Benin - Country economic memorandum (Vol. 2 of 4) : The economic and social sectors

Report No. 4686-BEN Benin Country Economic Memorandum (In Four Volumes) Volume ll The Economic and Social Sectors March 15,1984 Western Africa Region FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties Its contents may not otherwise be disclosed without World Bank authorization  CURRENCY EQUIJVALENTS Currency Unit = CFAF US$1.00 3 CFAF 355- CFAF 1,000 US$2.81 WEIGHTS AND MEASURES 1 meter (m) - 3.28 feet (ft) 1 kilometer (km) 2 0.62 mile (mi) 1 square kilometer (km ) = 0.386 square mile (sq. mi.) 1 metric ton (m ton) = 2,204 pounds (lb) 1 hectare (ha) 3 2.47 acres 1 cubic meter (m ) 1.308 cubic yards FISCAL YEAR January 1 - December 31 1/ The CFA Franc (CFAF) is tied to the French Franc (FF) in the ratio of FF 1 to CFAF 50. The French Franc is currently floating. Throughout the text the CFAF/dollar equivalents are established using the rate of CFAF 3551$.  FOR OFFICIAL USE ONLY ABBREVIATIONS AND ACRONYMS AfDB African Development Bank AGB Societe d'Alimentation Generale du Bgnin BCEAO Banque Centrale des Etats de l'Afrique de l'Ouest BBD Banque Beninoise de Developpement BCB Banque Commerciale du Bgnin BOAD Banque Ouest Africaine de Developpement CAA Caisse Autonome d'Amortissement CARDER Centre d'Action Regionale pour le Developpement Rural CATS Cooperative Agricole du Type Socialist CCCE Caisse Centrale de Cooperation Economique (France) CEB Compagnie Electrique du Benin/Communaut6 Electrique du Benin CNCA Caisse Nationale de Credit Agricole CPU ColMhge Folytechnique Universitaire DEP Direction des Etudes et de la Planification DPE Direction de la Planification d'Etat EC European Communities ECOWAS Economic Community of West African States EDF European Development Fund ENJ Ecole Normale Superieure FAC Fonds d'Aide et de Cooperation (France) FAS Fonds Autonome de Stabilisation et de Soutien des Prix de Produits Agricoles FED Fonds European de Developpement FLASH Faculte des Lettres, Arts et Sciences Humaines FNI Fonds National d'Investissement FRA/GTZ The German Association for Technical Co-operation FSA Faculte des Sciences Juridiques, Economiques et Politiques FST Faculte des Sciences Techniques GRVC Groupement Revolutionnaire a Vocation Cooperative IBETEX Industrie Beninoise des Textiles IDA International Development Association IFAD International Fund for Agricultural Development INSAE Institut Nationale de la Statistique et de l'Analyse Economique INEEPS Institut National pour 1'Enseignement de l'Education Physique et Sportive INSS Institut National des Sciences de la Sante INE Institut National de l'Economie INSJA Institut National des Sciences Juridiques et Administratives MDRAC Ministere du Developpement Rural et de l'Action Cooperative MPSAE MinistereduPlan de la Statistique, et de l1Analyse Economique OBEMAP Office Beninoise des Manutentions Portuaires OCBN Organisation Commune Benin-Niger des Chemins de Fer et des Transports PAC Port Autonome de Cotonou PAM Programme d'Alimentation Mondiale SBEE Socifte Beninoise d'Eau et d'Electricite SCO Societe des Ciments d'Onigbolo SNAFOR Societe Nationale pour le Developpement Forestier I This document has a restricted distribution and may be used by recipients only in the performance of 1 their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  ABBREVIATIONS AND ACRONYMS (continued) SOBEPALH Societe Beninoise de Palmier'a Huile SOBEMAC Societe Beninoise des Materiaux de Construction SOBETEX Societe Beninoise des Textiles SONACEB Societe Nationale de Commercialisation et d'Exploration du Benin SONACI Societe Nationale des Ciments SONACOP Societe Nationale de Commercialisation des Produits Petroliers SONAFEL Societe Nationale des Fruits et Legumes SONAGRI/SONAPRA Societe Nationale pour la Production Agricole SONAPECHE Societe Nationale d'Armement et de Pache  VOLUME II THE ECONOMIC AND SOCIAL SECTORS Table of Contents Page A. AGRICULTURE .................................................... I Introduction .................................................... 1 Production trends ...... ....................................... 2 Food Crops . ................................................ 2 Industrial crops ..... .0..... *.*........ .... ......................3 Provincial structure of production ...o ...................... 7 Future production strategy .................. ............... 9 Structure of production ........................................ 8 Farmer organizations .................... ........ . .... . . . ...9 Public enterprise in production and processing .... .......... 10 Producer incentives and support . ............ . ............. 14 Pricing and marketing policy ............. o.................14 Input and subsidy policies . .............................. 17 Agricultural credit ........................ .. ...... .. ..... 18 Applied research and seed production ........... ........... 20 Rural development institutions ................... .......... 21 Strategic choices for agriculture . . .................... o ..... 22 General Policy Orientation ................. . ............... 22 Summary of analysis ........................................ 23 Conclusions ..................................... .......... 24 B. LIVESTOCK .................................................... 35 C. FORESTRY ....... . .... . . ........... ....................... 37 Potential ...... 37 Output and demand ... ................................ * ....... 37 Main issues ... oo ............................................. 38 Institutional setting projects.......o.......................... 39 Ongoing projects .............. o..................................... 39  D. FISHERIES ................................ 40 Industrial fishing . . . .................................. . 41 Artisanal fishing .... . . .......................... 41 Coastal fishing ................................ 41 Issues .. .. . 42 E. THE INDUSTRIAL SECTOR ................................ 42 The Structure of the Sector ................................ 42 Major industrial operations ................................ 44 Agro-processing .......................... 44 F. ENERGY .......................... 55 Sectorial Institutions ........................ 55 Electrical Power .......................... 55 Petroleum .......................... 56 Seme Field .......................... 57 G. TRANSPORT ........................... 58 Sector Issues .......................... 59 Improving Efficiency of and Increasing Funds ................... 60 Postponing Paving Roads Prematurely . . 61 Assessing Needs . .. . ......................... 62 Developing small-scale Contractors ............................ 63 Reviewing TransBenin ........................................... 64 Avoiding Creation of Excess Capacity on Benin Route . . 64 Defining the Future Role of OCBN . . . 65 Avoiding Major Investments in Cotonou Port . . 66 Motivating and Upgrading Staff .. . 66 H. TELECOMUNICATIONS ............................................ 72 Network ...................................................... 72 Institutional Setting ........................................ 72 Issues ..................................................... 73 I. EDUCATION ................................................... 74 Structure ...................................................... 74 Organization ................................................... 75 Recent Changes ................................................. 76 Education reform ............................................... 76 Plans and issues ............................................... 78 J. HEALTH .. .................................................... 79 Plans ... 80  K. THE URBAN SECTOR ................... 81 The Urban conditions ..... 81 Urban Road Infrastructure ..... 81 Water Supply ..... 82 Sewerage ..... 83 Stormwater Drainage . ......................... . . 83 Electricity ............................ . 83 Housing ..................................................... 83 Sectoral Institutions ............................ 84 Urban Management: The Issues ............................ 85 Tables in the text 1 Foodcrop production, principal crops, 1974-76, 1979, 1980, 1981 2 Selected industrial crop production, 1974-76, 1979, 1980, 1981 3 Agricultural prices, food and cash crops, 1974-76, 1979, 1980, 1981 4 Provincial output as percent of national agricultural production, 1979-81 5 Evolution of Cooperatives, 1978/79-1981/82 6 Benin: Agricultural Performance, 1969-1981 7 Structure of Land Cultivated by Region and Main Crops, 1979-81 8 Net Income per Man/Day per Hectare; average, 1979-81 9 Evolution of average Annual Production by Province, 1969/71-1979/81 10 Evolution of Yields by Province, 1969/71-1979/81 11 Evolution of Percentage of National average Annual Production by Province, 1969/71-1979/81 12 Oil Palm Products, Industrial Production and Sales, 1976/77-1980/81 13 Volume, Price and Value of Exports of Cotton 14 Estimates of National Livestock, 1980 15 Demand for fuelwood and charcoal, 1978,1996 16 Fisheries Output, 1975-1979 17 Industrial sector Value added, 1977-1981 18 Production of oil products, 1975-1981 19 Planned investments in the manufacturing sector 20 Industrial Sector in Benin, 1981 21 Domestic consumption of fuel products 22 Transport Trends, 1978-1982 23 Cotonou Port Traffic, 1976-1985 24 Air Traffic, 1976-1980 25 Rail Traffic, 1976-1981 26 The Road Network 27 Road Construction and Road Maintenance Expenditures, 1975-1981 28 Schooling rates by sex and Province, 1971/72-1980/81 29 Student Enrollments, 1972-81 30 Educational Costs per education level   AGRICULTURE Introduction 1.01 Agriculture is the most important sector of the Beninese economy, employing about 70% of the active population, and accounting for about 40% of GDP. The sector is predominantly food crop oriented, producing yams, cassava, maize, sorghum, beans, and small quantities of rice. The main export crops have been cotton, oil palm products, and groundnuts. Farming patterns vary markedly by region. In the densely populated small southern provinces of Mono, Atlantique and Oueme, together containing only 10% of Benin's territory, 18% of available land is cultivated, average farm size is 1.2 ha, and two rainy seasons permit double cropping. Maize is the main crop, accounting for about 70% of the area cultivated. In the large northern region -- Atacora and Borgou provinces, accounting for 28% and 45% of Benin's territory, respectively -- with low population density and a single rainy season, under 4% of available land is cultivated. The cropping pattern is less intensive and the average farm size is 2.4 ha. Sorghum, millet and yams are the major food crops. The Zou Province containing 17% of Benin's territory in the center, is a transitional zone, displaying characteristics both of the south and the north. Benin agriculture is dominated by family type farms with an average of 3.3 active workers per farm. Most agricultural exploitations are individually owned, and while the number of agricultural cooperatives has increased from 215 in 1978 to 442 in 1981, individual farmers still acount for about 90% of the agricultural output. 1.02 Traditionally, Benin has been self-sufficient in food production and has generated significant foreign exchange earnings from the exports of cotton, oil palm products and, until recently, groundnuts. As much as 20% of Benin's food production is estimated to be clandestinelly exported to Nigeria. Since 1979, however, production has generally stagnated. At the same time, the urban population has grown at about 6% per annum, compared to 2.7% for the country as a whole, which together with Nigerian demand, has led to rising pressure on food prices. Production of food crops, which is in the private sector, has bid available resources away from production of traditional export crops, whose marketing is excl1isively government controlled. This has deprived Government of needed revenue with which to support the existing official marketing and processing infrastructure, as well as general budgetary support. 1.03 Government has been unable either to find new agriculturally based revenues to substitute for those lost on industrial crops, or to channel flows of food to the urban areas. Government has thus been forced to import growing quantities of food, and to support more of the recurrent cost of its official marketing and processing infrastructure from general tax revenue. These conditions have worsened government's overall fiscal situation. 1.04 The following sections contain an analysis of the trends in agricul- tural production that have occurred since 1970 in both food and export/ industrial crops. This is followed by an analysis of the structure of agricultural production, incentives and institutional support. Finally, consideration is given to major options open to Government, accounting for domestic and international market developments, and recommendations for a strategy for future agricultural development.  2 PRODUCTION TRENDS Food Crops 1.05 Food crop production represents more than 90% of total value of agricultural output and occupies around 84% of area cultivated (including maize 30%, sorghum and millet 13%, cassava 10%, oil palm 10%, groundnuts 9%, beans 9%, yams 7%, and others 12%). As shown in Table 1, production generally declined over the period 1970-1976, largely because of the institutional changes I that affected the sector after the change in Government structure. Since 1976, however, better weather and higher prices, due mainly to increased border trade with Western Nigeria, have resulted in a certain recovery in food crop output, although by 1980 and 1981, little further overall growth occurred. Since 1976, a shift in the structure of agricultural production has occurred. Increases in maize production (from 196 to 301 thousand tons), have been realized, partly due to introduction of new hybrid seeds; and in yams (from 590 to 667 thousand tons). The remaining crops performed less well. Cassava, whose output fell by 40% between 1966 and 1974, recovered slightly in 1976-79 but fell again in 1980 and 1981. Production of some traditional crops such as sorghum and millet continued to decline, as farmers substituted other food crops such as maize and yams with higher demand in domestic and neighbouring markets, for crops traded less widely. The decline in rice production is associated with institutional problems affecting the state enterprise in charge of rice production (para 3.12) Details of this evolution by crop and province are given in Table 6. Table 1. FOODCROP PRODUCTION, PRINCIPAL CROPS Index Average Level (1969-71 = 100) (000't) 1974-76 1979 1980 1981 1978-81 Maize III 159 140 149 302 Sorghum/Millet 133 132 121 125 59 Yams 93 133 132 127 689 Cassava 82 102 80 79 639 Beans 66 133 101 99 31 Rice 309 298 209 209 10 Weighted average* 94 129 116 115 iource MDRAC Weighting based on 1978 prices. 1/ The Government holds the viewpoint that the decline is essentially attributable to the removal of certain benefits enjoyed by extension workers which led to a general loss of motivation in their ranks.  3 1.06 Maize production increased in all provinces between 1974-76 and 1979- 81, most frequently at the expense of cassava in the south and traditional cereals and groundnuts in the north. In most provinces, the rate of increase has exceeded the rate of population growth. Similar increases in yam produc- tion have also occurred. However, prices have not fallen, which indicates that a viable trade network has developed. On the other hand, average yields have decreased since 1969-71 for most crops except beans and cotton which indicates a general extensification of agriculture and a general lowering of labor productivity. General shortages of modern inputs and degradation of improved seed have contributed to this problem. 1.07 In spite of increased agricultural production since 1974-76, output has been insufficient to meet the demand for food of the main urban centers (notably Porto Novo and Cotonou) after accounting for exports of about 20%. Imports of cereals have therefore increased substantially -- from 34,000 tons in 1976 to 110,000 in 1981. With rising world market prices, Benin's food import bill more than tripled in current terms over this period and in 1981 accounted for 6% of total imports. This has placed a particularly heavy burden on Government since Benin was already suffering balance of trade deficits. Government considered once to impose controls over export of food crops, but administrative enforcement would be difficult. Furthermore, it would be economically unwise to control food exports for Western Nigeria. 1.08 In conclusion, it is clear that Benin has enjoyed a comparative advantage as a food crop producer given its particular geographic location. Substitution of easily traded fooderops (maize) for more locally consumed traditional crops has occurred. However, while production is higher now than in the mid 70s, production has begun to stagnate with yields trending down- wards. This has occurred in spite of general price increases for foodcrops which have averaged 13.7% p.a. between 1975-1980 and up to 17.3% p.a. between 1979-81. This lack of productivity may in part be accounted for by poor weather but more significantly, because of shortages of modern inputs and new varieties, which have constrained producer responsiveness to price. This situation is also being exacerbated by reduced labor availability, caused by rural-urban migration and by migration to neighboring countries. Industrial crops 1.09 The general level of Benin's industrial crop production including cotton, oil palm products and groundnuts has been declining steadly since the mid 1970s. Table 2 illustrates that on a weighted index basis, production of industrial crop has fallen to about 85% of its 1969-71 level, and about 86% of its 1974-76 level. Moreover, most of this decline has occurred since 1980. The recent erosion of industrial crop production has been greater than the erosion of food crops production which indicates a relative shift of resources into food crop production away from industrial crops. This trend is motivated to a large degree by Government's official price policy since for cotton and groundnuts, Benin appears to have suitable climatic and environ- mental situation, capable of giving adequate yields. As shown in Table 3, on an index basis, prices for food crops which are set by the private trade have risen on average about 17% per year,(also this may be largely underestimated) whereas official industrial crop prices have risen by only 13% per year over  4 the decade. In the case of groundnuts, where production has increased marginally since 1974-76, a significant food market exists in parallel with the official market. However, even for this crop, production has recently begun to decrease. 1.10 The average annual value of industrial crop production 1978-80 valued at producer prices, is estimated in current terms to be about US$23 million and accounts for only about 10% of farm level agricultural income. The FOB value of this production is estimated to be about US$29 million for the comparable period. However, in the latest years for which complete data are available, 1977 and 1978, Government realized about US$10 million per year from the exportation of these crops of a possible US$22 million. This repre- sents only about 45% of the export value potential. This occurred because in this period, and since then, Government was unable to compete with private traders for commodities (particularly groundnuts), and followed a policy of using what supplies it could secure to support the country's processing facilities. As there are believed to be clear inefficiencies in Benin's processing this latter practice represents a considerable loss to the economy. As a result of its export policy, cotton has emerged as the most significant export commodity accounting for between 25-30% (1975-78) of Benin's recorded export revenues. 1.11 Oil Palm Products. Oil palm fruit has traditionally been gathered from wild groves, covering about 400,000 ha in the southern provinces. Yields of fresh fruit are estimated to be about 5t/ha which is low compared to yields obtainable on estates (12-15t/ha in West Africa and 25t-32t/ha in the Far East). Because of the oil's color and taste, consumers have shown a strong preference for that extracted from the fruit through artisanal processes. However, as the artisanal oil market has not been fully studied, it is not known how much of the estimated 50,000t per year national demand is being met from this source, although it is probably large. For a number of years, in the 1950s and 1960s, various efforts were made to improve artisanal productivity with only mixed success. In parallel, between 1962-75 about 28,000 ha of estate plantations were established with the object of increasing palm oil availability for local consumption, processing and export, supported by industrial scale oil extraction. IDA jointly supported this strategy with FAC in 1969. It was clear at that time, however, that potential yields of 8t/ha were about 50% lower than those possible in more favored zones in West Africa and during project execution even yields of 8t/ha did not materialize. Moreover, in 1968/69, neither the rapid growth in food crop production which subsequently occurred, nor the strong non-farm competition for labor, were anticipated. Organizational and management changes have also occurred which were originally identified as posing a risk to the viability of the estate production strategy. Thus, while estate production was viewed as economically attractive in the early 1970s, it has become obviously unattrac- tive today. 1.12 Recent production of oil palm products is shown in Table 9. These data show that palm oil extraction provided during the 1977-80 period an average of 3,800 tons of edible oil per year covering only 8% of national consumption, so that the original intent of producing significant amounts of palm oil from estates for the domestic edible oil market has not been realized. Instead, most estate produced palm oil is used in the local soap industry. However, since estate produced palm oil is being sold at only 74% of its import parity value, the local soap manufacturer (a public enterprise, SONICOG) is receiving an implicit subsidy.  5 1.13 It is clear that low productivity and lack of competitive advantage is due mainly from climatic factors that cannot be feasibly changed. This means that oil palm production should play a decreasing role in agriculture in the future and no new estate plantings should be undertaken when present palms become too old for useful production. On the other hand, as considerable fixed investment already exists, it would be economically rational to make short-term efforts to increase efficiency and production based on existing plantations and processing facilities, instead of building new mills or replanting. Table 2: SELECTED INDUSTRIAL CROP PRODUCTION Index Average Level (1969-71=100) (000't) 1974-76 1979 1980 1981 1979-81 Oil palm (fresh fruit) 116 73 76 88 356.0 Seed Cotton 71 72 46 39 18.7 Groundnuts 99 142 135 111 60.2 Tobacco 41 62 37 51 0.4 Weighted Average* 98 98 90 85 Source: MDRAC * Weighted by 1978 official producer prices. 1.14 Cotton. Cotton production was successfully introduced in Benin in 1963 under bilateral technical assistance. Output of seed cotton rose quickly from about 5,000 tons in 1965 to 50,000 tons in 1972, but subsequently fell, reaching a low of only 16,000 tons in 1976/77. This decline was partly a consequence of political changes in 1972 which brought about a fundamental restructuring of the management of the sector's supporting organizations, ended much of the expatriate technical assistance and reduced the priority given to cotton production relative to food crop production, despite efforts made under the joint IDA-FAC Zou-Bourgou Rural Development Project (FY72). Production has since partly recovered, rising to 18,000 tons in 1978/79 and to 25,000 tons in 1979/80. However, problems with input supply ( lack of timely availability of fertilizer and pesticides) and unremunerative producer prices caused it to decline again to 16,000 tons in 1980/81. 1.15 Seed cotton yields in Benin declined since the early 1970s from an average of about 0.8 t/ha 1969-71 to about 0.7 t/ha in 1979-81. This decline has been associated with worsening availability of modern production inputs, in a timely fashion, rather than climatic or environmental factors. Assuming that organizational constraints in inputs, marketing, and extension are over- come, future yields of seed cotton are expected to rise by 1988/89 to about  6 1.0 t/ha in the Borgou and Zou Provinces, with progressive farmers exceeding 1.1 t/ha. These yields are at the same level of yields obtained in Togo, although about 40% of the yields obtained in experiment station trials in Togo and Nigeria, under climatic and environmental conditions similar to those in Zou and Borgou Provinces. Therefore, cotton production in Benin has a sound agronomic basis and recent IDA financed projects in these provinces have been designed to capitalize on this advantage by relieving institutional and financial constraints that retard production. Table 3: AGRICULTURAL PRICES, FOOD AND CASH CROPS (Index: 1969-71=100) 1974-76 1979 1980 1981 Food Crops Maize 167 245 261 280 Sorghum/Millet 167 261 300 326 Yams 163 225 240 255 Cassava 125 234 281 328 Beans 121 211 229 256 Weighted Average* 149 233 258 321 Industrial Crops Palm fruits 100 114 114 114 Palm kernels 135 157 169 181 Cotton 127 160 189 235 Groundnuts 220 220 275 330 Tobacco 133 156 164 172 Weighted Average* 131 152 174 194 Source: MDRAC * Weighted by quantities produced 1978. 1.16 Groundnuts. Groundnuts are the third most important cash crop in Benin, and the only one whose production has consistently increased since early 1970s (table 2), owing to the existence of a strong open market for the crop. However, in spite of the existence of a strong open market for the crop, production declined of about 50,000 per year in 1981 from 62,000t in 1979-80. This has been due to degradation of improved seed and by lack of modern inputs. Owing to its practice of offering prices that are consistently lower than open market prices, Government has never been a significant buyer of groundnuts, acquiring only 8,000t (1976/77) or 13% of that years's production. In 1979/80, Government acquired only about 1,000t or 2% of that years's total production. What Government did purchase has been used mainly for oil extraction for local markets. There have been no significant official exports of groundnuts in the past four years.  7 1.17 As a traditional crop, groundnuts have been widely grown in Benin, with major concentration (55%) in the Zou Province. Yields have varied from between 400 kg and 600 kg per hectare, which, while low, are comparable to those in Togo and other West African countries. However, even with these yields, groundnuts have been competitive in crop rotations. With reform in the delivery inputs and improved seed, average yields are predicted to rise to between 650 (Atacora) to 850 kg/ha (Zou) with the most advanced farmers achieving 900-1100 kg per ha by 1988-89. Therefore, groundnuts should con- tinue to be an attractive crop for the foreseable future. Provincial Structure of Production 1.18 The structure of agricultural production varies considerably between northern provinces (Atacora and Borgou) which have between 800-1100 mm rain- fall in one season, and the southern provinces (Atlantique, Mono and Oueme) which have 1000-1500 mm rainfall in two seasons, permitting double cropping. The Zou province, which also has a bimodal rainfall, receives 1100-1200 mm per year. Farmers have adapted their production patterns to these environmental conditions and Government has, as a matter of policy, respected this in promo- ting agricultural production. As a result, a pattern of production has emerged in which Atacora and Borgou dominate overall production and together account for about 35% of the national food production by volume. Zou province contributes another 34% and the three southern provinces the remaining 31%. Borgou Province also dominates cotton production, followed by the Zou province. With the exception of maize production, 72% of which is produced in the three southern provincs, most cereals and yams are grown in the center and north. A summary of the percentage distribution of production for 1979-81 is given in table 4 below. Details on the provincial distribution of production and yields, and a brief synopsis of the agricultural situation by provinces are given later. Table 4: PROVINCIAL OUTPUT AS PERCENT OF NATIONAL AGRICULTURAL PRODUCTION 1979-81 (in percentage) Atlan- Oueme tique Mono Zou Borgou Atacora Overall Share 15 9 7 34 13 22 Maize 34 23 15 19 6 3 Sorghum -- -- -- 14 42 44 Yams 1 -- 1 35 22 41 Rice 9 -- 5 14 28 4 Beans 18 4 9 47 12 10 Cassava 23 15 12 35 5 10 Groundnuts 15 4 11 56 6 8 Cotton 1 - 5 36 54 4 Source: MDRAC and Bank estimates  8 1.19 The situation in 1979-81 is in contrast with that which existed in the early 1970s, when the three southern provinces contributed 51% of food production, Zou province 19% and Borgou and Atacora 30%. This evolution has occurred, in spite of the fact that yield potentials in most crops except cotton and sorghum are as good if not better in the south than the north (see Table 10). Although it is not fully documented, increasing urbanism, competition for labor both in southern Benin and in neighboring countries, and increased competition for available land, are important factors explaining the declining importance of southern agriculture. Moreover, the possibility that cotton could be widely grown with acceptable yields in both Zou and Borgou provinces gave these provinces a basis on which a financially viable system of input distribution, seasonal credit, and highly focussed extension services could be built. A secondary benefit of this has been that production of cereals, tubercules, and legumes was also stimulated. Similar services provided through extension and credit institutions not linked to an export crop have not performed as well, although a recently initiated project in the Atacora province (financed by IFAD) is expected to give positive experience using cereals production as a basis for development. The Zou province presents a special case in that it has enjoyed high yields in all crops, has had an export cash crop (cotton), and has been sufficiently close to urban areas where demand for food crops has stimulated production, without suffering from urban pressure. Future Production Strategy 1.20 Government's best future production strategy would be to continue building on traditional cropping patterns, focussing on attaining productivity increases. This would preclude attempts to introduce new or exotic lines of production into Beninese agriculture, for which the production risks are high, and the need for supplementary infrastructure great. Moreover, such attempts would divert scarce Government and farmer resources away from improving existing lines of production where risks are lower. In most traditional lines of food production, including cereals, legumes (groundnuts) and tubercules, regaining yield levels already achieved in Benin would continue to provide a good basis for viable production. Indicative world market price increases expected for these crops are expected to increase at least to 1995 (maize, sorghum, 1.3% p.a.; rice, 1.6% p.a.; groundnuts and oil, 2.1% p.a.) in real terms. With yields that are possible, this should continue to give Benin a comparative advantage in these crops. Low yields for oil palm fresh fruit limit the economic production of this crop in Benin, following a costly estate and cooperative production system, in spite of projected 1.6% p.a. growth in real prices for palm kernels and a 1% p.a. growth rate for palm oil. On the other hand, for cotton, whose yield prospects are good, the price is expected to rise by 2.1% p.a., production should continue to be economically justified. STRUCTURE OF PRODUCTION 1.21 Within Beninese agriculture, production is predominantly in the hands of traditional individual family farms. There are approximately 300,000 such holdings whose size varies from 1.1 ha in Oueme to 2.7 ha in Borgou. These exploitations currently account for 90% of the value of agricultural production. The principal areas in which Government is actively involved as  9 agricultural producer is in the case of oil palm estates, irrigated rice production, maize and sugar. In no case has this form of production proven to be economically viable, mainly because of high production cost and low work efficiency, rather than low yield potentials. Government has also organized several apex companies which have tried to establish and manage farmer coope- ratives for the product of specific crops, supporting these with input supply, technical advice, and marketing services. However, the only crop for which this formula has been successful has been cotton which is uniquely an indus- trial export crop with no alternative market to the official marketing channel. 1.22 Given the individualistic nature of production, the basic problems to resolve in stimulating production in the future will be in the areas of incentives, input supply and distribution, and organization of support for the sector. In addition, it is inevitable that the number of farms and their regional distribution will adjust so that there will be fewer farms overall and fewer farmers serving as the sole source of family income. Moreover, future agricultural growth will probably occur in the central and northern provinces, away from the influences of urbanism. These factors will set important constraints and demands on addaptive research and extension, marketing, input supply and credit, which will be discussed below. However, in spite of their individualism, Beninese farmers have shown considerable willingness to form cooperatives on the basis of maintaining individually owned land and the role of cooperatives will become increasingly important in the future. Farmer Organizations 1 .23 The cooperative movement was formally launched in Benin in 1971 with a program to create village groups (Groupement Villageois or GVs). These were volunteer associations whose objective was to coordinate the distribution of fertilizers and agricultural chemicals, extension services, and marketing services in the case of officially marketed crops such as cotton, groundnuts and to some extent oil palm products. In 1974, Government decided to emphasize food crop production over cotton production which removed the initial motivation for farming cooperating. However, Government retained its interest in official marketing and continued to promote farmers' groups as an instrument to facilitate this. In 1977, a new type of farmers' group was introduced, the "Groupement Revolutionnaire a Vocation Agricole" (GRVC). This type of organization required that farmers group their individually owned plots into blocks (such as had been done previously for cotton production) while making independent but coordinated decisions concerning use of labor, their equipment, and agronomic practices. In return, the GRVCs were to benefit from preferred access to input supplies, credit, extension services and marketing services as were the GVs. Between 1978/79 and 1981/82, the number of GRVCs increased from 190 to 473. A summary of the evolution in the numbers of farmer groups is given in Table 5. In 1981/82, the GRVs included about 15,000 farms or about 5% of the country's total who cultivated, globally, about 20,000 ha. Government has also experimented with forming farmer groups based on collective resource use and decision making. Between 1978/79 and 1981/82, 42 "Cooperative Agricole Experimental de Type Socialiste" (CAETS) were organized including 800 members and covering, globally, about 1,000 ha so that the collective approach has neither been widely accepted nor significant. Where CAETS have been formed, Government has provided large  10 subsidies in the form of free inputs and mechanized services. However, yields have not reached those obtained in traditional exploitations or GRVC's and this form of cooperative has proven to be uneconomic. In addition to the CAETS, Government has promoted the creation of youth clubs, with the object of having each club evolve into a GRVC. However, young farmers appear to prefer working on their families' holdings rather than within clubs and there are only about 200 of them with about 3,000 members today. It is clear from this experience that Beninese farmers can be encouraged to form associations provided that the objectives of the associations are focussed on production goals which respect technically and economically sound norms. Table 5: Evolution of Cooperatives 1978/79 - 1981/82 -----G.IR.V.C- ----------C.A.E.T.S.-- Provinces 1978/79 1979/8n 190/81 1981/82 1978/79 1979/8n 198/81 1981/82 Atacora 20 52 80 111 1 3 5 9 Atlantique - 20 25 24 - 1 1 1 Borgou 97 8 87 51 25 43 25 17 Mono 1 20 20 64 - 1 4 2 Oueme 24 111 112 138 - 1 1 - Zou 48 74 76 85 - 1 5 13 National 190 363 400 473 26 50 41 42 Source: MDRAC 1.24 GRVC/GVs provide a vehicle for effective and efficient promotion of agricultural production. They provide a means of collective responsibility for credit, and sharing of extension information. However, to date Government has established ridgid norms for the formation of these groups, covering such aspects as grouped field size, cropping pattern, and sale of product at official prices. In many cases, these norms have not been agronomically sound or economically justifiable, and farmers have been unwilling to participate. On the other hand, Government has given highest priority to the distribution of inputs, credit, and support services to the groups respecting official norms most faithfully. Thus, full benefit has not been derived from inputs that are available and conflict and mistrust has been created between Government services and farmers. It is clear, therefore that a more flexible, technically sound approach to the formation of farmer groups would allow for the formation of more groups and a subsequent promotion of agriculture. Public Enterprises in Production and Processing 1.25 Since 1970 public enterprises have played a growing role in Benin's agriculture, particularly in the processing and marketing stages. In many cases this has also been closely linked to primary production, either through self-contained farming or through single-product producer cooperatives. However, almost all public enterprises in agriculture have become over-staffed and costly and less able to generate sufficient income (either by design or by inefficiency) to finance their own operations and pay remunerative prices to  1 farmers. Therefore, a fundamental issue is now facing Government in reassessing in what areas of the production and processing chain it has a clear public interest in participating, and where this participation is econo- mically justified. (a) Cotton 1.26 Cotton production has been a principle concern of Government concern since the mid-1960s, owing to its role as a foreign exchange earner (up to 30% of official export receipts). Within the cotton producing zones of Borgou and Zou provinces about 38,000 farmers now produce cotton (35% of all farmers in these provinces) and this number is expected to increase to about 75,000 by 1988/89. This also underlies a Government concern for the crop. Government's level of participation in cotton production has been limited to primary and overseas marketing, and ginning, which has been justified given the economics of scale that exist in ginning and the volume and consistency of quality demanded by international markets. 1.27 Presently, the provincial rural development agencies, the CARDERs, are responsible for primary marketing of cotton, which includes assembling seed cotton to central points, paying farmers and transporting seed cotton to ginneries. Until recently, the management of ginneries was the responsiblity of another organization (SONAGRI), but this responsibility has been transfer- red to the CARDERs for the Borgou and Zou provinces. After ginning, cotton lint and cotton seed are sold to the newly established SONAPRA (Societe Nationale pour la Production Agricole) which undertakes selling of the commodity overseas. The present organization of Government's participation in cotton marketing is the result of a recent reorganization, which amalgamated into SONAPRA the functions of (i) SONAGRI, previously responsible for cotton ginning, handling and delivery to Cotonou, as well as input procurement and distribution; (ii) FAS, responsible for price stabilization and subsidy payments; and (iii) SONACEB, responsible for export marketing. This reorga- nization should resolve several financing and coordination problems which hindered efficient operations in the past, and Government has received the support of external financing agencies (IDA, CCCE, FAC, IFAD) in strengthening the new organizational arrangement. (b) Oil Palm Products 1.28 SOBEPALH 1/ Government manages about 30,000 ha of improved oil palm plantations under the auspices of SOBEPALH (Societe Beninoise de Palmier a Huile) of which all but 830 irrigated hectares and 1100 hectares in immature palms are subdivided into lots which have been sold or subcontracted to 40 "cooperatives" with about 16,000 reported members. These cooperatives have paid staff and are responsible for harvesting fresh fruit bunches and deli- vering them to road sides. Apart from their legal structure as cooperatives, however, these cooperatives do not make independent management decisions, except to pay labor for field work, so that they are, in effect, work units within SOBEPALH. Indeed, only a small percentage (10%) of the members of the 1/ In a reorganization which became effective towards the end of 1982 SOBEPALH was dissolved and its industrial activities were transferred to SONICOG.  12 cooperatives work in the estates; the rest being hired from outside. Quality of the fruit collected is not monitored and is frequently low. SOBEPALH itself collects fresh fruit bunches from cooperatives, performs crushing operations and markets oil and palm kernels. SOBEPALH itself employs 1,200 persons in addition to the 1,600-2,000 employed by the "cooperatives". This operation is complex, has proven to be difficult to manage, and has high costs. Even with recent increases in price payable to"cooperatives" for fresh fruit bunches, only about 33% of the 40 "cooperatives" harvest sufficient fruit to cover their financial costs. These losses are eventually born by SOBEPALH which prefinances these operations. Moreover, for its part, SOBEPALH has also experienced persistent losses, which it has covered by bank overdrafts, because of payments to cooperatives and uneconomical crushing plant operation. The principal cause of this situation can be traced to the low yield and therefore low throughput potential of the entire operation, which is beyond management's control. 1.29 A number of technically sound proposals have been made in a recent Bank consultant's report 1/, which would reduce the overall level of unprofitability of the present SOBEPALH system. It is clear, however, that Government has gone too far into direct production and has developed a system which is not economically viable given the labor and yield constraints that exist. It is equally clear that the cooperatives involved in this activity bear no resemblance to the legitimate producer cooperatives upon which efficient production might be based. Moreover, with an average of 400 members per cooperative who have the right to share in the proceeds of the operations, but who do not actually work in the plantation, a considerable unjustified expenditure is being incurred. Therefore, Government should consider a change in its approach. As palm oil has been a significant food, produced on a subsistence level in the traditional sector, a new approach might include support for small scale palm oil and kernel production facilities using appropriate technologies, adaptable to conditions facing legitimate producer cooperatives. Such a system might also include the distribution of improved planting material. This approach would not, clearly, be aimed at exportation of palm oil, but rather at increasing the efficiency with which the local market is supplied. A CCCE-financed study on the rehabilitation of the oil palm sector is on-going. As long as firm conclusions on the profitability of such a rehabilitation have not emerged, further investment in state plantations should be avoided. 1.30 SONICOG. SONICOG (Societe Nationale pour l'Industrie des Corps Gras) is a state owned enterprise responsible for extracting and exporting a number of vegetable oils and by-products, from cotton seed, groundnut, shea nut, and palm kernels. Of this, the processing of palm kernels for oil extraction is its largest income earner. SONICOG also manufactures soap. SONICOG purchases SOBEPALH's entire output of palm kernels as the raw material for the production of palm kernel oil and palm kernel cake. However, the bulk of its purchases (about 85%) comes from small farmers in the traditional sector. SONICOG's production of both palm kernel oil and palm kernel cake has dimi- nished in recent years due mainly to shortage of raw material as shown in Annex Table 4. Between 1975-1980, palm kernels delivered to SONICOG fell from 50,000 tons to 30,000 tons, about half of the 60,000 tons processing 1/ Review of the Oil Palm Sector, Booker agriculture International, July 1982.  13 capacity. As a consequence, SONICOG's net income dropped by almost 50% over this period. However, the enterprise still appears to have been profitable in spite of paying taxes and making contributions to the national investment fund. (c) Rice and other Crops 1.31 SONIAH (Societe Nationale d'Irrigation et d'Amenagement Hydro- agricole) was established in 1975, with the mission of undertaking soil improvements, irrigation works, and bottomland's drainage, principally for the production of rice. The organization has followed two lines of activity: direct mechanized production of rice, and marketing and milling of paddy purchased from the traditional sector. A full and conclusive analysis of SONIAH's financial and economic viability is not available. However, SONIAH's production performance has been poor, with paddy production falling from about 1,100 t in 1978/79 to 240 t in 1980/81. This is a minor contribution when compared to the 10,000 t produced in the traditional sector and the 32,000 t paddy equivalent (21,000 t rice) imported into Benin in 1980/81. Moreover, SONIAH offers traditional sector paddy producers a price of about CFAF 40 per kg, which is considerably below traditional market prices of CFAF 100- 120/kg. Thus, SONIAH has not been an attractive marketing alternative for traditional sector production. 1.32 In this activity, Government has extended its activities too far into the productin process. Although not conclusive for Benin, experience in other West African countries with similar climatic conditions casts serious doubt on the economic viability of mechanized irrigated rice production and production systems requiring major land developments, indicate that when rainfed, low investment methods of producing rice are available, as they are in Benin, these should be preferred economically over mechanized production methods. SONIAH's marketing interventions are clearly not required, as viable traditional marketing channels appear to exist. Indeed, SONIAH has in the past been supported in its marketing by the force of civil authority, which has produced a serious disincentive for traditional rice production. Therefore, Government should consider terminating SONIAH's activities. 1.33 Government has established a number of state farms with the principle objective of providing raw products for other state processing industries. Maize was a frequent crop grown. Information on the economic performance of these farms is fragmentary, but indicates that they have not given yields which were comparable to traditional sector yields, and have been very costly in terms of investment, personnel and operating costs. Government has recognized that these farms were of marginal efficiency and is studying the possibilities of closing some of them and converting others into experimental or training farms, the latter with a view to permitting the gainful use of the knowledge acquired in the past and to perfect the techniques of operation of such units. It has also proposed that they may be transformed into research or seed production farms. Such a transformation would be difficult to achieve, however, considering the specialized nature of these operations and the lack of flexibility characteristic of the state management of these enterprises. The least costly and least risky option in our view would be for Government to permit these farms to be liquidated.  14 1.34 Sugar. Societe Sucriere de Save (SSS) was formed by the Beninese and Nigerian governments in 1973 with the object of producing raw sugar from 5,200 ha of irrigated cane, and processing this into about 47,000 t oL granulated and cube sugar per year. As Beninese demand for sugar is less than 10,000 t per year, it was planned that most of the sugar produced would be exported to Nigeria. It proved to be difficult to find financing for this venture and plant construction began only in 1977, with delays retarding completion to 1983. During this time, investment costs rose 21% and SSS continually lacked working capital. There is serious doubt whether estimates of cane sugar yield can be met. Moreover, it is increasingly doubtful that ex-factory prices can be maintained above prevailing CIF prices, as would be required to assure the long term financial viability of the enterprise. Because of these factors-- high costs, low yields, and low CIF prices--it is clear that sugar production is not economically or financially justified in Benin, and that it would be less costly for the economy to continue to import refined sugar as it has to date, rather than attempt to achieve domestic self-sufficiency and export. However, it is recognized that with investment near completion, most of SSS establishment costs have been incurred. Thus, Government would have difficulty in not assuring at least a minimum level of production provided that revenues cover operating costs. Therefore, it would be advisable for Government to initiate production as soon as possible based on the plantation it now has at its disposal, to verify whether operations are technically viable. A decision to increase the area under cane production to compensate for the low yield should be made only after the technical viability and marketing arrangements have been finalized. The solution, envisaged by Government, of extending acreage planted to other sites than Save would be technically justifiable only if the yields were adequate and easy transportation of the cane foreseeable. PRODUCER INCENTIVES AND SUPPORTS Pricing and Marketing Policy 1.35 The Beninese Government has followed a policy of setting official farm level prices for all food and cash crops, though less regularly for food crops than for cash crops. It has also attempted to organize various marketing systems for assuring the purchase of crops at the official prices via official marketing agencies for specific cash crops and via the CARDERs for food crops. Prices have been set by the National Pricing Commission, which has sought to balance a variety of considerations including the desire to guarantee low food prices in urban areas, and to distribute revenue among the various agencies involved in the official marKeting process to cover their costs. To do this, the National Pricing Commission has set annual baremes for the crops involved. 1.36 For non-food crops such as cotton, where there have been no alter- native marketing possibilities, the state marketing system has been able to control the flow of product. Even so, there have been serious issues in setting proper baremes so that the true costs of the state agencies are covered and farmers are offered attractive prices. For food crops where alternative markets do exist, voluntary compliance with official marketing has not generally occured since farm prices offered have invariably been below prices in non controlled open markets. In these cases, Government has attempted to use non-economic incentives to encourage farmers to sell to  15 official agencies. These attempts have been generally unsuccessful, as farmers have either marketed their crops privately or limited production to family needs so that marketable surplusses for domestic consumption have not been produced. A direct result has been that little or no food enters the official food marketing channels (except that, which is directly imported by Government), and a costly marketing apparatus has had to be supported by the national budget. Possibly more significant, however, farmers have been less than enthusiastic about establishing contact with Government agencies or the officially sponsored cooperative movement which means that they have not benefited as they should have from services provided by these agencies. This has particularly limited the formation of cooperatives by the CARDERs, which hold one of the best potentials for rapid acceleration of development. Thus, in its attempt to secure more food stuffs, Government's pricing and marketing policy has had an overall disincentive effect on agricultural production. Future for Food Crop Pricing and Marketing 1.37 The demand for food in Benin is expected to increase in the future due to the rise in domestic consumption and export demand from neighboring countries. With a population growth rate of 2.7%, the prospects for marketing food crops produced in excess of farm family needs, such as maize, groundnuts, beans and rice, are favorable. This demand is likely to remain strong over the foreseeable future. Markets in Nigeria and Niger have been absorbing significant quantities of Benin-produced food crops (estimated at about 20% of Benin's production) and are expected to continue to do so. Because of this, trends in domestic prices in the long run are not expected to be less favorable than those projected for international markets. Thus, prices in real terms for maize, sorghum, groundnuts and rice are expected to rise at between 1.3%-2.1% per annum to at least 1995. Such price increases would provide a strong incentive for the production, particularly if productivity increases are realized because of improved input availability. Marketing channels of food crops are satisfactory throughout Benin. Traders are personally appearing in villages and local markets, buying up to 80% of local maize stocks, 50% of groundnuts and 70% of beans. Small quantities of root crops such as cassava and yams are marketed in the processed form as gari and flour. Therefore,in the absence of disincentives created by Government policy, farmers are likely to respond positively to the price signals they receive from the open market, leading to increased produc- tion of foodcrops in the future. 1.38 The degree to which Benin's urban areas are supplied from domestic production is and has been a legitimate concern of Government, which has attempted to limit food price inflation in urban areas so that salaries and wages levels could also be controlled. In addition to attempts at official marketing of domestically produced crops, Government has also imported food stocks for sale in Cotonou and Porto Novo through AGB (Alimentation Generale du Benin), a 100% state-owned enterprise which distributes consumer goods. This enterprise was created in 1978 to purchase imported as well as locally- produced general consumer goods for resale to private retailers through ten regional wholesale depots, and directly to consumers through 19 supermarkets. Four food products for which AGB held until recently an import monopoly -- wheat, rice, sugar and milk -- have dominated its activities. AGB sells these products at official fixed prices, which are generally below competing market prices. As a result, between 1980-81, due to  16 the fact the Government did not allow AGB to pass increased import costs on to consumers, the firm incurred considerable losses (estimated at about US$3.0 million) on its food importing activities. These losses represent a subsidy to consumers paid by AGB, which were not compensated for by Government payments or profits on other products sold. As a result, AGB has ceased importing wheat, and continues to lose about 15% and 5%, respectively, on rice and sugar imports. For example, in March 1982, AGB imported 5,000 tons of maize, at the request of the Government, to cover apparent urban shortfalls and to dampen prices on the local market. However, AGB's receipts did not fully cover its costs and further losses were incurred. In the case of rice, moreover, it is estimated that 30-40% of AGB's rice is eventually sold to Nigerian consumers which represents a serious leakage in the public subsidy. It is clear, therefore, that Government's urban feeding policy of officially importing food stuffs and selling at fixed prices is costly, and subject to possible abuse. Moreover, the threat that quantities of food stuffs might be available at subsidized prices in Cotonou and Porto Novo would be a disincentive to traders to make supplies available in these cities. The Bank's economic dialogue with Government has pointed out the deficiencies in this approach to the urban-area cereal deficit problem and the Government is willing to reconsider its policy. 1.39 The official creation on December 8, 1983 of the Office National de Cereales (ONC), however, testifies to the fact that the Government's policy on the marketing of foodstuffs still continues to vacillate between liberalisation and intervention, and requires detailed study. According to the decree establishing the ONC, its principal objectives are: (i) to undertake the first collection of cereals following harvests, (ii) to study the problems related to keeping cereal prices stable (iii) to steady the course of the products by assisting in their production and (iv) to guarantee availability to the population and cereal processing units by building up reserve stocks. IDA has expressed its fears about the results of the creation of this office, mainly those of financial insolvency, distortions of the price of cereals and the encouragement of clandestine exports and, finally, the negative incentive effect on producers arising from price controls. In order to avoid all such undesirable future outcomes it would be useful to clearly define the exact roles of the ONC before the office becomes operational. For this reason, - the Government intends, with the collaboration of IDA, to undertake a study on food crop marketing to, among other things, identify the factors determining the level of foodstuff supplies to Cotonou and Porto Novo and, from that, to define the precise limits on ONC's intervention. Future Cash Crop Incentives 1.40 On average, over the period of the 1970s, the rate of increase in prices paid for cash crops (notably cotton) was markedly below that obtainable by farmers for food crops. This largely explains why production of these crops lagged, in spite of input subsidies (para 1.41) and extension service activities. Since 1980, however, Government has revised its policy on seed cotton pricing to account for the rising attractiveness of food crop produc- tion. Government has recently agreed to covenants governing loans and credits from IDA, CCCE, IFAD and FAC to finance rural development projects in Zou, Borgou and Atacora Provinces to review seed cotton prices annually, setting them to reflect prices of competing crops, cost increases generated by Govern- ment's new policy to reduce input subsidies (para 1.42) and world market  17 prospects. World market prices for cotton are projected to rise by about 2.1% per annum to 1990 in real terms and remain constant thereafter. This means that for the near term at least, world market conditions will permit Government some latitude to continue to offer incentive prices to farmers for cotton. In 1983, as a result of improved incentives, farmers expanded cotton acreage and output despite the drought which reduced yields. Input and Subsidy Policies 1.41 Benin imports all of the fertilizer and agricultural chemicals used in agricultural production, and Government has traditionally subsidized the use of these inputs heavily. Until a recent reorganization of institutions all of those inputs were imported by a single Government agency (SONAPRA) and used almost exclusively in cotton production. Government was able to finance subsidies by paying low producer prices for seed cotton on one hand, and by taxing the revenue earned by exporting cotton. As food crop prices became more attractive, however, fertilizer was diverted to food crop production which Government was unable to control and from which Government could not generate revenues to cover input subsidies. Moreover, under the complex financial managements which were governing cotton marketing (in which three separate organizations performed different functions: export marketing, cotton price stabilizatioi, input subsidy management, interior cotton marketing and input procurement-) inefficiency and lack of accountability led to a rapid depletion of funds which should have been available to finance subsidies. As a result Government increasingly had to resort to extraordinary measures to pre-finance the importation of inputs, and cover the cost of subsidies from the national budget. 1.42 This situation which existed till 1982/83 has recently improved owing, on one hand, to the regrouping of the three agencies dealing the marketing of cotton within a single agency (SONAPRA) and, on the other hand, to the start-up of the Borgou and Zou rural development projects with their respective financing and the introduction of new policies on subsidy. During 1982, Government, in fact, made significant changes in its input subsidy policy and in the administration, management and financing of inputs. Beginnning with the 1982-83 agricultural campaign, Government will begin a six-year phased elimination of all subsidies on fertilizer and insecticides which for 1981/82 campaign covered about 75% of the cost. Over the same period, prices paid for seed cotton will be increased to compensate for the increased input costs. Government also undertook a major reorganization in the cotton and input sectors involving the regrouping within SONAPRA of functions previously performed by the three organizations FAS, SONACEB and SONAGRI in areas of input procurement distribution and finance, cotton marketing and price stabilization. IDA, CCCE, FAC and IFAD have supported this reorganization through the provision of funds for technical assistance, vehicles and equiment and working capital. 1.43 In the circumstances in which Beninese agriculture found itself in 1981/82 the most serious constraint on production was the lack of assured supplies of modern inputs of acceptable quality, delivered in a timely manner, 1/ Described in detail in Zou Province Rural Development Project, IBRD Report 3969-BEN  18 rather than the price of which these inputs were sold. Farmers were already becoming aware of the probable increases in productivity which fertilizer and pesticides could bring. Many already paid cash for what limited supplies of these inputs were available. Therefore, provided that institutional and financial reform was achieved, that permitted increased availability of inputs, a rising price of these inputs was not seen as a disincentive to increased use. This fact is illustrated by analysis of the situation for foodcrop production in the Atacora Province, which is one of the poorest areas in the country, where cash income is low. In this province, the increase in revenue between 1981/82 and 1986/87 attributable to fertilizer induced yield increases is expected to exceed the incremental cash cost of fertilizer resulting from the change in Government's policy by between 3.1 to 3.5 times for maize, paddy and groundnuts (traded crops) and 1.6 times for sorghum which is a home-consumed crop. It is normally expected that if product prices, cash costs and productivity gains are such that ratios of 2.0-2.5 are obtained by such a calculation,this indicates that sufficient incentive exists to use the input even at higher costs, even accounting for risks that may be involved in becoming committed to a cash transaction. In the case of cotton, where Government's pricing policy would at least compensate for reduced input subsidies, a similar ratio of increased revenue to increased cash cost between 1981/82 to 1986/87 would be 2.0. This is also believed to indicate that sufficient level of incentive would exist in spite of higher input parts since cotton production has always been a cash crop, so that risks would be low compared to food crops. Government is expected to continue its program of providing seasonal credit to fertilizer and pesticide users, which will minimize the working capital requirements of farmers, and thus further dampen any disincentive effect of rising input prices (para 1.42). Agricultural Credit 1.44 The national agricultural credit agency, Caisse Nationale de Credit Agricole (CNCA) was established in 1975, under the direction of a national committee composed of the ministers of MDRAC and Finance and Commerce,and representatives of state societies. CNCA extends credit to state enterprises, to CARDERs, and through the CARDERs to farmer cooperatives. CNCA also lends directly to farmers via provincial level agencies (CRCAM) (para 1.45). The primary source of finance for CNCA has been the Central Bank of West Africa (BCEAO) Other sources of finance include the West African Development Bank and CNCA's clients' deposits. CNCAs greatest volume of loans have been to state enterprises, while the CARDERs and farmers have received less. However, state agricultural enterDrises have a bad repayment record, and SONAFEL, SONIAH and SONAGRI (now part of SONAPRA) all have had overdue loans with CNCA. The CARDERs and farmers have had a better repayment record than state enterprises, although this record could also be improved. The principal reason for the poor repayment record of state societies is that the state societies themselves have not been profitable ventures. In addition, CNCA lacks technical appraisal and loan supervision capability to identify and monitor risky proposals. When credit is granted to farmer cooperatives, collective responsi- bility for the debts by individual members has tended to limit, but not eliminate, defaults. Owing to other problems associated with cooperatives, however, only a comparatively small number of farmers have grouped themselves into cooperatives, which has limited the volume of credit extended to them.  19 1.45 Each province has a Caisse Regionale de Credit Agricole Mutual (CRCAM), which behave as provincial branch offices of CNCA. CRCAMs have further decentralized their operations into permanent and temporary CLCAMs (Caisse Locale de Credit Agricole Mutual) which function at the community level. The level of lending by CRCAMs and CLCAMs is restricted on the one hand by limited CNCA resources and on the other by the slow development of vehicles for lending to farmers that have acceptable risk and administration cost, such as cooperatives. The provincial networks have begun to mobilize rural savings as a possible source of future resources, but it will require considerable time for this source to develop. The CRCAM/CLCAM system presently suffers from technical appraisal deficiencies similar to those of CNCA, but all have close working relationships with CARDER technical staff, particularly as it pertains to credit to various farmer groups. 1.46 The CNCA/CRCAM/CLCAM network extends three types of agricultural loans: short-term or seasonal credit (one year repayment period), medium term credit (five years), long term (ten years). Short term loans are made for production as well as social reasons, permitting farmers to adjust their con- sumption needs over agricultural growing season. CARDERs also have obtained short term credit from CNCA mainly to finance the bulk purchase of fertilizer and pesticides. These inputs are subsequently soid to farmers with short term (seasonal) credit at the start of the growing season and the credit is reim- bursed after harvest. Medium-term credit is granted to purchase implements such as ox-drawn equipment and oxen for animal traction. Long-term credit has been provided to certain investments such as coffee and cocoa plantations, where the return on investment lags several years after the provision of credit. However, only a few long-term loans have been extended to date. With their present level of resources and staffing, and particularly with the procedures that are mandated for processing loans, CRCAM/ CLCAMs are not in a position to deal with the volume of short term seasonal credit that is likely to be required to obtain maximum benefit from the rural development projects such as those recently initiated in the Zou, Borgou and Atacora provinces. For these projects, in the short run at least, Government has adopted the strategy that the CARDERs would deal with CNCA directly on assuming the risk of seasonal credit on behalf of all of the interested farmers. Establishing this direct link between the suppliers of inputs and extension advice (the CARDERs) and farmers would be efficient, and minimize risks of non-recovery. However, this has not precluded CRCAM/CLCAM participation in medium term credit operations, nor the possibility that new procedures for administering seasonal credit could not be found in the future. However, sufficient information on which to base further discussions of wider issues of credit policy and administration is not available. 1.47 Interest Rates. CNCA charges interest rates permitted by the BCEAO which reflect the cost of money to CNCA and its administration charges. No specific margin was included however, to cover losses on the loan portfolio due to defaults. Prior to 1980/81, Government did not charge an explicit interest rate on seasonal credit extended to farmers for fertilizer and insecticides. Instead, costs associated with seasonal credit administration and risk of losses was borne directly by SONAGRI and the CARDERs which managed and financed input distribution. In 1980/81, however, Government introduced an explicit interest rate charge for seasonal credit, which permitted the CARDERs to recover a 2% margin over the rate of which it borrowed from CNCA. In 1981/82, the rate charged to farmers was 11%, which is lower than the rate  20 of inflation--currently estimated to be 12% per annum. Improvements are being made in the administration of input distribution and seasonal credit in the CARDERs and Government intends to review the record of recoveries and adminis- tration costs when sufficient experience is gained to determine what increases would be required. These policy changes could have a significant beneficial impact on the financial viability of continuing and expanding rural credit programs and contribute to motivation of rural savings. Applied Research and Seed Production 1.48 Agricultural research in Benin, which was for many years under the direction of overseas research institutes, now falls under the supervision and direction of the DRA (Departement de la Recherche Agronomique) within the Ministry of Higher and Technical Education. The DRA is organized into a number of Research Units throughout the country, which focus on specific crops. Food crops research is centered in three stations in the Atlantique, Borgou and Atacora provinces. Cotton research is headquartered in Cotonou and rice research is conducted in the Zou Province. As a rule each research station operates decentralized trial sites, although some rationalization of out-station sites is required to remove duplication of efforts in zones with similar ecological conditions. 1.49 Results of past research work have been translated into proven crop- growing recommendations for maize, groundnuts, cowpeas and cotton, which have already been used as a basis for planning of major rural development efforts in the Zou, Borgou and Atacora provinces. However, a critical lack of assured operating funds and, in some instances, qualified personnel has slowed progress in expanding research efforts into critical areas including farm system's research, and threatens to disrupt continuity of progress in some long-term trials. Since 1980, applied research has been highlighted in Government's Borgou, Zou and Atacora provinces rural development projects. In each of these cases, with varying intensities, technical and material assistance to existing crop-specific research efforts, and a strengthening of pre-extension and demonstration trials have been provided for. This has dealt satisfactorilly with the critical level of research, which is expected to relieve constraints on productivity rapidly, rather than focusing on up-stream fundemental research for which Benin does not have the resources or the need. 1.50 Seed Multiplication. As a direct result of applied research efforts, growth in seed multiplication has been made possible. Intensified efforts since 1981 include the production of improved foundation and registered maize, groundnut, sorghum, rice, cowpea and cotton seed in the Borgou Province; maize, rice and groundnut seed in the Zou Province, and sorghum, maize, rice and groundnut seed in the Atacora Province. In all cases, with slight varia- tions in degree to suit the economics of the situation, farmers reproduce certified seed for their own use and for sale to others. In all cases, seed production is being closely supervised to assure quality and specialized extension workers supervise the last multiplication in farmers' plots. Government policy is to introduce charges for improved seed distributed to farmers, sufficient to cover seed farm costs. As a result of this newly expanded program, sufficient improved seed will be available in these strate- gically critical premises for the foreseeable future.  21 Rural Development Institutions 1.51 MDRAC. The Ministry of Rural Development and Cooperative Action (MDRAC) is responsible for directing and controlling agricultural and rural development policy and programming. In 1980, a separate ministry was formed which assumed responsible for state farms, livestock and forestry. MDRAC works at a policy and planning level through a number of directorates, and at the field execution level through six provincial CARDERs. The monitoring evaluation and planning function in the ministry has not been effective in the past which has led to poor technical analysis and preparation of investment proposals. With the revitalization the CARDERs of the Zou, Borgou and Atacora provinces under rural development projects financed by IDA, IFAD, CCCE and FAC, however, a systematic monitoring and evaluation capability has been initiated. Provincial monitoring and evaluation units are expected to develop and follow quantitative indicators of development, leading to periodic eva- luation of progress in the provinces involved. At the same time, Government has undertaken to strenghten the Ministry's Planning and Projects Directorate and this has also been supported by IDA, CCCE and FAC. This will assure that a source of strong technical and sectoral analysis will be available to complement macro-economic capabilities of the Planning Ministry, supported by UNDP and IDA. 1.52 CARDERs. The principle supporting institutions for agriculture and rural development at the provincial level are the CARDERs. These agencies were formed in 1976 and given responsibility for extension, research, coope- rative support, input distribution, infrastructure development and primary marketing. Financing of CARDERs has been only partially assured by the state budget. As a result, the CARDERs have also assumed responsibilities for production related enterprises: rice mills, state farms, ginneries and product marketing. In practice moreover, Government has viewed the CARDERs as an instrument of Government presence in the rural areas, and officials of the CARDERs have been expected to perform a number of civil and quasi-political duties which were not viewed as distinct from their technical duties. With the number of divers activities that the CARDERs were expected to play, and the heavy financial aspects associated with these, most CARDERs have found themselves in considerable debt to CNCA, and money deposit banks. Mana- gement of the CARDERs has also been relatively weak and technical competence of staff low since the CARDER's were unable to attract or keep qualified staff at the pay levels which they could afford to pay. 1.53 Since the late 1970s Government has sought to strengthen the technical competence of the CARDERs and revise the financial basis of their operations. Four major provincial rural development projects (Zou, Atakora, Borgou and Atlantique) have introduced more efficient extension methods based on the training and merit system, strengthened CARDER financial management, and reinforced support services in such areas as input supply, and field trails and demonstrations. Government has shown greater awareness of the disincentive for farmers to participate openly in cooperatives and follow extension service advice when CARDER agents play non-technical roles, and has agreed that key CARDER staff, particularly district level staff, be judged and employed strictly for their technical abilities. On the other hand, the CARDERs continue to be involved in food crop marketing programs and to be responsible for various industrial establishments which are of minor or no importance to a general rural development effort. These activities divert  22 management attention and resources away from the promotion of agriculture with a high potential for growth and should be stopped. STRATEGIC CHOICES FOR AGRICULTURE General Policy Orientation 1.54 According to recent policy statements, reflected in the 1982-87 Five- Year Plan, Government attaches high priority to promoting increased agricultural production and productivity. Special attention is given to foodcrops for which Government, correctly, has foreseen a growth in domestic urban consumption needs, and the opportunity for increased exports. However, traditional export crops, cotton and oil palm products, are also emphasized as these generate Government revenue to help cover the cost of official imports. Government also identifies the need to promote surplus production, over-and-above subsistence needs in order that the level of money income rises in rural areas. This is also a desirable goal, since it leads, ultimately, to increased rural savings and rising living standards. 1.55 Government correctly identifies critical features of its rural economy which would have to be taken into account in future planning. These include a flexible, economically motivated peasantry; and provincially specific production patterns which are well adapted climatically and environmentally. Government appears to recognize that production has resulted from spontaneous individual decisions accounting for traditional food preferences, market opportunities, and available production techniques and inputs. However, Government also identifies the low level of use of available land (15% of available arable land is believed to be used) as an issue for which a directly-managed solution should be found, rather than as a rational result of individual decision-making. Government also takes the same view of labor in agriculture--particularly in the southern province which are influenced by urbanization of Cotonou, Porto Novo and, significantly, Lagos, Nigeria. 1.56 Government's general approach, therefore, presents some ambivalences. For example, while recognizing the individualistic character of traditional agricultural production, Government continues to see "official" marketing (by which may be meant the official trading system, such as the ONC) as playing a non-negligeable role, since it appears to believe that "private" economic relations pose a "destabilizing" effect on the economy. While seeing the need to leave an increased cash income in farmers' hands, Government proposes to try to direct the use of these funds through a greater official presence in villages through CARDERs, possibly supported with increased efforts to mobilize rural savings via the CRCAM system. Lastly, Government places greatest emphasis on trying to superimpose a network of official economic relationships over the various parts of its agricultural economy, permitting it to direct results it wishes. But this has been impossible to do in the past, and has led to production levels which have stagnated in recent years. The question, therefore, is whether Government should modify or change its overall approach, relying on incentives and support for agriculture in order to be in a better position to realize its overall economic objectives.  23 Summary of this Analysis 1.57 The present analysis has concluded that Benin will enjoy a comparative advantage in producing traditional food crops into the future and should be able to realize growing economic benefit from cotton production. This appears to agree with Government's own assessment. At the provincial level, production in the small southern provinces, in which urbanism has reduced available labor supply, has grown less rapidly than in the Zou and northern provinces, and have lost their relative importance in food production. Production in the northern and Zou provinces benefited from association with a cash export crop, the institutional arrangements for which permitted a more rapid development of credit and input use and a money economy whereas similar institutions have not developed in the south. For agricultural development efforts in the southern provinces to be successful, significant increases in productivity would have to be possible, giving returns per man-day competitive with urban and Nigerian employment. More agronomic and economic information on these prospects would be required before consideration is given to new provincial-wide investments in these areas. Critical judgement should then be exercised when deciding on the regional allocation of investment to agricultural development. 1.58 This analysis has concluded that the promotion of voluntary cooperative groups of farmers offers a good potential for overcoming some of the constraints facing agriculture such as access to seasonal credit for food crop production. However, as farmers are individualistic, prudence will need to be exercised in the promotion effort so that farmers do not associate cooperatives with official marketing at low prices, or with increased social obligations to the state. Such prudence has not always been exercised in the past. Government appears to agree with this assessment, and has indicated its willingness to revise its approach. In this regard, collective agriculture has proven to be untenable even though Government continues to support the concept. 1.59 State enterprises in agriculture have generally not proven to be economically or financially viable when they enter directly into production, and such efforts should not be undertaken in the future. It is equally clear, moreover, that the creation of apex organizations to market the produce of individual farmers and groups have not been able to remain financially viable and pay competitive prices to farmers at the same time. This results from the heavy overhead cost that is generally associated with these efforts. Government now appears to agree that the private trade is more adapted to marketing agricultural products and that Govenment's official price policy has prevented Government from obtaining products. However, Government continues to believe that private trade should be replaced or supplemented with official trade to remove speculations and control food crop exports. This issue requires further investigation and dialogue with Gcverrnment, since in the past, political efforts to support official marketing have been a significant disincentive to production and marketing within Benin. 1.60 Sufficient incentive to production of food crops will probably exist for the foreseeable future because of market developments in Benin, Nigeria and other neighboring countries. Government's principle concern should be how to reduce the disincentive effects created by policies and practices which do not permit the full incentive effect to reach individual farmers. On the  24 other hand, Government has legitimate concerns about the degree to which its urban population is served by Beninese producers, in spite of high prices. This problem warrants additional study. However, it is clear from the experience with AGB, where a government policy of subsidizing urban food prices can disrupt the financial integrity of the organization managing the market, and where abuses are frequent, that official procurement and wholesaling has high risks. Moreover, while AGB activities in commodities that Benin does not produce in significant quantities, (for example, wheat) may be used to control prices, official marketing of traditionally produced and marketed fooderops such as maize, would be a disincentive to private traders who assure the majority of the overall food supply. This would exacerbate already high prices. 1.61 Government's policy with respect to subsidization of inputs has undergone significant changes in the past 12 months. Eliminating subsidies, instituting charges for improved seed, and initiating specific interest rates on a seasonal credit will have an overall effect of assuring supplies, and providing financial stability to the institution supplying these inputs. Moreover, Government has obtained substantial international assistance for the strengthening of the CARDERs in Atlantique, Borgou and Zou provinces, on the understanding that these institutions would focus their activities on technical rural development. Applied research and overall planning capabilities have recently been strengthened with international assistance. Conclusion 1.62 Overall, through a series of practical steps taken over the past two years, Government has established the preconditions necessary for a well founded expansion of agricultural production. The principal outstanding issues now is whether Government will permit the realization of the potential that is present by refraining from the re-introduction of centrally directed measures in marketing and for social restructuring which have been counter- productive in the past.  25 Agricultural Performance by Region 1. Agricultural performance has varied markedly among provinces, as summarized below. a) Oueme 2. Oueme, is located in the south-east of the country. As the largest southern province, covering about 4% of Benin's territory, Oueme was the most important agricultural region of Benin in the early 1970s. It included 35% of area cultivated and produced 32% of agricultural output, including around half of Benin's total production of maize, beans, and cassava. Since 1971 production and area cultivated have declined and by 1980 the share of Oueme in Benin's agricultural output was around 14%. This resulted from a reduction in rural male labor force as out-migration to nearby urban areas occurred. The decline was specially important in food crops--production of cassava and beans fell by more than 50%. Maize output fell by 25% between 1970-75, but has since recovered. As for cash crops, cotton production fell from 2,400 tons in 1972 to 180 tons in 1981. Groundnuts was the only crop whose production has increased during the last ten years owing to its high value. 3. Oueme is heavily specialized in maize cultivation (70% of area cultivated), and supplies 34% of the total national production of maize. The average farm size is 1.1 ha. The cooperative movement in Oueme has had some success but individual farmers still account for 99% of agricultural output. The number of GRVCs (Groupement Revolutionnaire a Vocation Cooperative) has grown from 24 to 138 between 1978-1981 but the development of CAETS (Cooperatives Agricoles Experimentales de Type Socialiste) based on communal ownership of land, equilment and remuneration, have had no success. No CAETS is currently functioning in the province. b) Atlantique 4. Atlantique Province covering 3% of Benin's territory represents around 9% of Benin's agricultural output. The province is specialized in maize (around 80% of area cultivated) and is the second most important supplier of this crop (23% of national production) after Oueme. Agricultural output grew substantially between 1970-75, due to substantial increases in production of maize (110%) and cassava (81%), but has since deteriorated, again owing to rural exodus of male labor. Production of most crops was in 1981 at lower levels than in 1975, except for maize whose output continues to grow. Yields for all major crops have decreased sharply and net income per man-day for most crops is below the national average (Table 8). The average farm size is 1.6 ha and area per worker around 0.6 ha. Tillable land is scarce. Roughly 80% of all tillable land is already under cultivation, and there is limited capacity to expand area cultivated. The province is fast becoming urbanized.  26 Table 6: EDO APRI=IL