at the level of individual policies, and these specific differences can give widely varying consequences for countries following the same general orientation. -5 - Table 1. CONTRASTING TYPES OF INTERNATIONAL ECONOMIC STRATEGIES AND EXAMPLES OF ECONOMIES USING THEM General strategy Examples of use 1. Highly open economies with little protection against imports and few restraints on foreign investment (a) with little government intervention to Hong Kong limit poverty or guide investment Chile after 1973 (b) with considerable domestic intervention Singapore from 1967 for social objectives 2. Export promotion, particularly for indus- Japan trial products, but considerable use of Republic of Korea from protection and of government control of the 1962 economy Brazil from 1964 Yugoslavia from 1964 Hungary from 1968 3. Regional groupings with free trade for European Economic industrial products inside the region but Community with protection of agriculture in each country and moderate protection from outside manufactures 4. Highly protected economies without any Brazil up to 1964 strong orientation toward exports, but with Colombia up to 1967 attempts to promote industrialization and India with some foreign investment 5. Relatively closed economies with tight Burma from 1962 restrictions on trade and investment Sri Lanka to 1976 China to 1977 The Soviet Union Note: The dates specified for changes in orientation are intended to mark initial years of redirection which sometimes involved immediate major changes (as in Brazil in 1964) and sometimes, instead, a slow process of change by erratic steps (as in Hungary after 1968). - 6 - 1.04 Perhaps the first point to note about these orientations is that three of them are associated with active participation in international trade but in different ways, with different connotations for the character of ec9no- mic growth. The range of possibilities is not limited to a two-way choice between "open" and "closed" systems; it is rather a broad spectrum allowing for many different mixtures. Some individual countries have stayed with a particular orientation for long periods, but at least three common paths of movement have been evident in the last 30 years. 1.05 One of the common patterns of movement is that the moderately indus- trialized developing countries started the postwar period with a strong orien- tation toward strategies of protection for import substitution, or type 4 in the definitions of Table 1, and then moved toward the more export-oriented categories of type 1 or 2. Some socialist countries of Eastern Europe similarly moved from closed systems towards more active concern for export promotion and efforts to bring in new technologies. At the same time, some of the least-developed countries, which started their postindependence periods with more open economies close to type 1, have moved toward more protection and control, as in type 4. The industrialized countries of Western Europe moved collectively from moderately protected national systems to the Common Market, with much less protection against each other, though with continued protection from the outside world. As these movements suggest, the particular mixture of advantages and disadvantages from each type of orientation change with the level of development of the country, and change also in response to the evolution of the world economy as a whole. - 7 - Exports and Growth 1.06 Differences in rates of growth of per capita income and of exports are shown in Table 2 for 17 countries and also for the averages of middle- and low-income developing countries for the period 1960-81. The countries included in this table were selected partly because they stand out for con- trasts in performance and partly because the more systematic reasons for such differences may provide useful suggestions for longer-term economic strategy in China. 1.07 Table 2 brings out an important association between trade policy and economic growth: high growth rates of per capita income are closely related to high growth rates of exports. Statistical tests of association are not valid for this group because the countries were selected partly to demonstrate that relationship, but it may be seen that the association holds both within different income levels and across the whole table.21 Among the four highest income countries, the one with the highest export growth, Japan, raised per capita income three times as fast as the United States and twice as fast as the Federal Republic of Germany. Japan, France and Germany achieved higher growth rates of exports than the United States and all had faster income growth. The only examples in the rest of the table of export growth compar- able to or exceeding Japan for the whole period are Hong Kong and the Republic of Korea. They are also the only two which raised per capita income faster 2/ For statistical tests of association between exports and growth, and debates about their validity, see Michaely (1977 and 1979), Balassa (1978), and Heller and Porter (1978). -8 - Table 2. INCOME LEVELS AND RATES OF GROWTH OF GNP PER CAPITA AND OF EXPORTS FOR SELECTED COUNTRIES, 1960-81 GNP per capita Growth rate Growth rate of 1981, measured of GNP per exports in US dollars capita, 1960-81 1960-70 1970-81 (% p.a.) (% p.a.) Germany 13,450 3.2 10.1 5.8 United States 12,820 2.3 6.0 6.5 France 12,190 3.8 8.2 6.6 Japan 10,180 6.3 17.2 9.0 Singapore 5,240 7.4 4.2 12.0/a Hong Kong 5,100 6.9 12.7 9.7 Uruguay 2,820 1.6 2.2 4.3 Yugoslavia 2,790 5.0 7.7 4.5 Chile 2,560 0.7 0.7 9.8 Brazil 2,220 5.1 5.0 8.7 Hungary 2,100 5.0 9.7 8.2 Korea 1,700 6.9 33.4 22.0 Thailand 770 4.6 5.2 11.8 China 300 5.0 /b Sri Lanka 300 2.5. 4.6 (-1.5) India 260 1.4 3.2 4.6 Burma 190 1.4 (-11.6) 1.3 Averages Middle-income oil importers 1,670 3.7 7.0 4.3 Low-income countries other than China and India 240 0.8 5.0 (-0.8) /a Refers to 1970-80, rather than 1970-81. 7i China's exports increased 23 percent from 1970 to 1980, but they include (unlike the other countries in this table), increasing earnings from oil exports. Excluding oil, exports increased at a rate of 20 percent in terms of value or 6 to 7 percent in volume (World Bank 1983a, Vol. II, pp. 413-416). Sources: All except estimates of exports for China from World Bank, World Development Report 1983, Annex, Tables 1 and 9. than Japan.3/ At the other extreme, Burma's closed-in strategy, intended to limit foreign influence, resulted in falling exports through the 1960s and in exceptionally low income growth. When averages for all middle-income oil importers and all low-income countries are compared, it is apparent that the middle-income group raised exports more rapidly than the low-income group; correspondingly, this group achieved distinctly faster income growth. 1.08 Exports of primary products may rise or fall because of changes in world demand without demonstrating any necessary link with national economic policies. The capacity of individual countries to increase both their own growth rates and their exports by their choices of economic policy is best demonstrated by ability to increase exports of industrial products (Keesing 1979, Riedel 1983). The most notable export performances in Table 2 are based solidly on success with competitive industrial exports. This is clearly true for Japan and Table 3 shows that it is also very much the case for Hong Kong, Korea, and Singapore. All of them greatly raised the share of manufactured goods in their total exports between 1960 and 1980, to 90 percent of the total for Korea and 93 percent for Hong Kong. 1.09 While these associations do not prove anything about cause and effect, they strongly suggest that a positive export performance is normally related to a positive growth performance and, further, that individual countries can achieve very different export results under the same conditions of world demand. The question about cause and effect is complex. All else 3/ Several of the oil-exporting countries left out of Table 2 had still faster rates of growth of exports and income, but in these cases--in contrast to Japan and Korea--the growth was due to good fortune with a natural resource plus agreements to control export prices rather than the creation of new productive capacity and heightened competitive strength. - 10 - Table 3. SHARES OF MANUFACTURED GOODS AND OF MACHINERY AND TRANSPORTATION EQUIPMENT IN TOTAL MERCHANDISE EXPORTS, 1960 AND 1980 Countries listed in same sequence as in Table 2, but Percentage shares in total exports excluding the four most Machinery and industrialized All manufactures transport equipment 1960 1980 1960 1980 Singapore 26 54 7 26 Hong Kong 80 93 4 19 Uruguay 29 38 0 4 Yugoslavia 37 73 15 28 Chile 4 20 0 1 Brazil 3 29 - 17 Hungary 66 62 38 32/a Korea 14 90 - 20 Thailand 2 29 0 6 China n.a. 47 n.a. 5 Sri Lanka 0 19 0 1 India 45 59 1 7/a Burma 1 n.a. 0 n.a. Average for all middle- income oil importers 17 54 2 14 /a Data for 1979 rather than 1980. Source: World Bank, World Development Report 1983, Annex, Table 10, pp. 166-167. being equal, a country which adopts a coherent policy of export promotion can expect a positive result for economic growth. But a country which is unable to formulate and implement coherent domestic economic policies is most unlikely to be able to achieve any remarkable success at export promotion and, on the contrary, a country which has firm lines of policy formulation and implementation is likely to find it easy to increase exports because it has - 11 - that underlying capacity. The fact that Japan and Korea have done extra- ordinarily well at exporting is not the ultimate explanation of their growth performance: the export success is a result of combining strong export incen- tives with everything else which has permitted them to achieve so much so rapidly. 1.10 The countries which stand out for growth in Table 2 are associated with two quite different types from Table 1: Hong Kong and Singapore are type 1 open economies, but Japan and Korea are clearly more of type 2, with a great deal of direct control and considerable protection. Some of the countries of types 1 and 2 have had relatively modest success with exports and growth, partly because of shorter periods of export orientation but also because of problems in the ways by which changes to more open systems were attempted. 1.11 An open economy does not guarantee growth, nor a closed economy prevent it. Both China and the Soviet Union achieved good rates of economic growth in the period from 1950 to 1980. Growth with closed economies was less difficult for them than it would be for most countries because of their immense size and diversified resources. But even then, the costs have been high in terms of severe restraint on consumption and housing, and in terms of technological backwardness. The examples of Burma and Sri Lanka also cited in type 5 are different in that they placed unusual emphasis on goals other than economic growth, specifically equality and avoidance of foreign influence. The results in Burma proved to be singularly poor in terms of reducing poverty, but Sri Lanka achieved strikingly high levels of social welfare. They are discussed separately below. - 12 - From Import Substitution to Export Promotion 1.12 The first of the common patterns of movement cited above, from protected industrialization toward an export orientation, was in most cases a response to increasingly serious obstacles raised by this kind of development strategy. The countries which were at one time strongly oriented toward import substitution--including Argentina, Brazil, India, Mexico, and Pakistan-- experienced an initial spurt of fairly rapid industrialization which was soon followed by deepening problems of inefficiency, external deficits and debt crises, inflation, and internal social conflict. Their problems have been convincingly detailed by many empirical studies, notably that of Little, Scitovsky and Scott (1970). A common thread in all these cases was that the reliance on protection to stimulate investment in the replacement of imports created a sheltered environment for the domestic firm, in which it could raise prices to cover high costs and was under no great pressure to innovate, improve quality standards, or reduce costs of production. The industrial sector as a whole required increasing imports of machinery and current inputs for production but was unable to achieve the exports to pay for its needs or the dynamism to keep up with external technological advance. 1.13 Protected industrialization fostered systematic disconnection between internal and external prices. Internal prices of manufactured con- sumer goods were raised above external prices, while those of capital equip- ment and primary products were held down. That pattern stimulated investment in production of consumer goods for middle- and upper-income groups, discour- aged development of capital goods industries, and taxed the rural sector to support industrialization. Favored access to imports of capital equipment and relatively open conditions for foreign investment brought in modern technology - 13 - for use in production of consumer goods, but rarely for learning how to pro- duce capital equipment. The imported technologies were often inappropriate for domestic conditions, particularly when designed in highly industrialized countries to save on the use of labor. This kind of industrialization turned the internal terms of trade against primary producers, aggravated rural poverty, and drew workers seeking jobs into the cities, but did not provide openings for productive employment at anywhere near the rate needed. 1.14 Concentration of industry on replacement of consumer goods imports might have had less unfavorable consequences if domestic production had aimed more at provision of basic products within the range of purchasing power of the majority of the population, with relatively stablized technology allowing domestic firms to handle production effectively. In practice, industrial investment was pulled toward the latest kinds of consumer goods typical of demand patterns in the most industrialized countries, emphasizing wide variety and rapid change of product characteristics (Felix 1968). That kind of modern "consumerism" acted to preclude economies of scale in production, to favor use of foreign machinery and equipment as well as direct production by foreign firms, and to increase requirements for scarce skills rather than unskilled labor (Stewart 1972). 1.15 Ragnar Nurkse long ago pointed out this problem of inappropriate attempts to duplicate consumption patterns of high-income countries at very early stages of development as a "demonstration effect" likely to be common in all developing countries (Nurkse 1955). But it may differ in significance in different countries and cultures. It is mainly a matter of leadership in con- sumption patterns by upper- and middle-income urban groups. Producers orient resources to the preferences of such groups to the degree that they dominate - 14 - the market. In Latin American countries, in particular, with extremely unequal distributions of income, the preferences of the upper 40 percent of consumers may dominate the national market for manufactured consumer goods. In addition, leading social groups identify themselves with European and North American life styles in terms of cultural traditions, education, and flows of information. The resulting pattern of market preferences could become a serious handicap for growth and for equity in any country, but it has been less marked in Japan (Felix 1983) and probably should remain so in East Asia generally, both because of Asia's greater equality of income within countries and because of its strong attachments to its own culture. 1.16 The distortions involved in the process of import substitution made it imperative to find a better-balanced kind of industrialization, but also made it difficult to change without severe disruption. Some of the changes that did take place were made through violent political overturns and severe repression; in Pakistan, the tensions helped incite a war and secession of the more rural region. Colombia and Mexico provide more peaceful examples, where governments came to recognize the nonfunctional character of the system and changed it more gradually, with a good deal less social trauma. Varieties of Export Orientation: Type 1 Open Economies 1.17 Economic analysis has a tendency to seek pure forms or universal principles which override differences in historical context. Arguments in favor of free trade in open economic systems have a strong logical basis and a good deal of empirical support: arbitrary restrictions on imports or obstacles to exports can be very costly. But few countries have achieved sustained high rates of growth with free trade, and these successful cases are - 15 - almost invariably countries with dynamic industrial sectors able to compete successfully in world markets for manufactured goods. The universal prescrip- tion for trade freedom does not work as well for countries with few competi- tive industries, which depend on primary products for their exports and rely on imports for most types of industrial products. 1.18 Relatively free trade with primary product exports can achieve good rates of growth under favorable conditions, but such trade does little to foster diversification of production, acquisition of technical knowledge, or the capacity to maintain growth during periods of contraction in world demand. A country with no industrial base beyond traditional clothing and housing activities is very unlikely to achieve much capacity for growth through reli- ance on an open economy: some combination of protection and direct promotion of new activities is almost surely necessary to begin to move. But once a country has been able to develop some industrial base and some capacity of either private entrepreneurs or public agencies to launch new industrial activities, it becomes both possible and highly desirable to link the indus- trial sector with the outside world. To delay too long such a turn toward external contact can condemn the country to slow growth and prolonged poverty; to make the turn without great care can lead to a mixture of stimulus and setbacks. 1.19 The discussion in this section is concerned with countries which have a considerable industrial base and clearly can, or should, gain from active international trade. Two of the examples cited as type 1 open econo- mies, Singapore and Hong Kong, demonstrate that industrialization and fast growth of income are indeed possible with such an orientation. Their open economy orientations have also been consistent with fairly high degrees of - 16 - equality and low rates of unemployment. None of these indicators of success have been achieved by the third example cited, Chile, or in the similar (though more protected) experience with reorientation toward market incentives and more open trade in Uruguay. In Chile, a study of consumption levels in 1978, comparing standards for different income groups to their levels ten years earlier, showed that living standards of the lowest fifth of the income distribution had fallen by 31 percent, while rising for the highest fifth of the distribution (CortAzar 1980, p. 6). Both Chile and Uruguay experienced great increases in unemployment after rejection of their prior closed-in types of economic systems. In Chile, unemployment went from 3.9 percent in 1970 to 16.6 percent by 1975, two years after the country's dramatic switch to a more market-oriented system. The rate of unemployment began to come down as the economy revived in a more trade-oriented system, but it was still far above the level of 1970 (at 11.4 percent) in the last relatively prosperous year of 1980 (PREALC 1982, p. 99; Mufioz 1982). In 1982, in much more adverse economic conditions for the world and for Chile, the rate of open unemployment reached 25 percent (Ffrench-Davis 1982). In Chile, massive popular demonstrations against the government have made clear that many people desperately want to change their economic and political system. Why have these two countries had such poor results from their orientations toward open trade and investment compared with the Asian countries? 1.20 One of the main differences is structural. The Asian countries cited'do not have strong raw material exports and a comparative advantage in primary production. Their comparative advantage has to be on the side of industrial exports. To export is to develop industry. A second crucial factor is that they have a strong base of human capital, including both wide- - 1 7 - spread labor skills and entrepreneurial capacity. For them, opening up and promoting trade in new products helps maintain a high rate of learning. The industrial exports of Hong Kong do not involve high participation of foreign firms, though those of Singapore do: transnational firms account for about 10 percent of manufactured goods exports from the former, but 82 percent of those from the latter (Nolan 1983, ch. 4). Both foreign and domestic firms in Singapore used employment-generating production processes in the early stages of industrial export promotion when the country had high unemployment, and then moved toward more labor-saving high technology as full employment condi- tions were reached (Tan and Ow 1982). Both Singapore and Hong Kong would have much lower incomes than they do today if they had opted for continued protec- tion rather than exporting as a basis for their industrialization. 1.21 Chile and Uruguay had a considerable level of industry when they turned toward more open economies but it was, to a high degree, dependent either on protection or on subsidies. The removal of protection was expected to drive some of the higher-cost firms out of production, and it did. Changing emphasis toward export promotion should have stimulated a more appro- priate set of industries and to some extent it did, but at the same time both countries repressed domestic real wage rates and domestic markets so that producers were under negative pressures from both external competition and constrained domestic markets (Foxley 1982). These negative pressures might have been offset by rapid development of new exports, and both countries seemed to be succeeding in that path in the last years of the 1970s (Balassa 1981b,.Keesing 1981). But then, in 1979, Chile took the fatal step of freezing an exchange rate which quickly became overvalued, and by 1980 its new exports were beginning to turn down. Then both countries were hit hard in - 18 - 1981-83 by falling world demand and increased protection against their exports, aggravated in Chile's case by forced contraction because of over- whelming external debt problems. 1.22 The main reason the degrees of poverty and inequality increased in Chile, even before the general economic downturn, was that the removal of protection, combined with tight restraints on domestic liquidity and low rates of investment, led to the great increases in unemployment noted above. Increased unemployment due to import competition is clearly a possible danger in a system newly exposed to such competition, but it is not.a necessary result. It has not been a problem in the Asian countries open to import competition. The special difficulties of Chile were related partly to the removal of protection but, more fundamentally, to a combination of high real interest rates and failure to keep the price of foreign exchange high enough to permit effective competition against foreign firms. The Chilean experience certainly does not prove that an open economy must, by its very nature, make poverty worse and set back industrialization: it only shows that it is possible to get such results if the details of economic policy are not struc- tured to avoid them. Varieties of Export Orientation: Type 2 Economies 1.23 The countries cited as examples of type 2 export promotion combined with protection and active management of the economy include both capitalist and socialist economies. They are certainly not open economies in the sense of Chile or Hong Kong. They include outstanding cases of successful growth and others which have had more difficulties. Some interpretations of their performances stress that continuing protection and direct intervention act as - 19 - restraints on efficiency and growth. The alternative interpretation suggested here is two-sided: some of the protectionism probably has held down effi- ciency and the growth of exports but, at the same time, both active economic management and particular rules limiting free trade and investment have con- tributed positively to national goals. These different interpretations are examined by considering three country experiences in particular--Japan, Korea, and Brazil--and by reference to a few selected aspects of experience of the other type 2 economies cited in Table 1. 1.24 Japan developed its powerful industrial base under relatively closed conditions, with foreign investment completely banned for many years. Even now, while it is much more open to import competition and partially open to foreign investment, a close network of relationships between the government and leading firms provides a high degree of organized control of trade and investment patterns. Japan has become the outstanding world export competitor for many reasons, including the basic factors of widely diffused human capital, hard work, entrepreneurial capacity, and avoidance of excess demand. But along with these fundamental factors, it has used a powerful system of thought-out investment in directions of advancing technological complexity in order to move from a phase of merely copying the known to one of moving forward faster than any other nation (Johnson 1982, Shinohara et al. 1983, Tyson and Zysman 1983). 1.25 Japanese capacity for rapid technological progress did not come from any significant participation by foreign investors. It came from a determined drive to build up national capacity through active support for research, heavy expenditure for acquisition of foreign technological information to be applied by the Japanese themselves, and direct study of foreign markets and production - 20 - processes. As Raymond Vernon emphasizes (Vernon 1977), it was not the exclu- sion of foreign investors which assured success: it was the whole-hearted commitment of the Japanese government and industrial leaders, both to rapid technological change and to exports, which made these achievements possible. 1.26 Protection of industries from import competition can serve either to maintain inefficient lines of production which inflict unnecessary costs on the economy as a whole, or as a means of helping new industries to get estab- lished on an efficient basis. Protection, as practiced in the United States in recent years, has been mainly of the former kind, and almost surely has adversely affected industry's ability to compete in the outside world. Protection in Japan has included both kinds but has focused mainly on creating new industries in dynamic fields (Shinohara et al. 1983, pp. 5-38). The organization most directly involved, the Ministry of International Trade and Industry (MITI), tries to select new activities for support according to two criteria: (1) products for which the income elasticity of demand is high, and (2) fields in which technological change is particularly rapid. MITI works directly with individual firms to provide help for financing and research, while a separate national planning agency formulates indicative macroeconomic plans to clarify likely market conditions. 1.27 The exceptionally rapid growth of Korea can, in part, be attributed to its early move away from excessive protection toward concentration on incentives for saving, investment, efficiency, and exports (Frank et al. 1975). These changes toward more effective use of market incentives have been a significant part of the picture, but at least two other factors have been of very great importance. One is that the government has made an enormous commitment of time and energy to somewhat the same style of active promotion - 21 - used by Japan. The government uses a considerable degree of direct control over allocation of credit and also a great deal of direct negotiation with individual firms to push them in desired directions (Jones and Sakong 1980, Shinohara et al. 1983). The system differs from that in Japan in that it involves rather more direct government control and less concern for consen- sus. The relative absence of open public discussion and freedom of choice for individual firms has involved a number of misdirections (Kim in Shinohara et al., pp. 76-77). 1.28 The third element in Korea's performance has been a highly fruitful learning process through concentration on industrial exports. Koreans them- selves initiate most of their export contacts, but then learn rapidly how to improve product design and maintain quality standards from the requirements set by buyers and from their advice based on plant inspection in Korea (Westphal et al. 1981). Improvements in productivity and in quality control have been, for the most part, carried through by Korean-owned firms, using a great deal of licensing and study of foreign equipment with limited direct investment by foreign-owned companies. Korea permits direct foreign invest- ment more readily than Japan but much less than in Brazil, Mexico, or Singapore. By the mid-1970s, more than half of all foreign firms were under requirements to maintain export targets, and a high fraction of them were limited entirely to production for export. Even under these conditions, majority-owned foreign firms accounted, in 1980, for less than one fourth of the country's exports of manufactured goods (Nolan 1983). Direct foreign investment has been the main source of technological improvements and exports only in one field, electronics, where changes in basic technology have, so far, been too rapid for Korean firms to keep up with, on their own. - 22 - 1.29 Westphal et al. (1981) give great credit, convincingly, to Korea's export orientation as opposed to production concentrated more on protected domestic markets. This orientation has encouraged rapid learning and achieve- ment of high quality standards, feeding back to the capacity to raise produc- tivity rapidly. They add one crucial consideration which may be distinctively Korean (or East Asian) and may account for much of this striking process of learning-by-doing. It was not simply a matter of choosing an export orienta- tion and providing the right incentives. The additional factor has been a rare capacity for learning: "the assimilation of technicaL know-how has been great" (p. 65). That is partly a consequence of widely diffused educational opportunities (an ingredient missing in much of Africa and Latin America), partly a result of freedom for personal movement among firms by people able to carry technical skills to new companies (a very restricted possibility in China), and partly a matter of the mysterious dynamism which used to be iden- tified with Western Europe and North America, but now seems to be an espe- cially marked feature of 'East Asia's reawakened old societies. 1.30 None of the semi-industrialized countries of Latin America have achieved anything like the spectacular growth of Japan or Korea, but Brazil did exceptionally well after making two major changes of economic orientation in the 1960s. Up to 1964, Brazil had a highly protected economy with few industrial exports: in 1960, they were only 3 percent of total exports (Table 2). In 1964, Brazil introduced fundamental policy changes aimed at increasing efficiency, including some reductions of import restrictions and a technique of making frequent small increases in the price of foreign exchange in order to encourage exports. The government did not, by any means, abandon protection and dismantle economic controls in the style of Chile or Uruguay, - 23 - but instead developed a complex system of direct promotion and control of industry, accompanied by active concern for efficiency and for exports (Mendonca de Barros, Jose, and Graham 1978). In many respects, this system was similar to that of Korea, though with somewhat less aggressive export promotion and much more reliance on foreign investment. 1.31 The initial effect of the new policies introduced in 1964 was to slow down economic growth and raise unemployment. Aggregate demand was kept under severe restraint in an effort to stop inflation and real wages were sharply reduced. But after three years of economic repression, Brazil turned more toward active expansionary policies and began to achieve exceptionally rapid growth. The new style of growth after 1967, with somewhat less protec- tion and more industrial exports, proved to be much more favorable than pre- vious policies for employment and for the reduction of poverty (Carvalho and Haddad 1982, Knight 1981). But it involved two major problems: a high degree of foreign ownership of productive facilities in the manufacturing sector, with consequent limitation on Brazilian capacity for learning and control (Gereffi and Evans 1981) and then, beginning in the mid-1970s, an excessive rate of growth of demand and imports. Import volume was allowed to rise even faster than the high growth rate of exports, in the same period in which prices of oil imports were greatly increased. The external debt went up fast as a result of the trade imbalance, and then even faster as Brazil accelerated its borrowing at rising interest rates in order to permit continuing current deficits. The external debt came to dominate policy options at the beginning of the 1980s as lenders drew back from further extensions of credit and Brazil was forced to accept agreements to restrain domestic demand. Growth practically stopped from 1980 through 1983 and inflation became much worse. - 24 - This costly impasse was not a necessary consequence of the orientation toward a more open economy; it was the result of failure to keep the growth of demand in line with the growth of productive capacity. 1.32 One of the crucial components of the better period of Brazilian eco- nomic performance, from 1967 to the end of the 1970s, was the technique of a "moving peg" exchange rate, keeping the price of foreign exchange rising at about the same rate as domestic prices. In the absence of such a technique, any excess of domestic over external inflation acts to make domestic markets more profitable than exports and can thereby undercut economic growth by driving industrial exports down. This negative effect proved particularly destructive in Chile when the exchange rate was frozen in the period 1979-81, and in Mexico at the same time. 1.33 Brazil's moving peg, and the similar policy adopted in Colombia from 1967, helped greatly to stimulate and diversify exports in both countries. But the technique relies on appropriate internal policies too. The Frei government in Chile introduced it there in 1964, but it failed to work nearly as well. In this case, the problem was chiefly that wages were raised much more rapidly than labor productivity, with the result that inflation began to speed up and the government then turned to generalized deflation, slowing down the whole economy (Ffrench-Davis 1973, pp. 51-64). More recently, Peru and several other countries have also found the technique difficult to handle. When inflation moves up rapidly and the pace of devaluation is accelerated correspondingly, the increasing price of foreign exchange feeds back to the process of inflation itself. A technique which is highly promising in the presence of stable rates of domestic inflation turns into an added problem when inflation rises seriously. - 25 - 1.34 Both Brazil and Colombia have been criticized for clinging to con- siderable protection and for applying restraints on foreign investment rather than going to the more open-economy model of Chile or the more forcefully export-oriented approach of Uruguay. Their remaining protection and interven- tion alters incentives from those operating in world markets and keeps costs of production for exporters higher than they are in countries where the producer has direct access to the world market for all inputs (Balassa 1981b, Ranis 1981). Brazil's restraints on foreign investment, previously mild, may now be acting to slow down transfers of technology; they have recently been tightened to exclude foreign firms entirely from certain fields, including small computers and health care equipment. Colombia does not use such active techniques of management and control of investment, but it participates in the Andean Pact which has tried to enforce restrictions on entry and location by foreign firms, to supervise their imports of supplies and their pricing, and to require eventual sale of the firms to domestic owners. 1.35 The type 2 strategies of Brazil and Colombia probably do hold down the rate of growth of their industrial exports, and their restraints on foreign investment surely do reduce its inflow in particular fields. But these are questions of balance among competing preferences rather than simple mistakes. The restraints on investment are intended to foster more domestic entrepreneurial capacity and greater national control of the economy. The restraints on imports probably act to keep alive some firms which are truly inefficient, but they are also aimed in part at keeping the industrial sector more diversified than it otherwise would be in the presence of strong primary exports. Diversification can be overdone, like everything else. Both Brazil and Colombia did do something to reduce the more costly forms of it in the - 26 - late 1960s and early 1970s by bringing down many of the particularly high rates of protection previously in effect, but Brazil has since added more protection than it had in the early 1970s. On the other hand, both countries. have increased the chances of stimulating dynamic new activities by providing stronger incentives for industrial exports and have considerably improved employment opportunities by these changes (Carvalho and Haddad 1982, Thoumi 1982). With their revised policies, the growth of income of the lowest 40 percent of the Brazilian population kept up with growth of national per capita income after 1967, and in Colombia the distribution of income began at last to be somewhat less unequal (Pfefferman and Webb 1979, Urrutia 1985). Export Orientation in East Europe: Hungary and Yugoslavia 1.36 Socialist countries of widely differing character have made a good many of the same kinds of moves toward improvement of efficiency through increasing exports and reducing barriers to trade, within contexts of exten- sive measures of continuing direct control of their economies. Of the East European countries, Hungary and Yugoslavia have been leaders in promoting greater economic flexibility and have achieved relatively good rates of increase in both their exports and income (Knight 1981). But they have, of course, encountered many problems en route, of which two are considered briefly here: high unemployment and serious external deficits and debt. To what extent have the countries' more open policies been responsible for these difficulties? 1.37 When a country moves from conditions of high protection to more active competition, the change increases pressures on firms to economize on input costs. Capitalist firms operating with high protection in the earlier - 27 - cases of import-substituting industrialization were biased toward excess use of capital-intensive methods: changes toward an export orientation in Brazil and Colombia were favorable for reduction of this bias and for change toward more employment-increasing techniques of production. Socialist firms under protection and unworried about costs are likely to use more workers than they really need. Concern for unemployment may override concern for costs. But the opening of greater competition and pressure for efficiency is likely to force each firm to reduce its use of labor, relative to output. If national output does not rise, unemployment must. A possible answer is to raise national output rapidly, as Korea did after adoption of its export orientation and Brazil did after 1967. But pressure to raise national output rapidly generates strong demands for iinmpo'rted equipment and supplies, and .isinR. exports may be accompanied by worsening external deficits that ssoonez or. Later force the country to put on the brakes. Is that perhaps a significant part of what happened in Hungary? The solution required rapid growth, which was achieved for a time, but it also required avoidance of excessive imports. On that count Hungary, Yugoslavia, and eventually Brazil as well, failed to exercise sufficient restraint. 1.38 Socialist countries may, for the reason suggested, be more prone to problems of unemployment when they turn to more open trade and they may, in addition, be handicapped in the resolution of this difficulty if increased trade takes place in a context of continuing direct control over allocation of inputs and finance. Increasing aggregate demand should stimulate employment in existing firms in socialist countries, as well as capitalist, but in a capitalist country, it should, in addition, provide opportunities to create new firms. If a socialist economy continues to use extensive controls over - 28 - inputs and financing, it may be extremely difficult for new groups to estab- lish new kinds of economic activities. Aggregate output may readily be stimu+ lated, but its character may-be restricted by lack of complementary new activ- ities helping to correct supply deficiencies and provide better alternatives to imports. If imports are simply blocked, productive efficiency is held down by that side; if imports are relatively free, an external deficit is likely to slow the system down before full employment can be restored. 1.39 Capitalist countries may still outperform socialist countries in creating or maintaining high rates of unemployment: Chile has outscored Hungary in that respect. Both kinds of system have infinite varieties of ways to do things better or worse. But for a socialist system, the conversion to more open trade does raise persistent structural problems which are of a special order. It is difficult to see how they could be handled merely by reducing restraints on trade: they raise fundamental questions about the nature of control over productive resources and the scope for independent action by individuals and groups. 1.40 External deficits on current account, financed by borrowing, can be either a source of improving capacity for autonomous growth or a recipe for loss of independence. When countries accumulate external debt on the scales of Hungary, Yugoslavia and Brazil, they are driven to share with external financing agencies some of the basic policymaking which determines the func- tioning of their economies. That may not be all to the bad. Hungary has been encouraged through negotiations with the IMF to make changes in its system of prices in directions which may be more positive for future growth than the country would have chosen on its own (Knight 1983). But if retention of: national control is a high priority goal, then avoidance of rising debt ratios - 29 - is important. The inefficient way to avoid them is to block import^s in, general, or imports which compete with domestic alternatives, regardless of the costs of the latter. The more efficient path is to increase the price,of foreign exchange and thus of all imports relative to the domestjic, .price l-ev'el. That kind of change both stimulates growth through exports and_raise.s, incentives to find efficient substitutes for imported inputs. 1.41 Socialist countries may also have a special problem of excess foreign borrowing by individual firms, if the management can gain from added productive capacity but is not checked by any personal concern for foreign exchange liabilities. This can be a problem with state-owned firms in capitalist countries as well, and has been so on colossal scales in Indonesia with Pertamina and Mexico with Pemex. In Hungary in 1950, it would have been unlikely because of the high degree of centralized control, but as the country liberalized its economic system and allowed greater freedom to individual firms, the possibilities of excessive borrowing and investment expenditure increased. The very real gains of greater flexibility for producers, better contact with worldwide improvements in-technology, and increased capacity for competitive exports may have brought with them a loss of control over the totals involved. 1.42 One answer to this kind of risk might be to reinstate a higher degree of central control over borrowing and investment, though this would imply a likely cost of interference with desirable flexibility of enterprise decisions.t Capitalist countries can, in principle, control the totals of borrowing without becoming involved in individual enterprise decisions by using monetary restraints and an exchange rate appropriate to hold down exces- sive claims on foreign exchange. They do not always succeed in practice, but - 30 - at least the possibility of controlling totals rather than details is clear. Socialist countries might need to go much further in controlling the actions of individual firms, or might instead rely more on capital markets allowed to function more freely. The range of possibilities is wide, but to avoid financial crisis, it is essential to find some way to limit the totals. European Industry and The European Economic Community 1.43 The highly industrialized countries of Western Europe, other than England, used considerable protection of industry in the early part of the twentieth century but changed that orientation markedly after the Second World War. Individual countries began, shortly after the war, to reverse the near- autarky they had practiced in the 1920s and 1930s, and then, in 1958, six of them joined to create the EEC, dedicated to removing all barriers to trade in industrial products among each other. This is not an open trading system: agriculture remains under jointly negotiated conditions of control, and pro- tection has been kept for industrial products coming from outside the group. Disputes over the controlled agricultural policies of the community have created intense conflicts of interest among the members, to a point threaten- ing survival of the Community itself. But on the side of trade in industrial products, the system has worked well. Restrictions against industrial products from member countries have been removed and the average level of protection against outside manufactures has been reduced to relatively low levels. The process might be compared to a total opening up of decentralized trade in manufactured goods among all the regional economies in China, accompanied by a partial opening of national trade with the outside world. - 31 - 1.44 In Europe, the creation of the EEC encouraged rapidly increasing industrial production in every member country: there has been no tendency for the more industrialized countries to drive the others back to a narrowed range of production. In comparison with the Chinese situation of widely dispersed industries selling mainly to assigned local or regional markets, the interest- ing question is how the previously separate industrial systems of the European countries, starting with many duplicating industries, managed to open up trade in a way which led to the expansion of industry in all of them rather than with setbacks to the initially less industrialized. 1.45 At the time of the creation of the EEC, it was widely expected that Germany with its exceptionally dynamic industry would dominate both France and Italy. France had been a close rival to England in the early stages of indus- trialization and adopted an open economy for a time in the mid-nineteenth century, but then gradually lost competitive strength in a long process of withdrawal into increasing protection. Italy was a lower-income country with a less-developed industrial sector than either of the other two. But all three, along with the three smaller countries joining the EEC, were determined to break out of previous restraints on trade in order to share in a more dynamic European Community. 1.46 For France, the possibilities of successful competition were greatly increased by the creation, at the beginning of the postwar period, of a system of national economic planning, oriented specifically toward restoring competi- tive strength through greater emphasis on efficiency in a more outward- oriented economy (Sheahan 1963). Planning was given force in the early post- war years by government control of a substantial share of investment financing and by an intricate system of negotiation between the government and leading - 32 - industrial groups. Even prior to the establishment of the EEC, the rate of investment and of economic growth proved to be far above any earlier period in the country's modern history. The ratio of exports to production increased in all industries simultaneously. That drive created the necessary productive base to meet German and Italian competition, but it was not in itself suffi- cient: the level of French industrial prices remained too high. The solution was to devalue the exchange rate sharply at the start of the EEC, and to apply macroeconomic and direct wage restraints to keep the devaluation from stimu- lating inflation. The combination worked. For the decade of the 1960s, France and all the members of the EEC shared exceptionally rapid economic growth without high rates of inflation. Estimates of the effect of the removal of trade barriers suggest that trade among the member countries grew approximately 50 percent faster in that decade than it would have in the absence of the Community (Balassa 1975). 1.47 The particular pattern of trade in industrial products among the EEC countries proved to be notably different from that suggested by the Heckscher- Ohlin theory of comparative advantage based on differing factor endowments. The countries did not separately specialize in any particular set of indus- tries. Instead, most trade took place among products within the same indus- tries: the products traded were either differentiated consumer goods answer- ing market preferences for variety or differentiated capital goods with specialized characteristics. Germany proved to be the most rapidly advancing producer of many lines of machinery, but French construction machinery, equip- ment for subways and railroads, and other specialized lines provided counter- part exports, keeping its capital goods industries advancing rapidly as well. - 33 - 1.48 Balances in industrial trade with mutual expansion were not auto- matic. Both France and Italy have had more trouble with inflation and more frequent trade deficits than the others. Imbalances have been corrected by combinations of macroeconomic policy and changes in exchange rates, using both appreciation in Germany and Holland and depreciation in France and Italy. These changes were not left up to uncontrolled market forces: all the coun- tries concerned have practiced intervention intended to keep competitive conditions approximately stable among themselves. These adjustments enabled the member countries to combine mutual growth with distinctly different national institutions and policies. Two Economies Closed for the Sake of Autonomy and Equality: Burma and Sri Lanka 1.49 Burma and Sri Lanka have been interesting contrasts to most other developing countries in that they firmly opted out of the common pattern of preoccupation with economic growth in order to emphasize social goals and, in Burma's case, also to protect a different way of life from change through contact with the outside world. Burma withdrew from its earlier postwar effort at import substitution in 1962, rejecting modernization itself as well as any form of external aid. Sri Lanka pursued a relatively isolated path from independence in 1948 up to the election of 1976, when the majority of the public voted for a party insisting on the need to give more attention to growth and to contact with the outside world. Both countries consider them- selves to be socialist, both are relatively small, and neither is under the direct influence of the Soviet Union. Perhaps the most important factor they have in common is their predominant religion: they are both Theravada - 34 - Buddhist countries, distrustful of the materialistic emphasis of the modern world. 1.50 The degree to which they withdrew from trade is striking. Foreign trade was equal to 20 percent of GDP for Burma in 1955 but down to 5 percent by 1975. For Sri Lanka, the ratio was cut from 32 percent in 1955 to 21 per- cent by 1975 (Morawetz 1980, Table 5, p. 349). In constant prices, Burma's exports in 1974/75 were only one sixth of their volume 15 years earlier. Both countries limited exports by high taxes on their principal food and raw mate- rial exports. Neither country took any steps to develop new industrial exports. The rate of growth of GNP per capita in Burma had been 4.3 percent per year from 1950 to 1960; it fell to 1.4 percent per year for 1960-81 (Table 2). Sri Lanka's rate of economic growth has been estimated at a higher rate for 1948-60 (2.8 percent per year), but World Bank data indicate a rate slightly below 2 percent for the period 1950-75 (Morawetz 1980, Table 1, p. 340). Morawetz explains the below-average growth rates of both countries in terms of negative price incentives for the primary sector, discouraging production and exports, combined with high public sector spending on subsidies for consumption and correspondingly limited resources for investment. 1.51 These two countries might be seen as demonstrating the retarding effect on growth of policies adverse to trade and investment. But the conse- quences of their choices should not be assessed without reference to their own explicit objectives. For Burma, a relatively egalitarian initial distribution of income seems to have been maintained and the country's life style kept intact to a degree which might not have been possible in a more dynamic economy, but it is difficult to find a great deal more about the experience that could be viewed positively. An extraordinarily high rate of infant - 35 - mortality (nearly four times as high as that of Sri Lanka in 1962), did fall slowly and educational attainments are slightly above international norms for the income level, but vocational school enrollments remained unusually low and unemployment stayed very high (Morawetz 1980). 1.52 Sri Lanka did distinctly better than Burma, perhaps in part because its open political system meant that the government had to maintain popular support by demonstrating achievement. The country came to stand out as a world leader in many dimensions of social performance. Highly equal access to education and health care and assured minimum food consumption made Sri Lanka in this period far above world norms for its level of income (Isenman 1980). Using a regression analysis for 59 countries to identify the norms to be expected at Sri Lanka's level of per capita income, Isenman (1980) showed that the country had the highest positive deviation of all 59 countries for perfor- mance in terms of life expectancy, adult literacy, and ability to achieve low infant mortality. 1.53 One of the key factors favoring social welfare in Sri Lanka was a policy similar to that in China; the price of the most important food, rice, was kept relatively low so that the poor could buy at least a minimum ration. In many countries, an unfortunate counterpart of such a policy has been that incentives to producers have been kept too low to maintain adequate output. Sri Lanka did import some of its rice supply, but it also used subsidies to producers sufficient to stimulate growing production and gradually reduced import dependence; rice output increased at the high rate of 5.8 percent per year during the 1960s (Isenman 1980, p. 240). On the negative side, this and other social subsidies seriously limited the resources available for invest- ment. Many new industries were established, but in most cases, on too small a - 36 - scale, protected from import competition and allowed to remain unconcerned about costs. This policy left productivity low and handicapped growth. Imports fell slightly in the course of the 1960s and then had to be cut more rapidly in the first half of the 1970s. When the main opposition party campaigned in 1976 on the basis of a promise to open up the economy and stimu- late trade, to restore more active contact with the outside world, the popular vote turned against the relatively closed system. 1.54 Sri Lanka's subsequent reorientation toward a more open economy was intended to combine stimulation of efficiency and growth with continued con- cern for equity. It might be regarded as a precursor of China's turn toward a more open economy shortly thereafter. But two particular aspects of the change in Sri Lanka serve chiefly to point toward misdirections that China would do well to avoid. One is the method chosen to reform the basic program of food support for the poor, and the other is the character of its export promotion. 1.55 The most important component of the pre-1976 method of supporting consumption of the poor was to keep down the price of rice through subsidies. Incentives to producers were maintained by paying them substantially higher prices than consumers were charged, parallel to the method used in China after the changes made in 1979-81 to raise prices to rural producers. The system worked to sustain growing production, but it also created an unmanageably large government deficit. In 1976, the method was changed to reduce the subsidies while still helping the poor: ration claims were issued to the lower-income half of the population, and prices were allowed to move up to an open market level so that all other purchases were at an open market price. The change effectively concentrated subsidies on the poor and brought the - 37 - costs of rice to everyone else up to the cost of production. That could have been a useful model to follow except for one feature. Ration claims were fixed in terms of market value at the initial set of prices, rather than quantities of rice. Inflation, devaluation and rapidly increasing market prices for rice quickly reduced the quantity which could be obtained by poor families. Since the government kept the nominal value fixed, many families were driven to drastically reduced levels of food consumption. The change in method could have been consistent with increased efficiency and continued protection of the poor; as actually implemented, it deepened poverty. 1.56 For export promotion, Sri Lanka introduced duty-free zones for production of exports. That drew in a number of foreign firms but, with no links to existing national industries, they offered few opportunities for use- ful learning. Domestic industries outside the zones might have been able to develop their own exports with a more favorable exchange rate, but the initi4l devaluation was soon swamped by inflation, due in large measure to a great increase in public sector investment. The nominal exchange rate was not moved to keep ahead of inflation; the effective rate actually appreciated. Domestic industry was subject to the impact of increased external competition under tlie handicap of the adverse change in effective exchange rates, rather than stimulated toward new exports. Sri Lanka's move to a more open system has included many positive features, but these two aspects could hardly be considered either necessary or desirable. Effects of Trade on Regional Inequalities 1.57 Questions of regional inequalities and interactions constitute a major field of study in themselves; personal ignorance of most of this litera- - 38 - ture makes it best to limit the following suggestions to a very small corner of the subject. However, it is of particular concern to large countries and it may be especially so for China. In large countries, the inherently uneven process of change, accentuated by opening up markets to increased trade, can set back incomes and wipe out local economic activity in the poorer regions for long periods. It may be of exceptional importance to China because the country has more dispersed economic activity subject to possible destruction than has been the case in other newly industrializing countries. Some of these activities may be so inefficient that it would be desirable to shut them down, but too much may be destroyed in the process. 1.58 Classic discussions of regional imbalance effects in developing countries, by Gunnar Myrdal (1957) and Albert Hirschman (1958), both pointed out a variety of ways in which centers of growth in the higher-income regions may destroy more traditional lines of production in lower-income areas by providing new products that are lower cost or preferred as to quality, by offering higher-pay activities in the industrializing areas that draw away the people with technical skills or above-average enterprise, and by the depress- ing effect on everyone's morale of regions increasingly devoid of varied opportunities. Myrdal leaves the impression that negative "backwash" effects on the poorer regions constitute the whole story, while Hirschman gives more attention to the accompanying positive possibilities gradually opened up by the spread of knowledge about new opportunities, by increasing demand from the industrial centers for the products of the poorer areas, and by political con- cern likely to-stimulate central government action in favor of the more bAck- ward areas. Both analyses present reasons to expect that poverty in the poorer regions may increase in absolute terms for considerable periods, even though it is bound to decrease eventually if national economic growth continues. - 39 - 1.59 Empirical investigation of actual trends in regional inequalities suggests a common pattern of change that could be seen to be reassuring for the long run: regional inequalities usually increase in the early stages of industrialization, stabilize after a time, and eventually decrease (Williamson 1965). But even if that generalization is accepted as valid, it is not espe- cially encouraging in respect of the timing: regional inequality increased in France through the nineteenth century right up to the First World War, increased through much of the nineteenth century in the United States and, after two decades of moderate increase in Brazil from 1920 to 1940, increased sharply in the next decade. 1.60 In Williamson's study, Brazil showed the highest degree of inequality for all countries considered, in the years around 1960, though with some signs of a turn toward lessening inequality. Subsequent studies suggest rather that differences among regions remained fairly stable through the 1960s, and that within the particularly impoverished Northeast (with 25 per- cent of total population), interpersonal inequality increased (St8hr and Thdtling 1977). In the 1970s, the situation finally did improve signifi- cantly, thanks to exceptionally rapid economic growth and to the new set of national policies favoring more labor-intensive production: per capita income in the Northeast increased 84 percent in real terms, compared to an increase of 70 percent for the country as a whole (World Bank 1983d). 1.61 The more recent empirical studies have done more to support Williamson's theme that inequality among regions increases in the early stages of industrialization than that it narrows down later on. Even where the national government attempts systematic measures to limit the process, it is difficult to prevent. Regional inequality increased in Malaysia between 1963 - 40 - and 1970, and in both Indonesia and the Philippines between 1971 and 1979 (World Bank 1983e, Table 2.5, p. 13). It is invariably the more rural regions that lose out in the early stages, both with respect to income and with shares of industrial production. For Indonesia's poorest region, the Eastern Islands, 47 percent of the people are considered to be in poverty and 26 per- cent in destitution, compared to 3 percent for the country as a whole. Cor- respondingly, with 7.5 percent of the country's population, the region had, in 1971, only 0.8 percent of national value added by medium or large industrial plants, and by 1980 that share had dropped to 0.3 percent (ibid, Table 2.12, p. 23). 1.62 At the time of Williamson's study, India was going through a severely negative early stage of increasing regional inequality, combined with a relatively low rate of aggregate economic growth. National policy has since devoted a great deal of attention to such inequalities and has included a wide range of measures to lessen them (Sekhar 1983). They have been particularly oriented to protection of household industry, and have kept it alive to a degree most unlikely to have resulted in a fully open trading system. The policies used have included three major types: (1) direct controls forbidding some investments in the more industrialized areas or requiring public firms to place investments in poorer areas, (2) financial incentives to encourage smaller-scale industry and location in poorer areas, and (3) allocations of public investment in infrastructure to favor such areas. 1.63 As discussed by Sekhar (1983), a good many of these efforts have had negative effects on efficiency, either blocking potentially desirable invest- ment decisions or raising costs by leading to location decisions too far from markets or appropriate sources of inputs. Direct controls forcing higher-cost - 41 - location for individual firms would seem to be a singularly undesirable way to help prevent increasing poverty in the more backward regions. Another measure used in India has been to pool freight charges to make delivered product prices equal for basic commodities no matter where purchased, so that interior regions far from the point of production do not need to pay higher prices than those prevailing in the more industrialized zones. When applied to basic commodities such as cement and steel, this can clearly lead to a great waste of energy and transport facilities. Financial incentives such as lower cost credit or tax advantages would seem less likely to lead to serious waste: at least they would not selectively favor location in poorer areas by firms using specific inputs with high transportation costs. On the other hand, they do not push firms into action in the poorer areas, and they do not preclude ten- dencies of managers to concentrate production in the higher-income areas for reasons of personal preference. Market forces naturally favor location near markets, but that does not provide any ultimate test of locational desirabil- ity: if policies to support poorer regions are effective, then the location of markets will change more to the poorer regions and so will the pattern of new investment. 1.64 Hirschman's suggestions about desirable policies to prevent deepen- ing poverty, for long periods, in the less-industrialized regions point toward a search for measures which would create for the region some of the protective advantages it would have if it were a separate country. The goal would be to slow down negative impacts on local economic activities without preventing the gradual development of positive stimulation radiating outward from the more dynamic areas. Some possibilities aimed at such balancing measures for China are suggested in paras. 3.10-3.17 below. They would all involve some - 42 - restraint on the free operation of market forces. So*me such restraint is almost surely necessary if the lower-income regions of the country are to be sheltered from negative local repercussions of the generally positive change toward more freedom for both international and internal trade. 2. ASPECTS OF THE CURRENT CHINESE SITUATION 2.01 China is such a world in itself that questions of trade and special- ization need to be considered jointly in terms of the nation's relationships to the outside world and of regional balance within the country. On both sides, the country is in the midst of a rapid process of change, subject to question and to new starts at every turn. The present chapter reviews leading aspects of recent developments in an attempt to characterize the current situation, particularly with respect to external balance, the character of specialization and trade, questions of migration and regional balance, and the relationships between internal and external prices. The following chapter then considers major options of national economic strategy. Direction, Balance, and Composition of External Trade 2.02 China's inward-directed socialism up to 1977 accentuated the normal tendency of all large countries to export only a small fraction of national product and to import a correspondingly small fraction of national supplies, but the ratio of exports of goods and services to ONP has actually been only slightly below that of the United States. The United States raised its exports of goods and nonfactor services from 5 percent of GNP in 1950 to 10 percent in 1981, while China raised its ratio from 5 percent in 1961 to 9 percent in 1981 (World Bank 1983c, pp. 156-157). - 43 - 2.03 Total exports and imports both increased rapidly from 1977 to 1980. Exports continued increasing thereafter, though more slowly, while imports fell slightly. The greater part of the increases since 1977 have been in trade with market economies; exports to them accounted for 80 percent of the total in 1975 and 95 percent in 1983 (Table 4). The close balance between exports and imports characteristic of China's trade in preceding years changed to deficits averaging $1.5 billion a year for 1978-80, but restrictions on spending and imports in 1981, combined with increasing exports, created a virtual balance in 1981 and trade surpluses in 1982 and 1983. Table 4. CHINESE EXPORTS AND IMPORTS, AT CURRENT PRICES, IN TRADE WITH CENTRALLY PLANNED AND WITH MARKET ECONOMIES, 1975-82 (US$ million) Exports Imports Total Central Market Market Total Central Market Market Balance tZ) (x) 1975 7,264 1,039 5,816 80.1 7,486 1,002 6,484 86.6 -222 1977 7,590 1,321 6,229 7,214 955 6,259 +376 1978 9,745 1,078 8,667 10,893 1,007 9,886 -1,148 1979 13,658 1,272 12,386 15,675 1,418 14,257 -2,017 1980 18,272 1,190 17,082 19,550 1,310 18,240 -1,278 1981 21,561 782 20,779 96.4 21,568 811 20,757 96.2 -7 1982 21,908 894 21,014 18,925 1,251 17,674 +2,983 1983 21,970 1,032 20,938 95.3 21,317 1,360 19,957 93.6 +712 Source: Ministry of Foreign Economic Relations and Trade and General Administration of Customs. 2.04 Exports to developing countries with market economic systems were 58 percent of all exports to market economies as of 1978. From 1978 to 1983, exports to industrialized countries increased 2.6 times in current value, - 44 - while those to developing countries increased 2.3 times. However, the share going to developing countries remained slightly higher, at 56 percent of the total. Two destinations dominate the total: Japan took 48 percent of Chinese exports to industrialized countries in 1983, and Hong Kong took 49 percent of the exports to developing countries. The third largest market, the United States, took only 7 percent of Chinese exports to industrialized countries in 1978, but that share increased to 18 percent of such exports by 1983. 2.05 In imports, the balance between trade with developing and with industrialized countries is the reverse: industrialized countries supplied 81 percent of total imports from market economies in 1978 and 74 percent in 1983. The overall trade surplus in 1983 thus consisted of a large surplus with developing countries, a small deficit with centrally planned economies, and a large deficit with the industrialized countries. Does this imbalance between trade with industrialized and with developing countries seriously matter? Probably not; the groupings themselves are fairly arbitrary. If trade with Hong Kong and Singapore were reclassified into the group of indus- trialized countries, this would not change reality, but it would considerably change the degree of contrast between balances with the two groups. 2.06 A major imbalance in 1983 was in trade with the United States. The deficit in trade with the United States was $1.0 billion, 20 percent of the deficit with all industrialized countries combined. The other two'imajor imbalances were in trade with Germany and with Japan. 2.07 The broad. features of the structures of exports and of imports are shown in Table 5. The largest change in exports between 1979 and 1983 was the decrease in the share of primary products other than oil from 34[tor23 per- cent. Perhaps the most encouraging aspect of the export pictureiis! that China - 45 - avoided any negative impact on industrial exports when oil exports increased. In Mexico, the effect of greatly increased oil exports in the 1970s was to raise domestic prices and demand at rates which undermined previous progress in developing industrial exports, but China was able to combine higher earnings from oil with a rapid increase in total industrial exports. By 1983, they accounted for 57 percent of total exports. Since oil exports are not likely to grow indefinitely, and may well begin to fall as Chinese incomes and demand continue rising, the central hope for continuing rapid growth of exports must lie on the side of industrial products. 2.08 The structure of imports shown in Table 5 includes a marked decrease in the share of machinery and equipment from 1979 to 1983. The share of other heavy industry was also cut back, but increased greatly in 1983. Concern for food supplies, especially grains, remains high despite success in stimulating output by new agricultural policies from 1979. China's imports of wheat accounted for 13 percent of total world wheat imports in 1981/82 (Hickok and Arguelles 1983, p. 47). Internal Trade and the Location of Industry 2.09 A particularly striking feature of China's economic strategy from 1949 was the-extraordinary dispersion of industrial production among separately diversified local areas. The system has been termed a "cellular" approach to industrial development, in which each local cell tries to come as close as possible to meeting its own needs for simple manufactured goods, building up from the smallest local unit to nearly complete diversification at the level ofithe province (Lyons 1983). It would be difficult to find any- where a morevthorough-going contradiction of the principles of specialization - 46 - and comparative advantage. All of Adam Smith's themes about the gains of efficiency through division of labor and trade were explicitly repudiated. That repudiation is now being reconsidered, consistent with the reorientation of national policy toward greater trade with the outside world (Chen 1983). Table 5. STRUCTURES OF EXPORTS AND OF IMPORTS, 1979-83 (Percentage shares) 1979 1980 1981 1982 1983 Exports Primary products Fuels, lubricants, and related 19.5 25.1 20.2 23.8 20.3 All other primary products 34.1 28.3 26.4 21.2 22.5 Total Primary 53.6 53.4 46.6 45.0 42.8 Industrial products Chemical and related 3.1 3.4 6.0 5.4 5.7 Machinery and transport equipment 3.4 4.7 5.1 5.7 5.5 All other heavy industry 4.4 4.8 8.7 11.5 11.0 Light industry and textiles 35.5 33.7 33.6 32.4 Total Industrial 46.4 46.6 53.4 55.0 57.2 Imports Primary products Food 14.4 16.1 16.5 22.3 14.4 Others 13.8 19.3 20.0 17,4 12.8 Total Primary 28.2 35.4 36.5 39.7 27.2 Industrial products Chemicals and related products 10.4 12.6 11.9 15.3 14.8 Machinery and equipment 25.9 26.9 26.6 16.6 18.5 Other heavy industry 31.0 17.1 8.9 13.4 27.3 Light industry and textiles 4.5 8.0 16.1 15.0 12.2 Total Industrial 71.8 64.6 63.5 60.3 72.8 Source: Ministry of Foreign Economic Relations and Trade and General Administration of Customs. - 47 - It would be of great interest to find out exactly how much, and in what ways, this reconsideration has led to changes in the patterns of internal production and trade.4/ It would also be of great interest to clarify the gains and losses both of the earlier orientation and the changes now underway. 2.10 As a result of the emphasis on dispersion of industrial production, there are no provinces as devoid of industry, as low-income rural areas so often are in market-oriented developing countries. In Peru, with per capita income approximately four times that of China, 13 departments out of the 24 in the country have either no industrial production at all or, at best, no more than 1 percent of the national total of industrial value added. The urban concentration of Lima-Callao accounts, by itself, for no less than 74 percent of the country's industrial value added (World Bank 1979, Vol. III, p. 383). In China, the provinces with the least industrial production per capita in 1982, Xizang, Guizhou, Yunnan and Guangxi had, respectively, 12, 33, 44, and 44 percent as much industrial output per capita as the national average (Table 6). No other province had less than 50 percent of the national average. 2.11 In some respects, the more appropriate comparison would be between dispersion within Chinese provinces, many of which have populations exceeding that of Peru. Some, but not all, of them have high degrees of industrial con- centration in the provincial capital. In Gansu, one of the least industrial- ized provinces, the municipality of Lanzhou includes 12 percent of the provin- 4/ It would be particularly helpful to establish data for trade by indivi- dual provinces, both with each other and with the outside world, for major categories of products, if possible for a series of years starting from 1977. This could be formulated in ways similar to the regional trade tables established for Indonesia in World Bank 1983e, Tables 2.8, 2.14, and 2.15, though it would be useful to aim at a matrix of interprovince trade by product categories rather than just totals for all trade. - 48 - cial population, but accounts for fully 55 percent of gross industrial output value (Statistical Yearbook of China 1983, pp. 95, 106 and 226). Still, the counterpart of even that high degree of concentration is that the rural counties in Gansu have somewhat more diversified activity than the rural departments in Peru: the concern of policy might well be to build up their share, which would allow for closing down particularly costly firms, but should mean serious effort to bring in new activities. 2.12 Given the common condition of concentrated poverty and absence of local employment opportunities in the rural areas of so many developing coun- tries, the intended dispersion of industrial production in China seems a highly appealing alternative. What is perhaps most striking of all, if one compares rural Latin America or Africa, is that local people have been able to initiate such diversified kinds of industrial production. That is the posi- tive side of the process, and it offers considerable hope for widely diffused progress in the future. At the same time, there is a clear and possibly very costly negative side. The whole approach allows for a great deal of ineffi- ciency in terms of inappropriate industrial location, impediments to achieve- ment of economies of scale, failure to concentrate scarce inputs of energy and raw material in the plants best able to use them, and failure to use domestic competition to select the most productive allocation of resources. 2.13 Some indications of the characteristics of Chinese industrial struc- ture suggest that costs and qualities of similar products differ widely among regions. For trucks and tractors, nearly all provinces have some output of their own, but many have such low rates of production that costs per unit must be exceptionally high (Lyons 1983, ch. 4). In the presence of active trade among regions, many lines of production would become more concentrated as - 49 - those with advantages of scale or superior organization proved able to under- sell higher-cost local producers. Such a process would not knock out all local industry: some is surely efficient and well adapted to local needs, some production units would respond to competition by copying or improving on the techniques which proved superior in the first instance, and all local production would have advantages of lower transport costs. But the normal expectation would be that removal of trade barriers would set back production and close higher cost units in many activities in the less-industrialized regions, at least temporarily increase problems of unemployment, and make it more difficult for local production groups to initiate new lines of production competing with established activities elsewhere in the country. 2.14 The arguments for encouraging increased trade among regions have ktrong logic on their side but in China they involve a special problem. A major feature of national economic policy has been the restriction of movement by people. To the degree to which this restriction is effective, it means that the people in any region which suffers loss of income and employment Dpportunities by increased access to industrial imports are simply stuck: there is no legal scope to move to the regions which gain. 2.15 In practice, not all people stay where they were born or assigned. But personal movement without permission remains difficult: urban jobs are normally allocated through public labor bureaus, and the economic system, as a whole, requires reliance on personal documentation to get ration coupons for basic necessities, public allocation of housing, and access to public agencies on many of the steps that, in most countries, would be handled privately. The important role of internal migration in most countries that allow at least some members of the family in the poorest areas to move to where incomes are - 50 - higher has been partially blocked. Failure to keep alive diversified economic activities in rural areas would have more serious consequences in China than it would in most other countries. 2.16 Limitations on migration are, in part, intended to prevent the massive urban unemployment experienced in many low-income countries, as people move into cities looking for jobs in much greater numbers than employment opportunities can be created. Given the enormous sizes of the major Chinese industrial cities and the tightly limited supply of housing, restrictions on migration to the largest cities are certainly understandable. But migration from rural areas to market towns and smaller cities could help equalize oppor- tunities without concentrating ever larger numbers in the major industrial centers. 2.17 The central government has, of course, done a great deal to provide help to lower-income regions. Such measures include provision of a guaranteed minimum food supply for each family, a tax structure allowing poorer locali- ties to retain a higher fraction of tax receipts for local use, and a policy of moving grain into areas with deficit production when adverse conditions threaten local famine. One might think that this provision of food to deficit areas would be natural in any society, but it is the opposite of the pattern that can occur in low-income market economies: localized crop failures in Ethiopia and in Bangladesh in the 1970s, which wiped out local purchasing power because there was so little to sell, led to shipments out of the very areas where famine was most severe (Sen 1981). The Chinese system works much better for supplying minimum food requirements. It also holds down inequali- ties among the people in each region. But it heightens inequality among people in different regions by preventing personal movement. - 5 1 - Table 6. RECIONAL INEQUALITY IN CHINA, 1982 Gross value of output Personal income Population per capita as percentage per capita as (percentage of of national average percentage of national Industry Agriculture national average total) 3 Municipalities 678 108 170 2.9 Beijing 451 287 179 0.9 Tianjin 493 292 147 0.8 Shanghai 977 566 180 1.2 Northeast 176 104 113 9.0 Liaoning 240 165 130 3.6 Jilin 116 101 89 2.2 Heilongjiang 112 141 110 3.2 North 80 97 94 9.7 Hebei 78 89 91 5.3 Shanxi 95 97 94 2.5 Nei Monggol 64 86 102 1.9 East 91 118 102 28.2 Jiangsu 150 126 109 6.0 Zhejiang 107 119 117 3.9 Anhui 53 81 92 5.0 Fujian 61 86 97 2.6 Jiangxi 53 79 93 3.3 Shandong 89 101 103 7.4 Central/South 69 95 101 27.1 Henan 53 70 76 7.4 Hubei 103 103 103 4.8 Hunan 64 85 98 5.4 Guangdong 82 106 137 5.9 Guangxi 44 70 84 3.6 Northwest 70 82 91 6.9 Shaanxi 71 71 87 2.9 Gansu 74 73 76 1.9 Ningxia 66 79 90 0.4 Qinghai 65 83 108 0.4 Xinjiang 64 97 115 1.3 Southwest 48 83 83 16.2 Sichuan 55 71 84 10.0 Guizhou 33 51 79 2.8 Yunnan 44 68 83 3.2 Xizang 12 56 106 0.2 National Average 100 100 100 100.0 Source: State Statistical Bureau. - 52 - 2.18 Table 6 above gives a picture of differences in product and income among provinces. If the listed cities are excluded from the comparisons, per capita personal income levels range from 1.8 to 1. If the cities are included, the range for personal income is 2.4 to 1 between Shanghai and Gansu. That is a fairly wide range, and it does not even take into account the great differences between poorer and less poor areas within each province. Personal incomes are relatively equal within each city, and concentrated personal incomes based on property ownership have been practically eliminated. But incomes are still nowhere near equal between the less and the more favored areas, either because of differences in the presence of varied economic activities or differences in the quality and scale of land available in the rural areas. The question this raises for national policy toward interprovincial trade is that limiting such trade keeps the national income lower than it could readily be, but opening it up, while keeping people from moving freely, could aggravate these differences and quite possibly increase poverty in the least-favored regions. Balance of Payments, Capital Flows, Foreign Investment 2.19 China has pursued one of the world's most resolutely conservative policies of external finance, keeping very close to balance on current account in most years. While this caution about external deficits may have cost opportunities for growth, it also insulated the country from the severe shocks of the worldwide debt and financial crisis of 1982/83. China did not need to go through the forced contraction-plus-inflation strains of the heavy debtors such as Brazil, Chile, and Mexico. - 53 - 2.20 As shown in Table 7, the previous policy of careful balance on external current account was relaxed temporarily at the end of the 1970s: moderate deficits were allowed from 1978 through 1980, but these departures were followed by a shift to a current surplus in 1981, with exports increasing and with reduced imports. Imports continued to be held down in 1982 and 1983, and the country has continued to run a surplus on current account. Table 7. BALANCE OF PAYMENTS, 1978-83 (US$ million at current prices) 1978 1979 1980 1981 1982 1983 ittq tq9' Exports Merchandise, f.o.b. 9,607 13,658 18,492 22,027 22,476 24,815 23,9 6 Nonfactor services 763 1,325 1,832 2,396 2,426 (52 Imports I" Merchandise 10,745 16,212 22,049 21,047 17,113 18,9955 2- ? 379v Nonfactor services 923 1,547 2,009 2,399 2,876 3 039 Resource balance -1,298 -2,776 -3,734 977 4,913 j,81 Net factor income -8 -77 -117 -200 283 1,124 Net current transfers 597 656 640 464 530 436 Current account balance -709 -2,197 -3,211 1,241 5,726 4,381 2.7O so Long-term capital I Direct investment - - 57 265 359 478 All other -830 822 1,703 366 -181 34 Net credit from IMF - - - 5 4 _ _ -_4_48 Net short-term capital -241 1,554 76 -959 -55 -326 Capital flows NEI -69 -30 -91 -48 -174 Errors and omissions 1,061 448 1,836 958 501 -4'6 -t Change in net reserves 788 - 597 - 391 -2,503 -6,291 -3,514 (- indicates increase) Sources: Bank of China and IMF and World Bank estimates. 2.21 Staying close to balance on current account in most years has meant that capital flows have not been of great significance, though China has borrowed fairly regularly when it has been possible to do so on favorable - 54 - terms. The largest source of low-interest lending has been Japan. Total debt outstanding at the end of 1983 was $6.4 billion; service payments in that year were equal to 5.5 percent of the exports of goods and services. Foreign exchange reserves including gold were equivalent to nearly 12 months of imports. 2.22 The balance of payments shows small inflows of direct investment through 1983. This understates the actual role of foreign firms in providing equipment and taking part in production because such activity takes a variety of forms which minimize foreign exchange transactions. These include export processing agreements under which foreign firms supply materials and technical specifications for assembly operations directed exclusively to exports, "compensated trade" under which foreign firms provide capital equipment and production rights in return for subsequent deliveries of a share of output, and "cooperative production" under which the Chinese firm gets licensing rights and technical advice while full equity remains Chinese (World Bank 1983a, Vol. II, pp. 433-435). 2.23 Joint ventures with foreign equity have been officially encouraged since 1979, but the number and the amount of investment involved have remained low. Outright foreign-owned production for sales on domestic markets has been allowed in a very few cases. The restraint in joint ventures and completely owned foreign investment has been due both to caution by the Chinese and to uncertainty on the part of foreign firms, particularly about such issues as the right to choose and discharge workers and the administrative complications of dealing with such unfamiliar conditions (ibid, pp. 436-441). The agree- ments concluded for cooperative production under full Chinese ownership have so far been of considerably greater importance than investment with foreign equity. - 55 - Export Performance, Exchange Rates, and Prices 2.24 China's export agencies have shown a phenomenal ability to raise sales in recent years, first very rapidly from 1977 to 1980 and then continu- ing right on through the weak conditions of the world economy in 1981-83. This achievement has been especially impressive for manufactured goods. What determines this capacity, what problems are involved, and what is the poten- tial for the future? 2.25 The process of export promotion is clearly not an ordinary matter of decisions by individual producers to enter export markets in response to incentives determined jointly by their own production costs, foreign prices, and exchange rates. Relationships between exchange rates and prices are significant, but they do not readily account for actual performance. Exactly what the key factors are is hard to pin down. The following account should be regarded simply as an hypothesis attempting to account for aspects of behavior which do not readily fit traditional western economic analysis. 2.26 Export performance seems to be determined more by social decision than by incentives at the level of the individual firm. The core of the question is not so much what exports do to the profits of producers as it is the objectives considered to be desirable at the level of national economic policy, and how these preferences are brought to bear on the provincial or local government agencies in the areas able to provide exports. Exchange rates (including the internal settlement rate) and compensation arrangements for the producing' firm play their roles primarily as means to eliminate any negative effects of exports on earnings, rather than as incentives for profits to be earned through exports. - 56 - 2.27 The firm producing exports may not make any more profits, or any less, when it delivers goods to meet export targets of the relevant trading agency instead of delivering them to the home market. An export target for the firm simply becomes another assignment to meet. If this is a correct interpretation, why did exports rise so much? The suggestion here is that they increased because of new policy preferences at the level of both the central government and the local communities concerned. 2.28 At the level of central government planning, decisions may be taken either to encourage exports for national objectives or (as in early 1984) to slow them down. When the decision is taken to encourage exports, incentives are provided in the form of favorable allocations of materials for production, for investment, and for such highly desired purposes as construction of more workers' housing. Exporters or their communities may also be allocated part of the foreign exchange earned to pay for imports of desired supplies. Export targets are then negotiated at levels which producers consider to be feasible and which planning authorities (and domestic purchasers) agree to,be consis- tent with adequate domestic supplies. At the level of the local community and firm, such increased allocations of materials may mean improved possibilities for investment, greater local employment opportunities, better housing, and more local tax receipts. Whether or not they increase profits for the firms would not seem to be the essential consideration. If they meant actual losses for the firms, that would be a serious problem, but they do not because the settlement process precludes such losses. 2.29 What determines the target for total exports and their composition? The level is not explained by a need to cover import requirements: the country has been running an export surplus. Furthermore, imports themselves - 57 - are a policy variable, not an unalterable rate which must be balanced by exports. Chinese officials asked about the matter seem to agree that total imports are set, by decision, at levels far below the preferences of potential importers. The restraint on imports might be explained by many different reasons, but it has not, in recent years, been a problem of an effective limit on foreign exchange availability. 2.30 What determines the particular products selected for exports? That seems to depend first on identification of market possibilities, and then by the question of the cost of compensating producers at the domestic price level. That should limit possible cases of exporting at a loss in terms of real national income, but it may not rule them out. The problem is not so much a matter of subsidies to the exporters as of arbitrary prices on raw material inputs: if the prices are below the value of the materials in world trade, then China's national income may be lowered by exporting products which use them as inputs. On the other side, opportunities for truly profitable exports may be missed if the individual firms who could make the sales cannot actively enter export marketing themselves. The system depends on successful search by the export corporations, both for external markets and for potential internal suppliers. That latter consideration may be a real problem: it must be difficult for sales agencies outside the firm to discover possibilities of production for export which could become profitable to all concerned if pro- duction volume were raised through exporting. 2.31 In March 1984, the government decided to cut back on export promo- tion to some;degree by reestablishing centralized export marketing for a number of specified commodities. The movement of the immediately preceding years toward greater freedom for local agencies to export was at least tempo- - 58 - rarily reversed. This change was explained as being due to a belief that some Chinese export agencies were undercutting others, driving down prices in foreign markets and reducing national earnings of foreign exchange. This could, of course, be a real problem for exports of primary commodities facing inelastic foreign demand, or even for sales of manufactures in particular markets in which foreign buyers play off different Chinese agencies against each other to negotiate lower prices. A centralized selling agency, or a system of minimum prices to be accepted by sellers of standardized products, could give better results than competing individual exporters in such condi- tions. But the reversion to greater restraint on diversified exporting might itself have costly results in the sense of limiting the ability of Chinese exporters to enter new markets and grow. 2.32 In world markets for manufactured goods, much of China's possibili- ties for successful new entry depend on taking advantage of its low costs of production for labor-intensive goods. Potential exporters must be allowed to undercut existing prices unless they can offer clearly superior quality. Even if the elasticity of demand for a product in world markets is relatively low, this is not the key question: the question is the ability of the Chinese seller to compete with many other exporters for shares of markets. Many empirical tests have shown high substitution among export suppliers in response to moderate price differentials. If the authority to export manufactured goods is highly centralized, that could greatly slow the development of new industrial exports through successful price cutting. A better alternative to avoid mutually harmful price competition would be to limit the exporter's earnings to conversion at the relevant exchange rate and not allow any sales at a loss. That limitation would discourage any price - 59 - cutting actually harmful to Chinese income. To encourage many exporters, subject to the requirements that none make sales at actual losses, would be more favorable to the growth of national income and to the process of learning from participation in external markets. 3. CONFLICTS AND OPTIONS 3.01 China's strategy of double-layer insulation--of the country from the outside world and of internal regions from free movement of goods and people among each other--is clearly being changed toward one allowing greater inter- action. This process should lead toward more flexibility and efficiency in production, increased interchange of ideas and knowledge of new technological possibilities, more rapidly rising national income, and enhanced freedom of personal choice of location and kind of work. But it could also involve a threat of increased poverty and reduced opportunities in the more backward regions, of greater unemployment as pressures to reduce costs are heightened, of a stimulus to "consumerism" for a minority, and, in general, of change away from a social order which has served--despite all the pains and costs involved --to reverse the country's long descent into weakness and poverty. 3.02 At the risk of oversimplifying these complex issues, it might be useful to divide them into two kinds of tensions: (1) emphasis on rapid modernization through more active and decentralized participation in world trade, as opposed to retention of controls, and (2) emphasis on increasing efficiency through greater use of internal comparative advantage, with prices and production more closely related to the structure of external prices, as opposed to the use of differential prices and controls to alleviate poverty and to foster development of the least-industrialized regions. - 60 - External Trade and Finance: Incentives and Controls 3.03 Increasing industrial exports and increasing imports of raw materials and equipment for production will both be necessary to achieve a significant rise in per capita income between now and the end of the century. Failure to continue the growth of exports would put a serious brake on the possibilities of economic growth. But there are a wide range of possibilities for different degrees of involvement in world trade, for different mixtures of financial incentives and controls, for the treatment of imports, and for policies with respect to foreign borrowing and foreign direct investment. 3.04 As a very general guide to expectations for the years ahead, several Chinese officials stated that foreign trade would quadruple in the period between now and the year 2000, in line with the expected quadrupling in national income. How appropriate is such a target? Is there a need to stipulate any particular relationship between exports and output, or imports and income? China has been subject to so many direct restraints on trade and is behind modern technology in so many areas that its income could surely be raised more rapidly in the near future if it reduced these constraints and increased the ratio of trade to income) Then, after a period of catching up, the ratio might reasonably be expected to decline, consistent with efficiency and continued rapid growth. There is no magic overall ratio which is just right: the most efficient ratio at any given time depends on the differences in costs and opportunities open at that time, and they are always changing. 3.05 Two kinds of problems may be involved in any general decision to reduce restraints and participate more actively in world trade. One is that imports may go up much faster than exports can be raised, and the other is that imports may knock out domestic lines of production which are worth - 61 - keeping alive for the future. It is not necessary to rule out an import sur- plus entirely: it can provide the resources needed for a higher standard of living with a given rate of investment, or a faster rate of investment for any given standard of living, or a combination of both together.5/ It is certainly desirable to keep the service costs of external debt from rising too rapidly relative to export earnings, but China is so far from any such diffi- culty that a target of holding imports below exports at present seems almost pointless. Or worse than pointless: an export surplus by a low-income country means that it is holding down its own standard of living while serving to raise investment and consumption in richer countries. 3.06 Still, while China could readily relax restraints on imports in order to raise consumption and investment, it surely should maintain active export promotion. The point is not so much to pay for imports as to accele- rate industrial growth and learning through greater participation in world markets (Yuan and Wang 1983). But the possibilities of success and likely costs depend on exactly what techniques are used and how hard they are pushed. It is possible to get into trouble with export promotion, as with everything else. 3.07 The present methods of export promotion include one clearly desir- able element: the use of an exchange rate for conversion of foreign exchange earnings which is high enough to keep industrial exports growing. How rapidly should exports be expected to grow? A possible guide would be that their rate of growth should outpace that of world industrial exports. They must, if 5/ Quantitative relationships among investment rates, growth of productive capacity and national income, and growth of consumption are examined carefully in World Bank 1984. - 62 - Chinese economic growth is to come anywhere near the stated targets. If they are not rising faster than world trade in manufactures, then the value of foreign exchange is too low and should be raised. 3.08 It might be suggested that the price of foreign exchange is already too high, both because of the country's current account surplus and because problems have arisen of excessive competition on the part of some Chinese exporters. As far as the current account is concerned, the surplus of the last few years can scarcely be considered a demonstration that the currency is undervalued. Imports are restrained both by a multitude of controls and by tariffs. All Chinese officials asked about the matter seem to agree, and are surely right, that a policy of open access to imports at an exchange rate of Y 2.8 per dollar would result in a dramatic rise of imports. But exports would also rise dramatically if more exporters could benefit directly from the 2.8 rate. It is, therefore, not easy to tell whether the currency is under- valued or overvalued. There is considerable room to relax import restraints now, within an existing export surplus and then beyond that, within reasonable limits of external borrowing. Exports of manufactured goods have been rising well, at least as measured against actual world market conditions. So the 2.8 rate seems reasonable enough at present, but it would be a shame to hold it down to this level if the growth of industrial exports begins to fall below the rate of growth of world trade in such products in the future. 3.09 The possibility that present export incentives are too strong has also been suggested by the sudden retreat toward more central control of exports in March 1984. As discussed above, the problem may well be related to excess incentives for primary products, for which internal prices are held down while they are pulled toward exports because the latter give favored - 63 - access to highly desired allocations of scarce production materials, housing, and import rights. For primary products, it may be desirable to cut down on such direct pressures to export, or--as the government has recently done-- centralize marketing for more of them. For industrial products, the problems are different, the goal of raising exports is much more promising for the long run, and allowing decentralized contact with external markets is extremely important. For them, a return to restrictive central control would seem to be dangerously adverse for China's growth prospects. 3.10 It is probably best to do everything possible to develop increasing exports of all kinds of industrial products which can be sold in external markets, over the hurdle of restrictions imposed by other countries. The ability of Chinese firms to keep on raising sales of labor-intensive clothing, textiles, and other basic consumer goods is very high, but is bound to be limited by the import restrictions of the industrialized countries. China could probably take over these markets entirely if allowed to do so, but it will not be. For consumer goods, the more hopeful direction is to increase exports of higher-valued, or income-elastic, products for which markets in the wealthier countries have grown very rapidly. But perhaps the most construc- tive direction of all would be to improve the variety and quality of exports of machinery and equipment to developing countries, where Chinese equipment may have the special advantage of being designed for use with more labor- intensive characteristics than that normally exported by more industrialized countries. Capital goods exports to the developing countries could provide a valuable complement of alternative and more appropriate technology for the buying countries, as well as an excellent growth possibility for China, but it will undoubtedly require major improvements in the quality of capital goods production. - 64 - 3.11 Both of the promising components of industrial exports will require more active contact by Chinese firms with foreign markets, and will certainly generate pressure by them for increased freedom of access to raw materials and high-quality equipment. They are also likely to exert pressure for freedom to borrow abroad to finance such imported inputs, and for freedom to switch away from Chinese suppliers whose products do not match the qualities available in external markets. Such pressures are to be expected precisely because increased flexibility and more dependable input quality will be necessary for them to compete effectively in external markets. If the process is to succeed, it means that detailed control of the enterprise and of foreign trade will need to be relaxed. That process got well under way in 1981, as part of a general program of enterprise-level reform providing stronger incentives for efficiency and allowing somewhat greater enterprise autonomy. It has led to improvements in efficiency, but also to problems of keeping enough control to avoid wasteful duplication of effort and imbalance between total investment by individual enterprises and central government investment in infrastructure (Byrd 1983). 3.12 Both exports of income-elastic consumer goods aimed at markets in the most industrialized countries and potential imports of similar goods in China raise problems of social goals. Such exports have been a dynamic compo- nent of growth for Japan and other East Asian countries, and could be espe- cially so for China because of its abundant labor supply and present advantage of much lower labor costs (Keesing 1983). Such goods would also find a domestic market and would act as an incentive to harder work. But at least for the immediate future, only a few consumers would be able to afford to buy them. It might, therefore, be desirable to limit such lines of production - 65 - exclusively to export markets rather than allow sales on internal markets, to avoid any near-term stimulation of domestic demand toward such products and away from traditional consumer goods. 3.13 It would clearly be necessary to permit greater freedom for imported industrial inputs. This need not imply taking away controls of imported consumer goods. But a sharp dichotomy between treatment of the two groups of imports has often been a source of distortion in developing countries: local firms may introduce production of substitutes for blocked foreign consumer goods, even at very high costs in terms of required inputs of machinery and current inputs. China may not have this problem to any great degree because of the relative equality of its income distribution. It should not readily be subject to the expensive foreign-oriented tastes of Latin American middle- income consumers. But some such effect could become a problem: Chinese consumers, like Japanese, could increase their expectations if encouraged to do so. The Japanese did it slowly, with restricted presence of foreign firms promoting foreign styles of consumption. China could do the same, keeping the internal market separate from the world market for many lines of production, at the same time as exports are promoted: a type 2 strategy and not a type 1 open economy. This partial separation would certainly have costs including reduced pressures on domestic consumer goods producers to cut costs and improve quality; it would not be a long-run ideal. But it could be a way of moderating the strains likely in the decade ahead. Foreign Investment 3.14 China's long isolation from foreign investment and relatively low use of external credit probably held down the rate of acquisition of technical - 66 - knowledge and of economic growth, but it had the offsetting values of an unusual degree of self-dependent industrialization and freedom-from pressure by external creditors. In its own way, the Chinese experience had some of the effects of Japan's period of early industrialization closed to foreign inves- tors. But a major difference is that Japan used this period to acquire, for its own use, as much knowledge as possible of advancing technologies in other countries, while China did not. The present Chinese move to reach outward and promote faster economic growth is starting from a position much further out of touch with external technology and markets. China has a greater need to consider the possible gains of cooperation with foreign firms at the same time as it must consider the desirability of keeping alive most of its widely diversified industrial enterprises. 3.15 Three quite different approaches to foreign investment by three countries which have achieved above-average economic growth and success in exporting were discussed above: the Japanese model of almost complete exclu- sion of foreign ownership, the Korean practice of allowing limited foreign investment while stipulating export requirements for most foreign firms, and the Brazilian approach of wide-open welcome for many years, followed by slowly increasing reservations after foreign firms came to dominate the more modern industries. All of these approaches proved to be consistent with rapid modernization. That of Japan takes a longer period of preparation and a most unusual degree of government and business cooperation. It should not be impossible for China but it implies somewhat slower modernization than would otherwise be possible. The other end of the spectrum, the Brazilian approach, works more rapidly but involves a high share of foreign influence in the modern industrial sector and may involve retarding effects on domestic entre- - 67 - preneurship as well. That leaves Korea as an example of a possible in-between path, which could be pulled toward either of the two more extreme alternatives depending on Chinese preferences for either (a) the highest possible degree of national control and domestic learning, or (b) the fastest possible rate of conversion to modern technology in the years immediately ahead. 3.16 The diverse cooperative arrangements with foreign firms noted earlier include many kinds of contact which could speed up the process of learning about foreign technology and markets, and speed up the development of capacity to compete in export markets, without involving significant foreign ownership. When the emphasis is placed on quick action to meet specific objectives such as offshore oil exploration, it would seem worthwhile to welcome direct participation by foreign firms, without imposing complex restraints. Similar possibilities for gain from direct foreign investment might be fairly common in the creation of new lines of production for export, particularly where measures are taken to ensure Chinese participation in positions which will facilitate learning. But allowing greatly increased foreign investment outside of well-defined specific needs is a much more doubtful matter, which could bring costly problems. 3.17 Direct foreign investment is especially problematic when it involves production for the internal market under conditions of protection from import competition. That has been the common, and in some cases the dominant, role of foreign investment in many developing countries. It may easily cost more than it is worth. The foreign firm may build up import requirements for production which would not otherwise have been necessary, and its profits may be the counterpart of little more than the arbitrary prices made possible by protection against imports. Where the foreign firm makes its earnings from - 68 - exports, these considerations are much less likely to be problems: if its earnings come from outside the country, they cannot be falsified by domestic protection, and they must reflect an excess of foreign exchange earned over that spent on inputs. 3.18 For countries in weak bargaining positions, or with indecisive governments, there is always the danger that the foreign firm can exert pres- sure to distort policy decisions in its favor on such issues as taxation and protection. Another problem which has been common in Brazil and Mexico is that foreign firms may initiate or expand their role by buying control of the best local firms, downgrading domestic entrepreneurs to secondary management levels and reducing the scope for independent growth. China would seem to be relatively safe from such problems: its bargaining power is high, it should not need to give excessive concessions to attract foreign investors, and it should not be subject to takeovers of independent domestic firms. The varied methods already in use to improve access to new technology without direct foreign ownership provide promising options which should serve to minimize difficulties in dealing with foreign investment. Low-Income Regions 3.19 The emphasis of national policy prior to 1977 toward local self- sufficiency and dispersed industrial production fostered a great many high- cost activities which reduced possible income for the country as a whole. Recent changes favoring increased exchange among regions and with the outside world have begun to shift the balance toward concentration of production on lines of comparative advantage. This has helped considerably to raise national income, but it works unequally among regions. Those regions which - 69 - are best placed for international trade and those which have the more advanced industrial bases are the ones which gain most from this kind of change. Poorer regions, with less efficient industry in the first place, may find their industrial sector forced to close down many activities. The diversity of opportunities for learning and employment may be reduced precisely where poverty is worst. 3.20 The least-industrialized regions within China are in much the same position as the least-industrialized countries in the world economy. Their industrial activities are not protected by tariffs and quotas, but they have been protected by three major factors: (a) most inputs are in scarce supply and have been assigned through the national planning process to firms in all areas, wherever production has been established; (b) many final products have been allocated from producers to distributors, with only minimal scope for product choice or for negotiation over prices; (c) the internal transport system, especially the road system, has been so little developed--consistent with the prior policy of local self-sufficiency--that this too has restricted the possibilities of moving goods among regions. 3.21 At least two of these barriers to internal trade are being pro- gressively relaxed. The number of basic commodities under full state control for allocation to producers has been reduced, and changes in allocation policies have been introduced to cut down allocations to higher-cost producers. Retail trade has been greatly loosened up by the rapidly increas- ing market shares of retailers outside the system of state stores and by increasing flexibility for the state stores themselves to order from varied sources. The third barrier, the limits imposed by the transport system, will be a problem for a long time to come. The net result is not likely to be any - 70 - sudden elimination of higher-cost firms in backward areas, but it could easily be one of increasing inequality and decreasing potential for growth in the least-favored regions. 3.22 More active internal exchange is bound to be favorable for Chinese economic growth, but the options for policy do not need to be limited to the extremes of "cellular" development versus completely open models. It should be possible to combine some of the advantages of both, though any such combi- nation would also include some of the costs of both. As a possible guide, China could consider the parallel of the European Economic Community in which each country wanted to gain the advantages of greater trade but not to lose its industrial base (see paras. 1.43-1.48). All produced similar lines of products for national markets in which incomes differed, and all feared the incursion of competition from similar industries in the other countries. They chose to open up free trade among themselves in industrial products, while keeping controls on agricultural trade and on imports from countries outside the EEC. That approach worked out well in that every country expanded its production of some lines of industrial goods for export, without destroying any major industries in any of the countries. 3.23 The reasons the EEC did not set back industry in any country involve the whole underlying complex of national investment and growth patterns, but there is one central economic variable which was essential: if any country began to find its products collectively becoming less competitive and losing market shares, the country was able to devalue its currency and make its industries more competitive. Italy and France have devalued fairly often relative to Germany and Holland. That has helped France to maintain a some- what higher growth rate than Germany and has helped Italy, despite its much - 71 - lower initial level of industrialization, to keep up its industrial growth in competition with all the others. Real wage levels among the countries continue to differ, but open trade in manufactures has benefited the least industrialized as well as the most. 3.24 China's least-industrialized regions cannot use currency differences to maintain a balance in trade. Is it possible to consider some substitute mechanism, such as imposing taxes on industrial goods brought into the region or providing subsidies to enterprises in the region? If taxes or subsidies were set at 10 percent of final product value for a lower-income region, this would mean that all the particular industrial activities with costs more than 10 percent above outside levels would be forced to contract. It would also mean that any activity with costs less than 10 percent above the outside levels would be able to expand and even to export to other regions. Present differentials in costs throughout the country could be reduced, but not elimi- nated. This technique (suggested by Little, Scitovsky, and Scott 1970) would provide an option consistent both with reduction of current cost differentials and with retention of diversified industrial activity and employment in all regions. 3.25 There would be little point in any such guidance of internal trade patterns if China's labor market conditions were anywhere near full employ- ment, in the sense of alternatives for productive employment for everyone in all regions. It would also be less necessary if people could move freely to places where labor productivity and wages are higher than average. Without these conditions, the central problem is much more acute and the desirability of some restraint on the likely backwash effects of increased trade is greater. That does not argue in favor of keeping barriers against internal - 72 - trade without attention to costs. It argues rather in favor of reducing the barriers but trying to limit the human costs of doing so. Relative Prices: Efficiency and Social Goals 3.26 The structure of prices has been shaped by three kinds of pressures which do not fit well with each other: (1) relationships established deliberately for specific social purposes; (2) inconsistencies in the sense of widely differing prices for similar products according to where they are produced, or arbitrarily low, controlled prices for particular raw materials maintained more out of reluctance to consider changes than for any clearly established purposes; and (3) newly active market forces which have developed as trade has become somewhat more open, rural households have been allowed to produce and market more freely on their own, and industrial enterprises have been freed (or put under pressure) to find their own markets for at least some of their output. The growing element of market forces should, in general, lead in the direction of raising efficiency and national income by inducing a shift of productive resources toward higher-value output, and by encouraging greater economy in the use of scarce inputs. But this process may lead to costly mistakes if the structure of costs is distorted by holding rigidly to long-established prices for basic materials, regardless of changes in relative scarcities. In addition to these problems of costs, market forces could run into direct conflict with the use of prices set for social purposes. To what extent should the use of differential prices for social goals be sacrificed for the sake of greater efficiency? What policies can combine the two kinds of objective at least cost? - 73 - 3.27 From the point of view of maximizing efficiency and current national income, the structure of prices of tradeable goods should match the structure of prices at which they could be exported or imported. If the internal ratio of prices between two products differs from the external ratio, there is at least a strong likelihood of a two-sided loss: (a) producers will devote too many resources to the production of the relatively high-priced good and will use more domestic resources to produce it than would be necessary to produce the alternative exports needed to pay for its import; and (b) production of the relatively low-priced good will be less than it would otherwise have been, stopping even though the marginal cost of production remains below the vaLue which could be earned by exporting it. Marked misalignment between the struc- ture of domestic prices and the structure of external prices implies missed opportunities for increasing national income. 3.28 Chinese officials discussing this question sometimes express exactly the contrary preference: domestic prices should be guided by national goals and insulated from the meaningless variations of external prices. That posi- tion is highly understandable from several angles, but may cost the country a good deal if not modified. It is true that external prices lack direct meaning in terms of social goals, but they present China with specific oppor- tunities, both to raise earnings by producing more goods of high outside value and to reduce costs by importing inexpensive or high-quality inputs to produc- tion. If Chinese producers must pay higher prices for inputs than other producers do (or if they cannot buy them at all), then the whole related set of costs in China will remain much higher than it could otherwise have been. It is the prices to producers which matter in this respect and if they are misaligned, it is the nation's production which suffers. But the objectives - 74 - of the Chinese officials who take the opposite view are readily defensible if taken in either of two carefully defined ways: (1) if they are applied to prices charged to consumers, as distinct from those received by producers to cover costs of production, and (2) if any application to producer prices is limited to clearly defined objectives such as providing extra incentives for entering new fields with special long-term promise, or undertaking new lines of research activity, or similar investments in reasonable long-range changes which might be impeded by current market conditions. Decisions on specific exceptions of type 2 must be matters of analysis of the particular case. It is the issue of type 1 pricing for social purposes which seems of special interest for the character of the Chinese economic system. 3.29 Perhaps the most important examples of price controls used for social purposes have been those keeping food, coal and other basic consumer goods relatively cheap compared to manufactured consumer goods. Holding down food prices relative to manufactured goods is a common pattern in many developing countries and it has common negative results: it discriminates- against rural incomes, aggravates rural poverty, and discourages agricultural production. China partly overcame the negative effects on production by reliance on collective effort combined with work-points linking individual incomes to individual effort. But when the reforms of 1979-81 added greater scope for family earnings by raising animals for sale and by cultivation of family plots, the result proved to be a rapid and significant increase in food production, demonstrating the prior waste of potential (Wiens 1983). With respect to incomes, the prior system had held the growth of rural income to not much more than half the rate of urban income, at 1.6 percent a year for 1957-79 compared to 2.9 percent for urban incomes (World Bank 1983a, pp. 87 - 75 - and 276-278). As of 1979, urban per capita incomes were about 2.2 times as high as rural (ibid, p. 86). The increased scope for independent earnings by rural families introduced in the late 1970s, combined with the dramatic increases in prices paid to food producers in 1979 and 1981, should lessen this imbalance. At the same time, the state shielded urban workers from the impact of higher food prices by using subsidies instead of higher market prices, which left relative internal prices to consumers far out of line with, relative costs and with external prices. 3.30 Does the objective of increased efficiency in a system with more open trade require that such subsidies be eliminated, that internal prices of food and other basic consumer goods be lined up with those in the outside world? Not necessarily. It does require that relative prices to producers be lined up reasonably well with relative external prices, and changes in exactly this sense have been implemented as part of the agriculture reforms instituted since 1979 (Wiens 1983). At the level of prices to urban consumers, subsidies remain in effect and are a significant part of the national budget. That need not be seen as a handicap to the economy, given the capacity to provide tax revenue to limit budget deficits. It becomes a question of equality and inequality rather than one of productive inefficiency. Still, it would be worthwhile to consider the alternative method introduced in Sri Lanka for rice supplies, discussed in para. 1.55. Instead of cheap rice for everyone, the government provided rationed quantities of free rice for the lower half of the income distribution, and allowed the market price to move freely so that higher-income groups (and those poorer people who chose to buy more than the ration) had to pay the open market price. The way the technique was applied gradually became adverse to the poor because rations were set in terms of - 76 - nominal values rather than quantities; their real value went down rapidly. That result was not inherent in the technique itself, which could be helpful to the poor as well as efficient. 3.31 At the other end of the range of consumer goods--the television sets and such that can be purchased only by a minority for the time being-- relatively high arbitrary prices make a good deal of sense as a form of progressive taxation. That could create problems if similarly high prices were paid to producers because it might pull too many productive resources into lines of production with relatively high costs. Again, concern for restraint of inequality might well support separately established domestic prices to consumers, but concern for efficiency would call rather for con- straining prices to producers to fit the external price structure. 3.32 Social-purpose control is also actively used in the particularly sensitive field of wage determination. Restraints on wages are used both to keep them higher than they otherwise would be in the state firms in regions with surplus labor, and to keep down the differentials among skill levels and occupations. In a context of increasing emphasis on profit-seeking by enter- prises, the classic danger is that firms will respond by cutting down employ- ment to the extent they are allowed to do so, and may substitute machinery and other inputs for labor even in areas in which there is a great deal of under- employment. Keeping real wages equal everywhere could block exports of labor- intensive products from labor-surplus areas. That causes an especially diffi- cult conflict between the goal of equal wages for similar work and the goal of stimulating employment. A partial response might be the system suggested above for use of subsidies to reduce labor costs in the lower-income regions. They could allow labor-intensive lines of production to become more competi- tive while at the same time limiting inequality. - 7 7 - 3.33 Restraint on the range of wage differentials among skills and occupations raises still another thorny problem of conflicting objectives. To keep the range narrow improves equality but limits incentives to acquire skills and may lead to a waste of high-skill people doing work in one enter- prise which does not need their special abilities while other enterprises need and cannot find enough such workers. As in so many other dimensions, the question is the balance to be sought between turning the wage structure loose to be shaped by market forces and maintaining limits to prevent widening differentials based on local advantages of position and bargaining power. In all countries, such differentials are, to a high degree, determined by custom and market power, with strong mixtures of control by professional associations and individual labor organizations. It should be possible to open up much greater freedom for personal mobility and decentralized wage setting, without entirely giving up the use of limits on the range of differentials to be allowed among occupations and skill levels. 3.34 All these suggestions about use of restraints on prices to limit or guide the impact of market forces involve some possible costs of reduced effi- ciency, administrative expenses, and distortions by judgment subject to bias. They could conflict with a dominant goal of raising efficiency and national. income as rapidly as possible. They need not prevent very significant increases in efficiency and income, but that depends to a considerable degree on exactly what methods are used. Questions of balance among conflicting objectives are matters of Chinese preferences; questions of-ehe-methods Lused are more nearly matters of logical consistency and of learning from the experiences of others. The conflicts are real, but care in the choice-of, techniques, exercised in a country with such a high degree of cohesion, could - 78 - in principle reconcile a high rate of growth with measures to lessen poverty and to ensure wide participation in new opportunities. - 79 - BIBLIOGRAPHY Baer, Werner and Gillis, Malcolm. 1981. Export Diversification and the New Protectionism. Urbana: University of Illinois for the National Bureau of Economic Research. Balassa, Bela, ed. 1975. European Economic Integration. Amsterdam: North Holland. Balassa, Bela. 1978. "Exports and Economic Growth: Fresh Evidence," Journal of Development Economics 5, pp. 181-189. Balassa, Bela. 1981a The Newly Industrializing Countries in the World Economy. New York: Pergamon Books. Balassa, Bela. 1981b. "Policy Responses to External Shocks in Selected Latin-American Countries," pp. 131-164 in Baer and Gillis 1981. Balassa, Bela. 1982. Development Strategies in Semi-Industrialized Economies. Baltimore: Johns Hopkins for the World Bank. Bhagwati, Jagish N. 1978. Anatomy and Consequences of Exchange Control Regimes. Cambridge, Mass: Ballinger. Byrd, William. 1983. "Enterprise-Level Reforms in Chinese State-Owned Industry," American Economic Review, vol. 73, no. 2, pp. 329-332. Carvalho, Jose and Claudio L. S. Haddad. 1982. "Foreign Trade Strategies and Employment Implications in Brazil," in Krueger 1982, pp. 29-81. Chen Jiyuan. 1983. "Evolution of Approaches to China's Regional Economic Structures." Cortazar, Rene. 1980. "Distribuci6n del ingreso, empleo y remuneraciones reales en Chile, 1970-78," Collecci6n estudios cieplan no. 3, junio 1980, pp. 5-24. Felix, David. 1968. "The Dilemmas of Import Substitution - Argentina," in Gustav Papenek 1968. Felix, David. 1983. "Income Distribution and the Quality of Life in Latin America: Patterns, Trends, and Policy Implications," Latin American Reseach Review, 1983, No. 2, Vol. XVII, pp. 3-33. Ffrench-Davis, Ricardo. 1973. Politicas Econ6micas en Chile, 1952-1970. Santiago: Ediciones Nueva Universidad. Ffrench-Davis, Ricardo. 1982. "El experimento monetarista en Chila: una sintesis critica," Colleccion estudios cieplan, no. 9, diciembre de 1982, pp. 5-40. - 80 - Foxley, Alejandro. 1982. "Experimentos neoliberales en america latina," Collecci6n estudios cieplan, no. 7, marzo 1982. Frank, Charles R. Jr., Kwang Suk Kim, and Larry E. Westphal. 1975. Foreign Trade Regimes and Economic Development: South Korea. New York: National Bureau of Economic Research. Gereffi, Gary and Peter Evans. 1981. "Transnational Corporations, Dependent Development and State Policy in the Semi-Periphery: a Comparison of Brazil and Mexico," Latin American Research Review XVI-3, pp. 31-64. Heller, Peter S. and Porter, Richard C. 1978. "Exports and Growth: an Empirical Re-Investigation," Journal of Development Economics 5, pp. 191-193. Hickok, Susan and Rosanna Arguelles. 1983. "China's Rapid Trade Growth and Impact on the World Economy," Federal Reserve Bank of New York, Quarterly Review, Winter 1982-83, vol. 7, no. 4, pp. 41-51. Hirschman, Albert 0. 1958. The Strategy of Economic Development. New Haven: Yale University Press. International Monetary Fund. 1983. Statistical Yearbook 1983. Isenman, Paul. 1980. "Basic Needs: The Case of Sri Lanka," World Development, vol. 8, pp. 237-258. Johnson, Chalmers. 1982. MITI and the Japanese Miracle. Stanford, Calif.: Stanford University Press. Jones, Leroy P. and Il Sakong. 1980. Government, Business and Entrepreneur- ship in Economic Development: the Korean Case. Cambridge, Mass.: Harvard University Press. Keesing, Donald. 1979. "Trade Policy for Developing Countries." World Bank Staff Working Paper No. 353. Keesing, Donald. 1981. "Exports and Policy in Latin-American Countries: Prospects for the World Economy and for Latin-American Exports, 1980-90" in Baer and Gillis 1981. Keesing, Donald. 1983. "Linking up to Distant Markets: South to North Exports of Manufactured Consumer Goods," American Economic Review, vol. 73, no. 2 (May), pp. 338-342. Knight, Peter T. 1981. "Brazilian Socio-Economic Development: Issues for the Eighties," World Development, Vol. 9, No. 11-12, pp. 1063-1082. Knight, Peter T. 1983. "Economic Reform in Socialist Countries." World Bank Staff Working Paper No. 579. - 81 - Krueger, Anne. 1978. Liberalization Attempts and Consequences. Cambridge, Mass: Ballinger. Krueger, Anne, ed. 1982. Trade and Employment in Developing Countries. Chicago: Chicago University Press. Little, Ian, Tibor Scitovsky, and Maurice Scott. 1970. Industry and Trade in Some Developing Countries. Oxford University Press for the OECD. Lyons, Thomas Patrick. 1983. "Economic Integration and Development in China, 1957-71." PhD thesis, Cornell University. Mendonca de Barros, Roberto Jose and Graham, Douglas. 1978. "The Brazilian Economic Miracle Revisted: Private and Public Sector Initiative in a Market Economy," Latin American Research Review XIII-2, pp. 5-38. Michaely, Michael. 1977. "Exports and Growth: an Empirical Investigation," Journal of Development Economics 4, pp. 49-53. Michaely, Michael. 1979. "Exports and Growth, a Reply," Journal of Development Economics 6, pp. 141-163. Morawetz, David. 1980. "Economic Lessons from Some Small Socialist Developing Countries," World Development, vol. 8, pp. 337-369. Mufnoz, Oscar. 1982. "Crecimiento y desequilibrios en una economia abierta: el case chileno, 1976-81," Colleccion estudios cieplan, no. 8, julio, pp. 19-41. Myrdal, Gunnar. 1957. Economic Theory and Under-Developed Regions. London: Duckworth and Co. Nolan, Sean. 1983. "Transnational Enterprises and Manufactured Exports from Industrializing Counties: Theoretical and Empirical Analysis." PhD thesis, Yale (in manuscript). Nurkse, Ragnar. 1955. Problems of Capital Formation in Underdeveloped Countries. New York: Oxford University Press. Papanek, Gustav, ed. 1968. Development Policy, Theory and Practice. Cambridge, Mass: Harvard University Press. Pfefferman, Guy and Richard Webb. 1979. "The Distribution of Income in Brazil" World Bank Staff Working Paper No. 356. PREALC (Programa Regional de Empleo para America Latina y el Caribe). 1982. Mercado de Trabajo en Cifras, 1950-1980, Santiago: Oficina Interna- cional de Trabajo. Ranis, Gustav. 1981. "Challenges and Opportunities Posed by Asia's Super- exporters: Implications for Manufactured Exports from Latin America," pp. 204-30 in Baer and Gillis 1981. - 82 - Riedel, James. 1983. "Trade as the Engine of Growth in Developing Countries, A Reappraisal." World Bank Staff Working Paper No. 555. Sekhar, A. Uday. 1983. "Industrial Location Policy, The Indian Experience." World Bank Staff Working Paper No. 620. Sen, Amartya. 1981. Poverty and Famines: an Essay on Entitlement and Dep- rivation. New York and Oxford: Oxford University Press. Sheahan, John. 1963. Promotion and Control of Industry in Postwar France. Cambridge, Mass: Harvard University Press. Shinohara, Miyohei, Toru Yanagihara, and Kwang Suk Kim. 1983. "The Japanese and Korean Experiences in Managing Development." World Bank Staff Working Paper No. 574. State Statistical Bureau, People's Republic of China. Statistical Yearbook of China 1983. Hong Kong: Economic Information and Agency. Stewart, Frances. 1972. "Choices of Technology in Developing Countries," Journal of Development Studies (October), pp. 99-122. StUhr, Walter and Franz T8dtling. 1977. "Spatial Equity - Some Anti- Theses to Current Regional Development Doctrine," Sixteenth European Congress of the Regional Science Association, pp. 33-53. Tan, Augustine, and Ow Chin Hock. 1982. Chapter 9, "Singapore" in Balassa, 1982, pp. 280-309. Thoumi, Francisco. 1982. "International Trade Strategies, Employment, and Income Distribution in Colombia," in Krueger 1982, pp. 135-179. Tyson, Laura and John Zysman, eds. 1983. American Industry in International Competition. Ithaca, New York: Cornell University Press. Urrutia, Miguel. 1985. Winners and Losers in Colombia's Growth in the 1970s. Oxford: Oxford University Press. Vernon, Raymond. 1977. Storm Over the Multinationals: the Real Issues. Cambridge, Mass: Harvard University Press. Westphal, Larry, E., Yung W. Rhee, and Garry Pursell. 1981. "Korean Industrial Competence: Where it Came From." World Bank Staff Working Paper No. 469. Wiens, Thomas B. 1983. "Chinese Economic Reforms: Price Adjustment, the Responsibility System, and Agricultural Productivity," American Economic Review, vol. 73, No. 2 (May), pp. 319-324. Williamson, Jeffrey D. 1965. "Regional Inequality and the Process of National Development: A Description of the Patterns," Economic Development and Cultural Change, Vol. 13, No. 4, pp. 3-45. - 83 - World Bank. 1979. Peru: Long-Term Development Issues. Washington: World Bank, 3 volumes. World Bank. 1983a. China: Socialist Economic Development. Washington: World Bank, 3 volumes. World Bank. 1983b. "China: Recent Economic Trends and Policy Developments," Report No. 4072-CHA, March 31. World Bank. 1983c. World Development Report 1983. Washington: World Bank. World Bank. 1983d. Brazil, An Interim Assessment of Rural Development Programs for the Northeast. Washington: World Bank. World Bank. 1983e. "Indonesia: Selected Issues in Spatial Development," Report 4776-IND. World Bank. 1984. "Possibilities and Options for Growth and Structural Change in China: a Preliminary Overview in Light of International Experience," discussion paper, Feb. 1984. Yuan Wenqi and Wang Jianmin. 1983. "We Must Review and Reevaluate the Role of Foreign Trade in the Development of the National Economy," Chinese Economic Studies, Vol. XVI, No. 3, Spring 1983, pp. 24-39. HG 3881.5 .W57 W67 NO.759 SHEAHAN, JOHN, 1923- iALTERNATIVE INTERNATIONAL ECONOMIC STRATEGIES AND . . A The World Bank Headquarters European Office Tokyo Office 1818 H Street, N.W. 66, avenue d'lena Kokusai Building Washington, D.C. 20433, U.S.A. 75116 Paris, France 1-1 Marunouchi 3-chome Telephone: (202) 477-1234 Telephone: (1) 47.23.54.21 Chiyoda-ku, Tokyo 100, Japan Telex: WUI 64145 WORLDBANK Telex: 842-620628 Telephone: (03) 214-5001 RCA 248423 WORLDBK Telex: 781-26838 Cable Address: INTBAFRAD WASHINGTONDC ISSN 0253-2115/ISBN 0-8213-0709-6