Agroindustrialization, Globalization, and International Development an Overview of Issues, Patterns, and Determinants

ELSEVIER Agricultural Economics 23 (2000) 195-205 www.elsevier.com/locate/agecon

Agroindustrialization, globalization, and international development An overview of issues, patterns, and determinants

Thomas Reardon a,*, Christopher B. Barrett b

a Department of Agricultural Economics, Michigan State University, East Lansing, MI 48824, USA b Department of Agricultural, Resource and Managerial Economics, Cornell University, Ithaca, NY I4853-7801, USA

Abstract ( This paper offers an overview for a special issue on agroindustrialization, globalization, and international development. It sets out a conceptual framework for understanding the links among these three broad phenomena and then discusses emerging issues and evidence concerning the factors conditioning agroindustrialization in developing countries and the subsequent effects on employment, poverty, and the natural environment. We conclude with a research agenda. © 2000 Published by Elsevier Science B.V. JEL classification: 01; Q 1

Keywords: Agroindustry; Agribusiness; Developing countries; Institutions; Technology; Markets; Trade

1. Introduction The premise behind this special issue of Agricul tural Economics, and of the conference from which We believe that two sets of opportunities for en this selection of papers was drawn, 1 is that fostering riched debate and joint work have been neglected, and closer contact between the parties of each of the two that in each case, increasing interaction will be highly sets above can enhance the quality of debate, schol desirable. The first set is in the discipline of agricul arship, and praxis. An increasingly integrated global tural economics, where there have long been strong, economy causes established agribusiness firms to look distinct traditions of research on agribusiness and on increasingly to foreign suppliers and customers in international agricultural economic development. The order to improve profitability. But the structural foun second set is in the development debate, where ad dations of developing country agriculture often differ vocacy and research on competitiveness and business radically from those of OECD economies. So the development, and on poverty alleviation, often take agribusiness community can learn much from those place in separate camps. In both cases, the separate groups are like two adventurers following roughly 1 The international preconference to the August 1999 annual parallel paths that do not cross. meeting of the American Agricultural Economics Association, held in Nashville, Tennessee. Other papers from the preconference will appear shortly in special issues of the journals Environment and Development Economics and the International Food and Agribusi * Corresponding author. Fax: +1-517-432-1800. ness Management Review. We draw on some of those papers in E-mail address: [email protected] (T. Reardon). this overview and encourage readers to read those issues as well.

working on development issues. Meanwhile, given and (3) concomitant changes in the farm sector, such steady long-term decline in raw agricultural commod as changes in product composition, technology, and ity prices, the increasingly widespread perception is sectoral and market structures (Wilkinson, 1995). The that stimulating value-added activities that build on 1990s brought relatively rapid and intense agroin production agriculture is a necessary condition for dustrialization in many low- and middle-income improving living standards among poor rural popula economies. In the face of a so recent phenomenon, tions in developing countries. Yet there is a dearth of there are inevitably many gaps in our understand experience and research on development of competi ing of how and why these changes occur, and what tive and economically sustainable agroindustrial firms agroindustrialization implies for development. in developing countries, especially in the context of Fig. 1 sets out a general conceptual framework profoundly changed global and domestic economic showing the "conditioners", i.e. the factors that condi conditions. Moreover, it appears that agroindustrial tion or influence agroindustrialization, and the effects ization is merely a necessary, not a sufficient condi of the latter on development indicators. The feedback tion. It may accentuate prevailing inequities, deepen loop among these forces begins conceptually from poverty among vulnerable subpopulations, or damage the leftmost column, which depicts meta-trends in the natural environment if not induced and monitored both developed and less-developed countries. Popu carefully. So the development community likewise lation, income growth and urbanization induce global has much to learn from the agribusiness community changes in consumer demand patterns, notably dis about how to stimulate agroindustrialization of the proportionate growth in demand for dairy, meat, hor sort that advances sustainable development objectives. ticultural, and processed grain products, as compared The objective of this special issue is thus not a sum to the demand for unprocessed staple foods, follow mary of a rich literature offering definitive answers or ing Bennett's law. Since demand for food is income even a comprehensive set of key questions, so much inelastic, these meta-trends also fuel disproportionate as it is to shed light on previously underemphasized growth in demand for non-food goods and services, topics. The nine papers that follow examine specific thereby inducing rural industrialization and non-farm issues and cases in an attempt to improve our under employment growth. Meanwhile, the meta-trend of standing of what factors condition the process and ef market-oriented economic reforms, often embodied fects of agroindustrialization in developing countries. in structural adjustment programs and multilateral Our tasks in this introductory essay are to define terms, trade liberalization, reduces cross-border distribution to present a conceptual framework for understanding costs and barriers and gives increased currency to the links among these three broad phenomena, to syn profit-minded activities by private sector firms. This thesize emerging issues and evidence concerning the trend has fostered increasing integration of goods and factors conditioning agroindustrialization in develop capital markets around the world, linking farmers in ing countries and their subsequent effects on employ the developing world to OECD countries' consumers ment, poverty, and the natural environment, and to and corporations in unprecedented ways. Increased identify key areas in need of further research. integration also raises issues about the international ization of product standards (e.g., quality grades and food safety standards) either by means of non-tariff 2. A conceptual framework barriers to trade or restrictions on national sovereignty in imposing such barriers under trade accords such "Agroindustrialization" comprises three related sets as North American Free Trade Agreement (NAFTA) of changes: (1) the growth of agroprocessing, distribu or the World Trade Organization (WTO). Mean tion, and farm-input provision activities off-farm, un while, rapid technological change is transforming the dertaken by what we shall call "agroindustrial firms" conduct and structure of production and commerce which are called agribusiness firms in the agribusiness in all sectors, and throughout the world, enhancing research literature; (2) institutional and organizational productivity and enabling customized production change in the relation between agroindustrial firms and marketing processes, all with lower transaction and farms, such as increasing vertical coordination; costs. T. Reardon, C.B. Barrett/Agricultural Economics 23 (2000) 195-205 197

1 2 3 4 Meta- Trends Developing Country Development Agroindustries Indicators

I. Income and I. Globalization and Increase in scale and I. Growth in incomes population growth liberalization oncentration and output per capita (Bennett's & (GATT/WTO/FTAs, spatial, sectoral, firm) (aggregate, regional, Engel's Laws) market opening) t s~bsectoral) t . Product/subsector 2. Urbanization~ and 2. Organizational I omposition change 2. Changes in poverty female employment institutional change horticulture, processed and inequality (processed foods) (vertical coordination, ·etail, nontraditional) contracts, G&S, t ~ property rights, etc.) t 3. Employment and 3. Political economy 3. Extroversion of real wagt change (neoliberal arkets and ownership SAPs, capitalism) t multinationalization 3. Technological and export orientation) 4. Natural resource ~ change depletion/degradation 4. Modern technology (biotechnology, t or protection (e.g., information information, storage, . Increased use of (effluvia, water use, and biotechnologies) transport, drying, etc.) coordination/control ag extensification or intensification) echanisms t 5. Increased capital 5. Sociocultur1effect intensity in production (change in diet, and processing traditions, decision- making authority, etc.)

Fig. I. Flow diagram showing links among globalization, agroindustrialization, and development.

These meta-trends foster more specific changes the rise and spread of new contractual arrangements in the global agrifood economy (column 2). For ex between processing firms and farms, of quality and ample, while agriculture remains one of the most safety standards, and of intellectual property rights. 2 protected tradable sectors on earth, the Uruguay These organizational and institutional changes and Round, various regional trade agreements, and uni increased global competition in the agrifood economy lateral liberalization efforts by some countries have are closely linked with rapid technological change, opened domestic agrifood markets to considerable in from biotechnology advances dramatically affect ternational competition. Many countries (e.g., Kenya, ing farm-input industries (e.g., seed, chemicals) and Malaysia) redefined national food security objectives the farm-to-table distribution channel, to informa to move away from previous strong commitments tion, transport, and robotics technologies that affect to food self-sufficiency. These changes create op processing, storage, shipping, and inspection activi portunities for agroindustry while also raising issues ties. Agroindustrialization is both an agent of and a of competitiveness (and derivative questions of pro duction scale, institutional arrangements, vertical 2 We define institutions as in Hoff et al. (1993): "By an economic coordination, and technological innovation). At the institution we mean a public system of rules that define the kinds of same time, there have been profound changes in the exchanges that can occur among individuals and that structure their incentives in exchange. Economic institutions include markets and organization and institutions of the agrifood economy property rights, systems of land and animal tenure, obligations of associated with reduced state regulation of farmers' mutual insurance within lineage groups, and other systems of ex production and marketing choices, globalization, and change that are determined by implicit contracts or social norms." 198 T. Reardon, C.B. Barrett/Agricultural Economics 23 (2000) 195-205 response to globalization and induced institutional and 1970s, entry into the sector was more commonly and technological change. in the form of plantation-cum-processing-for-export These broader patterns inevitably affect the evolv enclaves by firms such as Unilever or Del Monte, ing character of agroindustries in developing countries the past decade has witnessed widespread developing (column 3), through changing relative factor and prod country market entry by input supply firms such as uct prices, foreign direct investment (FDI) by global Monsanto or Pioneer, processors like Coca-Cola and and regional companies, and the transfer of technol Nestle, traders such as Cargill, and retail distributors ogy, organizational structures, and institutions. Among like Carrefour, McDonalds, or Walmart. the broad patterns observed over the past decade are Growth in downstream components of the agrifood increased concentration in agroindustrial sectors, and channel and multinationalization have set off signif an increase in the average size of processing firms and icant changes in the agrifood system organization farms struggling to reap available economies of scale. and institutions in developing countries. The most The various forces already discussed have exposed de prominent and widespread changes include the rise veloping country firms to competitive pressures from of contractual exchange in the place of spot markets, which they had been partly or wholly shielded until and increased attention paid to product quality and the past decade or two. safety, both in fact and as perceived by urban or In part, this induces a shift in product composition foreign consumers. These same forces have sparked toward those subsectors in which developing country rapid change in off-farm technologies, often with the producers and upstream and downstream firms hold effect of increasing capital/labor ratios relative to tra global comparative advantage and toward which local ditional, more artisanal methods of processing and consumers' preferences are moving: shellfish, horti distributing agrifood products. cultural products, processed foodstuffs, and non-staple These changes inevitably affect development in products. They can also redefine traditional domestic dicators, such as those in column 4. Technological comparative advantage, as where industrial produc change, renewed access to private foreign capital, new tion of shrimp has replaced tidal and deep sea capture organizational forms and institutional arrangements methods or biochemical innovations provide cheap to enhance coordination can all stimulate growth in substitutes for natural spices (e.g., vanillin for vanilla). output and income per capita. But as any introduc These changes have sharply increased the value-added tory trade text makes plain, the aggregate net gains share of processing and distribution within the agri disguise distinct winners and losers. Regions with food chain, and of non-staple subsectors relative to greater access to ports, serviceable infrastructure staples, including a significant increase in several re (e.g., roads, electricity, telephone service), and pro gions in non-traditional raw and processed agricultural ducers of non-traditional products enjoying increased exports. These changes often attract or are driven by global demand commonly enjoy advantageous terms rapid "multinationalization" of the off-farm portions of trade shifts while producers of unprocessed sta of developing country agrifood systems. Previously ple commodities in rural hinterlands too often lose semi-subsistence sectors often lacked a critical mass out. Changing technologies and scale economies in of private, modern input supply, and post-harvest processing and distribution, where none may have processing and distribution service providers. Many existed in production, can lead to differentiation be foreign entrepreneurs and firms have stepped in to fill tween large and small firms and farms. The net effect these gaps, often through acquisition, mergers, and on employment and poverty cannot be predicted joint ventures. Their presence is manifested by an in generally but depends substantially on the ex ante crease in the share of foreign control and/or ownership spatial and sectoral distribution of the poor, the na of domestic firms and the markedly increased presence ture of the particular technologies introduced (e.g., of multinational firms in the agrifood sectors of the do they substitute grading and sorting machines for most middle-income countries. 3 Where in the 1960s human packers?), and the indirect effects created by overall economic growth. For example, it may 3 The trend is less evident in low-income countries, but seems be that labor/output ratios fall but aggregate output underway. rises sufficiently to swamp this effect and stimulate T. Reardon, C.B. Barrett/ Agricultural Economics 23 (2000) 195-205 199 aggregate employment, while lowering food prices, 3. Trade and marketing issues thereby pushing up real wages and generating dis tributionally progressive growth even though the International trade in processed agrifood products direct benefits appear to be concentrated on elite sub and in fresh fruit and vegetables and oilseeds has populations. grown rapidly over the past several decades. The Changes in the agrifood sector can significantly growth has been uneven across LDC regions, with affect consumer welfare, by altering the quality and greatest growth in exports from Latin America, and quantity of their diets, by increasing convenience and exports and imports in Asia, but with stagnation and product variety, but also by imperiling cultural norms loss of market share of Africa (Dfaz-Bonilla and Reca, surrounding traditional foods or local firms. At issue 2000; Yumkella et al., 2000). Intra-regional trade has is whether agroindustrialization leaves small farm increased, thanks in part to the formation of regional ers, landless laborers, or artisanal service providers free-trade areas such as Mercosur (Farina, 2000). In behind, or even impoverishes them, and whether some cases growth in regional product markets has foreign firms or consolidated privatized industries been accompanied by rapid growth in regional capital owned by local elites crowd out small-scale rural en markets and inter-country flows, as in East Asia (Wei trepreneurs. No clear pattern yet exists as to whether and Cacho, 2000). the gains in opportunities and market access for Increasing product trade has been driven partly by small rural firms and farms and laborers outweigh the trade liberalization, at the global scale via GATT and losses. WTO and at the regional level by regional trade agree Finally, agroindustrialization, like any change in ments such as Mercosur, NAFTA, and at the country production patterns, can leave its mark on the natural level by structural adjustment programs that have em environment, and thereby on the welfare prospects phasized currency devaluation, removal of tariff and of future generations in the developing world. Some non-tariff batTiers to trade, reduction of export taxes, agroindustries can concentrate and increase unde eased FDI restrictions, and reduced government in sirable effluvia, as with tanneries that have polluted volvement in agrifood marketing and trade. Kenya's Lake Nakuru to the point that its famed However, the pace of policy change has varied flamingos are dying en masse and evacuating to across countries and products. For example, OECD other soda lakes in the Rift Valley. Increased de nations have been relatively slow to reduce tariffs on mand for standardized products and reorganization processed agrifood products, thereby concentrating of the input supply-to-output marketing channel can developing country agrifood export gains chiefly in also lead to increased chemical and water use by raw, rather than value-added products. Dfaz-Bonilla small farmers intensifying production for markets, and Reca document the resulting pro-wealthy country often with undesirable environmental consequences bias in international agrifood trade due to persistent (Pingali, 2000), and to expansion of cultivated area trade distortions. into fragile forest, rangeland, or watershed margins, Rising incomes and urbanization have fed a shift in as discussed by Barbier (Barbier, 2000). On the diet from staples toward non-staple foods (fruits, veg other hand, agroindustrialization can reduce environ etables, dairy products, meat, edible oils) and drawn mental pressures by inducing adoption of improved more women into employment outside the home, in production technologies, fostering the transmission creasing demand for processed products and restau of consumer demand for environment-friendly pro rant food. This trend too has stimulated growth in both duction practices, or generating useful processing domestic and international agrifood trade. by-products, such as animal feeds that promote Consider the case of Brazil, now the fourth largest sedentarized livestock production, thereby reducing agricultural and agroindustrial producer in the world overgrazing and competition with wildlife for natural and hub to the soybean oil and orange pulp and juice forage. As with its poverty and employment effects, industries. The 1995 formation of the Mercosur union agroindustrialization' s environmental effects remain by Argentina, Brazil, Uruguay, and Paraguay came generally unclear, depending considerably on local immediately on the heels of structural adjustment conditions. programs that substantially reduced trade distortions, 200 T. Reardon, C. B. Barrett I Agricultural Economics 23 (2000) 195-205 impediments to FDI inflows, and government food 3.1. Sectoral organization and institutional issues procurement and price control policies. Private do mestic and foreign firms entered en masse, and agri Agroindustrialization both prompts and responds to cultural output, external trade, and real wages grew the changing organization of agrifood sectors in de rapidly over the last half of the decade. However, veloping countries and in the institutions governing increased competition constrained profit growth and production and exchange. This issue highlights several led to rapid consolidation in the agrifood processing such changes. and retail sectors, where supermarkets and fast food Village-level cooperatives are resurgent, after ape chains become prominent players and firms increas riod of disfavor in the 1970s and 1980s, as a means of ingly adopt Mercosur-wide strategies (Jank et al., overcoming liquidity constraints, information asym 2000). The emergence of powerful downstream play metries, and minimum efficient scales of production ers in the agrifood chain has also affected upstream or marketing that can otherwise impede smallholder organization, manifest in the imposition of product participation in rapidly growing agrifood sectors (Jaf quality and safety grades and standards and new pro fee and Morton, 1995; Candler and Kumar, 1998; duction technologies on farmers in subsectors such as Holloway et al., 2000). In South Africa, some farmer dairy (Dirven, 2000). cooperatives have recently listed on the Johannes FDI has also been extremely important in influenc burg stock exchange. Widespread contract farming ing the pace and nature of agroindustrialization else has also emerged in many developing countries as a where (Gopinath and Bolling, 2000). LDCs currently means to reduce risk and ensure throughput volumes account for over 37 and 14% of global FDI inflows of known price and quality for downstream proces and outflows, respectively (World Investment Report, sors and distributors, as documented in Africa (Little 1998), up sharply over the course of the decade, partic and Watts, 1994; Jaffee and Morton, 1995), Latin ularly in South and Southeast Asia and Latin America, America (Schejtman, 1998; Key and Runsten, 1999), and especially in China. Developing country markets and Asia (Gandhi et al., 2000). Contracts may reduce now represent one-quarter of US firms' FDI in food coordination costs within the agrifood chain, but there processing globally. is growing evidence that contracts are not necessarily But while FDI can relax capital constraints for do the institutional panacea for small farmer involvement mestic agroindustrial firms, as in Argentina, Malaysia in agroindustrialization. Contract farming typically or Slovakia (Gopinath and Bolling, 2000; Gow et al., displaces decision-making authority from the farmer 2000), it can also foster industrial concentration and to the downstream processor or distributor, turning large flows of repatriated profits, as in the Brazilian farmers into quasi-employees. Given high per unit case just cited. Developing country governments and costs of contracting with smaller farmers, who also firms are responding in various ways, by regulating commonly have greater problems meeting stringent acquisitions and joint ventures, and through strate quality and safety requirements, contract farming in gic response. For example, Wei and Cacho (2000) some settings favors more capital-intensive medium note that in China international product quality stan and large farmers, lessening the income and employ dards introduced by Unilever and Nestle, each a new ment impact of agroindustrialization for the poor entrant to the Chinese market through acquisitions (Key and Runsten, 1999; Reardon et al., 1999; Farina, and joint ventures, spurred indigenous firms' adop 2000). Enforcement of contract compliance is also a tion of international quality standards and imitation of sticky issue in developing countries, where contract the multinationals' management and marketing tech law is less well developed and judicial systems are niques. The innovating group of domestic firms grew often less efficient or predictable. Gow et a!. analyze competitive, increasing their market share at the ex the use of self-enforcing contractual arrangements pense of the foreign firms and of domestic firms forced or "internal" private enforcement mechanisms in the out of the market. Wei and Cacho make similar ob Slovakian sugar sector, and show its favorable effects servations about other subsectors in China, including on output and efficiency. beverages, instant noodles, fast food, chocolate, and One reason for the rise of cooperatives and contract snack foods. farming is the increased importance and changing T. Reardon, C. B. Barrett/Agricultural Economics 23 (2000) 195-205 201 nature of agrifood grades and standards (G&S). Rear thing resembling share tenancy contracts with small don et al. (2000) show that G&S have grown more cotton farmers (Escobal et al., 2000). In much of numerous and influential over the past two decades in the world, non-profit non-governmental organizations developing economies. Most G&S are established by (NGOs) are providing such services amid growing multinational firms or consortia, less commonly by fascination with the opportunities of "microenterprise multilateral organizations, so small developing coun development," and increased demand in subsectors try firms and farmers are almost invariably "G&S such as Ghanaian cassava (al Hassan, 2000). Whether takers." In some sectors, such as horticulture, meat, NGO entry into these activities truly helps small dairy, and fisheries, G&S increasingly specify pro farms and rural firms, especially when foreign donors duction, processing, and distribution processes rather heavily subsidize foreign NGOs, or whether they than outcomes. Unnevehr emphasizes the growing mainly crowd out indigenous private service providers importance of "farm-to-table" quality management nevertheless remains an open question. systems for food safety, based on documented pro Property rights regime heavily influence agroin duction practices demonstrated to prevent and control dustrialization in developing countries. Escobal et al. hazards (Unnevehr, 2000). But the introduction and note that until recently agroindustrial firms could not evolution of process G&S requires investments in own land in Peru (in order to force contracts with training, equipment, infrastructure and monitoring local farmers). The change in land laws has induced systems, and not all firms or governments can afford vertical integration by firms and land acquisition by these expenses (Stephenson, 1997). The rise of G&S processors. Many states have been recognizing pri may therefore imply scale-biased growth in some vate rights to commercial entities as well. Widespread sectors. Furthermore, G&S can create new non-tariff privatization of previously state-held firms has of trade barriers, sparking serious trade disputes. Un ten brought (lagged) efficiency and output gains, as like conventional trade barriers, however, G&S may Li and Rozelle (2000) document among township resolve consumer uncertainty about product qual enterprises in China. ity, thereby creating trade and increasing both con Finally, intellectual property rights (IPR) are a sumer and exporter welfare (Thilmany and Barrett, fundamental topic in sectors undergoing rapid techno 1997). logical change. Lesser et al. (2000) discuss the links Agglomerations of agroindustrial firms and farms among IPR, agricultural biotechnology, "upstream" are not always organized, as in cooperatives or through agroindustry multinationals such as Monsanto, na contract farming schemes. Sometimes clusters emerge tional research organizations, such as Brazil's Em spontaneously for various reasons, including exter brapa. Licensing arrangements between public sector nal economies of scale, reduced transactions cost, agricultural research organizations with elite local and sharing of information, workers, and equipment germplasm and multinationals offering faster access (Porter, 1990; Krugman, 1991). Clusters need not be to new biotechnologies may stimulate local agricul homogeneous. Palm sugar and soybean processing in tural growth and provide needed income for agricul Indonesia rests upon clusters of small firms (Sandee tural ministries. But such agreements require complex and Burger, 2000), while dairy clusters in Argentina, institutional preparatory steps, including appropriate Chile, Colombia, and Uruguay include farms of patenting and other IPR laws, biosafety regulations widely varying scale (Dirven, 2000). Clustering need and institutional policies, and a hospitable macro not protect small producers, however, as Dirven doc economic environment for foreign multinationals. uments in the cases of dairy clusters in Argentina and Chile. 3.2. Technology issues Increasingly competitive and rapidly changing mar kets put a premium on management skills. In some The rapid pace of technological change is not con developing countries, this has led to the emergence fined to biotechnology, an area drawing increased of private management service firms, as in coastal scrutiny with respect to food safety and environmen Peru, where firms exchange management services tal concerns (Batie and Ervin, 2000). There have been for labor supervision and land collateral in some- substantial improvements in transport and storage 202 T. Reardon, C.B. Barrett/ Agricultural Economics 23 (2000) 195-205 technologies, such as controlled atmosphere storage of agroindustrialization. First, many organizational methods, refrigerated and faster and larger ships, and and institutional changes (e.g., clustering, cooper chemical applications to reduce fungus formation atives, private management companies) explicitly (OECD, 1996). Dehydration techniques for vegeta foster smallholder participation in agroindustrializa bles, small-scale processing machines for cassava tion. Second, an increase in prevailing capital/labor (al Hassan, 2000), vacuum packing for milk, and ratios does not necessarily imply that aggregate rural other packaging technologies have also improved. employment declines. If access to dynamic external These changes have sparked dramatic growth in some markets increases, output increases may stimulate suf countries' agrifood sectors, including apples and ficient growth to increase employment and real wages, pears in central Chile, vacuum-packed milk in Brazil, thereby contributing directly to the welfare of the and shrimp in Ecuador. poorer subpopulations in most developing countries, Further upstream in the chain, the use of equip which tend to depend partly or wholly on wage labor ment and chemicals, in order to improve product income. Third, in some cases - such as Mexico's quality and reduce labor demand and output variabil tomato agroindustry in Mexico, described by Barron ity, has expanded considerably in the middle-income and Rello (2000) - agroindustry relies on cheap countries over the past decade. Of course, increased labor to be competitive in global markets, and thus chemical use remains very controversial because of even large firms maintain relatively high labor-output adverse potential environmental effects (Dasgupta et ratios, with employment-increasing effects on poor al., 2000; Pingali, 2000). Even in low-income coun households, as in the Mexican case. tries, the adoption of higher-yielding cross-bred cows has greatly increased smallholder milk output in Ethiopia (Holloway et al., 2000) and India (Candler 4. Conclusions and a research agenda and Kumar, 1998), demonstrating that technological change need not crowd out the poor by inducing sub The pace of agroindustrialization has been nothing stitution of capital for labor. Yet remarkably little is short of remarkable in parts of Latin America, Central known about possible farm size biases to adoption and and Eastern Europe and Southeast Asia over the past use of chemicals and capital inputs needed for farm 5-10 years, and is apparent even in lower-income set ers to produce crops and livestock having the quality, tings of Africa and South Asia. Developing economies cost, and risk characteristics desired downstream in undeniably need improved technologies throughout the agroindustrial chain. Reduced government input the agrifood production, processing, and distribution distribution and credit services have diminished ac chain, skills transfer, foreign capital, and increased cess to these inputs by smallholders in many areas export earnings. So identification of those factors of Africa (Reardon et al., 1999) and Latin America that help to stimulate more rapid and widespread (Schejtman, 1998) may constrain their sharing in the agroindustrialization can provide valuable informa gains from agroindustrialization. tion to developing countries designing or reform At first glance, the induced technological and insti ing policies so as to ensure that they participate tutional changes associated with agroindustrialization in the rapid changes taking place, and are not left in the developing world might seem to increase cap behind. ital/labor ratios and favor medium-to-large farmers The necessity of agroindustrialization is almost in over small farmers. At a minimum, increasing pro disputable. Yet a plethora of questions remain as to cessed products' share in the agrifood sector implies how to get the right kind of agroindustrialization, the an increase in capital/labor ratios, as Ehui and Del sort that stimulates employment, reduces poverty and gado (1999) estimate in models for Africa. There is real food prices, stimulates real wages, improves food also increasing evidence, found in various papers in safety, quality and consumer choice, and protects this special issue, of small firms and farms going out the natural environment. Too often the process of of business under the new competitive pressures. agroindustrialization leads to industrial concentration, However, three factors complicate the assess exclusionary practices that crowd out undercapital ment of the likely employment and poverty effects ized indigenous firms and small farmers, substitution T. Reardon, C. B. Barrett/Agricultural Economics 23 (2000) 195-205 203 of imported equipment and managers for domestic at odds with most of our training and instincts. Yet workers and entrepreneurs, enrichment of local urban the sheer complexity of modeling institutions, spa elites at the expense of the rural poor, and depletion tial arrangements, technological change, international or degradation of the host country's natural resources. flows of goods and capital, and general equilibrium So the key questions surround the conditions under employment and price effects limits the ability of which agroindustrialization is most likely to yield traditional, deductive modeling to generate results ro broad-based, environmentally sustainable growth that bust to the fast-changing features of the agroindustrial creates wealth and improves human well-being. What landscape. are the distributional implications - within and The phenomenon of agroindustrialization is rapidly between sectors and regions - of different agroin attracting attention worldwide. The broader commu dustrialization pathways, as well as the follow-on nity of agricultural economists have much to contri political economy consequences of the distributional bute to understanding the phenomenon and har outcomes? Does agroindustrialization bolster the do nessing its energies for good ends. The papers that mestic constituency in favor of openness to foreign follow offer some fine examples of the interesting trade? Does increased multinationalization of the issues and important findings just now beginning agrifood sector suppress indigenous development of to emerge. technologies appropriate to local cultures and agroe cosystems? The papers in this special issue examine specific issues and cases and offer suggestive answers to several key questions. Still, much remains to be Acknowledgements discovered. One of the challenges lies in harmonizing lan The authors thank the following institutions (listed guage across the distinct subdisciplinary cultures of in alphabetical order) for generously supporting the agribusiness and international development (Cook preconference from which the papers in this spe and Chaddad, 2000). For example, the role, if any, of cial issue were drawn: AAEA Foundation; Cornell spatial clustering and agglomeration effects in deter International Institute for Food, Agriculture, and De mining firms' competitiveness is inherently an indus velopment; Farm Foundation; Food and Agricultural trial organization question. One must understand how Organization of the United Nations (with individual the performance of one firm affects that of others. Yet grants from the Agriculture Sector in Economic De the sort of general equilibrium effects familiar from velopment Service, the Farm Management and Pro village economy models in the development literature duction Economics Service, and the Rural Institutions also lie at the heart of this particular issue. Similarly, and Participation Service, Rural Development Divi while the development literature is replete with work sion); International Food Policy Research Institute; on technology adoption, many of the innovations International Livestock Research Institute; Michigan permeating agroindustrial sectors relate to improved State University (with individual grants from the Cen product standards or production processes, rather ter for Latin American and Caribbean Studies, Office than to improved inputs or discrete technologies of of International Studies and Programs, the Institute the sort on which the technology adoption literature of International Agriculture, the Department of Agri focuses. Technological and institutional changes in cultural Economics, the Center for Advanced Studies the agrifood sectors of many developing countries are in International Development, the African Studies perhaps better understood using contemporary man Center, and the Asian Studies Center). We also thank agement theory than conventional production theory. the FAO Regional Office for Latin America and the Real progress will therefore depend partly on our Caribbean for funding Reardon's sabbatical work on ability to bridge the literatures and the languages of this topic, Joyce Knuutila for outstanding editorial as different scholarly traditions. sistance, Joyce Cacho, Mike Cook, Liz Evans, Nancy The other main challenge will be to build theory Herselius, and Keely Martin for help in organizing inductively, from the rich mass of case study evidence the conference, and Stan Johnson for his enthusiastic beginning to emerge. Such methods of theorizing are support of this special issue. 204 T. Reardon, C. B. Barrett I Agricultural Economics 23 (2000) 195-205

References transactions costs, cooperatives and milk-market development in the East-African highlands. Agric. Econ 23 (3), 279-288. al Hassan, R., 2000. Issues of response of rural producers to Jaffee, S., Morton, J. (Eds.), 1995. Marketing Africa's High-value emerging markets for Africa's processed cassava: the case of Foods: Comparative Experiences of an Emergent Private Sector. Ghana. Int. Food Agribusiness Mgmt. Rev., in press. Kenall/Hunt Publishing Company, Dubuque, IA. Barbier, E.B., 2000. Links between economic libralization and Jank, M.S., Leme, M.F.P., de Paula, S.P.L., Filho, P.F., Nassar, rural resource degradation in the developing regions. Agric. A.M., 2000. Multinationals and intra-industry concentration: the Econ. 23 (3), 299-310. case of Brazilian agribusiness exporters. Int. Food Agribusiness Barron, M.A., Rello, F., 2000. The impact of the tomato agro Mgmt. Rev., in press. industry on the rural poor in Mexico. Agric. Econ. 23 (3), Key, N., Runsten, D., 1999. Contract farming, smallholders, 289-297. and rural development in Latin America: the organization of Batie, S.S., Ervin, D.E., 2000. Biotechnology and the environment: agroprocessing firms and the scale of outgrower production. issues and linkages. Environ. Dev. Econ., in press. World Dev. 27 (2), 381-402. Candler, W., Kumar, N., 1998. India: the dairy revolution. Krugman, P., 1991. Geography and Trade. MIT Press, Cambridge, The impact of dairy development in India and the World MA. Bank's contribution. The World Bank Operations Evaluation Lesser, W., Schmit, T., Ruiz, L., 2000. Forthcoming. National Department, Washington. breeding programs as technology sellers: Marketing elite Cook, M.L., Chaddad, F.R., 2000. Agroindustrialization of the germplasm for GMOs by developing countries. International global agrifood economy: bridging development economics and Food and Agribusiness Management Review. agribusiness research. Agric. Econ. 23 (3), 207-218. Li, H., Rozelle, S., 2000. Saving or stripping rural industry: an Dasgupta, S., Mamingi, N., Meisner, C., 2000. Agricultural trade analysis of privatization and efficiency in China. Agric. Econ. liberalization, and chemical pesticide use in Brazil. Environ. 23 (3 ), 241-252. Dev. Econ., in press. Little, P.D., Watts, M., 1994. Living Under Contract: Contract Dfaz-Bonilla, E., Reca, L., 2000. Trade and agroindustrialization in Farming and Agrarian Transformation in Sub-Saharan Africa. developing countries: trends and policy impacts. Agric. Econ. University of Wisconsin Press, Madison, WI. 23 (3), 219-229. OECD, 1996. Export Fruit Boom from the South: A Threat Dirven, M., 2000. Clustering and declustering effects of for the North? Organization for Economic Co-operation and globalization: based on case studies of the milk sector in Chile, Development. OECD Publications, France. Colombia, Uruguay, Argentina, and the Netherlands. Int. Food Pingali, P., 2000. Environmental consequences of agricultural Agribusiness Mgmt. Rev., in press. commercialization in Asia. Environ. Dev. Econ., in press. Ehui, S., Delgado, C.L., 1999. Economy-wide impacts of Porter, M., 1990. The Competitive Advantage of Nations. technological change in the agrifood processing sectors in Sub Macmillan, London. Saharan Africa. Paper presented at the AAEA Preconference Reardon, T., Rello, F., Schejtman, A., Stamoulis, K., 1999. on Agroindustrialization, Globalization and International Evaluating the employment impacts of agroindustrialization: Development, Nashville, Tennessee, August 6-7, 1999. hypotheses and illustrations from Latin America. Paper Escobal, J., Agreda, V., Reardon, T., 2000. Institutional change presented at the AAEA Preconference on Agroindustrialization, and agroindustrialization on the Peruvian coast: innovations Globalization and International Development, Nashville, impacts and implications, Agric. Econ. 23 (3), 267-277. Tennessee, August 6-7, 1999. Farina, E.M.M.Q., 2000. Issues and challenges for the Mercosul Reardon, T., Codron, J.-M., Busch, L., Bingen, J., Harris, C., 2000. food industry in the context of agroindustrialization and Agrifood grades and standards in developing countries: issues globalization. Int. Food Agribusiness Mgmt. Rev., in press. of inclusion and exclusion. Int. Food Agribusiness Mgmt. Rev., Gandhi, V., Kumar, G., Marsh, R., 2000. Agroindustries for rural in press. and small farmer development: a study of the issues and lessons Sandee, H., Burger, K., 2000. Clustered small enterprise under from India. Int. Food Agribusiness Mgmt. Rev., in press. economic distress: the impact of the Indonesian crisis on small Gopinath, M., Bolling, C.H., 2000. Effects of FDI in developing industry clusters, with reference to agroindustries. Int. Food countries: the case of food and agriculture. Int. Food Agribusiness Mgmt. Rev., in press. Agribusiness Mgmt. Rev., in press. Schejtman, A., 1998. Agroindustria y pequeiia agricultura: expe Gow, H.R., Streeter, D.H., Swinnen, J.F.M., 2000. How private riencias y opciones de transformaci6n. In: CEPAL/GTZ/FAO contract enforcement mechanisms can succeed where public (Eds.), Agroindustria y Pequeiia Agricultura: Vfnculos, Poten institutions fail: the case of Juhocukor a.s .. Agric. Econ. 23 (3), cialidades y Oportunidades Comerciales. UNO, Santiago. 253-265. Stephenson, S.M., 1997. Standards and conformity assessment Hoff, K., Braverman, A., Stiglitz, J.E., 1993. Introduction. In: as nontariff barriers to trade. Policy Research Working Paper Hoff, K., Braverman, A., Stiglitz, J.E. (Eds.), The Economics No. 1826. Development Research Group, The World Bank, of Rural Organization: Theory, Practice, and Policy. Oxford Washington. University Press, New York. Thilmany, D.D., Barrett, C.B., 1997. Regulatory barriers in an inte Holloway, G., Nicholson, C., Delgado, C., Staal, S., Ehui, S., grating world food market. Rev. Agric. Econ. 19 (1), 91- 2000. Agroindustrialization through institutional innovation: 107. T. Reardon, C. B. Barrett I Agricultural Economics 23 (2000) 195-205 205

Unnevehr, L.J., 2000. Food safety issue and fresh food product Familiar. In: CEPAL (Eds.), Las Relaciones Agroindustriales y exports from LDCs. Agric. Econ. 23 (3), 231-240. la Transformacion de la Agricultura. CEPAL, Santiago, Chile. Wei. A.. Cacho, J., 2000. Establishing linkages between global World Investment Report, 1998. United Nations Conference on ization, developing country economic growth and agro Trade and Development, Geneva. industrialization: the case of East Asia. Int. Food Agribusiness Yumkella, K.. Roepstorff, T., Vinanchiarachi, J., Hawkins, T., 2000. Mgmt. Rev., in press. Issues and challenges for Africa in the context of globalization Wilkinson, J., 1995. Agroindustria: Articula~ao com os Mercados and agroindustrialization. Int. Food Agribusiness Mgmt. Rev., e Capacidade de Integra9ao Socio-economica da Produ91io in press.