MADIA DISCUSSION PAPER 11 8436 MANAGING AGRICULTURAL DEVELOPMENT IN AFRICA FOREWORD The MADIA study and the papers comprising this MADIA Discussion Paper Series are important both for their content and the process of diagnosis and analysis that was used in the conduct of the study. The MADIA research project has been consultative, nonideological, and based on the collection and analysis of a substantial amount of concrete information on specific topics to draw policy lessons; it represents a unique blend of country-oriented analysis with a cross-country perspective. The conclusions of the studies emphasize the fundamental importance of a sound -macroeconomic environment for ensuring the broad-based development of agriculture, and at the same time stress the need for achieving several difficult balances: among macroeconomic, sectoral, and location-specific factors that determine the growth of agricultural output; between the development of food and export crops; and between the immediate impact and long-run development of human and institutional capital. The papers also highlight the complementarity of and the need to maintain a balance between the private and public sectors; and further the need to recognize that both price and nonprice incentives are critical to achieving sustainable growth in output. The findings of the MADIA study presented in the papers were discussed at a symposium of senior African and donor policymakers and analysts funded by USAID in June 1989 at Annapolis, Maryland. The participants recommended that donors and African governments should move expeditiously to implement many of the study's valuable lessons. The symposium also concluded that the process used in carrying out the MADIA study must continue if a stronger, more effective consensus among donors and governments is to be achieved on the ways to proceed in resuming broad-based growth in African agriculture. The World Bank is committed to assisting African countries in developing long-term strategies of agricultural development and in translating the MADIA findings into the Bank's operational programs. Stanley Fischer Edward V. K. laycox Vice President Development Economics Vice President and Chief Economist Africa Regional Office MADIA DISCUSSION PAPER 11 MARKETS, MARKETING BOARDS, AND COOPERATIVES IN AFRICA ISSUES IN ADJUSTMENT POLICY UMA LELE ROBERT E. CHRISTIANSEN _ s~~~~~~~~_N -1 I THE WORLD BANK _ WASHINGTON, D.C. _ _w Copyright © 1989 All rights reserved The International Bank for Reconstruction Manufactured in the United States of America and Development/THE WORLD BANK First printing December 1989 1818 H Street, N.W. Washington, D.C. 20433. U.S.A. MADIA Discussion Papers are circulated to encourage discussion and ment. at the address shown in the copyright notice above. The World Bank comment and to communicate the results of the Bank's work quickly to the encourages dissemination of its work and will normally give permission development community: citation and the use of these papers should take promptly and, when the reproduction is for noncommercial purposes. with- account of their provisional character, Because of the informality and to out asking a fee. Permission to photocopy portions for classroom use is not present the results of research with the least possible delay, the manuscript required. though notification of such use having been made will be has not been prepared in accordance with the procedures appropriate to appreciated. formal printed texts, and the World Bank accepts no responsibility for The complete backlist of publications from the World Bank is shown in errors. The findings, interpretations, and conclusions expressed in this the annual Index of Publications, which contains an alphabetical title list and paper are entirely those of the author(si and should not be attributed in any indexes of subjects. authors. and countries and regions. The latest edition is manner to the World Bank, to its affiliated organizations, or to members of available free of charge from the Publications Sales Unit, Department F, The its Board of Executive Directors or the countries they represent. World Bank, 1818 H Street. N.W. Washington, D.C. 2043 3, U.S.A.. or from The material in this publication is copyrighted. Requests for permission Publications, The World Bank, 66. avenue d lena. 7 5116 Paris, France. to reproduce portions of it should be sent to Director. Publications Depart- Uma Lele is the manager of Agricultural Policy in the Africa Technical Department at the World Bank. Robert E. Christiansen is an economist of Agricultural Policy in the Africa Technical Department at the World Bank. Library of Congress Cataloging-in-Publication Data Lele, Uma J. Markets, marketing boards, and cooperatives in Africa: issues in adjustment policy / Uma Lele and Robert E. Christiansen. (MADIA discussion paper; I 1) Includes bibliographical references. 1. Farm produce-Africa, Sub-Saharan-Marketing 2. Marketing boards-Africa, Sub-Saharan. 3. Cooperative marketing of farm produce-Africa, Sub-Saharan. 4. Managing Agricultural Development in Africa (Organization) I. Christiansen, Robert E. II. Title. mI. Series. HD9017.S82.L44 1989 381'.41'0967-dc2O 89-22763 ISBN 0-8213-1327-4 Contents Introduction ........ 4 The Genesis and Evolution of Intervention In Agricultural Marketing. 5 Agricultural Marketing at Independence. 5 Growth of Marketing Parastatals in the Postcolonial Period. 6 The Economics of Market Intervention .7 Risk in Agriculture .7 Instability in Export Agriculture .9 Price Stabilization. 9 Scale of Investments .10 Inadequacy of Financial Markets .10 Public-Sector Intervention in Marketing . 1 Marketing Costs .II Instability Of Agricultural Institutions .16 Lessons from Experience with Parastatals .17 Cooperatives .18 Lessons from Experience with Cooperatives .20 Agricultural Marketing Reforms: Overview of the Experience of MADIA Countries .21 Summary of Conclusions and implications .24 Bibliography .26 Appendix .29 Notes .30 We have benefited from comments on earlier drafts of this paper by Bela Belassa, David Steeds, Roy Southworth, Barbara Harriss, Thomas Hobgood (USAID), Emmy Simmons (USAID), loan Atherton (USAID), Steven Sinding (USAID/Kenya), Shlomo Reutlinger, Manmohan Agarwal, Alex Duncan, Gavin Williams, and ludith Heyer, as well as from our exchanges with Kevin Cleaver and Michael Westlake. Valuable research assistance was provided by Ann Mitchell, Kundhavi Kadiresan, and Beth Porter and typing by Christina Dhanaraj. An earlier version of this paper was presented at a conference sponsored by USAID/Kenya and the Government of Malawi on Agricultural Growth and Market Town Development" held in Lilongwe, Malawi, on May 16- 19, 1988. 3 Introduction The structural adjustment efforts under way since the early ization. Donor agencies, on the other hand, seek increased 1980s have emphasized the liberalization of agricultural private-sector marketing but, facing government resistance prices and markets and have led to a vigorous debate to privatization, envision cooperatives as a reasonable about the appropriate roles of the private and public "private" sector alternative to parastatals. Perhaps the most sectors. Among the issues raised in this debate are the interesting aspect of these different views is that neither nature and causes of the weaknesses in public-sector party seems to take account of the experience of earlier marketing activities that have dominated agriculture, the efforts to promote cooperatives-i.e., the fundamental types of public sector interventions in pricing and market- importance of a limited role for the state in organizing and ing that may be justified (including whether to tax export- supporting cooperatives and of genuine grassroots partici- crop production and to stabilize producer or consumer pation, and the problems caused by the desire of govern- prices), their effects on efficiency or equity in production ments to control cooperatives to keep political power and consumption, and the cost of such interventions centralized and to use them as a means of patronage relative to alternative means of achieving the same objec- distribution. tives. Advocates of market liberalization argue that the Thus economic decentralization sought by donors failure of public marketing institutions to operate efficiently through the use of market mechanisms cannot be seen in is evidence of the need to privatize a wide range of isolation of the ensuing distribution of political power. marketing activities and to press the public parastatals that Finally, the conditions needed for the private sector to may remain to undertake solely profit-making activities. operate efficiently, including free entry, information, and Those who support some forms of interventions ask for factor mobility, access to credit and transport of private evidence on the causes of parastatal losses and point to traders and small producers must also be considered. The the need to perform certain inherently loss-making but paper documents that it is not so much the principle of socially desirable functions. They also point to the absence privatization as the sequencing, phasing, and pace of of markets for risks or information in support of their privatization and price liberalization that have been the argument. problamatic issues. Donors-who indiscriminately sup- The role of cooperatives is also raised frequently in the ported the growth of the public sector in the 1970s-have context of examining appropriate marketing institutions and not shown adequate appreciation of the preconditions arrangements. In the 1960s, many newly independent necessary for the private sector to operate competitively or governments as well as donors promoted cooperatives as a of the steps needed to be taken by the public sector to potential source of decentralized grassroots participation in ensure the competitiveness of the private sector. agricultural credit, input, and commodity markets. As To illustrate these points and to explore their policy cooperatives began to gain political and economic power, implications, the paper examines the causes of state however, governments often perceived them as a threat to intervention prior to independence as well as post- their own central political control. In response, governments independence experience with marketing parastatals and sought to increase their control of cooperatives and, in so cooperatives in Cameroon, Kenya, Malawi, Nigeria, Senegal, doing, undermined their effectiveness. By the 1970s, most and Tanzania.' This analysis is followed by an overview of cooperatives had become state-dominated entities that did the conten- and outcomes of the marketing and pricing not represent the interests of their membership-espe- liberalization programs. cially not of the small-scale farmers. Despite this experi- The focus is on the pricing and marketing of traditional ence in the adjustment dialogue, cooperatives have come export and food crops, which in the six MADIA countries to be regarded by donors as a "middle ground" between constitute well over 90 percent of the area harvested, value the roles of the public and private sectors. Governments added, and employment created in agriculture. The paper are under pressure to abandon the single channel public is not concerned with dairy, livestock, or horticultural marketing network, yet they are fearful of losing control of crops-areas in which government intervention has been politically and economically important functions. They see less obtrusive and in which the private sector has played an * cooperatives as a more acceptable alternative than privat- important role. 4 The Genesis and Evolution of Intervention in Agricultural Marketing Agricultural Marketing at Independence who were relatively more efficient (Lele and Agarwal 1989), Many of the restrictive practices that characterize agricul- enabled European farmers to maintain a monopoly on tural marketing in the MADIA countries have their origins in production of such crops as tea and coffee in Kenya. Lele the structure of the agricultural sector during the colonial and Agarwal (1989) have shown how postcolonial policies era. The evolution of these practices reflected a prejudice with regard to smallholders and large-scale African produc- against the trading function in general and was influenced ers who replaced European farmers have varied between by economic relationships between classes and the public Kenya and Malawi. During the colonial period, marketing sector's need for revenues (neither of which factors was boards in Kenya emerged only in the areas where Euro- unique to Africa). Thus Bauer referring to the Nowell pean production was dominant. Bates (1989) argues that Commission's investigation of agricultural marketing in West although the efforts of European farmers to gain an Africa, noted that: economic advantage over African farmers in Kenya took The Commission criticized the seemingly chaotic and several forms, one of the most durable was the operation unorganized system of marketing; and the collective of agricultural boards or parastatals, e.g., the Land and marketing agencies which it proposed had the Agriculture Bank, the Wheat Board, the Maize Board, the appearance of simplicity, efficiency and neatness. Coffee Board, and the Kenya Cooperative Creameries, each Moreover, the establishment of such agencies would of which were formed at the initiative of large-scale farmers result in the creation of large-scale organizations and and designed to enhance the profitability of production. of important official positions (Bauer 1967, p.265). Bates cites the Maize Board as representative of this In part, this preference for "organized" marketing stemmed phenomenon. from a misunderstanding of the operation of markets. Again, The basic instrument of their efforts was the Kenya Bauer referred to the observations of many colonial Farmers' Association (KFA), an organization first administrators in Nigeria and Ghana: formed in the 1912/13 crop year to dispose of the The number and variety of intermediaries have been maize crop. The main obstacle to securing a high much criticized by official and unofficial observers. price for the maize crop was competition. Particularly They are condemned as wasteful and are said to be troublesome were the small- scale African producers responsible for wide distributive margins both in the .... Without the regulatory power of the state, there sale of merchandise and the purchase of produce was little the KFA could do to compel cooperation by (Bauer 1967, p.22).2 its members, much less to restrict sales by non- Although an explanation based on the desire to achieve members, and thereby secure higher prices (Bates order and efficiency had appeal at one level, it is also clear 1989, pp. 16-17). that marketing restrictions were used extensively during the colonial period to create economic rents through trade. In To ensure food supplies during the Second World War, West Africa, the diversity and small-scale nature of food- the Kenyan government sought to guarantee the profitabil- crop trade, combined with the large number of producers, ity of foad production by empowering the KFA to act as an made it difficult to control or regulate. Consequently, agent of the government with power to regulate the marketing boards were confined to export crops-an area movement of produce and product prices. The KFA took that European trading companies dominated (Harriss 1981). charge of maize production, storage, and transport on Thus, agricultural parastatals in West Africa were created to behalf of the government's maize board. The association gain control of the acquisition and distribution of agricul- formed by the large-scale farmers was thus legally accorded tural (primarily export) products-some argue at the monopsony powers and it was mandated to secure prices expense of the producers (Keene, Monk, and Associates, high enough to guarantee a good profit to its members. In Inc. 1984, p.i). In her review of West African trade, however, the case of Kenya, therefore, it can be observed with some Harriss observes that investments in marketing and trans- irony that "the dominance of the colonial state meant that, port infrastructure prior to the Second World War had the when independence came, the government of Kenya was result of encouraging domestic agricultural production even endowed with public institutions for the promotion and though they were undertaken primarily in the interest of regulation of agriculture that were unusually rich and European trading companies. The establishment of some of sophisticated" (Bates 1989, p. 22). the West African marketing boards was also motivated by Certainly this long history of state intervention (Harriss declining world market prices for export crops and by the 1981, p.15), the African governments' perceptions of the growing distrust of private trading companies by African "success" of public-sector marketing institutions to foster producers. The internally inconsistent objectives emanating commercial agricultural production, their ability to mobilize from the desire of the British government to maintain resources to maintain operations of the colonial govern- popular support for colonial regimes and to mobilize ments, their desire to maintain control over the marketing revenues during the war thus played a part (Harriss 1981, of politically strategic commodities such as maize, and pp. 23-24). doubts about the efficacy of the traditional private sector in In East Africa as well, the structure of European economic conducting organized trade all contributed to the postinde- interests determined the nature of government interven- pendence prejudice against private trade and to the tion in marketing. Colonial agriculture was divided into continuation of public control and regulation. This predilec- estate and smallholder subsectors. The restrictions tion for state-dominated marketing structures manifested imposed on production of export crops by smallholders, itself in each of the MADIA countries. 5 Growth of Marketing Parastatals in the public sector. Donor staff also operate from distant head- Postcolonial Period quarters in developed countries rather than on the basis of Although the growth of public-sector marketing in the 1970s insights developed from prolonged field experience in the had obvious appeal for governments in extending their countries which they assist. control as well as benefits to larger segments of the Again, the World Bank review offers interesting insights: economy and society, it could not have been accomplished ... there has never been an explicit Bank policy either on such a scale without significant donor support. Between in favor of or against parastatal organizations, it is 1970 and 1987, the value of external assistance constituted evident that a marked change has occurred in the way between 35 percent and 70 percent of government expen- these organizations are regarded. In the late 1960s ditures in MADIA countries (Lele and lain 1989). In general, and 1970s, parastatals, many of them inherited from Tanzania and Senegal, the countries that received the most the colonial powers, were accepted fairly uncritically. foreign assistance, were also those with the greatest growth Where deficiencies of hardware or management were of parastatals. In Tanzania, nearly 400 parastatals handled identified, appraisal missions typically concluded that production, processing, transport, and marketing of goods these could be remedied by project investments or and services by the end of the 1970s; and the prices of by the introduction of technical assistance personnel nearly 1,000 commodities were controlled. A recent World at strategic points (World Bank 1988, p. iii). Bank report documents that, over the period 1974-1985, Privatization programs also need to identify clearly who Sub-Saharan Africa received more World Bank support for the likely actors and beneficiaries are, and what the parastatals than either Asia or Latin America, involving sociopolitical and employment implications of privatization some 48 food-crop projects, 45 export-crop projects, and 17 will be vis-a-vis the activities of the indigenous population, livestock projects in Africa that had some marketing the public sector, etc.-and precisely how much efficiency components (World Bank 1988, p. 11). gain can be expected over what time period. These often The concern of donors about these parastatals in the are not articulated explicitly, and their implications for the course of liberalization in the 1980s must thus be viewed in pacing of privatization policy are not adequately consid- the context of the vested interests donors helped to create. ered. For example, the promotion of nonindigenous traders In many ways, the aid community tended to accept the such as the Indians and the Lebanese and enterprising parastatal structures for the same reasons that they local ethnic groups (e.g., the Kikuyu, the Chaga, the Ibo, or appealed to governments-they were easier to work with the Bamileke) who are already relatively more dominant than the private sector because they were controllable and commercially are often unacceptable to governments on limited in number. As the World Bank's review observes: political grounds. Relatedly, governments feel the political Parastatal projects were easy to design and appraise. need to develop entrepreneurial talent among those indige- They often had a legal monopoly of trade, so that nous ethnic groups that have had less exposure to the assistance was clearly being provided to the only commercial economy. Due to the political sensitivity of the orgar.ization available-and one that was approved racial and ethnic dimensions of privatization these are and supported by government. Centralized decision rarely discussed in policy dialogue. Similarly, it is important making meant that the principles governing the to distinguish between the small-scale "market women," project could be sorted out fairly easily with the and the handful of large-scale indigenous traders or military responsible manager, and ties to a parent ministry officials, and the external multinational corporations that meant that the government's endorsement of the might be the beneficiaries of privatization and might appraisal design could be readily obtained (World exercise monopoly power. The internal growth linkages, Bank 1988, p.iii). employment, and income distribution effects of the partic- Parastatals also seemed able to complete tasks that the ipation of the large-scale oligopolistic operators may well indigenous private sector could not, given the stage of be quite different from those involving a large number of Africa's development. The weakness of the local private small actors. sector combined with the clear unwillingness of African Finally, taxation of agriculture was explicitly encouraged governments to let non-indigenous communities (i.e., by donors and external policy advisors until the early 1970s Indians in East Africa and Lebanese in Senegal) achieve a in recognition of the need to mobilize public-sector prominent position in politically sensitive or economically revenues to modernize and industrialize economies (Lele powerful arenas worked in favor of parastatals. Because the and lain 1989). Given the perceived inelastic demand for promotion of parastatal operations was the focus of policy primary commodity exports, taxing agriculture to modernize the steps needed to develop the legal and institutional and diversify economies was considered to be the logical framework (e.g., free entry, information, and mobility) approach. Export pessimism even led donors (together with essential for the development of competitive markets were the recipient African governments) to shift their own rarely taken (Lele 1989b). It takes time to create these investment resources and policy attention away from export conditions. Although it requires relatively little financing on crops toward food production, and toward industrialization the part of donors, it does require a substantial micro-level in a period when foreign aid was rising. Moreover, the 1973- knowledge of the political, sociocultural, technological, 74 food crisis focused attention on food shortages and infrastructure and financial circumstances to intervene reinforced the anti-export crop bias. Only in Kenya and in effectively to create conditions in which the private sector estate agriculture in Malawi did this export pessimism not can operate competitively. Unfortunately, such knowledge is lead to explicit or implicit taxation of agriculture, with the frequently inadequate among donor staff, who have tended result that both countries experienced sustained growth in to dominate the design of projects in circumstances of the production of export crops (Lele 1989a). Malawi, limited internal African capacity for planning and imple- Cameroon, and Senegal taxed smallholders explicitly mentation, and who by-and-large have experience in the through low producer prices for tobacco, coffee, and 6 groundnuts, while Nigeria and Tanzania did so implicitly on grounds of income distribution. Rice prices have since through overvaluation of currency-all with adverse effects fallen, and the economic viability of many previously funded on export production (see Table 1). By the late 1970s, schemes has come into question, although donor financing Nigeria and Tanzania had begun to provide budgetary for them has not necessarily stopped. Several conclusions subsidies to export crop parastatals as a means of compen- emerge from this situation. First, the role of donors in sating for the export taxes obtained through currency facilitating distortions in African agriculture must not be overvaluation, but these remedies did not fully redress the overlooked given the large share of government expendi- deleterious effects of currency overvaluation. tures that are financed by external aid. Second, it is not so A fact that is frequently overlooked, however, is that while much the quantity of public expenditures on agriculture in export agriculture was taxed, food crop operations were Africa that has been inadequate as their quality and often highly subsidized through price stabilization opera- allocation among regions and crops, between capital and tions and fertilizer subsidies. Thus when resource transfers recurrent expenditures, between physical infrastructure and to and out of agriculture for both food and export crops are communications vis-a-vis that devoted to increasing public considered, in most MADIA countries there was no net sector employment in parastatals, etc. taxation of agriculture. For instance, in Fast Africa in the The Economics of Market intervention early 1980s, the losses or overdue payments of the grain- marketing operations alone amounted to 5 billion shillings Risk In agriculture in Kenya and 2.8 billion shillings in Tanzania. These costs of Some of the most influential agricultural economists maize operations were substantially larger than the involved in formulating U.S. policy for the modernization of revenues that governments extracted from export agricul- its agriculture in the 1940s justified price intervention on ture in either of these countries (Lele 1989a, p. 138). the ground that risks in agriculture are greater than in other Similarly, while the government of Malawi taxed smal- sectors.3 Much like their European counterparts working in Iholder tobacco it subsidized smallholder maize and colonial Africa, they considered the assurance of a guaran- fertilizer. In West Africa too, fertilizer subsidies provided teed market essential if farmers were to invest in modern through the public sector in Nigeria reached I billion naira purchased inputs. More recently, the private-sector annually in 1987. Finally, public-sector investments were oriented International Finance Corporation (IFC) -in a also made in large-scale irrigation for food crops through report examining the reasons for the weak performance of River Development Basin Authorities in Nigeria and the agricultural production schemes that it has funded- through various parastatals in Senegal (SAED) and Came- similarly ventures an observation that there is a need for roon (SEMRY). Each of these had a public-sector marketing IFC to recognize that uncertainty is much more important in component, and their operations were subsidized by agriculture than in industry. Production and prices vary over governments and supported by donors. For instance, a much wider range, and often both move in the same donors supported rice irrigation schemes-in part due to direction" (IFC 1989, p.vi). the high prices of rice projected in the 1970s, and in part Rainfed agriculture in Africa of course tends to be far to support the sociopolitical objectives governments riskier than the irrigated agriculture practiced widely in pursued in remote northern parts of West African countries Asia. Furthermore, information regarding the likely failure of Table 1 Ratio of producer to international prices, 1970-86 (calculated at nominal exchange rates) Kenya Malawi Tanzania Smaliholder Smaliholder Estate Smailholder Year Coffee Tea Dark-Fired Burley Flue-Cured Tobacco Cotton Coffee 1970 0.91 0.60 0.22 0.4 0.57 0.43 0.72 1971 0.90 0.67 0.25 0.39 0.68 0.50 0.61 1972 0.98 0.63 0.23 0.40 0.63 0.46 0.57 0.57 1973 0.96 0.60 0.22 0.54 0.86 0.44 0.35 0.43 1974 0.97 0.55 0.23 0.62 0.84 0.42 0.32 0.43 1975 1.01 0.63 0.22 0.47 0.66 0.47 0.51 0.36 1976 0.85 0.57 0.21 0.48 0.70 0.40 0.41 0.30 1977 0.92 0.70 0.26 0.60 0.76 0.42 0.45 0.35 1978 0.94 0.64 0.26 0.50 0.74 0.47 0.55 0.39 1979 0.93 0.66 0.24 0.45 0.65 0.37 0.51 0.29 1980 0.98 0.76 0.23 0.46 0.40 0.35 0.52 0.41 1981 0.84 0.62 0.19 0.73 0.56 0.33 0.61 0.53 1982 0.83 0.56 0.24 0.51 0.50 0.30 0.73 0.52 1983 0.90 0.98 0.23 0.27 0.39 0.38 0.67 0.47 1984 0.80 0.66 025 0.30 0.38 0.27 0.65 0.47 1985 0.88 0.76 0.21 0.26 0.34 0.36 1.07 0.53 1986 0.79 0.69 0.22 0.43 0.45 0.32 1.11 0.33 Sources: International prices from World Bank BESD, Commodity Statistics, for Coffee: "Other Mild Arabica"; Tea: "Average Auction (London)"; Tobacco: "United States All Markets"; Cotton: "Egypt (Liverpool)"; Groundnuts: "Nigerian (London)."Nominal exchange rates from IMF 1987. Producer prices for Kenya: Ministry of Agriculture 1987; Malawi: tobacco prices from Government of Malawi 1988, cotton (grade A) from World Bank 1986a, groundnuts [1970-771 from World Bank 1981 and (1978-86) from Government of Malawi 1988; Tanzania: World Bank 1986b. 7 rains is poorer in Africa than elsewhere, as the develop- Africa that provides quantitative information on crop ment of early warning systems is still at an earlier stage storage behaviour between seasons of rural households, (Lele 1985). Yet improvement of early warning and produc- wholesalers, retailers, or rural communities. Pinckney's work tion information systems has not received the high priority on Kenya-the only exception to this general rule- it deserves in donor support of agricultural sectors in Africa. suggests that such storage might be significant. However, Not surprisingly, weather-induced risks are frequently other evidence suggests that private year-to-year storage correlated with policy-induced risks. For example, domestic may be limited (even where price interventions do not food crop failures lead governments to increase commercial discourage such activity) because of the high costs of or aid-related food imports at short notice,4 and their storage (a minimum of up to 20-25 percent of the value of public distribution to remote areas involves high transport output), the risk that price increases the following year will costs. Such import and distribution tends to be undertaken not cover storage costs (especially given the opportunity by public sector marketing parastatals rather than left to cost of relatively scarce working capital), and the resources the private sector lest all imported commodities be sold in required to meet day-to-day survival needs in rainfed urban areas of population concentration. Moreover, high- agriculture, where savings may be relatively limited. There income public sector entities often are not reimbursed by has been relatively little donor support for the improve- governments for these distribution costs, although govern- ment of storage by rural households and the rural trading ments typically earn revenues on the food aid they receive. sectors prior to introducing market liberalization-reflecting In years of internal supply abundance, which is also a the poor sequencing and phasing of market reforms. In any frequent occurence, marketing parastatals are faced with case, whether private traders will undertake storage on a either defending a market price (which in turn leads to scale necessary to alleviate the supply shortfalls encoun- substantial losses from ensuring year-to-year storage) or to tered by a large share of urban and rural households in exports. Alternatively, letting the market operate in a years of shortages is an issue that needs empirical period of supply abundance-a step that many marketing investigation. boards adopt out of necessity dictated by the shortage of International trade can of course reduce the need for working capital-results in the collapse of the market price internal stocks, and it is frequently argued by donors that and farmer confidence, as in Kenya (Pinckney 1986). African governments have tended to be unwilling to pursue Risks confronted in rainfed agricultural production can be an outward-oriented trade policy. However, if trade orien- divided into those associated with variability in yields and tation is considered in terms of the proportion of GNP the resulting production instability, and those resulting originating in that sector, most small African countries have from variability in prices. Since market prices and yields high proportions in trade (with exports and imports should normally move inversely, they should lead to stable constituting between 40 to 60 percent of GNP). Besides, incomes. However, as the above-mentioned IFC report food imports have constituted a larger share of total observes, this does not always occur; both agricultural imports and internal food needs than in some of the larger supply and demand tend to be highly inelastic in the short Asian countries, for example India and Indonesia-even run, and futures markets do not exist. Besides, under compared to the period prior to Asia's Green Revolution, conditions of subsistence agriculture, where only a small when the susceptibility of Asian countries to droughts and proportion of the total food production is marketed, food shortages was significant. India imported only 10 fluctuations in marketed production tend to be proportion- percent of its total grain consumption in the mid-1960s, and ately greater than those in overall production-i.e., the virtually all of those imports were financed by food aid. In elasticity of marketed surpluses with respect to food Tanzania and Senegal, in contrast, food imports in some production tends to be greater than one. Thus, in the years have constituted up to 40 percent of export earnings. absence of some price support, a given increase in in Africa as a whole, even though food aid has been rising marketed surplus will in all likelihood result in a greater rapidly, commercial food imports nevertheless have than proportionate drop in market prices, especially in accounted for 60 percent of the total food imports in the situations of poor infrastructure (e.g., feeder and trunk mid-1980s. roads, and railways) as well as lack of information, credit, At the same time, the capacity of African countries to transportation, and storage facilities. A joint MADIA-SSATP finance such growing food imports has been decreasing study on the role of rural roads emphasizes the problems due to the neglect of export sectors by governments and created by the poor planning and inadequate maintenance donors alike. Since the mid 1970s both have focused on a of rural roads in the MADIA countries. A sharp drop in narrow conception of food security-one largely involving production tends to result in proportionately greater drop food production efforts in the areas of limited demon- in marketed output and price increases unless stocks can strated potential and without a consistent strategy. Even in be released to bring prices down or the need for imports the case of countries with a relatively successful record on is anticipated in a timely manner. Thus, in absence of an exports (e.g. Kenya), whose ability to finance food imports effective early warning system, a government planning has been relatively stronger, importing has been unreliable, mechanism that ensures adequate imports (although infor- given the dependence of these countries on the availability mal border trade may operate), and without some minimum of scarce foreign exchange. support for prices as one element of a larger modernization Yet food aid is equally unreliable-even when available. strategy, it is unlikely that modernization of agriculture will Improved early warning systems to detect potential food occur. shortages, increased internal production of food, and easier An important research question is the extent to which the access to financing for food imports (as for instance, through private sector undertakes year-to-year stocking, which can a more liberal access to the IMF food facility, and a more reduce the need for public-sector storage and thus mini- reliable and predictable availability of food aid) may mize the cost of government interventions (Lele and encourage liberalization of grain markets by increasing the Candler, 1989). There is virtually no empirical literature on sense of security governments enjoy with regard to food 8 availability in the countries. Some year-to-year price and latter case, instability in the production of food crops and supply stabilization is likely to be necessary-together with a in food prices may cause a substantial and proportionately policy of international trade, and the encouragement of greater instability in the production of annual export crops storage by households and traders as a means to address of low value through an acreage response-since the area the problems of glut and shortages. Donors must thus under export crops typically tends to be smaller in relation consider not only the principle of privatization, but its pace to that under competing staple food crops (see Rai Krishna and pattern, as well as its role relative to public stocks and in Southworth and Johnston 1967). The MADIA study has international trade at early stages of economic also stressed the need to distinguish between crops that development. do and do not require further processing. The former must Controls over interdistrict movements are another symp- provide a stable supply of raw materials to domestic tom of government insecurity-linked tc their inability to industries to preclude losses by the processors through command surpluses in the public sector in periods of underutilization of capacity caused by production fluctua- shortages to influence market prices and ensure supplies. tions leading to higher processing margins and lower Command of internal production for public distribution has producer prices. a high political value, as grains supplied under food aid Because the structure of export crop marketing requires often are not an adequate substitute for domestic con- some vertical integration and offers fewer opportunities to sumption due to strong consumer preference for local crops producers for alternative selling arrangements, owners of (e.g., the white maize in Kenya, the flint maizes in Malawi, processing operations may have more control over producer etc.) although the urban consumers are of course not by any prices for export crops than in the case of crops for means the poorest segments of the population. The domestic consumption.5 There has been a tendency to political cost governments perceive in not being able to object to taxation of agriculture without separating out the ensure the basic staple foods to such urban populations issue of price stabilization. A high level of taxation should tends to be higher than economists assume; it is strong be opposed, but year-to-year price stabilization at close-to- enough to lead governments to measures to control border prices may be necessary. At times, governments supplies in the public sector. Restrictions on interdistrict have also tended to cross-subsidize food crop operations movements of grain facilitate such control by depressing by taxing export crops. For instance, Malawi has supported prices in the areas of surpluses, although, of course, such smaliholder maize operations by taxing smallholder tobacco restrictions tend to make markets less integrated than they production while it has left estate production of tobacco would otherwise be. Markets become more integrated virtually untaxed (Lele 1989b). Whereas donors have criti- when movement restrictions are removed. However spa- cized Malawi's subsidization of maize prices, they have tially integrated they are, markets do not necessarily ensure objected less to its discriminatory policy of taxation of price stability in the face of unstable supplies and inelastic tobacco, which makes some compensatory relief on food demand referred to earlier. However, since free trade does prices necessary. not ensure the ability of governments to command grain at Price stabilization reasonable prices in poor crop years, maintenance of an adequate buffer stock may be a crucial incentive to Whereas Keynes (like the U.S. agricultural economists, encourage governments to liberalize markets. In this European settlers in Africa, and the IFC) considered wide context, it is noteworthy that in India liberalization of the and rapid fluctuations in the world prices of primary internal grain trade occured after the Green Revolution- commodities to be "one of the greatest evils in inter- once the government was secure about its ability to national trade" (Kanbur 1985), recent theoretical literature command adequate stocks in the public sector to exercise questions the need for price stability from the viewpoints influence on market supplies and prices even in poor crop of both consumers and producers. According to Newbery years. Grain market liberalization was not the cause of and Stiglitz (1981), stabilizing prices may increase income India's Green Revolution (Lele 1973). In reforming policy the variability. We have shown above, however, why this may not complex interactions among production, stocks, and trade always be the case in developing countries. They also argue must be more fully appreciated. that long-run effects of stabilization may be opposite to short-run effects, e.g., price stabilization may encourage Instability In export agriculture production of the commodity to a point where in the long Whereas instability in food crop prices by and large reflects run income decreases, rather than leading to a diversified changes in domestic supply (although in countries such as production strategy which reduces the risks of price Nigeria, erratic imports of rice have also caused consider- variability. They also emphasize, and Behrman concurs, that able price fluctuations), that of export crop prices is the distributive and efficiency effects of price stabilization externally induced. Elsewhere, the NMADIA study has dem- need to be distinguished and that policies to stabilize onstrated the considerable instability of international consumption and income or to improve financial and market prices for primary commodities (Lele 1985), as well futures' (commodity) markets may be more desirable than as the need to make a distinction between perennial those to stabilize prices (Behrman 1987). Others argue that export crops of relatively high value (i.e., tea, coffee, and instability in earnings due to instability in volumes may be cocoa), and annual food and export crops. In the case of greater than instability in earnings due to prices (Wahab perennial export crops, domestic production may be 1985). profitable, when internal prices are not distorted, even at Gains to consumers and producers from price instability the lowest level of international prices. In the case of depend on the relative magnitudes of the supply and annual export crops of comparatively lower value (i.e., demand variances, assumptions about the benefits of risk cotton, groundnuts, and tobacco), however, price competi- reduction, and log-linearity and multiplicativity of distur- tion from domestic food crops may be severe-especially in bances (Gilbert 1986, p. 635). Although some literature circumstances of growing internal demand for food. In the concludes that gains to consumers from price stabilization 9 may be positive, questions remain as to the size of gains to groups than for their high-income counterparts because the producers. Furthermore, the cost of holding the necessary former spend proportionately more of their expenditures stocks is believed to exceed the benefits in the form of on food. This is true even in rural areas. Small producers more stable prices. already tend to give high priority to meeting household Despite these theoretical considerations, why do govern- food requirements through their own production and ments like to stabilize producer and consumer prices? income strategies e.g., by opting for stable but lower yields Virtually every European, North American, and Asian on food crops through mixed cropping, varying planting country that has successfully modernized its agriculture has dates (including "late" planting) to reduce crop failure, guaranteed a minimum producer price for major commod- stressing the production of drought-resistant crops, mixing ities as a means to reduce risks in the adoption of modern livestock with cropping, and earning some off-farm income. technology-a factor not adequately considered in the Where export crops enable producers to be sure of higher theoretical literature, However, as the theoretical literature and more stable income, however, as in the case ut tea and notes, many countries suffer from high budgetary costs of coffee in Kenya or cotton in West Africa, they tend to producer price support programs that have become diffi- allocate a significant share of their acreage to higher-value cult to eliminate because of the producer lobbies devel- crops. Despite the adoption of these strategies, the various oped around them. This leads to greater allocation of MADIA papers document the growing dependence of rural resources to the agricultural sector than market forces households on the market for food purchases out of would warrant-the agricultural surpluses in the EC and the necessity rather than choice, as population pressure is United States being examples. The equity effects of these tending to reduce the area under cultivation (Lele 1989; programs are also less desirable than originally believed. In Lele and Stone 1989). In such situations, price stabilization the United States, for instance, large corporate producers of is also a means of stabilizing income on welfare grounds, agricultural production tend to benefit more from price and governments will continue to control prices on welfare supports and quotas than their smaller family-farm counter- and political grounds-as has been the case in East Africa parts because benefits tend to be proportional to the size through parastatals and in West Africa through trade policy of operations. Moreover, technological change tends to toward rice. become institutionalized in areas of greater profits and lower risks made possible by support programs, tending to Scale of investments perpetuate production increases. Marketing and processing of export crops involve scale Nevertheless there are several reasons why stabilization economies and frequently require lumpy investments in remains popular with governments. First, short and medium processing facilities, transport, and communications. In term effects are of greater interest to them than the likely many African countries, political leaders (e.g., Malawi's long-term adverse effects on resource allocation. Second, President Banda) often became the entrepreneurs, so there forward markets take time to develop and are not a was not an adequate entrepreneurial class with the neces- practical way to deal with risk for countries at early stages sary capital to undertake such operations. Although the of development. And finally, transport and capital markets entrepreneurial class is broader now than previously, there are still poorly developed, are major differences among countries-e.g., between This being said, even in the African context, efficiency of Malawi and Nigeria. Even in Nigeria, however, it is not clear price interventions must be weighed against the costs and why there is not more private investment in modernizing benefits of achieving the same objectives by other means- the processing of oil palm, or in road maintenance capacity. e.g., by improvement in the transportation network, infor- There is abundant evidence to suggest that infrastructu- mation systems, and increased access to credit by small ral investments facilitate the development of markets by farmers and traders-that can increase the competitiveness improving factor mobility and market information and by : of private trade. reducing transportation costs and risks. Despite the bene- - To the extent that some price stabilization may be fits of transport infrastructure, however, underinvestment, attempted, ways must be found to increase cost effective- especially in maintenance, has been characteristic of many of ness of these operations-for example, by increasing the the countries because {i) the benefits are long-term and range within which prices are stabilized rather than having difficult to measure; (ii) the links between market develop- a fixed producer and consumer price, combining minimum ment and transport networks are not fully understood; and stockswith trade (as done by Indonesia), and ensuring that (iii) the concentration of power in central governments surplus profits skimmed off by stabilization schemes are means that local governments do not have the administra- protected for use in the period of low prices by increasing tive and financial capacity to maintain roads ILele, Oyeiide, the control that producers rather than governments exercise et al. 1989). Public-sector investment is urgently needed in on the use of stabilization funds. transport, electrification, and communication. With respect to low-income consumers, the case for price Inadequacy of financial markets stabilization tends to be stronger, in part because the The inadequate development of capital markets poses a ability of governments to organize alternative means of major constraint on intensifying production. Covariance of stabilizing producer income tends to be limited-even risk over large geographical regions with uncertain rainfall though low-income producers themselves diversify their increases the risks of default on repayment and reduces the production and income among food and export crops, scope for development of private financial markets in many livestock, and non-agricultural activities. They also use other parts of semiarid regions-especially because savings tend means of risk minimization including late planting of crops, to be low in situations of low average labor productivity. A production of drought resistant crops, mixed cropping, etc. shortage of cash at critical periods is a constraint in The lowest income consumers spend up to 60 percent or purchasing fertilizers and hiring additional labor for land more of their income on food. Negative income effects of preparation, weeding, and harvesting. Unlike Asia, where food price increases are greater for these low-income informal rural financial markets provide up to 60-70 percent 10 of rural finance, interseasonal capital transfers among and the total expenses incurred in selling the crop either households to facilitate input purchases are less developed internally or externally) is difficult, as data on these costs in Africa. Reflecting the shortage of capital, many recent are incomplete, frequently unreliable, and generally not field studies report very high real interest rates in private comparable among countries6 (see Table 2). In addition, capital markets in Africa (see, for instance, Kelly 1988 on data for public and competitive private sector operations Senegal, and Scarborough 1989 on Tanzania). Also in are rarely available in Africa to permit thorough comparative contrast to Asia, labor contracts are not yet sophisticated analysis of the private and public sectors, the legitimate enough to make up for the financial market failure. costs of marketing, and the avoidable inefficiencies in the Financial markets have of course begun to develop, and public sector. they have become more sophisticated over time in some It is generally recognized that unit marketing costs tend countries and more so (e.g., Kenya) than in others (e.g., to be higher in Africa than in Asia because unit transport Tanzania and Malawi). Nonetheless, as in the case of costs are higher and distances are greater (Lele, Christi- commodity markets, the tendency of African governments ansen, and Kadiresan 1989). While wages in the public to monopolize control of capital in the formal sector sector are commonly believed to be high in Africa, recent through the financial institutions-often with large financial evidence indicates that this perception is not entirely support from donor agencies-has tended to inhibit the correct (Lincauer, Meesook, Suebsaeng 1988). Public sector development and competitive functioning of the private wages are higher than in the private sector for unskilled capital market. The shift to privatization that has occurred labor, but for skilled labor they tend to be lower. Neverthe- during the course of structural adjustment has been so less, the normal costs of marketing, storage, and processing swift that there has not been adequate time allowed for may well tend to be higher in the public than in the private assisting with the development of financial markets or for sector due to excessive employment and other inefficien- assessing the merits of an appropriate role for the public cies in handling, transport, storage, etc. sector (see, for instance, Lele, Christiansen, and Kadiresan Kenya. The performance of Kenya's parastatal sector is 1989). the most interesting because it is mixed. Kenya offers Public-Sector Intervention in Marketing successful examples of semiautonomous agencies (e.g., the In this section, we review the performance of the public- cooperatives) that have beAenhrity and coffee and dairy sector marketing operations. In particular, we evaluate the smallholder production. At the same time, some parastatals extent and causes of inefficiency and their implications for and cooperatives have had a decidedly mediocre record.7 future policy. The most spectacular of the successes is that of KTDA, Marketing costs which is also the most studied crop-marketing organization One of the most pervasive themes of liberalization has in Africa (Paul 1982, pp. 60-62). The causes of this success been a concern about the excessive marketing costs of deserve to be reviewed. parastatals. In this context, a clear distinction needs to be Tea required further processing and adequate supply of made between problems arising from financial conditions raw material was critical to building processing capacity- or from management weaknesses, and those linked to thereby providing a rationale for KTDAs monopoly on political intervention. purchasing the material. The private sector was uninter- Assessing the size of marketing costs (defined here to be ested in investing in processing units for small farmers, the difference between the value of payments to producers since units-given that tea leaf needs to be processed Table 2 Marketing expenses per ton of grain purchased by marketing boards in Malawi, Kenya, and Tanzania, 1972/73-1985/86 (Converted to US$ using real effective exchange rates) Malawi Kenya Tanzania Smaliholder Marketing Smaliholder Marketing Smaliholder Marketing Payments Costs Payments Costs Payments Costs Year (US$) (%) (US$) (%) (US$) (%) (US$) (%) (US$) (%) (US$) (%) 1972/73 120 68 55 32 NA NA NA NA NA NA NA NA 1973/74 115 59 79 41 NA NA NA NA NA NA NA NA 1974/75 129 60 87 40 NA NA NA NA NA NA NA NA 1975/76 208 60 138 40 NA NA NA NA NA NA NA NA 1976/77 163 57 125 43 NA NA NA NA NA NA NA NA 1977/78 182 54 156 46 NA NA NA NA NA NA NA NA 1978/79 198 57 151 43 NA NA NA NA NA NA NA NA 1979/80 221 48 238 52 NA NA NA NA NA NA NA NA 1980/81 192 45 232 55 94 68 90 32 NA NA NA NA 1981/82 148 45 180 55 66 73 61 27 NA NA NA NA 1982/83 133 53 118 47 30 68 60 32 44 59 244 41 1983/84 116 52 105 48 38 58 99 42 72 71 189 29 1984/85 120 56 95 44 65 62 162 38 26 60 221 40 1985/86 145 61 91 39 89 66 97 34 NA NA NA NA Source: Malawi data from ADMARC 1972-87, except for marketing costs for 1974/75-1978/79 from World Bank 1986a. Kenya data from Coopers and Lybrand 1987. Tanzania data from World Bank 1986b. Real effective exchange rates from P. Seka/MADIA. it reasonably quickly after harvesting to maintain high quality Table 3 of the processed product-for smallholders needed to be NCPB financial management, 1980-86 spread out in remote areas of production. Returns would be obtained too far in the future, when smallholder Bank Loans Finance Overdrafts From Government production developed, while investments in processing Costs & Loans CSFC Agency were needed up front (Lele and Meyers 1987). (The same Year ('000 MK) (f000 MK) ('000 MK) ('000 MK) arguments-ensuring adequate supply to processors-have applied for the development of a monopoly of coffee 1980/81 75,000 85,734 711,333 224,843 processing developed in the cooperative sector in Kenya). 1981/82 119,000 140,285 1,316,273 229,627 Lamb and Muller (1982) argue that KTDA does not provide 1982/83 209,000 303,638 1,316,921 252,371 a model of either rural development or public enterprises 1983/84 197,000 144,743 1,345,677 165,933 that can be neatly transferred to other countries, but they 1984/85 320,000 200,561 1,235,677 120,869 do point to four features that are instructive. First, KTDA 1985/86 371,000 262,015 2,235,677 31,050 was able to maintain its organizational autonomy. Its ability Note: CSFC is the government-owned Cereals and Sugar Finance to do so was largely due to financial success, to the Corporation. powerful domestic political constituency that tea growers Source: Coopers and Lybrand 1987. became, and (to a not insignificant extent) to the support the result of a buildup of stocks related to unusually high provided by external sources of financing (the Common- levels of production and increased indirect overhead costs, wealth Development Corporation and the World Bank's i.e., administrative costs. A report by Coopers and Lybrand concessional IDA credit)-especially since initial revenues Associates concludes that these large increases in overhead would not have been adequate to pay back commercial costs "cannot be explained by corresponding increases in debt provided to KTDA. Second, the KTDA exerted consid- either the volume of business activity or the annual level of erable control over the use of its resources, with the result general price inflation in any of the years, with the that resources were applied to tasks that were critical to exception of 1985, suggesting a significant absence of each stage of development. "This stands in marked contrast effective management and control of expenditure within the to many other examples of institutional development, NCPB in the past" (Coopers and Lybrand 1987, Annex 8, p. where the drive for bureaucratic growth often simultane- 5). Thus, despite an increase in the margin per bag-which ously creates both organizational redundancy and under- rose from Ksh 40.25 in 1980 to Ksh 133.55 in 1986-deficits deployment to key functions" (Lamb and Muller 1982, p. 57). increased sharply (Coopers and Lybrand 1987, Annex 8). Third, KTDA had an effective network of accountability built A more comprehensive examination of public sector into all aspects of its operations. This prevented unrealistic enterprises in Kenya (Grosh 1988)-covering 38 parastatals, assessments of potential performance and provided staff including the NCPB-acknowledges that many of them have with necessary incentives. Excellent technical support from been poorly managed. Grosh argues, however, that neither KTDAs two financers, CDC and IDA, played an important the donors' solution of widespread privatization nor the role in enhancing KTDAs professionalism. Finally, economic government's efforts to tighten control over managers is incentives to producers were provided in the form of a likely to constitute a genuine solution to the problems of direct link between producer and world market prices. This many public enterprises. She reaches several conclusions link also served as an incentive for improving the quality of that are consistent with our own observations and relevant tea (Lamb and Muller 1982). for an understanding of the factors affecting parastatal In sharp contrast to the success of KTDA are the performance in countries other than Kenya. First, there is a problems of Kenya's National Cereals and Produce Board wide difference in performance, ranging from "excellent to (NCPB). These are of special interest because, as will be abysmal," among Kenya's public enterprises. When seen later, they are not unique to Kenya. Similar problems assessed in terms of profitability and efficiency, approxi- are faced by Malawi and Tanzania, suggesting that issues mately half of the enterprises have performed well since related to intra- and inter-year price stabilization in the independence.8 Second, "firms which have attempted to case of cereals for the domestic market have a generic subsidize their consumers or suppliers have usually run quality; this is different from crops involving exports, where into financial problems. Their poor financial condition then the principle of residual producer policy seems to apply. causes inefficiency" (Grosh 1988, p. 48). This implies a need Maize marketing and the government's role have been an for adequate financing to account explicitly for the costs of intensely political issue in Kenya. Numerous commissions any subsidies in grain price-stabilization operations. Third, of enquiry established over the years have recommended many of the problems of public enterprises have occurred that the government relinquish its monopsony position in since 1978 and are due to the stringent economic environ- maize marketing and function as a buyer and seller of last ment of the past decade. One of the critical weaknesses resort. Until the latter half of the 1980s, the government these enterprises face in dealing with this environment is virtually ignored this advice. Perhaps due to pressure from their chronic undercapitalization, which often leads to donors, it has now allowed inter-district movement of grain excessive financing charges (see Lele, 1989a; Lele, 1989b). In and has given millers the right to purchase maize directly addition, foreign exchange shortages have caused shortfalls in the market. These measures have been undertaken as a of imported capital goods that also impair efficient opera- result of the escalating marketing costs in the 1980s, tions. Fourth, public enterprises are often required to particularly finance costs (see Table 3). undertake investments or provide services that are moti- Overall NCPB deficits increased by over 100 percent in vated by distributional rather than commercial criteria five years-from Ksh 312 million in 1980/81 to Ksh 647 (Lele, van de Walle, and Gbetibouo (1989) make a similar million in 1985/86. As in other countries, however, the point with respect to cotton parastatals in West Africa). Yet deficits stem from more than one source. In part, they were criticism of their poor performance rarely includes a critical 12 analysis of the legitimate costs of doing developmental Figure 1 business (e.g., construction of feeder roads to expand ADMARC Marketing Costs Per Ton, 1972/73-1986/87 smallholder production in cotton areas). In this context, it is Malawi kwacha per metric ton interesting to note, for example, that the tea roads in Kenya a were financed by the government, whereas in Cameroon, I financing for cotton roads has come largely from SODE- 0 - - Totalcosts COTON-a problem often overlooked in analyzing SODE- COTON's financing. Malawi. In Malawi, the operation of private-sector mar- keting has been weak-in large part because government t restrictions (since 1975) have prevented Indian traders from 140 - residing or working in rural areas, while the lack of ,, - experience and working capital on the part of Malawians Igo, has not filled the void adequately. Not only has the private ,0 / sector been weak, but, in contrast to Kenya, there has not s*o been an effective network of either foreign or domestic 40 4 voluntary agencies operating at the local level. Similarly, 20 grassroots institutions to represent the interests of low- ____._,_,_,_,_,_, _____X_'_' __ income rural households are severely underdeveloped; the 1WN 1174 1a0 1676 177n tI71 1S73 *a0 1*n1 9012 *213 1664 1663 IS 1*s7 government has not actively encouraged efforts like the Figure 2 Harambee or the cooperative coffee and dairy institutions ADMARC Marketing Costs, 1972/73-1986/87 in Kenya. In respoise to this lack of private sector activity, Malawi kwacha (millions) the government expanded the operations of ADMARC (the * marketing parastatal) to fill the void-a step which donors supported actively in the 1970s through financial support, 7- but without simultaneously helping to develop the Mala- L Total costs wian private sector. I + Direct costs As can be seen from Figure 1, ADMARC's nominal O Finance costs marketing costs per ton increased steadily over the 1973- / 1987 period, although costs per ton increased only slightly in real terms. For example, ADMARC's average total market- ing costs per ton of purchases for 1972/73-1978/79 were K 85.73. For 1980/81-1986/87, these costs were K 172.89-an increase of about 102 percent (see Figure 2 and Table 2).9 Between 1980 and 1986, the GDP deflator increased by approximately 100 percent. Over this same period, however, there were sustained increases in transport charges, employee salaries, and finance charges. o s 174 laS tS77 t1.7 1S'7S tSS0 1S tSS2 U1t3 1SI4 10 S U 17 In the case of transport costs, the increase was due to devaluation of the currency and to increasingly insecure external transport routes (see Table 4). The increase in mance, suggesting that in evaluating a parastatal's perfor- employee salary costs was due largely to a rise in numbers. mance, role, and policy implications, a broader range of Head-office senior staff increased from 428 to 742 between issues needs to be considered than is typically included. 1980/81-1982/83 and 1985/86-1987/88, while junior staff Although the rise in the number of employees per ton of increased over the same period from zero to 1,224. During purchases seems excessive, in part this represents an effort this time, the average number of staff members in the field by ADMARC to increase the number of seasonal markets in increased by 58 percent, from 16,095 to 24,089, but average order to provide less expensive access (in terms of annual purchases during the same period increased by only transport costs) to purchased inputs, sales points for about 5 percent (Deloitte, Haskins, and Sells 1987, Annex 3). outputs, and food for households that are dependent on The increase in finance costs is also noteworthy. For the the market to meet their food deficit. An evaluation of the 1979/80-1986/87 period, the average finance charge was K cost effectiveness of markets must therefore include all the 26.96 as compared to K 9.68 per ton for 1972/73-1978/79 benefits they offer and not simply attribute costs of maize (see Appendix). The steady increase in overdrafts and purchases. The location of markets of course does make a finance charges between 1978/79 and 1986/87 can be seen difference to the efficiency with which these operations are in Table 4.'° The increased finance charges were due largely performed. Donors have argued that in Malawi the location to the increase in producer prices and the resultant was influenced by political factors. Although they maintain decrease in liquidity. Some of these crop price increases that considerable overlap in the catchment areas of these were mandated by donors as part of the structural adjust- markets is likely because of the political decisions sur- ment process. Perhaps more important than the increased rounding their placement, approximately 75-80 percent of finance charges, the liquidity problems led to delays and the rural population is within eight kilometers of a seasonal limits on the amount of crop purchases, which caused market. There is no consensus on this issue, however, as farmers to lose confidence in the ability of ADMARC to the few voluntary agencies operating in Malawi (e.g., function as a buyer of last resort." These problems were OXFAM), have cited some hardships as a result of the not foreseen in the donor assessment of ADMARC's perfor- closing of those markets (see Lele). In the long run, the 13 Table 4 ADMARC's expenses related to crop trading, 1978/79-1986/87 Selling Direct Administrative Long-Term Over- Finance Year Expenses Expenses Expenses Borrowing drafts Costs 1978/79 ('000 MK) 3,189 12,916 4,106 31,244 - 2,996 (MKperTon) 14 58 18 140 13 1979/80 ('000 MK) 3,528 17,707 4,821 37,884 979 4,246 (MK per Ton) 19 97 26 208 5 23 1980/81 (000 MK) 3,584 20,114 5,874 50,995 6,419 6,530 (MKperTon) 19 105 31 265 33 34 1981/82 ('000 MK) 3,459 19,114 6,363 55,587 5,714 6,236 (MK per Ton) 16 89 29 258 26 29 1982/83 ('000 MK) 2,415 18,899 6,758 69,261 14,002 9,269 (MK per Ton) 8 62 22 228 46 31 1983/84 ('000 MK) 6,745 21,524 6,624 66,043 8,308 6,782 (MK per Ton) 23 73 23 226 28 23 1984/85 ('000 MK) 10,107 33,375 8,901 54,387 18,828 6,464 (MK per Ton) 27 88 24 144 50 17 1985/86 ('000 MK) 8,672 33,786 9,091 53,456 29,172 4,054 (MK per Ton) 23 90 24 143 78 11 1986/87 ('000 MK) 10,272 37,961 11,905 81,011 15,721 11,591 (MK per Ton) 42 157 49 335 65 48 Source: ADMARC 1972-87. establishment of an effective network of private traders and on a residual basis, i.e., after taking into account the cost of grassroots institutions, including local government, cooper- operating the parastatals. In the case of maize, the National atives, and voluntary organizations, could provide produc- Milling Corporation (NMC) was required to purchase maize tion inputs and food supplies on a regular basis at a at prices that both encouraged production and sought to reasonable cost. The major donors did not, however, take protect consumers. These price levels made little allowance active interest in such broadbased institutional develop- for the costs of marketing and processing, however. For ment until recently. In any case, it will take at least ten years example, between 1972-75 and 1978-81, maize producer for an effective network of institutions to develop even if it prices increased by 24 percent, but the margin for the is politically permitted by government and effectively conversion of maize to sembe was reduced by about 55 promoted by donors. The latter will depend on whether the percent, and the official consumer price of maize was wholesalers of external aid such as the World Bank and lowered by 23 percent (World Bank 1983). Indeed, in 1981/ USAID develop effective networks with the small domestic 82 government-set retail prices for preferred cereals, and external agencies that can more effectively work with drought staples, and pulses were below NMC's retail costs the micro agents in villages than their larger counterparts. (World Bank 1983; Bryceson 1985; Kaberuka 1984). Otherwise there is a danger that voluntary agencies will Government involvement extended beyond pridng deci- suffer from oversupply of external finance and limited good sions to the appointment of parastatal managers. ideas, as happened with integrated rural development The split responsibility between the Ministry of projects in the 1970s. Until effective grassroots institutions Agriculture and the State House for appointment and emerge, agencies such as ADMARC have an important role supervision of parastatal management is no less an to play in making fertilizer and other inputs available to important factor in explaining inadequate control over farmers-especially those without access to credit or parastatals. The Executive Chairman of the Board of purchasing power. The benefits of this role are difficult to Directors is a Presidential appointee, and the Minister measure. The important issue is how to minimize the costs appoints Board members. The parastatal general of these public sector operations while promoting develop- manager is bureaucratically essentially equal to Prin- ment of other institutions, rather than whether such cipal Secretary of the Ministry. The Ministry does not functions ought to be performed by the government. have the jurisdiction to dismiss a general manager, Tanzanla. Parastatals in Tanzania have been negatively even in cases of flagrant violations of management affected by a range of government policies. For instance, standards, but can only recommend action to the producer prices of export crops were typically determined Board of Directors or the President. Further, when 14 appointments are made centrally, frequently criteria commodities. NMC alone was responsible for over two- other than commercial or managerial acumen seem to thirds of total losses, representing 31 percent of its sales. enter in the choice of a general manager. Since the The parastatals' overdrafts had reached Tsh 5,127 million State House makes CEO appointments across the and accounted for 80 percent of the loans of the National spectrum of the roughly 400 parastatals, its span of Bank of Commerce, the country's only commercial bank. control is far beyond levels which would be consid- It is important to stress that administrative costs resulting ered desirable under reasonably well working systems, from the growth of employment, although excessive, were a of management. It is no wonder, therefore, that the relatively small part of total losses, e.g., only I percent of degree of control needed on a daily basis to ensure sales in the case of NMC. The costs of financing and sales managerial efficiency in the agricultural sector is not (which reflect purchases plus transport and processing exercised and major decisions even about the future costs) accounted for 97 percent of losses in 1980/81. Thus, of the organization itself... remain pending for years even a significant reduction in administrative expenses (World Bank 1983, pp. 84-85). would have had little effect on losses. 'Inefficiency in day-to-day operations, including the weak Senegal. In Senegal, low, variable, and declining rainfall financial and physical performance, were attributed to weak has caused the production of groundnuts to stagnate. At the management and poor technical skills, lack of competition, same time there has been an increase in the area planted externally imposed pricing policy, and the government's to sorghum and millets as population pressure on the land requirement that parastatals provide a complete set of has increased. Elsewhere, Lele, Christiansen, and Kadiresan social services for their employees-including education (1989) have demonstrated the devastating effect of the and medical facilities, work and private transportation, elimination in 1980 of ONCAD (the marketing parastatal) on prepared food services, provision stores, mechanical work- the network of input distribution and output purchases in shops, and sports teams.'2 Given this interference, the the vital Groundnut Basin. The collapse in input supply has productivity of parastatals declined significantly between been complicated by the reluctance of the private sector to 1974 and 1981. The volume handled by agricultural parasta- supply production inputs on credit, although the private tals increased by 18 percent, whereas parastatal employ- sector has been active in the purchase of groundnuts for ment increased by 37 percent, leading to a decline in labor crushing. Market prices of groundnut oil have been increas- productivity of 14 percent. Although the NMC increased its ing relative to what the processing companies (e.g., SONA- labor productivity noticeably between 1974/75 and 1977/78, COS) that have replaced ONCAD will offer for groundnuts. processing volume decreased sharply between 1977/78 and Government disbanded private Lebanese traders in the 1980/81, with the result that the same number of employ- mid-1960s, establishing OCA and later ONCAD, each of ees handled roughly half the volume. Only two parastatals which operated as a vertically integrated agency that decreased employment between 1974 and 1981. For all supplied inputs on credit and purchased output. Each of parastatals, a decline in volume of 17 percent was accompa- these suffered from highly variable repayment rates of nied by a decline in employment of only I percent during credit as well as highly variable marketing margins for the same period (see Figure 3). groundnuts offered by small farmers (see Table 5). As a result of these inefficiencies, most crop parastatals accumulated large losses. In 1980/81, only the coffee and sugar parastatals showed a profit; the remaining nine Table 5 showed combined losses of Tsh 692 million (US$84 million), Groundnut producer prices and marketing board margins in which is equal to 21 percent of the value of their processed Senegal, 1968/69-1983/84 Net Producer Price Marketing Bd Margin Marketing % of Sale % of Sale Sale Price Figure 3 Year (F/Kg) Price (F/Kg) Price (F/Kg) Change in Tanzania's Parastatal Production, Manpower, and 1968/69 16.50 67% 8.30 33% 24.80 Productivity, 1974-81 1969/70 16.50 54% 13.85 46% 30.35 Percent change 1970/71 17.60 46% 20.74 54% 38.34 240 1971/72 22.00 64% 12.12 36% 34.12 120- s > 7 1972/73 22.00 53% 19.90 47% 41.90 lAo - 1973/74 25.50 61% 16.40 39% 41.90 tOC - 9 s < 1974/75 40.00 76% 12.61 24% 52.61 lIC - 1975/76 40.00 75% 13.00 25% 53.00 140 - 1976/77 40.00 72% 15.80 28% 55.80 lao- 1977/78 40.00 73% 14.90 27% 54.90 100 - 1978/79 40.00 71% 16.00 29% 56.00 30 - > > N E9 / fi A 1979/80 43.00 68% 20.10 32% 63.10 50 ~~~~~~~~~~~~~~~~~~~~1980/81 45.00 54% 38.00 46% 83.00 40 1 N981/82 60.00 68% 28.00 32% 88.00 20 - V s: _ \ N N / < > z NK 1982/83 60.00 67% 30.00 33% 90.00 20 - S t Z S m z L r L r E L 1983/84 50.00 51% 48.50 49% 98.50 -20 Notes: -40o - S 1 These are conservative official estimates. The actual price received by -80O the farmer is much lower when the value of losses and impurities -oo - = generally imputed in official withholdings is taken into account. 2 These are prices paid by the processing plants to marketing boards. Production Manpower Productivity Source: Jammeh 1987. i5 The excessively high marketing costs that characterized stressed that many of the policies that contributed to this ONCAD were attributable to mismanagement and corrup- inefficiency were beyond the parastatal's control. tion, including rapid expansion of staff (Jammeh 1988). For The relative importance of the cost of increased employ- example, ONCAD's staff tripled from 400-500 full-time staff ment, debt forgiveness, and unpredictable marketed vol- in 1966 to 1,800 in 1968, 2,097 in 1974, and 2,964 in 1979. umes in explaining the financial problems of ONCAD is not During this period salaries accounted for over half of the known, as the data for ONCAD are not available to make agency's operating expenses. This excessive growth in such judgments. Nevertheless, it is clear from the recent personnel could not be justified by increases in ONCAD's experience with privatization (discussed later) that although marketing functions or by increases in groundnut transac- private traders tend to be active in output marketing, the tions but was due to political pressure to increase employ- salient issue is how to develop an integrated delivery ment. The impact of these increased operating costs on system that is not monopolistic but that ensures repayment marketing margins can be seen (see Table 5) in terms of the of credit provided for inputs-given that the private traders steady increase in margins charged per kilogram of product tend to be unwilling to provide credit for inputs, and the dating from 1969/70 to 1973/74-after which margins producers unable to purchase inputs without credit. declined from 1974 to 1978. On balance, in each of these four countries there are More recently, the government has favored a strong examples of government intervention that contributes to diversification program to reduce the risk of relying solely the inefficiency of parastatal operations. The solution, on groundnuts. Investments in agricultural production and however, does not lie exclusively with privatization. Instead, those in industry and construction increased. Investments at least two steps are necessary to strengthen parastatals: within agriculture shifted away from groundnut production (i) greater independence from central government control, toward other crops, rural development projects, irrigation, and (ii) explicit subsidy operations performed on behalf of and groundnut processing. As a result, operating capacity of the public sector for which criteria and incurred costs are processing firms grew from 695,000 tons in 1976 to 895,000 made overt. With these changes, a parastatal will be able to tons in 1987, while the actual capacity used as a percentage achieve greater operational efficiency, offer greater compe- of total operating capacity fell from 95.5 percent to 59.2 tition to the private sector, and provide the government percent during the same period (see Figure 4). with development functions (e.g., price stabilization) at an Marketing costs were lowest during the period in which explicit cost. private traders were allowed to compete with the cooper- Instability of agricultural institutions atives (lammeh 1988). In addition, the monopsonistic A crucial factor for improving the performance and produc- parastatal had several policies that served as disincentives tivity of farmers, especially those with small-scale opera- to producers, including onerous procedures for collecting tions, is the quality of institutional support.13 The instability credit provided for inputs. The result of these procedures of agricultural institutions has contributed to undermining was to cause serious delays in payments by cooperatives, farmers confidence in them and, therefore, their willingness thereby encouraging producers to seek alternative market- to rely on factor and output services. Examples from ing channels, even if on disadvantageous terms. Finally, Tanzania and Senegal illustrate this point. abuses and irregularities in the parastatal's input distribu- In Tanzania, marketing institutions have been in a state of tion program were common: shortages or late arrival of nearly continuous change since independence, with most often damaged inputs, diversion of needed supplies onto changes directed toward achieving greater central control the parallel market for personal gain, regional imbalances over agricultural production and marketing. As Candler favoring the Groundnut Basin, and patronage. In some (1986) shows, in virtually every year between 1961 and the cases, inputs requested by cooperative officials would early 1980s, there was a major change in marketing arrange- remain in warehouses while being charged to members. ments in Tanzania. Following independence in 1961, Tanza- Therefore, although inefficiency was a problem, it must be nia experienced extensive growth in cooperatives, which Figure 4 the government encouraged as a means of counteracting Senegal's Groundnut Processing Capacity and Utilization, the dominance of Asian traders. In 1963, the government 1976-87 created the National Agricultural Production Board (NAPB) Thousand tons as the coordinating institution for cooperative grain trade. _____________________________________ - By 1966, the cooperatives were supplying almost all of NAPB maize purchases (Bryceson 1985, p. 56). In 1967, the Arusha no . Z Capacity MS Capacity used Declaration called for the creation of multipurpose cooper- 700 , ative societies to replace the marketing cooperatives, as the former were accused of promoting capitalist relations. In tooo. ffi > 7 > ffi ffi 53 ffi S z 7 z a the following year, guidelines were implemented that effectively transferred control of the cooperatives from 500- ffi > ffi > ffi ffi < fi ffi fi / / / t < individual boards to state-appointed managers. This was part of an increased government role in marketing, partly in order to collect revenue, that became especially important zoo. :o > / \ / / 1 A / oN / N / 7 z N after the abolition of the direct tax in 1969 and was reflected in the creation of new crop authorities. In 1971, a 200 - ffi > ffi < ffi > ffi > 2 ffi R ffi > ffi > ffi M 7 M decision was taken to treat Ujamaa villages as multipurpose 100 / > / > / \ / >1 / 1 / >1 t \1 z N z S cooperatives, thereby further confusing the role of existing cooperatives. Over the next five years, numerous decisions a. is- 8 s7 s ols olls 1166 18- 1616os Il. were taken that profoundly changed the nature of cooper- 1376 1677 1tna 1660 S| O *| U 8-1X 0) atives. By 1976, all cooperative unions had been abolished 16 and replaced by village cooperatives and crop parastatals input supplies. When ONCAD was abolished in 1980, its (Candler 1986, pp. 6-10). smallholder credit program was eliminated and to date has Dissatisfaction with NMC mounted quickly, due mainly to not been replaced by a program of comparable coverage. the tremendous year-to-year fluctuations in the volume The institutional instability experienced by Tanzania and NMC handled due to climatic factors, frequency of late or Senegal obviously has undermined smallholder confidence missed payments for purchases, major financial losses that in the government and contributed to unreliable sources of resulted from being required to sell at prices below cost, supply for parastatals and agro-processors. This, in turn, has large storage losses (estimated by some at up to 30 affected the availability of working capital for parastatals, percent), and corruption. In the face of rising parastatal interest costs, liquidity problems, and problems of excess losses-financed by overdrafts on the National Bank of processing capacity. It is this set of problems that has Commerce that were so large as to be considered a threat attracted the attention of donors interested in improving to the NBC's financial stability-and continued dissatisfac- the performance of agricultural institutions. A common tion with performance, a commission was appointed in 1980 solution has been to restructure the parastatals, privatize to study the situation, and this led to the reestablishment several of the functions performed by the marketing board of cooperatives in 1982. Since 1986, cooperatives have been (e.g., input distribution and output marketing), and elimi- reinstituted, crop parastatals have been replaced by nate or scale back other functions (e.g., food security and commodity boards, and the private sector is being allowed buyer/seller of last resort operations). a greater role. Lessons from experience with parastatals Another example of instability in Tanzania comes from so from exaerine th paa tah u the experience with cotton marketing. The Cotton Lint and Do parastatals have a role to play-given that they suffer Seed Marketing Board (CLSMB) was created in 1955 to from numerous problems? The World Bank's review of coordinate cotton-sector activities and market seed and agricultural marketing experience argues: lint, while cooperatives performed ginning and oil refining; Current disenchantment with parastatals should not together, these accounted for much of the success of be allowed to obscure the important role they have smallholder cotton during the 1950s and 1960s. The CLSMB played in developing countries in Africa and Asia. was replaced in 1973 by the Tanzania Cotton Authority They have been the dominant market force for export (TCA), which centralized all cotton activities and then took products and large scale grain marketing, and were over the cooperatives' functions when they were dissolved often inherited from the colonial power. Sometimes, in 1976. In 1984, TCA transferred responsibility for some parastatals have been used to wrest control of the ginneries and oil mills to the regional farmers corporation, marketing system from ethnic minorities-in which whose shareholders were Ujamaa villagers. In 1985, the TCA case political considerations may well have overrid- was dissolved and replaced by the Tanzania Cotton Market- den concerns for efficiency. In other cases, govern- ing Board (TCMB), but without responsibility for ginneries ments seem to have felt that it was easier to replace and primary marketing, which were returned to the newly an oligopolistic marketing structure than to regulate it. restored cooperatives (Lele, van de Walle, and Gbetibouo In yet other cases, governments have seen parastatals 1989, p. 25). as a way of ensuring government "control" of domestic These reorganizations inevitably led to complete confu- food supplies. They have been unwilling to leave this sion on the part of producers as to which agency was to vital function, especially stockholding, in private hands serve them and resulted in delays in payments to produc- (World Bank 1988, pp. 7-8). ers. Furthermore, the separation of responsibility for credit, A USAID study that examined the problems with parastatals input distribution, and farm-gate purchasing contributed to summarized the debate about parastatals as follows: extremely low repayment rates, which aggravated the Some economists argue that agricultural parastatals financial difficulties of crop parastatals. More recently (1986- are unwarranted governmental intrusions in the 87), markets for inputs and products, and should therefore ... excellent weather and structural adjustment policy be dismantled forthwith to make way for a laissez- changes, including an improved producer price (espe- faire market economy. Others, noting the volatile cially relative to maize), led to a bumper crop... nature of production, the inadequate distribution and However, institutional factors constrained the manage- marketing systems, arbitrary and rapidly changing ment of the resulting supplies. TCMB purchased less policies, and the political instability of many of these than two-thirds of the crop due to lack of funds, nations, argue that because the risks to entrepreneurs transportation problems, and weaknesses of the are so great, private interests will not organize cooperatives (Lele, van de Walle, and Gbetibouo 1989, effective markets, so parastatals are necessary to fill p. 25). the void. They contend that parastatals per se are not In Senegal, the groundnut sector was hurt by several the problem, rather, the ways in which they are institutional changes, the most important of which was the (mis)managed and the uses to which they are put, institutional arrangements for services. In this regard, the especially to reward rent-seeking (Keene, Monk and changes in Senegal were reminiscent of those in Tanzania Associates 1984, p. 12). because they discouraged private trade and created a Accepting these views creates a much more difficult parastatal (ONCAD) as a substitute. ONCAD encountered problem for policymakers. Instead of simply encouraging financial difficulties when it was required by the govern- privatization, it is necessary to explore solutions to the ment to forgive the debts of drought-affected producers but problem of political intervention in parastatals to ensure was not reimbursed. At the same time, mismanagement of competition among different forms of institutions. We will funds and overexpansion of its staff due to political return to this subject after examining the experience of the pressures contributed to problems with the reliability of MADIA countries with cooperatives. 17 Cooperadves The importance of agricultural marketing cooperatives in domestic consumption or sold easily in rural markets- the context of recent policy reforms is that, as stated earlier, making a centralized marketing facility easier to organize. donors and governments see them as a second best Moreover, crops that require processing provide scope for alternative to parastatals. Although this perception contains economies of scale where the value added is usually some elements of truth, it is also a reason for many of the substantial. In Kenya, the major crops other than coffee that problems that have plagued so many cooperatives in the are marketed through cooperatives are (with the exception MADIA countries. The conclusion that emerges from the of milk, which is considered a cash crop) predominantly experience with cooperatives is that their all-too-common export crops. failure is due to the nature and extent of public sector Promotion of food cooperatives continues to receive involvement in their management and operation. The enthusiastic support from a broad range of interests, motivations for public sector intervention are similar to despite their limited success. To date, with the exception of those for involvement in marketing parastatals: control of maize in certain surplus areas, cooperatives have marketed marketing channels, financial resources, and political power. only'small amounts of food crops. In Africa, cooperative Therefore, although public sector support is typically crucial management has rarely been found to be efficient when to the success of cooperatives, the nature of intervention as commodities are bulky and low-value (e.g., maize), or have it is commonly practiced serves to undermine the benefits complicated and expensive processing requirements (e.g., of and the incentives for grassroots participation.'4 cotton and sugar). To understand the nature of the conflict between the In Tanzania, strong grassroots cooperatives were encour- ingredients required for successful cooperatives and the aged before and immediately after independence, with the appropriate role for the public sector, it is necessary to result that they grew in political strength to the point of distinguish between what is necessary and what is being willing to challenge the authority of the ruling party- excessive. first TANU and later CCM.'6 Subsequently, however, their The broader concept of co-operation... acknowledges political power came to be perceived as a threat to the the interaction between economic and socio-political existing authorities and caused the government to increase power and, therefore, recognizes the frequent need its control over the cooperative movement. As the members for structural change or for political mobilization for lost control over the operations of the cooperatives, the co-operatives to be able to benefit the poor. Accord- cooperatives became less effective in furthering the inter- ing to the broader view, even with such change or ests of their members. For several reasons, including the innovation, but particularly in the absence of either, evolving political philosophy of government and complaints paternalism and external assistance in the form of of increased corruption, the government abolished cooper- leadership, management and finances are inevitable atives in 1976 and replaced them with parastatal crop as a step towards more voluntary and self-reliant authorities and, as part of villagization, with small village cooperatives in the long run (Lele 1981, p. 58). societies (Bryceson 1985; FAO 19871. The measurable consequences of this change in terms of agricultural A cooperative that is so centralized in its management production were so disastrous that in 1982 parliament and decision-making authority as to assure the state passed legislation in 1982 reestablishing cooperatives. sufficient control to mitigate any perceived threat to the Although the lessons of the early 1970s with regard to the state's power will typically fail to meet the needs of a consequences of too much government intervention in the voluntary and active membership. As a consequence, operations of cooperatives are clear, it is by no means cooperatives have the best chance of success when the certain that the new cooperatives will be successful. The political leadership of the country is secure and willing to attitude of the government in the short run is more both tolerate groups with divergent interests and at the favorable than in the past; yet there is also evidence that in same time provide technical and training support for the the long run the government views cooperatives as a tool cooperatives' operations.'5 of social policy-that membership in cooperatives is not Among the MADIA countries, Kenya has provided the voluntary but compulsory. in addition, there are several most encouragement for cooperatives by developing a more technical factors that will hinder the prospects of highly decentralized system that articulates and responds success for cooperatives. For example, after a hiatus of a to producer interests (Alila 1985). As a result, cooperatives decade, the cooperatives were poorly prepared for taking represent about 50 percent of small-farm households. The over responsibility for agricultural marketing; the agricul- coffee processing and marketing cooperatives are by far the tural economy is weak, which will make it difficult for most active, accounting for half the membership in agricul- cooperatives to maintain financial viability; there is some tural cooperatives and 71 percent of the turnover (Lele and evidence of overstaffing; and only a limited amount of Meyers 1987). technical assistance is available to cooperatives (FAO 1987, Experience suggests that cooperatives dealing with pp. 17-18). export crops have been more successful than those concen- Hanak and Loft's (1987) comparison of the performance of trating on subsistence-related food crops. Many export cooperatives in Kenya and Tanzania is especially helpful for crops require further processing and cannot be used for understanding the ingredients of successful cooperative 18 activities. They point out that, at independence, a grass- sion occurred, in areas where tont'nes (informal savings and roots cooperative movement was more advanced in Tanza- lending clubs) were already established. As a result, the nia than in Kenya; whereas in Tanzania small farmers had credit unions have been able to formalize an already well- been allowed to produce export crops, in Kenya production established institution (tontines) in ways that are beneficial of export crops was confined to estates. Kenya, however, to members. For example, before a newly formed cooper- turned out to be more tolerant of cooperatives than ative can become a member of the Cooperative Credit Tanzania. Referring to the efforts of Nordic donors to Union League (the umbrella organization that provides improve the performance and management of cotton management support, training, audits, and insurance ser- cooperation in Tanzania, Hanak and Loft argue that the vices), it must operate as a pre-cooperative savings club to improvement in performance was dramatic-especially in establish member confidence. Although credit unions are comparison with cooperatives in other parts of the country. formal, registered cooperatives, they are rarely subjected to The difference in performance in the two countries regulations or supervision by government cooperative stemmed from the role of government. The Kenyan and agencies or central bank authorities. It is the presence of an Tanzanian governments and the Nordic donor group had a effective umbrella organization-providing technical sup- shared perception of cooperatives as an inherently appro- port and supervision-in conjunction with the absence of priate institutional form. Both countries propped up weak political interference that explains much of the success of cooperatives and developed mechanisms for external the credit unions. regulation. Kenya, however, built on the more successful The experience of the credit unions is in contrast to that cooperatives, whereas Tanzania retained many cooperatives of the cocoa-marketing cooperatives in the Center and despite their poor performance. The Nordic donors con- South provinces, which have been exposed to considerable tinued to support these poor performers despite "the government intervention, including the appointment of Tanzanian government's clear transgression of the by laws of government-seconded civil servants as managers and the International Co-operative Alliance in its 1976 action" senior staff. In general, government seems to regard these because of the general wave of support for Tanzania. In marketing cooperatives as parastatals, reserving the right ol Kenya, by contrast, the government's own intentions to approval for budgets of individual cooperatives. This focus on the less successful agricultural areas was essential pattern of state intervention continues despite attempts at to the progress of management capabilities in the Kenyan reform. coffee, dairy, and pyrethrum societies and unions (Hanak Prior to independence, there was an active independent and Loft 1987). cooperative movement. By 1960, there were 217 primary After twenty years of work with cooperatives, however, the centers and higher-level cooperatives for marketing cocoa. Nordic donors have little to show for the effort outside of In 1963, most of these cooperatives were dissolved, how- the coffee/dairy belt. Hanak and Loft argue that Nordic ever. According to one source, their independence was evaluations tended to overlook the fundamental issue of perceived as a threat to the state (GTZ 1986, p. 10). Ten the viability of cooperatives in situations where they handle years later, the central government undertook a series of high-bulk and low-value crops grown in dispersed areas, or reforms directed at cooperatives: crops with complicated and expensive processing require- Reforms of the period were conceived largely in a top- ments. Hanak and Loft conclude that, "unless the Nordic down manner, faced with the necessity, as the Minis- donors openly begin to reassess the blanket assumption try saw it, of rapidly establishing a cooperative that cooperatives are a good thing everywhere, there are "presence" at all primary cocoa markets to protect the dangers that future promotion will again turn to areas where individual seller from manipulation by the private co-operatives do not stand a chance of success" (Hanak and cocoa buyers (GTZ 1986, p. II). Loft 1987). Although these reforms offered the potential of farmer Cameroon's experience with cooperatives also shows control over primary cocoa-marketing organizations, the examples of extremely good and extremely weak perfor- state continued to dominate the management and opera- mance. At present, there are about 500 formally registered tion of cooperatives. The pattern of continued government cooperatives in the country (with total membership at involvement in cooperatives, including secondment of large approximately 400,000) dealing mainly with cocoa and numbers of officials, represented a failure to distinguish coffee. In addition, there is a system of approximately 230 between structures imposed by the state in response to (as of 1987) savings and credit cooperatives (or credit the need for development and those for promoting volun- unions) with a membership of 62,000; these are the most tary participation at the household and local level. successful cooperatives in the country. The characteristic In assessing the condition of a sample of cocoa-marketing that seems to explain the variation in cooperative perfor- cooperatives, a recent report (GTZ 1986, p. 17) concludes mance is the extent of state involvement: that the structure of the present system suffers from There seems to have been almost an inverse relation- fundamental problems, as evidenced by the precarious ship between support provided to cooperatives by financial situation of the cooperatives-eleven of the fifteen governmtween cies and cooperative performances examined were technically bankrupt. The failure of cocoa i.e., cooperatives which received most government cooperatives was attributed at one level to inadequacies in attention performed extremely poorly, and those the financing system on which the cooperatives depend, to which were insulated (often by efficient cooperative excessive costs within the cooperatives due to weak unions) to a high degree from government interven- management, and to poor management advice provided by tion performed well (personal communication with state agencies charged with supporting the cooperatives. At the authors). a more fundamental level, however, it can be argued that the real problem with the cocoa-marketing cooperatives is The first credit union (1963) began, and subsequent expan- that the membership is neither capable of nor interested in 19 influencing the operation of the cooperatives. For example, government to reform rural cooperatives. As was the case the cooperatives are too large to facilitate communication earlier, the philosophy behind the reforms advocated by or to allow a sense of community.'7 Another dimension of donors and some government agencies was that ". . . a unit the size problem is the absence of intermediate levels of must be found whose members trust one another, who will farmer control, meaning that the gap between centralized police themselves so that free-riders do not take over, who management and the village-which is the natural admin- will feel directly responsible for their affairs and for their istrative and organizational unit-is not bridged. The production, and who will hold larger cooperative structures centralized administrative style of cooperatives stifles responsible to them" (Waterbury 1986, p. 82). Such a initiative on the part of the membership. The result of this cooperative organization would threaten several rural structure is that cooperatives are "largely the creation of interest groups, among them clearly the marabouts. There- their managers, and these in turn are for the most part an fore a compromise was sought. Features of the compromise emanation of the state" (GTZ 1986, p. 58). included larger cooperatives (in Sine-Saloum, average Similar experiences with state involvement in coopera- membership rose from 200 to 1,730), little change in tives have characterized Senegal, which has alternately personnel and patrons who manage cooperatives, and the encouraged independent cooperatives and tried to control absence of grassroots participation in organizing the new them through increased state involvement. In Senegal, the cooperatives. As a result, the new cooperatives were government in 1960 legislated a cooperatives statute (based unlikely to be any more successful than earlier ones. on pre-independence French efforts) aimed at establishing a nationwide network of agricultural cooperatives. The new Lessons from experience with cooperatives program, which was sponsored by Prime Minister Dia, was Two conclusions emerge from the experience with market- intended to "..... provide a genuine feeling among its ing cooperatives in the MADIA countries. First, to serve members that the coop was theirs, totally preempt credit their membership effectively-and thus be effective mar- and marketing from private traders and moneylenders, and keting institutions-cooperative must strike a delicate thwart efforts by the marabouts to capture the new institu- balance of state support and the absence of state interven- tions" (Waterbury 1986, p. 81). The program was, however, tion. judging from the often disappointing performance of inimical to so many rural interests, most notably the cooperatives, this blend is difficult to achieve. Support marabouts, that it contributed to Dia's demise and was needs to be in the form of managerial assistance, training eventually termed "boy-scoutism" by Senghor himself. After of officials, and ensuring compliance with bylaws. The 1964, the marabouts, party members, extension agents, and delicate mix of adequate support and avoidance of exces- officials of ONCAD had made their own local arrangements, sive intervention hinges on how the objectives of the with the result that cooperatives fell under the influence of cooperative are defined and achieved. Typically, in situa- local patrons. tions where government sees cooperatives as policy tools Although the nature of the cooperatives changed, the and thus seeks to control operations, the nature of the institutions were still encouraged by the government; by cooperative is corrupted and benefits for members are 1970, there were 1,870 cooperatives in Senegal, with over eroded. half (1,060) in the Groundnut Basin. Throughout the 1970s, Second, given the general ingredients necessary for their cooperatives were under the jurisdiction of ONCAD, an success (from the point of view of members), cooperatives affiliation that effectively converted the cooperatives into are not well suited to quickly supplementing or supplanting state-run entities within the agricultural bureaucracy and the operations of marketing boards or the private sector. A hurt their image and performance. The dissolution of policy that seeks to maintain control of marketing functions ONCAD, the increasing politicization of cooperatives, inter- by substituting cooperatives for marketing boards is not nal corruption, and pressure from donors caused the likely to be successful. 20 Agricultural Marketing Reforms: Overview of the Experience of MADIA Countries The reform policies agreed to by donors and African Table 6 governments, chiefly in the form of structural and sectoral Agricultural Pricing and Marketing Policy Reforms in Sene- adjustment loans made by the World Bank and supported gal, Cameroon, and Nigeria by other donors, have contained numerous measures Senegal relating to pricing and marketing policy, including the reform of marketing parastatals. Although it is not possible Output prices: Producer prices for groundnuts and cereals were to describe in detail the specifics of these reform policies increased. Further, the price of imported rice was to for each of the MADIA countries, an overview will give the be increased to encourage domestic production. flavor of the relevant provisions. Input prices: The fertilizer subsidy provided by SONAR was A primary goal of adjustment programs has been to replaced by a limited subsidy financed by donors A, primary goal of adjustment programswhich is to be gradually phased out by 1989. Regional increase producer prices, both through adjustment of the fertilizer prices are to reflect transport costs. exchange rate, where necessary (i.e., Nigeria and Tanzania), Institutions: The groundnut seed and fertilizer marketing parastatal and through changes in official prices, mainly for export (SONAR) was abolished in 1986 and responsibility for crops (see Table 1). (It has not been possible to adjust marketing groundnuts was transferred to the ground- exchange rates in Senegal and Cameroon due to their nut crushing firms, private traders, and cooperatives. membership in the franc zone.) Reducing the role of the Seed stocks were reduced and free distribution public sector in agricultural marketing has also been an limited. Further, administrative barriers to the internal issue. Donors (chiefly the World Bank and USAID) have also marketing of cereals were removed. addressed the pricing and marketing of inputs-primarily Privatization: The elimination of SONAR shifted responsibility for the policies governing the importation, domestic distribu- marketing to the private sector and cooperatives. tion, and subsidization of fertilizer (see Tables 6 and 7).18 Cameroon It has been demonstrated elsewhere that the impetus for Output prices: As part of the fertilizer subsidy removal program, subsidy-removal programs has been generated by concern government increased the producer price for coffee. about budget deficits (Lele, Christiansen, and Kadiresan Input prices: The fertilizer subsidy that was targeted at coffee 1989; World Bank 1986b), the failure of subsidies to reach producer is t be graduallyeliminatedby1992 their intended beneficiaries, and the growth of the public pucer is to be abyl1992. sector."9 This comparatively narrow focus precluded analyz- Institutions A numbeR of ptarasrtatalestructured; cooperatives are ing the role of fertilizer subsidies in the broader context of being promoted. the need for production intensification in African agriculture Privatization: As part of the fertilizer reform package, the importation and the feasibility of the private sector's rapid replacement and distribution of fertilizer was privatized. of the public sector. An analysis of these broader issues concluded that there is not a single solution to achieving Nigeria the widespread use of fertilizer (Lele, Christiansen, and Output prices: Coincident with the abolition of marketing boards, the Kadiresan 1989). Fertilizer subsidies may be justified in producer prices for exports crops increased. specific cases, i.e., where there are clear market failures, or Input prices: Efforts to eliminate the fertilizer subsidy met with some where fertilizer demand is constrained by low but positive success, however, more recently an implicit subsidy is returns, or by the low income of its users. Subsidies can be paid due to currency overvaluation. most effective if they are combined with the relaxation of Institutions: Export crop marketing boards were abolished. supply constraints stemming from market imperfection, e.g., Privatization: Increased private sector activity in export crop trading inadequate supply of foreign exchange or the absence of has been accomplished and plans are underway to financial markets. encourage private sector fertilizer trading. With respect to reform of agricultural marketing policies, provisions have included restructuring of the institutions responsible for input and output marketing, reducing the role of the public sector in marketing, and encouraging sis on promoting the private sector without first insuring private-sector activity (see Tables 6 and 7).20 This emphasis that the prerequisites for a successful private sector are in has prevailed even though little empirically based work is place represents a serious shortcoming in the reform available on the structure and performance of private process. The lack of detailed agreements with governments agricultural markets in Africa. At present, these markets are on precisely which areas they are willing to privatize has generally confined to food crops-since governments posed many problems. Moreover, the capacity of the control the distribution and marketing of export crops. Even private sector to utilize resources has been limited, and where the private sector is allowed to operate freely in most donor officials have relatively little of the practical marketing and agro-processing (e.g., cotton and rubber in experience and knowledge required to strengthen the Nigeria), trade by licensed buying agents is often alleged to operations of the private sector. Instead, there is a ten- be uncompetitive. Also under way are operations to dency to assume that the private sector is active and strong improve the legal and institutional framework in which the and can quickly meet the challenge of agricultural private sector operates, as well as efforts to improve credit, marketing. information, and transport infrastructure. The Agricultural Sector Adjustment Operation (ASAO) in In view of the extent to which donors historically sup- Kenya provides an example of some of the difficulties that ported the growth of the public sector, the current empha- arise from structural adjustment agreements that aim to 21 Table 7 In addition, the privatization of grain trading was imple- Pricing and Marketing Policy Reforms in Malawi, Kenya, and mented with the intention of providing farmers with an Tanzania alternative marketing channel and increasing the efficiency Malawi of crop procurement. Although the first buying season in which private traders participated under the new scheme Output prices: Smailholder producer prices increased with the (1987 coinied wth aargeinuxdof ref and there exception of tobacco. (1'987! coincided with a large influx of refugees and there- Inputprices: Fexcetionzer subsfdy oac slatefore increased demand for maize, implementation pro- Input prices: Fertilizer subsidy was slated for removal but govern- ceeded smoothly.2' Nonetheless, several issues need to be mentabadone th reovalproram n 187.addressed before privatization can be said to have been Institutions: The marketing parastatal (ADMARC) has been successful. The increased demand for maize raised the reformed to increase efficiency and focus more r p a exclusively on smallholder agriculture. market pnce above the official price, thereby causing the volume of ADMARC's purchases to fall considerably below Privatization: The use of private traders to purchase selected teluev of A RCe s. purchases ic fs d b smallholder output has been encouraged. There are the level of recent years. Although purchases mcreased plans to privatize the distribution of fertilizer. somewhat in 1988, there is concern about ADMARC's ability to fulfill its mandate to achieve food security under these Kenya circumstances. Output prices: Because smallholder producer prices have been kept The two most serious problems facing private traders- close to international levels, recommendations have transport constraints and lack of access to credit-have yet focused on other aspects. to be addressed. Without reliable transport and access to Input prices: Kenya has not had a fertilizer subsidy since the mid- credit, private-sector performance will fall far short of 1970s; its control of distribution and handling margins expectations and will probably come to be dominated by for a range of crops has been liberalized. the few traders who have access to these requirements. Institutions: Implementation of parastatal reform programs along Finally, the privatization of grain marketing in Malawi-and with further restructuring of the public investment and the anticipated privatization of fertilizer marketing-has left expenditure programs related to parastatals. ADMARC in the position of being held to commercial Privatization: Efforts to improve the efficiency of fertilizer importation operational standards (i.e., no losses) while having respon- and distribution have been addressed through import sibility for loss-making development functions without the liberalization and privatization. anticipated government reimbursement. For example, Tanzania ADMARC is expected to ensure national food security by Output prices: Substantial increases in official producer prices for holding stock at the Strategic Grain Reserve, yet it is food and export crops. expected to pay producer prices that preclude any taxation to finance the costs of its food-security and price-stabiliza- Input prices: The explicit fertilizer subsidy was removed in 1984, to functhenstso A te capitaliti and regulaco- but has been offset by pricing of grant-aid fertilizer bon functionS.22 Adequate capitalization and regular com- and currency overvaluation. pensation for loss-making operations by the government Institutions: The National Milling Corporation (NMC) has been may, however. turn out to be a problem elsewhere in East restricted to a food security role and agricultural Africa. cooperatives have been introduced. Finally, as a part of the liberalization program, approxi- Privatization: An increased role for the private sector in agricultural mately 6 percent of seasonal markets were closed. The marketing has been encouraged. World Bank hoped to have nearly 200 markets closed in 1987/88, but it settled for about 125. At present, some 75- 80 percent of the rural population is within eight kilometers of a seasonal market location. Given the political impor- reform marketing conditions and parastatals. The ASAO tance of seasonal markets, their closure was a contentious involved numerous conditions and subconditions related to issue. Not surprisingly, by 1989/90 the closures had been restructuring parastatals-issues that donors and the more than offset by the new markets that ADMARC opened. Kenyan government have disputed for well over a decade In summary, the privatization of smallholder-output (Lele and Meyers 1987). The difficulty exists in part because marketing in Malawi has been generally successful, although donors have confined their analysis to narrow technocratic the early performance of traders suffered somewhat from or economic-efficiency criteria, with little explicit analysis of the rapid pace of implementation, and traders' activities are the political factors that affect decisions. Recent MADIA hindered by shortages of working capital and vehicles studies (of cotton and sugar in Kenya, and of fertilizers (Christiansen and Stackhouse 1989). generally) provide examples of the political reasons why In Tanzania, various marketing reforms have been imple- and how governments have pursued certain courses of mented since 1984, including the reintroduction of cooper- action. They show the degree of detailed analysis that atives, the abolition of crop-purchasing authorities (1984), a needs to be undertaken on both an issue-by-issue and a more tolerant attitude toward private traders (e.g., the country-by-country basis. Several examples will help to elimination of restrictions on buying and transporting food illustrate the point. grains), a more clearly defined and restricted role for the In Malawi, policy reforms affecting the smaliholder mar- grain-marketing parastatal (NMC), and legalization of pri- keting parastatal ADMARC have concentrated on divestiture vate-sector purchases from and sales to the NMC, cooper- of assets not related to smallholder agriculture and enforc- ative unions, primary societies, and farmers (Scarborough ing the requirement that ADMARC operate on commercial 1989). A preliminary assessment shows that all these steps criteria (a condition that has caused concern within the have had a very positive effect-as could be expected in a government and ADMARC about the ability to stabilize country that had been operating substantially below its prices and act as the buyer/seller of last resort). production possibility. Once the effect of these initial 22 reforms is absorbed, however, Tanzania will encounter many marketing is already largely in the hands of the private of the problems concerning the appropriate roles of the sector. According to Lele (1989), the government's attempts private and public sectors that already confront other to intervene in grain markets have met with little success countries that did not deviate from growth as much as because support prices have been well below market Tanzania. For instance, traders face a pervasive shortage of prices, and the capacity of the Nigerian public sector for credit, information, and transportation. Any small increase managing price-support programs is weak. Export-crop in production causes substantial bottlenecks in movements marketing was a government monopoly until the end of of marketed output. The public sector role in improving the 1986, when the relevant commodity boards were abolished capacity of the private sector to undertake these tasks as part of the structural adjustment program. Subsequent effectively, i.e., through the dissemination of market infor- privatization of export-crop trading, combined with currency mation, the provision of credit to traders, an 1 tne improve- devaluation in October 1986, led to substantially higher ment of transport, is less developed in Tanzania than in producer prices for these crops.23 However, privatization Malawi, for example. also led to problems of quality control, as a result of which In Senegal, policy reforms have sought to enhance the Nigerian cocoa now sells at a discount. Interviews with economic environment through the liberalization of input cocoa buyers in London suggest that the poor-quality cocoa and output markets. Although it is still early in the process offered by Nigerian exporters interested only in short-term of state disengagement from the marketing of fertilizer and profit is a serious problem. groundnuts, distortions may be emerging in the private The privatization of fertilizer procurement and distribu- market. Following the liquidation of the marketing parasta- tion has been an issue between the World Bank and the tal (ONCAD) in 1980, groundnut marketing was turned over Nigerian government for several years. The issue is linked to the groundnut-crushing firms, especially the Societe to the high rate of subsidy on fertilizer, which has in general Nationale de Commercialisation des Oleagineux du Senegal (SONA- precluded private trade. At present, there is a proposal to COS), while management of the national seedstock became privatize the commercial supply companies of the multist- the responsibility of the Societ Nationale d'Approvisionment du ate ADPs, which have operated as wholesalers. There is also Monde Rural (SONAR). Cooperatives became the agents for a proposal to privatize retail fertilizer sales through SONACOS, acting as intermediaries between the farmers encouragement of private traders and farmers' coopera- and the groundnut-crushing firms. In 1986, due to financial tives. But cooperatives are weak in Nigeria, and the losses among the crushing firms and a growing parallel response of the private sector is yet to be seen. trade in groundnuts, donor pressure led to the transfer of In summary, the lesson that emerges from experience marketing to three agents: the cooperatives, the crushing with privatization and liberalization efforts in the MADIA firms, and private traders. However, SONACOS continued to countries confirms the experience of other developing regulate the system and to provide marketing finance, countries: The ability of the private sector to function although cooperatives were free to trade and sell at retail. effectively in the area of agricultural marketing depends on Critics of the reforms claim that the crushing firms still the environment in which traders must operate. This control the market and that financing is still only available environment is defined by many of the elements whose through SONACOS (jammeh 1987). For example, groundnut absence in the past served as a rationale for public-sector production, which was domestically marketed by nearly intervention in the first place. These elements include: 2,500 cooperatives, is now channeled through a declining number of buying points-750 in 1988/89. Moreover, there * the presence of an entrepreneurial class able to are allegations that private traders (the organismes prives undertake risk; stockeurs) exercise monopsony power over the farmers (EIU * competitive markets; 1989). * adequate infrastructure, including transport and com- Another aspect of Senegal's experience with privatization munication networks that allow the efficient move- has been the effort to privatize fertilizer distribution- ment of information, goods, and services; largely because of concern about the corruption and weak * efficient markets for inputs and outputs (i.e., there are performance that characterized marketing parastatals in no market failures), including financial services; and Senegal. Even as this privatization initiative was being * food security. pursued, the attractiveness of fertilizer use diminished due to subsidy removal, the reduced availability of credit, and Unfortunately, the public sector, due to its distrust of increased climatic variation. As a result, the private sector private markets and of ethnic domination, has concentrated has been reluctant to make the necessary investment to on substituting for, rather than supplementing, the private become involved in fertilizer distribution. The lack of sector. The result of this policy in most cases has been a private sector participation combined with the public lack of competition in the provision of marketing services sector's inability (due to financial constraints) to function as that has contributed to inefficiency in marketing systems. a seller of last resort has meant that the distribution of This inefficiency, combined with a change in donor atti- fertilizer has suffered, which has contributed to the decline tudes toward the public sector, has caused reform policy to in fertilizer use (Lele, Christiansen, and Kadiresan 1989). encourage increased private-sector involvement in market- Efforts to privatize cereal marketing, in which traders were ing. But too little attention has been paid to insuring the allowed a more active role, have been successful in the adequacy of the environment in which the private sector Groundnut Basin, but less so in areas such as Casamance, must operate. Reliance on markets alone will not necessar- where trade is less well developed. The stabilization of ily ensure competitive pricing and marketing of crops where grain prices will remain an important issue in Senegal. scale economies exist, food security, or timely availability of In Nigeria, marketing reforms have concentrated on inputs. privatization of export-crop marketing, since food-crop 23 Summary of Conclusions and Implications Although public sector intervention in agricultural market- Despite their political nature, marketing organizations ing is closely linked to the nature of agricultural production need to perform legitimate economic functions including: and the processing requirements of crops, its implementa- tion is frequently based on political objectives. The * reducing the inherent riskiness of agriculture for small- marketing arrangements that many African countries inher- scale farmers; ited from the colonial era were heavily influenced by the * ensuring markets and input supply to promote price economic interests of expatriate farmers and traders. stability; During the colonial period, the public sector intervened in * providing revenues for the public sector; agricultural markets to impose order on a seemingly chaotic * supporting large-scale investments in processing that marketing system, to create economic rents for European the private sector is unwilling or unable to attempt; trading companies and estates, and to generate revenues * addressing the constraints imposed by inadequate for the public sector. To preserve the benefits derived from financial markets; state-dominated marketing structures, many of the inde- * creating demand for inputs; and pendent African governments retained the marketing * assuring supply of food and inputs to low-income boards and parastatals bequeathed by the colonial govern- households in remote regions, that might not other- ments. Thus, newly dominant groups came to direct wise be reached. marketing policy and institutions. In this new context, The experience with public sector intervention in agricul- market intervention increasingly came to be seen as a tural marketing in the MADIA countries indicates a clear means of ensuring food security, enabling the government need for institutional pluralism in fostering competition. to perform development functions, stimulating agricultural The private sector can provide increased competition and production, maintaining control over politically strategic can perform some tasks more efficiently than parastatals, commodities, and providing a source of political patronage. but if the private sector is to operate effectively, the public Government interventions in the operations of marketing sector must first ensure that certain requirements have boards and cooperatives have however, often adversely been met. These requirements include: affected the efficiency of these agencies. Although there is a tendency to assume that the failure of many parastatals is * stimulating the development of an entrepreneurial due to their inherent inefficiencies, the sources of these class capable of undertaking risk; inefficiencies often lie beyond the control of the parastatals * encouraging free entry into markets; themselves-e.g., in pressure from the government to * creating adequate infrastructure, transport, and com- overstaff as a form of political patronage or to perform munication networks for the efficient movement of development functions without remuneration. goods; and Many of the policy reforms that pertain to agricultural * promoting efficient financial markets that are able to marketing undertaken by African governments during the support commodity markets. 1980s have emphasized the need to improve parastatal performance through a combination of restructuring, greater With respect to the role of cooperatives, the experience emphasis on commercial criteria, and privatization. This of the MADIA study countries indicates that successful emphasis frequently requires that part or all of the agency cooperatives have two-often opposing-requirements: in question be privatized and that the losses of the The first is the capability to independently and effectively operations that remain in the public sector be minimized or represent the interests of their memberships. It is abund- eliminated. Since extensive public sector control has been antly clear that cooperatives cannot be used as substitutes crucial to maintaining control over important marketing for parastatals-with the public sector controlling their functions, most African governments have been reluctant to operations-since cooperatives by their nature require allow anything other than selective and closely regulated active and democratic grassroots participation. (Govern- private sector involvement in agricultural marketing. Given ments are of course often fearful of the political power of the importance of parastatals-and even cooperatives-as such cooperatives and therefore reluctant to encourage institutions that can extend the political and economic grassroots arrangements). Second, cooperatives need pub- power of governments, it is unlikely that governments will lic support to deal with the complex organizational, completely relinquish the right to intervene in agricultural technological, and financial requirements of modern coop- markets. erative management. 24 In Africa, privatization has not been preceded by the while depoliticizing parastatal operations so as to enhance strengthening of the private sector or the establishment of competition while development requirements are met. This legal and other institutions (e.g., standardization of weights means defining the appropriate role of the public sector in and measures, collection and dissemination of market terms of the circumstances in which public support and information, availability of credit to traders, transporters, regulation is required to ensure a competitive environment, wholesalers, and retailers). Thoughtful and long-term donor and intervention is needed to provide services that the assistance to the private sector is required in transport, private sector is unwilling or unable to provide. Assuring communication, information, and credit to contribute to the agricultural growth with development in Africa will continue decentralization of economic and political power. The issue to make it necessary: of timing will be critical to the development of efficient and (I) to assist producers in confronting the risks associated effective marketing systems and will determine whether the with rainfed agriculture as practiced in Africa, private sector will be competitive or will merely replace (2) establish an environment where capital and techno- public sector oligopolies while continuing to serve the logical inputs are readily available, and same vested interests. Thus far, donors have tended to be (3) act as a buyer and seller of last resort even while naive about the appropriate extent and pace of privatiza- protecting consumers, particularly low-income con- tion, especially given that the interests in public sector sumers, from wide price fluctuations. operations, which they supported, have become entrenched. A limited amount of market intervention will necessarily be The implication of these findings for donors is that, part of any overall agricultural strategy. Progress will in all although the perception of parastatals as politicized and likelihood be slow: Donors must recognize and take into inefficient is correct, it is not sufficient to sponsor reforms account major differences among and within countries in that in effect expect the private sector to address even a attempting to play a useful role in developing appropriate majority of agricultural marketing needs. Policies must be marketing institutions and arrangements that include both devised that continue to encourage the private sector even the private and public sectors. 25 Bibliography ADMARC. Various dates. "Annual Accounts and Report for the Year Cleaver, Kevin and Mike Westlake. 1987. "Pricing, Marketing and Ended 31st March." Blantyre, Malawi: Ministry of Agriculture, Agricultural Development in Kenya." MADIA Working Paper. 1972-1987. Washington, D.C.: World Bank. Adams, D.W. 1978. "Mobilizing Household Savings Through Rural Clough, Paul. 1985. "The Social Relations of Grain Marketing in Financial Markets." Economic Development and Cultural Change Northern Nigeria." Review of African Political Economy 34 26, no. 3 (April): 547-60. (December): 16- 34. Alila, Patrick 0. 1985. "Administration of Cooperatives for Rural Collinson, M. 1972. Farm Management in Peasant Agriculture. Boulder, Development in Kenya." In Challenging Rural Poverty, edited CO: Westview Press. by Fassil G. Kiros, 169-182. Trenton, NI: Africa World Press. Coopers and Lybrand Associates. 1987. "NCPB Reorganization Avramovic, Dragoslav. 1987. "Commodity Problems: What Next?" Study." Vol. 6. Washington, D.C. World Development 15, no. 5 (May): 645-655. COPAC. n.d. "Cooperative Information Note: Republic of Kenya." Babin, D. 1988. "The World Bank and the IMF: Rolling Back the No. 2 Revised. State or Backing its Role?" In The Promise of Privatization, _. n.d. "Cooperative Information Note: Republic of Tanzania." edited by Raymond Vernon. Washington, D.C.: Council on No. 5 (Ist Revision). Foreign Relations. Foreign Relations. ~~~~~Cox, Pamela M.l. 1984. "Implementing Agricultural Development Balcet, I.C. and Wilfred Candler. 1982. Farm Technology Adoption in Policy in Kenya." Food Research Institute Studies 19, no. 2:153- Northern Nigeria. Washington, D.C.: World Bank 176. Bates, Robert H. 1981. Markets and States in Tropical Africa: The Political de Wilde, IC. 1967. Experiences with Agricultural Development in Tropical Basis of Agricultural Policies. Berkeley: University of California Africa. Baltimore: Johns Hopkins University Press. Press.Arc.Blioe on Hpn nvriyPes Press. . .Deaton, Angus. 1987. "The Demand for Personal Travel in Devel- _____ 1983. "Patterns of Market Intervention in Agrarian Africa." oping Countries." Infrastructure and Urban Development Food Policy 8, no. 4. (November). Discussion Paper. Washington, D.C.: World Bank. _ . 1989. "Politics and Agriculture in Kenya." in The Politics of Deloitte, Haskins, and Sells. 1987. "Agricultural Development and Agricultural Policy in Africa, edited by Uma Lele and Ellen Marketing Corporation Organisation and Management Hanak. Washington, D.C.: World Bank (forthcoming). Review." Lilongwe, Malawi. Bauer, PT. 1967. West Afrcan Trade: A Study of Competition, Oligopoly Economist Inteligence Unit. 1989. Country Report: Senegal, and Monopoly in a Changing Economy. New York: Augustus M. Gambia, Guinea Bissau, Cape Verde. No. 1. Kelley. 1 Eicher, Carl K. 1982. "Facing Up to Africa's Food Crisis." Foreign Behrman, Jere R. 1987. "Commodity Price Instability and Economic Afairs 61 (Fall): 151-174. Goal Attainment in Developing Countries." World Development frank. 1 "a M e adPsntSt 15, no. 5 (May): 559-573. Ellis, Frank. 1983. "Agricultural Marketing and Peasant-State 15, no. 5 (May): 559-573. Transfers in Tanzania." Journal of Peasant Studies 10 (July): 214- Bienen, Henry. 1986. "Politics and Agricultural Policy in Nigeria." 242. in The Politics of Agricultural Policy in Africa, edited by Uma Lele and Ellen Hanak. Washington, D.C.: World Bank FAO/World Bank. 1987. "Agricultural Cooperatives in Tanzania." (forthcoming). Agricultural Sector Review Mission, Working Paper 5. Rome: Bingen, R. James. 1987. "An Orientation to Production Systems FAO (July). Research in Senegal." MSU International Development Papers. Gaviria, Juan. Forthcoming. "The Role of Transportation in Reprint no. 16. Agricultural Performance in Cameroon." MADIA Working Bowbrick, Peter. 1988. "An Economic Analysis of the Impact of Paper. Washington, D.C: World Bank. Private Traders on Agricultural Marketing." Lilongwe, Malawi: . Forthcoming. "The Role of Transportation in Agricultural Planning Division, Ministry of Agriculture. Performance in Malawi." MADIA Working Paper. Washington, Bryceson, Deborah Fahy. 1985. "The Organization of Tanzanian D.C.: World Bank. Grain Marketing: Switching Roles of the Co-operative and German Agency for Technical Cooperation (GTZ). 1986. "Rehabil- the Parastatal." In Marketing Boards in Tropical Africa, edited by itation of the Cocoa Marketing Cooperatives in the Central- Kwame Arhin, Paul Hesp, and Laurens van der Laan, 53-78. Southern Region of Cameroon." Provisional Final Report London: KPI. (November). . n.d. "Second Thoughts on Marketing Co-operatives in Gilbert, Christopher L. 1986. "Commodity Price Stabilization: The Tanzania: Background to Their Reinstatement." Plunkett Massell Model and Multiplicative Disturbances." The Quar- Development Series 5. terly journal of Economics (August). Candler, Wilfred. 1986. "Tanzania: Notes on the Development of Government of Kenya. Various dates. Economic Surveys. Nairobi, Marketing and Price Policy: 1961-1986." MADIA Working Kenya, 1970- 1985. Paper. Washington, D.C.: World Bank. Government of Kenya, Ministry of Agriculture. 1987. Mimeo- Christiansen, Robert E. and V. Roy Southworth. 1988. "Agricultural graphed sheets. Nairobi, Kenya. Pricing and Marketing Policy in Malawi: Implications for a Government of Malawi, Ministry of Agriculture. 1988. Mimeo- Development Strategy." Paper presented at the "Sympo- graphed sheets. Malawi. sium on Agricultural Policies for Growth and Development" Grosh, Barbara Ann. 1988. "Improving the Economic Performance at Mangochi, Malawi, October 31 to November 4. of Public Enterprises in Kenya: Lessons from the First Two _ and J.G. Kydd. 1987. "The Political Economy of Agricultural Decades of Independence." Ph.D. diss., University of Policy Formulation in Malawi, 1960-1985." in The Politics of Califomia-Berkeley. Agricultural Policy in Africa, edited by Uma Lele and Ellen Gulhati, Ravi. 1988. "The Political Economy of Reform in Sub- Hanak. Washington, D.C.: World Bank (forthcoming). Saharan Africa." Economic Development Institute of the _____ and Lee Ann Stackhouse. 1989. "The Privatization of World Bank. Washington, D.C.: World Bank. Agricultural Marketing in Malawi." World Development 17, no. 5 (May): 729-740. 26 Hanak, Ellen. 1989. "The Politics of Agricultural Policy in Tanzania." Kelly, Valerie Auserehl. 1988. "Acquisition and Use of Agricultural in The Politics of Agricultural Policy in Africa, edited by Uma Lele Inputs in the Context of Senegal's New Agricultural Policy: and Ellen Hanak. Washington, D.C.: World Bank The Implications of Farmer's Attitudes and Input Purchasing (forthcoming). Behavior for the Design of Agricultural Policy and Research and Michael Loft. 1987. "Danish Development Assistance to Programs." MSU International Development Papers. Reprint Tanzania and Kenya, 1962-1985: Its Contribution to Agricul- no. 18. tural Development." MADIA Working Paper. Washington, Kooten, G.C. and Andrew Schmitz. 1985. "Commodity Price D.C.: World Bank. Stabilization: The Price Uncertainty Case." Canadian Journal of Harrison, Kelly, Donald Henley, Harold Riley, and lames Shaffer. Economics 2 (May). 1987. "Improving Food Marketing Systems in Developing Kuhn, lohannes and Heinz Stoffregen. 1975. "How to Measure the Countries: Experiences from Latin America." MSU Inter- Efficiency of Agricultural Cooperatives in Developing Coun- national Development Papers. Reprint no. 9. tries. Case Study- Kenya." Rome: FAO. Harriss, Barbara. 1981. "Some Lessons from the History of Internal Lamb, Geoffrey and Linda Muller. 1982. "Control, Accountability, Trade in the Interior of West Africa." Discussion Paper no. and Incentives in a Successful Development Institution: The 105. ;nstitute of Development Studies (December). Kenya Tea Development Authority." World Bank Staff Hesp, Paul and Laurens van der Laan. 1985. "Marketing Boards in Working Papers no. 550. Washington, D.C.: World Bank. Tropical Africa: A Survey." In Marketing Boards in Tropical Africa, Lele, Uma. 1985. "Agricultural Growth, Domestic Policies, the edited by Kwame Arhin, Paul Hesp, and Laurens van der External Environment and Assistance to Africa: Lessons of a Laan. London: KPI. Quarter Century." In Trade, Aid, and Policy Reform: Proceedings Higgins, G.M. 1982. Potential Population Supporting Capacities of Lands of the Eighth Agriculture Sector Symposium, edited by Colleen in the Developing World. A technical report of the project, Roberts. Washington, D.C.: World Bank. "Land Resources for Populations of the Future." Rome: _ . 1987. "Agriculture and Infrastructure." Paper prepared for FAO/International Institute for Applied Systems Analysis. "Symposium on Transportation and Structural Adjustment," International Labor Organization. 1988. "A Review of Co-operative Transportation Department, World Bank, Baltimore, MD, Development in the African Region: Scope. Impact and May 6-8. Prospects." Report prepared for the 7th African Regional _ . 1988. "Comparative Advantage and Structural Transforma- Conference. tion: A Review of Africa's Economic Development Experi- International Monetary Fund. 1987. International Financial Statistics ence." In The State of Development Economics: Progress and Statistical Yearbook. Washington, D.C.: International Monetary Perspectives, edited by Gustav Ranis and T. Paul Schultz. New Fund. York: Basil Blackwell, Inc. International Fertilizer Development Center. 1986. "Cameroon _ . 1977. "Considerations Related to Optimum Pricing and Fertilizer Sector Study." Muscle Shoals, Alabama: Inter- Marketing Strategies in Rural Development." In Decision- national Fertilizer Development Center. Making and Agriculture, edited by Dams and Hunt. Lincoln: International Finance Corporation. 1989. "Lessons from IFC's University of Nebraska Press. Experience in the Agricultural Production Sub-Sector." . 1981. "Cooperatives and the Poor: A Comparative Perspec- Washington, D.C.: International Finance Corporation. tive." World Development 9: 55-72. Iammeh, Sidi C. 1985. "The Evolution of Marketing and Pricing . 1979. The Design of Rural Development: Lessons from Africa. Policy in Senegal." MADIA Working Paper. Washington, D.C.: Baltimore: Johns Hopkins University Press. World Bank. _ . 1971. Food Grain Marketing in India: Private Performance and - 1987. State Intervention in Agricultural Pricing and Marketing in Public Policy. Ithaca: Cornell University Press. SenegaL Ph.D. diss., Johns Hopkins University. _ . 1989a. "Sources of Growth in East Afcican Agriculture." The _ and C.G. Ranade. 1986. "Agricultural Pricing and Marketing World Bank Economic Review 3, no. I (January): 119-144. in Senegal." MADIA Working Paper. Washington, D.C.: World _ . 1989b. "Structural Adjustment, Agricultural Development, Bank. and the Poor: Some Lessons from the Malawian Experi- and Uma Lele. 1988. "Building Agricultural Research ence." World Development. Forthcoming. Capacity in Senegal." MADIA Working Paper. Washington, _ and L. Richard Meyers. 1986. "Agricultural Development D.C.: World Bank. and Foreign Assistance: A Review of the World Bank's Johnson, D. Gale. 1947. Forward Prices for Agriculture. Chicago: Experience in Kenya, 1963 to 1986." MADIA Working Paper. University of Chicago Press. Washington, D.C.: World Bank. Jolly, C.M., M. Kamuanga, S. Sall, and J.L. Posner. 1988. "Farm Level _ and Rahul lain. 1989. "Synthesis: Aid to African Agricul- Cereal Situation in Lower Casamance: Results of a Field ture-Lessons from Two Decades of Donor Experience." In Study." MSU International Development Series. Reprint no. Aid to African Agriculture: Lessons from Two Decades of Donor 27. Experience, edited by Uma Lele. Forthcoming. Washington, Jones, William 0. 1987. "Food-Crop Marketing Boards in Tropical D.C.: World Bank. Africa." Journal of Modern African Studies 25 (September): 375- and Wilfred Candler. 1981. "Food Security: Some East 402. African Considerations." In Food Security for Developing Coun- Kaberuka, D.P. 1984. "Evaluating the Performance of Food Market- tries, edited by Alberto Valdes. Boulder, CO: Westview Press. ing Parastatals." Development Review Policy 2: 190-216. - , Ademola Oyeiide, Vishva Bindlish, and Balu Bumb. 1989. Kaldor, Nicholas. 1987. "The Role of Commodity Prices in Nigeria's Economic Development, Agriculture's Role, and World Bank Economic Recovery." World Development 15, no. 5: 551-558. Assistance, 1961-88: Lessons for the Future. Washington, D.C.: Kanbur, S.M. Ravi. 1983. "How to Analyze Commodity Price World Bank. Stabilization?" Discussion Paper 35 (August). - , Nicolas van de Walle, and Mathurin Gbetibouo. 1989. Keene, Monk and Associates, Inc. 1984. "Agricultural Parastatals." "Cotton in Africa: An Analysis of Differences in Perfor- prepared for USAID, PPC/PDPR (September). mance." MADIA Working Paper. Washington, D.C.: World A report ~~~~~~~~~~~~~Bank. 27 _____ and L. Richard Meyers. 1987. "Growth and Structural Pinckney, Thomas C. 111. 1986. "Production Instability and Food Change in East Africa: Domestic Policies, Agricultural Security in Kenya: Measuring Trade Between Government Performance, and World Bank Assistance, 1963-1986." Parts Objectives." Ph.D. diss., Stanford University. I and II. DRD Discussion Paper nos. 273 and 274. Washing- Quinn, Victoria, Mabel Cuiligo, and J. Price Gittinger. 1988. ton, D.C.: World Bank. "Household Food and Nutritional Security in Malawi." Paper _ and Steven Stone. 1989. "Population Pressure, the Environ- presented at the "Symposium on Agricultural Policies for ment, and Agricultural Intensification: Variations on the Growth and Development" at Mangochi, Malawi, October 31 Boserup Hypothesis." MADIA Working Paper. Washington, to November 4. D.C.: World Bank. R.R. Nathan Associates. 1987. "The Impact of the Fertilizer Subsidy _____ and Manmohan Agarwal. 1989. "Smallholder and Large- Removal Program on Smallholder Agriculture in Malawi." Scale Agriculture: Are There Trade-offs in Growth and Washington, D.C. Equity?" MADIA Working Paper. Washington, D.C.: World Ruthenberg, H. 1960. Farming Systems in the Tropics. Oxford: Bank. Clarendon Press. - Robert E. Christiansen, and Kundhavi Kadiresan. 1989. Scarborough, V. 1989. "The Current Status of Food Marketing in "Issues in Fertilizer Policy in Africa: Lessons from Develop- Tanzania." Unpublished paper. ment Programs and Adjustment Lending, 1970-87." MADIA Schultz, T.W. 1945. Agriculture in an Unstable Economy. New York and Working Paper. Washington, D.C.: World Bank. London: McGraw Hill. Lindauer, David L., Oey Astra Meesook, and Parita Suebsaeng. Siamwalla, Ammar. 1982. "Security of Rice Supplies in the ASEAN 1988. "Government Wage Policy in Africa: Some Findings Region." In Food Security for Developing Countries, edited by and Policy Issues." The World Bank Research Observer 9 Alberto Valdes. Boulder, CO: Westview Press. (January): 1-26. Southworth, HW. and B.F Johnston. 1967. Agricultural Development Lipton, Michael. 1983. "Is Increased Agricultural Marketing Good and Economic Growth. Ithaca: Cornell University Press. for the Rural Poor?" Development Research Digest 10 (Winter): 55-59. Speirs, Mike. 1989. "Peasants, Merchants and the State-Some Reflections on Cereals Policies in Burkina Faso." Economics Macbean, Alastair and Ductin Nguyen. 1987. "International Coin- Institute, Royal Veterinary and Agricultural University, modity Agreements: Shadow and Substance." World Develop- Copenhagen (February,. ment 15, no. 5: 575-590. Cpnae Fbur) MacKentz15, RichardoB. and 575-590. Tullock.1981.The New World of von Pischke, ID. 1980. "The Political Economy of Specialized Farm MacKenzie, Richard B. and Gordon Tullock. 1981. The New World of Credit Institutions." World Bank Working Paper no. 446. Economics. Homewood, IL: Richard D.lrwin, Inc.Wahnt,D.:WolBnk Maizels, Alfred. 1987. "Commodities in Crisis: An Overview of the Washington, D.C.: World Bank. MainIssus." orl Devlopmnt 1, n. 5:537-49.Wahab, Indra. 1985. "Effects of Price Stabilization on Export Main Issues." World Development 15, no. 5: 537-549. Revenue Instability of the Individual Countries." Trade and Meagher, Kate. 1988. "The Market in the Lived Economy: A Report Development 6: 17-46. on the Dynamic of Official and Parallel Market Activities in Waterbury, John. 1986. "Agricultural Policy Making and Stagnation Arua District, Uganda." University of Sussex (November). in Senegal." in The Politics of Agricultural Policy in Africa, edited Mellor, lohn W 1978. "Food Price Policy and Income Distribution by Uma Lele and Ellen Hanak. Washington, D.C.: World Bank in Low Income Countries." Economic Development and Cultural (forthcoming). Change 27 (October). Wellisz, Stanislaw and Ronald Findlay. 1988. "The State and the Morris, Michael. 1988. "Etude sur la Commercialisation des Invisible Hand." The World Bank Research Observer 3 (January): Cer6ales dans la Region du Fleuve Senegal: Methodologie." 59-80. MSU International Development Series. Reprint no. 25F Williams, Gavin. 1985. "Marketing Without and With Marketing Mosley, Paul. 1984. "The Politics of Economic Liberalisation: Boards: The Origins of State Marketing Boards in Nigeria." USAID and the World Bank in Kenya, 1980-84." Working Review of African Political Economy 34 (December): 4-15. Paper no. 30. University of Bath (December). Wolf, Thomas. 1986. "State Intervention at the Cabbage-Roots: A Newbery, David M.G. and Joseph E. Stiglitz. 1981. The Theory of Case Study from Kenya." IDS Bulletin 17, no.1: 47-50. Commodity Price Stabilization: A Study in the Economics of Risk. World Bank. 1988. "Agricultural Marketing: World Bank's Experi- Oxford: Clarendon Press. ence." Report no. 7353. Washington, D.C.: World Bank. Newman, Mark D. 1987. "Grain Marketing in Senegal's Peanut World Bank. 1986a. "Improving Agricultural Marketing and Food Basin: 1984/85 Situation and Issues." MSU International Security Policies and Organization-A Reform Proposal." Development Papers. Reprint no. 14. Report no. 6083- MAn. Washington, D.C.: World Bank. ____ 1987. "Tradeoffs Between Domestic and Imported Cereals World Bank. 1986b. Wo68d Devepment Report, 1986. New York: Oxford in Senegal: A Marketing Systems Perspective." MSU Inter- University Press. national Development Papers. Reprint no. 15. World Bank. 1981. "Malawi: The Development of the Agricultural _, R Alassane Sow, and Ousseynou Ndoye. 1988. "Regulatory Sector." Report no. 3459-MAI. Washington, D.C,: World Bank. Uncertainty and Government Objectives for the Organiza-Setr"Rptno359M.Wahgo,DC:WrlBnk tion and Performance of Cereal Markets: The Case of World Bank. 1986c. Tanzania Agricultural Sector Mission: Technical Senegal." MSU International Development Papers. Reprint Papers. Rome: FAO/World Bank Cooperative Programme no. 24. Investment Centre. Paul, Samuel. 1982. Managing Development Programs: The Lessons of World Bank, Eastern Africa Projects Department, Southern Agricul- Success. Boulder, CO: Westview Press. ture Division. 1983. "Tanzania Agricultural Sector Report." Report no. 4052-TA. Washington, D.C.: World Bank. 28 Appendix Total and per ton ADMARC marketing costs, 1972/73-1986/87 1972/73 1973/74 1974/75 1975/76 1976/77 1977/78 1978/79 Average Total Crop Purchases (M. Ton) 185,269 162,140 170,300 117,900 181,300 192,176 223,391 ('000 Kwacha) 16,477 14,343 17,179 20,458 23,042 25,477 30,522 Direct Costs' ('000 Kwacha) 6,595 8,307 9,291 9,710 12,603 15,364 16,105 (Kwacha/M. Ton) 35.60 51.23 54.56 82.36 69.51 79.95 72.09 63.61 Administrative Costs ('000 Kwacha) 858 1,448 1,281 1,859 2,527 3,427 4,106 (Kwacha/M. Ton) 4.63 8.93 7.52 15.77 13.94 17.83 18.38 12.43 Finance Costs ('000 Kwacha) 132 98 1,032 2,015 2,522 3,075 2,996 (Kwacha/M. Ton) 0.71 0.60 6.06 17.09 13.91 16.00 13.41 9.68 Total Marketing Costs ('000 Kwacha) 7,585 9,853 11,604 13,584 17,652 21,866 23,207 (Kwacha/M. Ton) 40.94 60.77 68.14 115.22 97.36 113.78 103.89 85.73 1979/80 1980/81 1981/82 1982/83 1983/84 1984/85 1985/86 1986/87 Average Total Crop Purchases (M. Ton) 182,380 192,149 215,796 303,841 292,849 377,342 374,245 242,000 ('000 Kwacha) 28,238 29,900 28,859 41,940 46,048 74,262 88,544 90,585 Direct Costs' ('000 Kwacha) 21,235 23,698 22,573 21,314 28,269 43,482 42,458 48,233 (Kwacha/M. Ton) 116.43 123.33 104.60 70.15 96.53 115.23 113.45 199.31 117.38 Administrative Costs ('000 Kwacha) 4,821 5,874 6,363 6,758 6,624 8,901 9,091 11,905 (Kwacha/M. Ton) 26.43 30.57 29.49 22.24 22.62 23.59 24.29 49.19 28.55 Finance Costs ('000 Kwacha) 4,246 6,530 6,236 9,270 6,782 6,464 4,054 11,591 (Kwacha/M. Ton) 23.28 33.98 28.90 30.51 23.16 17.13 10.83 47.90 26.96 Total Marketing Costs ('000 Kwacha) 30,302 36,102 35,172 37,342 41,675 58,847 55,603 71,729 (Kwacha/M. Ton) 166.15 187.89 162.99 122.90 14231 155.95 148.57 296.40 172.89 Notes: Direct costs are the sum of "total selling expenses" and "total buying and direct expenses," which include transport and packing costs, auction floor charges, insurance, marketing costs, depot and storage costs grading, ginning and milling costs, fumigation costs, and seed distribution costs. 2 ADMARC data on volume of crop purchases, before 1979/80 were converted from short tons to metric tons by multiplying by 1.102. 3 Finance costs are listed as "interest payable" on the ADMARC Profit and Loss Accounts and include interest paid on long-term borrowings, bank overdrafts, and other. Sources: ADMARC 1972-87, except for total purchases for 1986/87 from Deloitte, Haskins, and Sells 1987, and marketing costs for 1974/75-1977/78 from World Bank 1986a. 29 Notes 1. Another MADIA paper has examined input pricing, specifi- comparing actual prices with opportunity costs-for tradable cally fertilizer pricing policy, in great detail. See Lele, Christi- goods this opportunity cost is measured by the relevant inter- ansen, and Kadiresan 1989. national parity price; and (4) returns to suppliers. This indicator 2. The same phenomenon operated in India during the colonial also compares actual prices with opportunity costs (Grosh 1988, and postcolonial period (see Lele 1971). PP 40-45) 3. See Schultz 1945 and Johnson 1947. 9. Total marketing costs plus the value of payments to produc- 4. The availability of imported food, and therefore the reliabil- ers equals the total expenses. Direct costs are the sum of selling, ity of supplies of food for farmers dependent on the market, can buying, and direct expenses from ADMARC's trading accounts. be jeopardized by shortages of foreign exchange and the vagaries These costs include transport, packing, and storage costs; auction of donor food supplies (Lele 1985, p 200). Shortages of foreign floor charges; insurance; marketing costs; grading, ginning, milling, exchange are common, given the volatility of primary commodity and fumigation costs; and seed distribution costs. Administrative markets, on which most African countries rely for much of their costs are comprised of expenses incurred by ADMARC's head foreign exchange earnings and employment. For a detailed offices, such as salaries and travel expenses of central office staff, discussion of the degree of dependence on agriculture as a legal and professional fees, rent, and insurance costs. Finance costs source of foreign exchange nd employment, see Lele 1985. consist of interest payable on long-term loans and bank overdrafts. 5. An additional production-oriented argument for price stabi- 10. For the period 1978/79-1980/81 the average size of lization has been made in the context of correcting the domestic ADMARC's bank overdraft was K 2.5 million, or K 12.6 per ton of terms of trade for agriculture in order to encourage long- term purchases, as compared to K 21.2 million, or K 64 per ton of growth. While improvement in the terms of trade in favor of purchases, forthe 1984/85-1986/87 period. agriculture may induce some immediate aggregate supply response, a sustained response cannot depend on continued 11. The problems that ADMARC has suffered in this regard can p rice increases: .... . upward price adjustments cannot be have serious consequences for farmers'confidence in the market- expected to continue once the initial price distortions have been ing parastatal and, therefore, in their willingness to rely on corrected, especially in view of the limited international market markets for food supplies. For an account of the impact of delays corrcte, esecillyin vew f th liite intrnaiona maket in ADMARC's purchases, see R. R. Nathan Associates (19871. prospects for many of Africa's traditional export and food crops. Once the price distortions are corrected, important issues 12. The oil palm projects in Cameroon show a similar pattern regarding long-term agricultural growth relate more to the stability of social services to employees that extended to the whole of the pricing environment" (Lele 1985, p. 209). township (Lele, van de Walle, Gbetibouo 1989). 6. Despite the massive donor assistance to parastatals, effort 13. For more detailed accounts of the role of institutions in on the part of donors to collect data that would help to identify agriculture, see Lele, van de Walle, and Gbetibouo 1989; Lele, the true sources and causes of parastatal inefficiency have been Christiansen, and Kadiresan 1989; jammeh and Lele 1989; and weak. Although there are exceptions, e.g., a World Bank effort to Lele, Oyejide, et al. 1989. establish a capacity in the Marketing Development Bureau in 14. Providing a universally accepted definition of a cooperative Tanzania. MDB had developed by far the most complete account- is difficult. The International Labor Organization (ILO) cites the ing picture of the major parastatals as a basis for providing following somewhat narrow definition: governments with advice on price and marketing development ... an association of persons, who have voluntarily joined policy. together to achieve a common end through the formation of 7. See Lele, van de Walle, and Gbetibouo 1989 for an analysis a democratically controlled organisation, making equitable of the problems with the Cotton Lint and Seed Marketing Board contributions to the capital required and accepting a fair (CLSMB) in Kenya. The mediocre performance of CLSMB can be share of the risks and benefits of the undertaking in which attributed to a number of factors, ranging from complex techno- the members actively participate (ILO 1988, p.l51. logical problems and poor management to the historical lack of In addition, the 23rd Congress of the International Cooperative political clout of cotton producers relative to those of tea and Alliance, held in 1966, adopted the following six principles of coffee. Despite considerable evidence of the government's cooperatives, adherence to the first four of which is necessary for commitment to support cooperatives, some argue that more membership in the Alliance: recently government has put the interests of powerful economic groups ahead of its general policy of support for cooperatives (i) voluntary and open membership; (iil equal rights of (Wolf 1986, p.50). voting for members, i.e. one member, one vote; (iii) limited 8. Grosh uses four different measures of performance to interest on share capital, if any; (iv) equitable distribution of evaluate public enterprises: (1) financial rate of return or profita- the economic results arising out of the operations of the bility; (2) efficiency measured in terms of good service, low unit society; (v) constitution of provisions for the education of cost of margins, rational resource allocation, and returns to members, officers, employees and the general public; (vi} employees; (3) returns to consumers, which is measured by co-operation between co-operatives (ILO 1988, p.19). 30 15. With regard to the impact of state intervention on the 19. Among the MADIA countries. Nigeria is an extreme case as operation of cooperatives, the ILO argues that: regards the costs of a fertilizer subsidy-in 1985 the subsidy cost There is ample evidence that co-operatives are becoming US$240.9 million, which was 32.1 percent of the agricultural increasingly subject to government intervention and in budget and 3.7 percent of the total budget (Lele, Christiansen, some countries are completely controlled by the State. The and Kadiresan 1989). danger of increased government intervention or control is 20. For an excellent survey of the literature and the evolution that emerging co-operatives may not be geared to the of issues concerning privatization see van de Walle (1989). problems of their members but to pre-established govern- 21. See Christiansen and Stackhouse 1989; Lele 1989b; and ment policies: in other words, that co-operatives may Bowbrick 1988. become a government instrument and not an instrument of 22. Historically ADMARC has financed its loss on the maize self-help of the people (ILO 1988, p.27). trading account through a subsidy from the profits earned on 16. Part of the reason for continued state support of coopera- tobacco and groundnut trading. Recent price increases for these tives after independence was that expansion of the cooperative commodities precludes this method of financing in the future. movement provided a means of pushing Asian traders out of the 23. For example, the producer price of cocoa increased by 180 rural sector (Bryceson n.d., p.5). percent from 1,600 naira per metric ton in 1986 to 4,500 naira 17. The average size of a credit union is approximately 268 immediately following the devaluation. It increased further to members while that of a cocoa marketing cooperative is about 6,500 naira per metric ton in mid-1988, and to 12,000 naira by the 800. end of 1988. 18. For a detailed account of the level of fertilizer subsidies, fiscal cost, the cost of nutrients relative to output prices in the MADIA countries, see Lele, Christiansen, and Kadiresan 1989. 31 list of MADIA Discussion Paper Series Lele, Uma. "Agricultural Growth, Domestic Policies, the External Environment, and Assistance to Africa: Lessons of a Quarter Century." MADIA Discussion Paper No. 1. Washington, D.C.: World Bank, 1989. Lele, Uma. "Managing Agricultural Development in Africa: Three Articles on Lessons from Experience." MADIA Discussion Paper No. 2. Washington, D.C.: World Bank, 1989. Lele, Uma and L. Richard Meyers. "Growth and Structural Change in East Africa: Domestic Policies, Agricultural Performance, and World Bank Assistance, 1963-86." MADIA Discussion Paper No. 3. Washington, D.C.: World Bank, 1989. Lele, Uma and Steven Stone. "Population Pressure, the Environment, and Agricultural Intensification: Variations on the Boserup Hypothesis." MADIA Discussion Paper No. 4. Washington, D.C.: World Bank, 1989. Lele, Uma. Robert Christiansen, and Kundhavi Kadiresan. "Fertilizer Policy in Africa: Lessons from Development Programs and Adjustment Lending, 1970-87." MADIA Discussion Paper No. 5. Washington, D.C.: World Bank, 1989. Lele, Uma and Manmohan Agarwal. "Smallholder and Large-Scale Agriculture: Are There Trade-offs in Growth and Equity?" MADIA Discussion Paper No. 6. Washington, D.C.: World Bank, 1989. Lele, Uma, Nicolas van de Walle, and Mathurin Gbetibouo. "Cotton in Africa: An Analysis of Differences in Performance." MADIA Discussion Paper No. 7. Washington, D.C.: World Bank, 1989. Idachaba, F S. "State-Federal Relations in Nigerian Agriculture." MADIA Discussion Paper No. 8. Washington, D.C.: World Bank, 1989. Lele, Uma. "Structural Adjustment, Agricultural Development, and the Poor: Lessons from the Malawian Experience." MADIA Discussion Paper No. 9. Washington, D.C.: World Bank, 1989. Gaviria, Juan, Vishva Bindlish, and Uma Lele. "The Rural Road Question and Nigeria's Agricultural Development." MADIA Discussion Paper No. 10. Washington, D.C.: World Bank, 1989. Lele, Uma and Robert Christiansen. "Markets, Marketing Boards, and Cooperatives in Africa: Issues in Adjustment Policy." MADIA Discussion Paper No. 11. Washington, D.C.: World Bank, 1989. Jammeh, Sidi, Mathurin Gbetibouo, and Uma Lele. "Building Agricultural Research Capacity in Senegal." MADIA Discussion Paper No. 12. Washington, D.C.: World Bank, 1989. Lele, Uma, Mathurin Gbetibouo, and Paul Fishstein. "Planning for Food Security in Africa: Lessons and Policy Implications, 1960-88." MADIA Discussion Paper No. 13. Washington, D.C.: World Bank, 1989. Lele, Uma, Bill Kinsey, and Antonia Obeya. "Building Agricultural Research Capacity in Africa: Policy Lessons from the MADIA Countries." MAADIA Discussion Paper No. 14. Washington, D.C.: World Bank, 1989. 32 THE MADIA STUDY Although many generalizations have been made about the agricultural crisis in Africa, relatively few detailed country and cross-country studies of African agriculture based on systematic data analysis have been conducted. Similarly, although foreign aid has constituted a large part of total government expenditures in Africa for close to fifteen years, there has been little analysis of the role of external assistance in African countries that goes beyond political criticism of official assistance or the alleged self- serving objectives of donors. The impetus for the study "Managing Agricultural Development in Africa" (MADIAN was to begin the process of filling this gap and to explain the nature and sources of the agricultural crisis, particularly the extent to which it originated in resource endow- ments, historical and contemporary events, external and internal policies, and the economic and political environment. The MADIA study involved detailed analysis of six African countries- Kenya, Malawi, Tanzania, Cameroon, Nigeria, and Senegal. In addition to the World Bank, seven donors, USAID, UKODA, DANIDA, SIDA, the French and German governments, and the EEC participated in the study. The analysis of country policies and performance during the last 20-25 years was carried out with the benefit of substantial input from the governments and nationals of each of the countries represented. The study had three main areas of focus: I1) the relationship between domestic macroeconomic and agricultural policy and agricultural performance, 12) donors' role in the development of agriculture, and 13) the politics of agricultural policy. The MADIA study was the result of encouragement and support from many people. Anne Krueger, former Vice President for Economic Research Staff in the World Bank, encouraged the establishment of these studies on aid and development in 1984. Gregory Ingram, former Director of the Development Research Department, provided unstinting support for the study. During the reorganization of the World Bank in 1986, the strong support from Benjamin King, then acting Vice President for Economic Research Staff, proved invaluable. Barber Conable, President of the World Bank, and Mr. Edward V K. Jaycox, Vice President for the Africa Region, have played a key role by ensuring support for the study's completion, as did Stanley Fischer, the Vice President for Development Economics. Yves Rovani, Director General of the Operations Evaluation Department, was particularly helpful as the MADIA study drew heavily on the works of OED. A special debt of gratitude is owed to the World Bank's Research Committee, which provided the initial funding for the study, and to the MADIA Steering Committee. In particular the strong support of the chair of the Steering Committee, Stephen O'Brien, has been of critical importance. Finally, without the active and continued encouragement of many African policymakers and donor officials, including numerous colleagues in the World Bank, this study would not have provided new perspectives. This support has taken the form of numerous reactions to written and oral presentations, and refinement of the analysis to identify the areas of consensus and continuing controversy. CD 0 The World Bank Headquarters f7 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. a Telephone: (202) 477-1234 C Facsimile: (202) 477-6391 Telex: WUI 64145 WORLDBANK RCA 248423 WORLDBK Cable Address: INTBAFRAD WASHINGTONDC X CD European Office 7 66, avenue d'iena 75116 Paris, France X Telephone: (1) 40.69.30.00 (D Facsimile: (1) 47.20.19.66 Telex: 842-620628 0 Tokyo Office O Kokusai Building 1-1, Marunouchi 3-chome Chiyoda-ku, Tokyo 100, Japan n 0 Telephone: (3) 214-5001 0 '0 Facsimile: (3) 214-3657 CD Telex: 781-26838 (D CD 0 0-82 13-1 3 27-4