Executive summary

0.1 Rising incomes, growing health consciousness, and urbanization are among the major factors contributing to changing dietary patterns, both in industrialized and developing countries. While there are differences in the pace and specific features of these dietary changes, there has been a common shift toward increased consumption of fresh and processed fruits and vegetables, of protein-rich meats, fish, dairy products, and vegetable oils, and of prepared 'convenience' foods. Compared with cereals and other staple food products, these foods have relatively high unit values and face relatively high income elasticities of demand.

0.2 Changing dietary patterns, together with technical advances in food-related logistics and multilateral reductions in trade barriers, have contributed to a rapid expansion in world trade in the above noted highvalue foods (HVF) over the past quarter century. During the 1980s, a period in which the aggregate value of world trade in cereals, sugar, and tropical beverage crops actually declined, the above categories of HVF experienced average annual growth rates in world trade ranging from 4% (e.g. fresh vegetables; fresh meat) to more than 11% (e.g. dairy products; shell fish). World trade in HVF is now considerable. In 1988/89, it totalled $144 billion, representing 5% of total world commodity trade and a value equivalent to world trade in crude petroleum.

0.3 Developing countries have actively participated in this international HVF trade, both as importers and exporters. In 1990, exports of HVF by middle- and low-income countries totalled $52.5 billion. For comparison, the aggregate exports of such countries for coffee, cocoa, tea, sugar, cotton, and tobacco was only $26.3 billion-- roughly one-half. Over the past decade, developing countries have outperformed industrialized countries in export growth for several categories of HVF (e.g. fresh/processed fish, fresh/processed fruit, oilseeds, and feedstuffs). Despite these patterns, most analyses of developing country agro-industrial experience and agricultural trade problems and prospects continue to focus on the traditional beverage and industrial crops.

0.4 In 1990, there were twenty-four middle- and low-income countries whose HFV exports exceeded $500 million. The majority of such countries are either Latin American (10) or Asian (8), with most of these countries also being among the leading developing country exporters of manufactured goods. Only a few African and Middle Eastern countries have developed significant levels of HFV exports. Two Eastern European countries—Hungary and Poland-- are among the leaders as a result of their significant exports of meat products and fresh and processed fruits and vegetables.

0.5 Although many developing countries have developed some exports of HVF, there is a relatively high level of concentration of this trade among a few countries. For example, only four countries--Brazil, Argentina, China, and Thailand-- account for 40% of the total HVF exports for all middle- and low-income countries. Only ten countries account for two-thirds of the total trade and twenty countries for 90%. Major cases of success in developing country HVF exporting are thus less widespread than might be indicated by aggregated data for export levels and growth.

0.6 This paper provides a synthesis of notable "success stories" of demand-driven production, processing, and marketing of HVF among developing countries. It examines in comparative perspective the development, organization, and performance of entire commodity systems rather than the experience of individual projects or companies. It focuses on cross-cutting issues and common patterns rather than elaborating on the microeconomic and historical details of individual cases. By identifying common technical, institutional, policy, and other factors which have contributed to commodity system development and international competitiveness, the paper seeks to contribute to the design of improved strategies for supporting food market development and export diversification in developing countries and in the formerly centrally-planned economies.

0.7 A total of fifteen commodity system "success stories" from nine countries were reviewed, drawing upon consultancy, academic, and international agency studies, trade journals, USDA country and commodity reports, and official and international agency production and trade data. Two case studies benefitted from prior field work conducted by the author. Abbreviated case studies (of 2 to 5 pages) are presented in an Appendix with the main text providing a synthesis of the operating contexts, performance patterns, and organizational features of the focal cases. Important cross-cutting issues include: the respective roles of the public and private sectors, the importance of foreign capital and technology, the institutional links between producers, processor/exporters, and foreign market distributors, and the conjuncture of market, technical, policy, and other factors which contributed to the export booms experienced in these commodity systems.

0.8 The fifteen cases covered had aggregate export earnings of $11 billion in 1988/89, representing more than 20% of total HFV exports by all middle- and low-income countries. The nine countries included in the sample together account for 50% of developing country HVF exports. The cases selected cover a range of horticultural, fish, meat, and oilseed products and include commodity systems whose 'take-off' or boom periods occurred in different decades. Although the majority of cases experienced such export booms in the 1970s and/or 1980s, several of the commodity systems pre-date this period, with two dating back to the 19th Century. The specific cases examined are:

(1) Mexican fresh tomatoes,
(2) Kenyan specialty and 'off-season' vegetables,
(3) Israeli fresh citrus,
(4-6) Chilean temperate fruit, processed tomatoes, and fish products,
(7-8) Brazilian frozen concentrated orange juice and soybean products,
(9-10) Argentine beef and soybean products,
(11-13) Thai poultry, tuna, and shrimp,
(14) Chinese shrimp, and
(15) Taiwanese high-value processed foods.

Economic and institutional issues in the marketing of high-value foods

0.9 Consistent with recent work on agro-industrial development (Austin (1992)), the paper argues that major problems in food marketing and the range of institutional, infrastructural, and technological solutions to such problems are best understood when examined from a commodity systems perspective. This perspective emphasizes that modern food marketing is demand-driven, that farm-level production and downstream marketing activities are highly interdependent, and that such activities must therefore be coordinated, whether through market, cooperative, or administrative means.

0.10 The paper identifies a series of intrinsic technical and economic characteristics of food commodities/ raw materials, food production, and marketing infrastructure and services which can lead commodity system participants (e.g. farmers, processors, traders) to experience major problems related to production and market risk, inadequate or asymmetric information, logistical bottlenecks, and high transaction costs. Among the noted intrinsic characteristics include: a) the bulkiness, perishability, and heterogeneity of food commodities/raw materials, b) the yield uncertainty, seasonality, and extended gestation periods associated with the production of many foods, and c) the public good aspects, externalities and/or economies of scale associated with many types of marketing-related infrastructure and marketing services.

0.11 The paper then examines a range of technical, institutional, and other measures which can counter the incentive, risk, transaction cost, and logistical problems raised by the intrinsic technical and economic properties of food commodities, production, etc.. Many of these counter measures are essentially market or quasi-market responses on the part of private firms and individuals; others entail government interventions which stimulate, re-direct, constrain, or supplement private activity.

0.12 Drawing from the industrial development literature, the paper also examines the issue of commodity system competitiveness and its contributing factors. From Porter (1990), it is noted that a commodity system can achieve a competitive advantage either by 1) having lower costs of production and delivery, or 2) differentiating its product(s) through its quality or through accompanied technical or marketing services. A lower cost structure allows a commodity system to underprice its competitors or obtain higher returns when international prices are at or near the competitors' costs. Product/service differentiation facilitates the attainment of premium prices and/or the ability to fill profitable niches in the market.

0.13 Five sets of factors are discussed as being determinants of commodity system competitiveness. These are: 1) the size and patterns of food demand, 2) macroeconomic and sector policies, 3) natural resources and human capital, 4) physical, technical, and social infrastructure, and 5) micro-marketing and logistical activities and the coordination of production with downstream requirements. The first and second of these factor sets determine or strongly influence the incentives for specialized food production and marketing activities. Factor sets (3) and (4) determine the capacity to respond to these incentives, while factor set (5) determines the efficiency of this response and the quality of the resultant product(s).

Synthesis of HVF Commodity Success Stories

0.14 Among the nine focal countries, only three--China, Thailand, and Brazil-- have achieved levels of agricultural performance above the norm for middle- and low-income countries over the 1965 - 1989 period. Over much of this period, each of the other focal countries has experienced rates of growth in agricultural GDP and total food production which trail aggregate growth rates for middle- and low-income countries. However, in certain commodity systems, including those examined in this paper, the focal countries have experienced high rates of production and trade growth, either spanning several decades or covering the past decade. Double-digit rates of growth in production and/or trade have been experienced in one, two, or more decades for each of our focal cases. The commodities in question have come to account for rising shares of national food exports.

0.15 Most of the focal commodity systems experienced their initial export booms during the 1970s or 1980s in response to favorable international market opportunities. Some cases featured phenomenal export growth, as in the growth of Brazil's soybean/soybean product exports from only $71 million in 1970 to $2.2 billion in 1980, and the growth in Thailand's canned tuna exports from zero in 1980 to over $500 million by the end of the decade. Large increases in world market shares were recorded in several of the focal cases. In such cases as Mexican tomatoes, Chilean temperate fruit, Thai tuna, Chinese shrimp, Brazil FCOJ, and Brazilian and Argentine soybeans/soybean products, the focal commodity systems are (among) the leaders of world trade.

0.16 In the majority of cases, the focal commodity systems have (or had) a production cost advantage over major rivals due to a combination of relatively low labor costs, low land costs, government-built or subsidized infrastructure, and/or relatively high yields. These production cost advantages have been at least partly off-set by higher processing, packaging, and/or transport costs, especially in comparison with industrialized country competitors.

0.17 In many of the focal commodity systems, either initially or more recently, firms have sought to compete in international markets by differentiating their products and/or marketing services. This has included efforts to provide especially high-quality products, to supply products matching special manufacturer requirements or consumer tastes, and/or to supply a broad range of products. In several cases (including Thai poultry and shrimp, Chilean fish products, and several areas of food processing in Taiwan (China)), firms (or an entire industry) have made a successful transition from being low-cost, low-price suppliers to suppliers of highquality, value-added products obtaining premium prices. In the case of Taiwan (China), rising labor and other costs rendered such product up-grading essential for industry survival.

0.18 Each of the focal commodity systems faced highly favorable international market conditions during their initial export boom years and for many years thereafter. Not only did income growth and changing life-styles contribute to generally increasing demand, but several of our focal commodities experienced sudden market undersupply periods ("market vacuums") lasting for several years as a result of trade embargoes or climatic, political, or other problems experienced by traditional suppliers. Furthermore, most of the commodities covered in this study are foods about which industrialized country consumers are well aware and for which market distribution channels pre-dated the arrival of the focal country firms. Hence, market development costs were far lower than they would be if an exotic product were being introduced.

0.19 Most of the focal commodity systems have also featured favorable trends in domestic market demand. In more than half of the cases, export development followed upon many years of domestic marketing experience, during which infrastructure and institutions were built up. In these and in several other cases, export booms were accompanied by a rapid growth in the domestic market which provided an outlet for blemished or local grade produce, an outlet for animal/fish parts or products which can not be profitably exported, and an overall fall-back position in case of unforeseen barriers to export. In several cases, levels of domestic per capita consumption are (among) the highest in the world.

0.20 The paper provides a brief review of the macroeconomic context in which commodity systems emerged and later developed. For only two of the focal countries--Taiwan (China) and Thailand-- has the macroeconomic environment been generally favorable for investment and export development over most of the past quarter century. While China featured strong government controls on investment and trade until the 1980s, each of the other focal countries has experienced extended periods of currency overvaluation, high inflation, and low growth and investment. Still, for many of the focal cases, export booms took place just after or parallel with the adoption of macroeconomic and trade reforms which improved incentives. The only exception to this was with Argentine soybeans/soybean products, whose initial export boom (in the early-tomid-1970s) accompanied the implementation of more stringent trade controls and higher export taxes by a new administration.

0.21 In thirteen of the fifteen focal cases, the private sector has played a dominant or exclusive role in commercial production, processing, and marketing. This private sector has generally consisted of both local and foreign/joint venture companies with the former continuing to account for the majority of trade in all but one of the cases. A dominant or major commercial role for government agencies has occurred only in the cases of Chinese shrimp and Israeli citrus. In the former, state enterprises were instrumental in the initial development of aquaculture and in the continued development of trade. In the latter, an export monopoly marketing board replaced a formerly competitive private and cooperative trade and retained its monopoly until 1991. In several other cases, government agencies have undertaken some trade, either in conjunction with domestic price support and stockholding programs or as part of government to government international deals. The magnitude of such trade has generally been quite small.

0.22 While state enterprises have played prominent commercial roles in few of the focal cases, in virtually all cases governments have played important facilitating roles. This has included the provision of social overhead and marketing infrastructure (e.g. ports, railways, roads, cold stores, auction/terminal markets), programs in agricultural and food technology research and training, factory and/or product inspection and certification, and, in about half of the cases, some form of market information scheme. In many cases, governments have negotiated favorable terms for international market access and offered some form of trade promotion assistance. Each of these common areas of government involvement pertain to facilities or services which either have public good properties or give rise to externalities.

0.23 In a majority of cases, government interventions have extended beyond the above roles to include more active microeconomic interventions. This has frequently involved one or more types of subsidy or investment credit, although in the majority of cases such bonuses were available to producers, processors, and traders of other commodities as well. Price supports and/or price or quantity controls have been periodically applied in many of the focal cases. Data available for several of the focal Latin American case studies indicate that the aggregate effect of direct and indirect government interventions has been one of net taxation, suggesting that levels of production and trade would have been higher in the absence of interventions.

0.24 The paper examines organizational patterns in the focal commodity systems, including competitive structures and institutional links between producers, processors/exporters, and foreign market distributors. In the vast majority of cases, while domestic marketing systems have remained decentralized, export-oriented processing and trade has tended toward high rates of concentration with between three and ten firms accounting for the bulk of capacity and trade. In some cases, such concentration patterns have derived primarily from economic factors (e.g. economies of scale, differential capabilities and performance, etc.); in other cases, government interventions have determined or directly contributed to such patterns. Concentrated (although not monopolistic) trade structures have apparently facilitated improved quality control, marketing logistics, and, in some cases, an ability to influence world commodity prices.

0.25 Another common organizational pattern has been the prominence of contract farming and/or vertical integration in the linkages between farm-level production and downstream processing and trade. Open market buying and selling of raw materials has become only a supplementary, market-clearing mechanism in many of the focal commodity systems. In many cases, the leading firms have developed their own programs of applied research, extension, input delivery, and credit as a supplement or replacement for markets and government programs for such services.

0.26 Contract-based or intra-firm trade have also been significant features of the export operations in the majority of the focal commodity systems. Such long-term marketing ties have helped to maintain market access, lower logistical and transaction costs, and facilitate flows of information and technologies which have enabled suppliers to better meet changing consumer and buyer tastes and requirements.

0.27 Foreign capital, technology, training, and/or management skills have played an important role in the development of most of the focal commodity systems. In virtually all of the focal Latin American cases, credit from foreign distributors or direct foreign investments in production and/or processing/marketing facilities played a major role in initial subsector growth. Similar agroclimatic conditions between the focal countries and parts of the United States facilitated the transfer of crop varieties and of cultivation techniques.

Lessons and Policy Implications

0.28 One of the lessons of the "success stories" experiences is that agricultural export diversification by developing countries need not focus on exotic tropical commodities and need not depend upon low labor costs for international competitiveness. The collective experience suggests that developing countries can compete against industrialized country suppliers in the markets for a wide range of high-value and highvolume commodities on the bases of both cost and quality. Although such quality-based competitiveness did not occur overnight, with the liberal adoption/adaption of foreign technologies and advice and with the buildup of local skills, infrastructure, research capabilities, and experience, many of the focal commodity systems succeeded to match or exceed competitive quality standards within a decade after initial international market entry. A niche market strategy may well be necessary for very small countries which lack the capacity the potential to serve mass-market distribution systems. Larger middle- and low-income countries should aspire to compete in larger, faster growing markets.

0.29 A second important lesson is that successful export diversification often depends upon prior or parallel development of domestic markets. None of the focal commodity systems has developed as an export-oriented enclave and relatively few have relied upon export markets for the bulk or even the majority of their sales. This complementarily between export and domestic market development is much more important for high-value food commodities than it is for many traditional developing country export crops (e.g. tropical beverages, tobacco) given the limited domestic market for the latter. This suggests that government and donor programs and other interventions geared towards agricultural export diversification should not only seek to build upon existing domestic marketing experience, but should also incorporate investment and other components to support further domestic market development.

0.30 A third lesson is that governments and donors should facilitate foreign and joint venture investments and international transfers of production and processing technologies. This implies a streamlining of procedures for foreign investment, and a reduction in tariff and non-tariff barriers for the importation of planting materials, irrigation and processing equipment, and other technology-embodied inputs into agribusiness operations.

0.31 A fourth lesson is that while favorable macroeconomic conditions and policies have provided stimuli for new investments and expanded trade for most of the focal commodity systems at certain points in time, much of the initial supply response to favorable international market conditions and most of the sustained production and trade expansion experienced in the focal cases can be attributed to a combination of a) microeconomic developments (e.g. joint ventures, vertical integration, investment incentive programs, etc.), and b) investments in human capital and support structures (e.g. infrastructure, research). This implies that efforts geared toward promoting agricultural export diversification will need to extend beyond policy reform programs to include human capital and brick and mortar investments as well as microeconomic initiatives.

0.32 A final lesson is that both the public and private sectors have important roles to play in the development of high-value food exports. Among the prerequisites for profitable and sustainable trade are the availability of transport and telecommunications facilities and the maintenance of law and order. These goods and services are normally provided by the public sector. In the focal commodity systems, the role of the public sector has extended beyond these basic public goods and services. Probably the most important means of government support has come through the development of public 'knowledge systems', comprising research, training, extension, quality control, and market information services. While such public services were commonly supplemented (or replaced) by private initiative owe the commodity systems reached relatively advanced levels of development, these public services were generally crucial in the initial development or adoption of production and processing technologies and in the subsequent improvements in productivity and quality which enabled the participating firms to successfully compete in international markets. In many of the focal cases, government agencies have also conducted export promotion activities, although these were generally less important than efforts to negotiate favorable or at least equitable access to important international markets.

0.33 The collective experience of the focal commodity systems leads to the conclusion that the commercial production, processing, and marketing of high-value foods should be left to the private sector. While in several of the focal cases public enterprises have directly participated in trade, most such enterprises have adopted a production push- rather than market-orientation, have been averse to risk-taking, and have been operationally inflexible. Though proving adept at selling commodities to undersupplied markets, such state enterprises were not effective in marketing differentiated products in competitive markets. In contrast, the private sector (including trading cooperatives) has demonstrated far greater ability to anticipate, respond to, or cultivate changes in market demand and service requirements and has been far more innovative in developing products and inter-organizational linkages to facilitate expanded trade. These differential properties of public and private trading enterprises are particularly important in the development of high-value food exports given the importance of service, quality, and product form differentiation in major international markets for these goods.