Senegal - Second Agricultural Credit Project

Document of The World Rank FOR OFFICIAL USE ONLY Report No. 3514 PROJECT PERFOPMANCE AUDIT REPORT SENFCAT SECONTD ARICUTTTTTAT PTTT PROTET (CREDIT 404-SE) June 25, 1981- Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their ofticial duties. Its contents may not otherwise he disclosed wanout wors Bank aumornauen. WEIGHTS AND MEASUKS 1 meter (m) = 1.1 yards I kilometer (km) = U.62 mile 1 hectare (ha) = 2.47 acres ABBREVIATIONS BCEAO Banque Centrale des Etats de 1-Afrique de 1-Ouest BNDS Banque Nationale de Developpement du Senegal CEP Centre des Etablissements Publics CPSP Caisse de Perequation et de Stabilisation des Prix CVCCEP Commission de Verification des Comptes et de Controle des Etablissements Publics EEC European Economic Community FED Fonds Europeen de Developpement (European Development Fund) FMDR Fonds Mutualistes du Developpement Rural IFDC International Fertilizer Development Centre IRAT Institut de Recherches Agronomiques Tropicales et des Cultures Vivrieres IRHO Institut de Recherches pour les Huiles et Oleagineux ISRA Institut Senegalais de Recherche Agricole ONCAD Office National de Cooperation et d'Assistance pour le Developpement PA Programme Agricole RDA Regional Development Agency RMWA Bank's Regional Mission in West Africa SATEC Societe d-Aide Technique et de Cooperation SISCOMA Societe Industrielle Senegalaise de Construction et de Materiels Agricoles SODEVA Societe de Developpement et de Vulgarization Agricole TVA Tennessee Valley Authority FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT SENEGAL SECOND AGRICULTURAL CREDIT PROJECT (Credit 404-SE) TABLE OF CONTENTS Page No. Preface.............................................................. Bsc Dat **h **t ***** .*******......................** * * * * * Highlights ............................................................ iii PROJECT PERFORMANCE AUDIT MEMORANDUM I. SUMMARY .................................................. 1 The Project Design ......................................... 1 Project implementation............................... 3 Credit Component......................................... 3 Studies Component ...................................... 4 Pr je t m act.............................................. TI. MAIN ISSUES ................. 5 The Project Design ......................................... 5 The Failure of the Cooperative System ......-*** 6 Project Beneficiaries .................................. 8 D)ificulties ini MeasburingI Project .impjact........... 9 The Issue of Disincentive Producer Prices .............. 11 PROJECT COMPLETION REPORT INTRODUCTION ................................................ 12 A. FORM ULATuI N .............................................. 13 B. IMPLEMENTATION............................................... 14 C. AGRICULTURAL LIPACTe******* ********....................... 19 D. INSTITUTION BUILDING COMPONENT............................... 22 E. THhE STUDIESTh COPOJNELNT................................................... 2 F. ECONOMIC REEVALUATION........................................ 28 G. INSTITUIOUNAL PERFORANCE.............................................. 2 H. BANK PERFORMANCE......................................... 29 I. CONCLUSIULN ................................................ 3 Annexes 1-6 3271 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.  PROJECT PERFORMANCE AUDIT REPORT SENEGAL SECOND AGRICULTURAL CREDIT PROJECT (Credit 404-SE) PREFACE This is a performance audit of the Second Agricultural Credit Project in Senegal for which Credit 404-SE was approved in June 1973 in the sum of US$8.2 million and closed in August 1979 after cancellation of US$40,000. The audit report consists of an audit memorandum prepared by the Operations Evaluation Department (OED) and a project completion report (PCR) dated March 15, 1980. The PCR was prepared by the West Africa Regional Office based on a country visit in September 1979. An OED mission visited Senegal in January 1981. Discussions were held with officials of the Ministry of Planning, the National Development Bank of Senegal and the Extension Service Organization. The information obtained during the mission were used to test the validity of the conclusions of the PCR. The audit memorandum is based on the mission's discussions, on interviews with Bank staff and on a review of the Annrainl Rennrt (Nn. 128a-SE) dated June 4, 1973, the President's Report (No. 1232a-SE) of June 7, 1973. the Credit Agrement riated June 95 1973 the PrCR- and cnrrPonnndence with the Borrower and internal Bank memoranda on project issues contained in relevant Rank filese A copy of the draft report was sent to the Borrower on April' 229, 1981 for comments. None were received. The audit finds that the PCR covers adequately the project-s salient features and the PPAM generally agrees with th conclusins. Tn. addIt4#Ion to summarizing the objectives and results of the project, the PPAM expands upon benefits and duration of an agricultural credit project, the failure of the coopertL.ive syLII, aiLu LLI e DUC UL priLc.i1 po.lices, UCaUse UL LuL importance to this as well as other agricultural credit projects in the same i Li va1 DL L he valuable assistance provided by the Government of Senegal and the project staff met during the preparation of this report is gratefully acknowledged.  - ii - PROJECT PERFORMANCE AUDIT REPORT BASIC DATA SHEET SENEGAL SECOND AGRICULTURAL CREDIT PROJECT KEY PROJECT DATA Appraisal Actual or Actual as Z of Estimate Estimated Actual Appraisal Estimate Project Costs (US$ million) 11.5 10.3 89.6i. Credit Amount (US$ million) 8.2 8.16 99.5 nDate Bard Anroval 06/26/73 06/19/73 - Date Effectiveness 10/30/73 10/30/73 Date Physical Component Completed Dec. 1976 Jan. 1979 157.1/2 Propotionthen comlete (1)100TO Closing Date 06/30/77 08/15/79 153.1/2 Economic Rate of Return (%) 22 15-20 80 Number of Direct Beneficiaries u5,000 mo,30u !4.0 CUMULATIVE DISBURSEMKNTS FY73 FY74 FY75 FY76 FY77 FY78 Appraisal estimate (US$ million) 2.2 4.9 7.8 8.2 - - Actual (US$ million) - 1.9 4.3 7.1 7.6 8.16 Actual as 2 of estimate 0 38.8 55.1 86.6 92.7 99.5 Date of final disbursement 05/31/79 MSTTIh DATA Date No. of Manweeks Performance Types of Mission (Month/Year) Persons in Field P~r' ~rolpu±A/5 Identification 05/72 3 6 Preparation U5/72 2 2 - Preparation 07/72 2 6 - - Appraisal 11/72 5 25 - - Supervision I 12/73 2 1 1 2 MTE Supervision II 09/74 3 3 3 3 FMTP Supervision TIT/O 12/74 1 0.5 - - Supervision IV 02/75 3 4.5 3 1 FM Supervision V 11/75 1 2 3 2 FM Supervision VeL2 06/762 - Supervision VIIL 10/76 1 0.8 - Supervision U.ITT/ 11/7 1 2 - Supervision IK 02/77 1 0.5 - Supervision X 05/77 3 3 2 1 FM Supervision Xi - 02/78 3 3 - - Supervision XI F' 05/78 1 0.2 - Completion 09/79 1 2 Total 62.5 OTHER PROJECT DATA Borrower Republic of Senegal Executing Agencies SODEVA, BNDS, ONCAD Fiscal Year Borrower: July 1 - June 30 Name of Currency: Communaute Financiere Africain Franc (CFAF) Currency Exchange Rate Appraisal Year Average (1973) USS 1.00 - CFAF 230.2 Intervening Years Average (1974-78) US$ 1.00 - CFAF 236.0 Completion Year Average (1979) US$ 1.00 - CFAF 220.0 Loan/Credit Numbers Ln. 1113/Cr. 549-SE Amount (US$ million) 14.0 Date Board Approval May 22, 1975 /1 Total project costs slightly less than estimated owing to the lower unit costs of agricultural equipment distributed. /2 Calculated from date of Board Approval. 13 1 problem-free or minor problems, 2 * moderate problems, and 3 * major problems /4 1 - improving, 2 - stationary, and 3 deteriorating P - financial. M = manaserial. T - techniral, E - economic, and P nolitical 7W Back-to-office Report.  - iii - PROJECT PERFORMANCE AUDIT REPORT SENEGAL SECOND AGRICULTURAL CREDIT PROJECT (Credit 404-SE) HIGHLIGHTS The project which was the continuation of the first agricultural credit proiect (Cr 140-SE/Ln 584-SE of February 1969, OED Report No. 1319 of October 1976) was to provide medium-term loans to farmer cooperatives for the purchase of about 120,000 animal-drawn implements (mainly seeders, hoes. ploughs and carts) and draft animals. Its secondary objective was to help strenc,-hPn the extension qprvice orvnni7.tinn (qODEVA) and the innut distri- bution and marketing organization (ONCAD). The credit component (85% of the total nrniert cost- was exnected t-n hpnpfit nhnnt- 59 000 fqrmr. (ahout 15% of the farming population) and lead to a 4% increase in the annual produc- finn of oroundnut. Cont-r- tot,-he first prointo' w.hich is regarded as a faiflure, d'isf project must be considered a success with respect to its agricultural credit jnnn, -t Both the, impnlemntsi- delive,red aid t-he numnber ofcredit benefici-4- ries exceeded appraisal estimates. Success of the credit component was in pa.rt Adue to goodA ci matic c ndition, o nseqr,untly higher, agri-c.1 ultral rodruc,- tion and better groundnut farmgate prices in the first project year. During project executio, .. was, howeve, rea..lz e Lu1 LLtUUv..o 'J.aru.uu existing institutions and the cooperative system in general were far more seL.Lus LLIaL1 assume11d at appLaiJa. 11e prUjeLL i.L. ntU 1 .inLiLuti building efforts and was unable to prevent the collapse of ONCAD, the disin- tegration of the cooperatives and of the entire marketing andu creuit system. The ERR has been reestimated in the 15-20% range compared with 22% calculated at appraisal. The following points may be of special interest: - difficulty of projecting accurately costs, benefits, and duration of an agricultural credit project when these factors depend to a large extent on uncontrollable external factors, especially weather (PPAM paras. 17-20); - difficulty of improving several large institutions through a single agricultural credit project kPFAM paras. 17-20); - cooperatives were regarded by farmers as the local off-shoots of government agencies rather than as institutions able to repre- sent them and defend their interests (PPAM paras. 21-25 and PCR Annex 6); - iv - - nrnhIPml nf hPnPfit nt-trihtioinn hv int-liuJing mr.nnnal inntitas nn financed by the project in the ERR calculation (PPAM paras. 9A-17 And PrR nAraa.- A* -~~~~~~~ inlec of - fa gt prc on famer - atttud (PPAM paras. 33-35); and - consultants' recommendations concerning financed management were nottakn-ntoaccun ( r LQsn- PROJECT PERFORMANCE AUDIT MEMORANDUM SENEGAL SECOND AGRICULTURAL CREDIT PROJECT (Credit 404-SE) I. SUMMARYl/ Background 1. Since 1964 the Senegalese Government has operated a groundnut and Millet Productivity Program,-kon^- te Pogam Agrcl (PA), Wh4Ih aims at increasing the production of groundnuts, the country's main cash crop and export earner, and of millet for subsis tence and dAomestic supplies. Under PA, fertilizer, improved seeds and animal-drawn farm implements have benMAeaalbe ofreso r A. 1 QAOL L Lit;j.anLexttCen.SiVon service ias a~lso beeni developed to promote improved cultivation methods. 2. The Bank Group first assisted PA in 1969 when it made a combined credit and Loan (Cr. 14-tE/Ln. 5o4aS;) totalling US$9.5) million for financing farm inputs in the groundnut basin and for strengthening PA institutions, in particular the National Development Bank of Senegal (BNDS), the extension service organization (SODEVA) and the input distribution and marketing organi- zation (ONCAD) during the 1969-72 period. The project's goals were ambitious and aimed at expanding the area cultivated under the groundnut-millet rotation by 24% and increasing the average yield per ha by 30. within the three-year project period. The objectives of the project's directly productive compo- nents were not achieved because of an exceptional sequence of severe droughts and an inappropriate price ratio between groundnuts and fertilizers. The demand for farm inputs was significantly lower in the project period than in previous years and much below appraisal estimates. Thus the loan portion of the credit was cancelled in 1971 on the ground that it could not be used within a reasonable period. The institutional reform component was also regarded as a failure; although some improvements were made in the performance of BNDS and SODEVA, the reform of ONCAD failed because of insufficient govern- ment support, poor performance of consultants and inadequate IDA supervision in view of the particular problems involved./ The Project Design 3. The Second Agricultural Credit Project was designed to be a follow- up of the first project with two exceptions - the project area was extended to cover the whole of Senegal, not only the groundnut basin, and credit 1/ Adapted from the PCR. 2/ Project Performance Audit Report No. 1319 dated October 20, 1976. for fertilizer was not included under the project since BNDS was expected to be able to finance any additional requirements. The basic objectives of the project were to provide over the 1973 - 1976 period: (i) medium term credits ploughs and carts), and of draft animals; (ii) technical assistance to SODEVA ad A X /.I A .t ; f± ± . *U.p11I dU4 .1 N.J4 Cnn4.J-VT A 4~ - A ./ V~.L &0 - LJ studies. 4. The credit component, representing 85% of the total project cost, was expecteu to ueuefit abouu J,uuu larmers wouuL £/ or Lne LaLm popu- lation) and lead to a modest incremental annual production of about 35,600 tons of groundnuts (about 4% of the total production) and 7,800 tons of millet and sorghum. It was expected that 75% of the implements would be used for replacement and 25% would De acquired by farmers not owUng any implement. 5. The technical assistance component (10% of the total project cost) was limited to: (i) helping ONCAD in the fields of accounting and data proces- sing with emphasis on counterpart training and (ii) assisting SUDEVA in re-orienting its activities towards a more farmer-oriented extension service and strengthening its equipment and infrastructure. The following studies were envisaged to be done under the project: (i) the feasibility study of the Sine Saloum projectL', (ii) a seed storage study, (iii) a monitoring and evaluation study, and (iv) a fertilizer use study. 6. The project was appraised in November-December 1972; the credit was signed in June 1973 and was declared effective in October 1973. Since the date was far into the 1973/74 agricultural season, IDA allowed retro- active financing for material credits given by BNDS for that season. Total project costs for the 1973-1976 period were estimated at US$11.5 million. The IDA credit (US$8.2 million) was to finance the foreign exchange cost and 45% of local currency expenditures. 7. Although a small part of the IDA credit was assigned to the insti- tution building component, improvement of existing institutions was expected to be implemented through a number of covenants; in particular since the delivery of farm inputs had been badly handled by ONCAD during the first project, assurance was obtained that ONCAD would within twelve months ensure that suppliers of fertilizer and implements undertake directly their delivery to cooperatives. In order to avoid procurement problems experienced during the first project, an exception was made to international competitive bidding, and assurance was obtained that contracts for supply of implements would be awarded on a negotiated basis mostly with SISCOMA, the main local supplier. ONCAD's and BNDS's accounts were to be audited by independent auditors. BNDS was to make ordinary-term loans at a rate of 2 percentage points above the 1/ Now under implementation under Bank financing (Credit 549-SE and Loan 111 3-giPT discount rate of the Central Bank. In addition, the level of subsidies on Cyonnut andt mille 1 fertiliz.er was expectedo to be disnncsse each, year wit-h IDA with a view to its progressive reduction. Project Implementation 8. In general, the project proceeded on schedule except for the stud- .Les, which delayed the closing date from. the original June 19771 to August 1979. Total project cost at completion was US$10.4 million (90% appraisal estimate) reflecting the lower cost or rarm implemens distrIDuted under the project and of studies. Credit Component 9. The medium-term credits were supplied on schedule by BNDS and the credit portion of the project was fully disbursed by the end of 1976. There was an upward trend in the distribution of implements throughout the project period. Implements delivered (about 150,000 units) exceeded appraisal esti- mates (120,000 units). Assuming that equipment is replaced every ten years, the stock of equipment with the project could be estimated at the end of 1976 at 47% above the without project situation. In addition, 6,700 pairs of oxen and 345 horses were supplied on credit (14% above appraisal estimates). About 78,000 farmers benefitted from credit, compared with 55,000 estimated at appraisal. 10. The success of the credit component was essentially due to good climatic conditions, good agricultural production and increase in groundnut farmgate price in the first project year. Farmers' demand for implements remained constantly higher than supply and was not affected by the deterio- ration of the institutions (para. 11) nor by the inability of ONCAD to deliver on time the amounts requested.. 11. Credit repayment rate remained high (72 to 90%) during the project period but sharply declined after 1977/78. The main reason for the decrease in repayments is the failure of the cooperative system (varas. 21-25) and the ONCAD-s inability to play adequately its role in marketing, inputs dis- tribution and cooperatives control. The Government policy of cancelling debts anytime the rainfall is slightly below normal also did not help the recovery rate. As a result of increasing defaults by cooperatives, BNDS was not able to set up the large revolving fund which was anticipated at appraisal and the financing of the PA had to be sianificantly reduced in 1980/81. Moreover, large discrepancies exist between ONCAD's accounts, cooperatives' accounts and debts as recognized by the farmers themselves. It is worth noting, however, that most of the cooperatives' indebtedness is not related to the farm implements provided under the project. but to seeds (70% of the indebtedness) and fertilizers. Institution Building Component 12. In addition to groundnut marketing, the distribution of inputs and recovery of farm credit, ONCAD's role was to manage the groundnut seed stockpile, maintain cooperative accounts and promote the cooperative move- ment. A major reorganization of ONCAD was financed with the first IDA credit, and it was considered at appraisal that ONCAD had significantly improved its performance. It was expected that during the second project, ONCAD would disengage from its inputs supply role and concentrate on its marketing functions, on ensuring that farmers received the guarantee farmgate price, and on the repayment of short- and medium-term credit by cooperatives. It was also anticipated that ONCAD would promote a more self-supporting cooperative system than the one prevailing at appraisal. In order to improve ONCAD's accounting system, two financial experts were hired at the end of 1973, but government with IDA agreement cancelled the consultants' contract at the end of 1974, and thus the technical assistance to ONCAD was ended. As for the first credit project. the consultants' recommendations were not imnlemented. The overall ONCAD situation sharply deteriorated since 1976/77. Poor manage- ment exreRe Rtaffina_ arck of overall financial nnntrol and had handlina of the seed stockpile and of input distribution finally resulted in August 1980 in ONCAD' limidation, which in turn antailed the r1ngin of RTCOMA the main agricultural implements supplier. 13. Financing of extension equipment, vehicles and office materials for SODEFVA was~ done a expec'ted at apraisa@4 Tec.hnical ,aitancewa provided to strengthen the monitoring and evaluation unit of the agency. The .imr%actl of t-h~ p"v^Ctti r.4nrvr1c qO1nV1Aniozat%hly % positive. 14. Compliance with project covenants was uneven; discussions between accounts were audited by independent auditors, but because of delays in yL.. LL.LL .. O UO.~l~. -t~ WO'S .llIj WCO U,&WL M-J. %AS 11"a. * LO aLLAJLLLaC LUA. LCU% Procurement procedures were consistent with the provisions of the Credit t,,6 ~ ~ L~~L L L. L JI.UL L_ LI VV1CZ.~.UJ %LL.oLie gg fo n u-gem n . ep t t om tm n ,ATA%AA U U .±~LO tsLJU s.IjL.L upplyJ and was for the most part responsible for delays encountered in delivery of agri.cutural implemeLt. rinaly, counary LU L "LU.L M6LtmU1fLL CLU aL LIC request of the government, BNDS increased its on-lending rate only two years after the Central Banks discount rate was raised in january 17/3. Studies Component 15. The Sine Saloum feasibility study was completed by mid 1974. The seed storage study was implemented in cooperation with the European Develop- ment Fund. The study was intended to cover the storage of all grounanuts, and its original objective of finding ways to transfer responsibility for seed storage from ONCAD to cooperatives was not achieved. The monitoring and evaluation study was completed in 1977/78 and helped SODEVA to do useful evaluation work in the ongoing Sine Saloum Project. The fertilizer study, - 5 - including field trials, was done during the 1975-1978 period; the study showed that savings can be made by the country and the farmers by changing the current fertilizer formula. The Senegalese Research Institute, however, is reluctant to accept the consultants' findings on the grounds that the trial period is too short to make final recommendations. Project Impact 16. Unlike most agricultural credit projects, the Senegal Second Agri- cultural Credit project made a genuine effort to isolate the benefits gen- erated by several factors and inputs. Using the factor analysis method. SODEVA and its consultants studied the relationship between the rate of equipment use and the yield increase per hectare. It was found, in parti- cular, that the use of hoes, plows and seeders allowed for early land prepa- ration and timely seeding. and that a 1% increase in the use rate of hoes, for example, increased the groundnut yield by 3.57 kg/ha in the South and 1.46 kg/ha in the North. In its calculation of benefits. the PCR also in- cludes the effect of non-Bank financed fertilizer on the grounds that both fprtili7pr and acriii1tiirql tninment qrp nqrt nf f-hp tpchnical nq(ckaiYt recommended by Extension Services. By extrapolating the results of the mnnitnrina stuidiesc in the Sine qn1nii rpainn to the whole project arA, t was estimated that the incremental production derived from the credit component of the project waS 9,000 tonso fo%r prou-nnutsa by year- 3A ndA 11 On t-n f- millet and sorghum (3% of the total groundnut production and 1.3% of the ___-CtL_U Unde th abv supion,bd sn the 1979 prices for inputs and outputs, the estimated rate of return (ERR) of the prjcta comp,leono 4is 29%. TT.,4-- +- ,-.o, 4 - - z,r 1 -~-4 - -,A actual costs in 1973 terms, the ERR is estimated at 18%. TT MATMT TOCTTVC The Project Design 17. This project was the continuation of the First Agricultural Credit Project, which had consiuderably overestimated its physical objectives (expan- sion of the cultivated areas by 24% and a 30% increase in yields) and its possible institutional impact. The first project unforunuately coincidedu with one of the worst droughts of the century, and half the credit funds could not oe disbursed. As Ior its impact on the institutions, Uns was relatively minor: the troubles plaguing ONCAD were effectively diagnosed, but little concrete action could be taken to remedy the situation. 18. The second project drew a number of lessons from the first, limiting its physical and economic objectives to a modest increase in production and no expansion of the cultivated areas. It so happened that favorable weather and good economic conditions during the first two years of the project period made it possible to exceed the physical targets. In terms of institutional impact, the second project provided only a little assistance to ONCAD and SODEVA, out this was offset by a number of covenants intended to contribute to the gradual improvement of those two institutions. 19. During the project execution, it was realized that the troubles plaguing ONCAD -- and the cooperative system in general -- were far more serious than was realized at appraisal. Contrary to expectations, ONCAD's already bad situation as regards management, accounts organization and input distribution grew steadily worse as the project progressed. The government's pricing and subsidies policy and its interference in ONCADs affairs aggra- vated the situation even further. The Bank made a praiseworthy effort to remedy the situation during project supervision, but with a single agricul- tural credit project as its only lever, it was not able to make more than a slight dent in this enormous institutional problem, which had been grossly underestimated at appraisal time. The Bank was unable to prevent the col- lapse of ONCAD, the cooperatives and the entire marketing and credit system. 20. The lessons to be learned from this experience confirm the obser- vations already made in connection with other projects of the same kind in the same part of the world: (a) It is difficult to project accurately the costs, benefits, and duration of an agricultural credit project when these depend mainly on uncon- trollable external factors, especially future weather conditions. It is clear that much flexibility is needed in the design and supervision of projects of this kind. Possible changes in project costs and timing are, in such cases, an integral part of the project risks and should be explicitly recognized as such in risk and sensitivity analyses. (b) It is difficult to improve several large institutions through a single agricultural credit project. The audit estimates that since physical progress was good, while the extent of the institutional problems was much more serious than foreseen, it was wise to allow the project to reach comole- tion and then tackle Senegal-s serious institutional problems through a Structural Adjustment Loan which was approved by the Board in December 1980. The Failure of the Cooperative System 21. The Droiect illustrates the weakness of Senegal's coonerative system, a weakness shared by several sub-Saharan countries. This system, modeled on those of other colintripq nd not adanted to qeneal's own soial and traditional context, is now in a state of collapse. The cooperatives, rrt- i -n -h QiY-P- di - - - rn- i1- frn- ; nott-q dni icrf-qD - mnrc- mnxY= -- : they were introduced by the government to facilitate its groundnut and cereal mnrVavina activities andl credit ditiuInoperatinos ApproxIm4-tel- 1,900) cooperatives were set up in Senegal, each one comprising several hundred members, frnm sratieandl illa o A s-up ii was ri ed by the members, administrative and accounting supervision was provided by ONCAD, -7I which was also responsible for supplying inputs and seeds, marketing prod- ucts, and recovering debts from the cooperatives. In addition to OuCa., tne General Assembly of Cooperatives, chaired by the prefects, and the Regional Credit Commission, chaired by the governor, were responsible for supervision. These two agencies were mainly in charge of checking the cooperatives appli- cations for short- and medium-term credit, limiting these to one-third the average value of the annual production marketed over the past three years. 22. During talks with the farmers and agricultural services, the audit mission came to realize that the system, in its design as well as its opera- tion, was much more like a government agency than a cooperative organization. The power and responsibility of the cooperative members were practically non-existent. In principle, seed, equipment and fertilizer were distributed to the members according to need, and the members were jointly responsible for the cooperative's debt. In reality, the inputs, of which the supply had always fallen short of demand, were often distributed arbitrarily among the cooperatives by the supervisory bodies, and within the cooperative itself by the chairman. Moreover, since the administrative and accounting services of the cooperatives were disorganized, certain farmers were debited for inputs that they had never received and sometimes had never asked for. Other farmers would buy inputs and resell them. Furthermore, the cooperatives and ONCAD were the only intermediaries through which the farmers could procure inputs, it being impossible for them to purchase directly from the suppliers even for cash. 23. The price structure was extremely complex and incomprehensible to the farmers. The producer price of groundnuts rose from CFAF 29.5 in 1973/74 to 41.5 in 1974/75, representing a definite inducement at the start of the nroiert- but the onrice then remained stable throuah 1979/80. while the price of inputs steadily increased. The levies on harvest proceeds imposed by government for the Equalizing Fund, by BNDS for the guarantee and investment fund (five times the cooperative's share capital) and by ONCAD for seed storage and distribution were not fnl1v understood by the farmers. Nor were the subsidies (on fertilizer and interest rates) regarded as such by the hnef4cInarin To sum un thp fnrmrc' nvprridina sentiment wag one of helplessness before these State-operated agencies. The cooperatives were regarded hv farmers as the local off-shoots of those agencies rather than as institutions able to represent them and defend their interests. 24. It was in this steadily deteriorating context that the project was executed The pr ' thvpa orn,-veare (1971/74 1974/75 and 1975/76) were marked by good rainfall and a steady growth in production (the groundnut harvest reaching nn all-time high of 1.A millinn tnns in 1975/76) with repayments of both short- and medium-term credit remaining at acceptable levels. T- 1077 /79 _-4-t-_ fall ff Arnat-4nly arinc to the Arniaht and it was then that the cooperative system demonstrated that it carried by the idea of self reliance, cooperative members throughout the country dILe LgLIeI ii ~.Ubta '. ~p.y LI~LL tL.A± .L9A L tjCAO C +-4 - 8 - members started to question their debts to ONCAD and BNDS. With the accounts in disorder, it was impossible to establish the true indebtedness situation. In 1978/79, despite a good harvest, the recovery rate fell to 54%. With the drought and low production figures during the two crop years, 1979/80 and 1980/81, the cooperatives were even more unwilling to repay their debts. (the proportion of unpaid debts rose to the equivalent of 50% of the annual value added in all rainfed farming). resulting in disintegration of the cooperative system, the fall of ONCAD, the closure of SISCOMA and the weakening of BNDS. In retrospect, the drought provided the trigger for this institutional disin- tegration, the underlying cause being the poor basic design of the coopera- tives and the omnipresent interference of government monopolistic agencies in the rural areas. 25. Through a Structural Adjustment Loan and the Sine Saloum project, the government and the Rank are nnw aiming toward a mnre afft-tive Apaian for the cooperatives. The basic idea is to give far more autonomy to the farmprR in manapina their nwn affnir%% and to reduice ovnernment intr-O ntinn in the cooperatives. The basic unit is no longer the vast cooperative that nruqileA durinc the Rank's twn earlier prnirets bu vh1ilag "%Y1MCnF" grouping together the 30-40 families found in the same village. It is at wvlllon 1al thaf- fhe farmers hava anmo t'han'on f paretcate andA that decisions can be taken by the community as a whole. Nevertheless, the verifi- ratlon of the arronts and tipht-a nf t-he farmer rnnor.tiuac nw in nvn&r=aa will still require at least another two years, and agricultural credit is for the -resent almost non-existent in Seea. fAn agIcltrl1reI restruc- turing study is being carried out by CCCE_1/). The farmers seem to have decidedA ton reduce dr-astical 1c utiva~ntion of ---.,'Anutsn fthe only. cas crop possible in most of the country) in favor of food crops (millet and sorghum), wth w1 .h ih theym can -at- less- mefet onffarM ----fmm.-4o need .io U D WALA U mosL agAcULULA LLdL LUject i L sL oLUUUL, aLhUUgh difficult to establish, that the better-off families in each cooperative have Uenu the ones tu outaiu tne greatest uenefits frum tue project. noVLUuvUL$ with the extended family that is typical in Senegal, it has not so far been pussue LU grant eut LU anyone otuer LILL L[LC nCUU U1 LIe [UU5enU.U kCnlr de carre). The cash crop cultivation system characteristic of Senegal's grounanut Dasin is based on the following principles: the lands owned by the chef de carre (35% of the family lands) are those closest to the village and the first to be tilled and sown (at the first rainfall). The lands of the sons of the extended family (40%) are tilled and sown at the second rainfall and receive less fertilizer. Lastly, the marginal and most remote lands (25%) are most often cultivated by women, receiving little fertilizer and being the least well maintained, since their potential is limited by the 1/ French technical and financing agency. fact that they are sown so late (at the third rainfall). Women hardly ever have access to equipment and their work is essentially manual. In an average year, the yield gap is approximately 30% between the lands of the chef de carre and those of the women. 27. Family labor and equipment are the chief bottlenecks to early sowing. But the principle of granting credit only to the head of the household seems to have had two negative consequences: (a) to limit the equipment avail- able to each extended family, and (b) to perpetuate inequality among family members. The experience of a similar project in a neighboring country should be mentioned, whereby credit was granted to individuals and not only to the head of the household; the result was better income distribution within the extended family and better productivity for the group as a whole. It seems that Senegal's context is also suited to such a development, a fact overlooked by the project: in certain villages, the attention of the audit mission was drawn to the fact that credit is not accessible to family members or to women, and this is resented as being unfair.1/ Difficulties in Measuring Project Impact 28. It is always difficult to assess the economics of an agricultural project that provides credit for the purchase of inputs. The standard method applied in the PCR to calculate the rate of return does not mean that it is strictly accurate. The first difficulty encountered is to compare the "with nroiect" with the "without project" situation. The without project situation does not necessarily correspond to a situation without inputs, to the extent that- the counitrv can nrovide alternative financing. The "with project" would then have to be compared with a "local credit or farmer self-financing" AuP1onment. Nevertheless- we may assume that this reservation does not apply in the present case, since at project completion it was clear that no local financing system wa ennable of comnletely taking over from the Bank. 9 Tn aln1uAting 1hnPfits_ thp PCR (Annex 4) includes the effects of both the Bank-financed agricultural equipment and the non-Bank financed fertilizer on the grounds that the technical packae recommended by the extension services includes both: agricultural equipment and fertilizer. The PCR assumption is that the nrnipnt aced as catalyst for providing the right inputs for crop intensification. In addition, in order to calculate the ecooi effec- ^f t-ha prnirt 1-hp PCR iisps indicators drawn from SODEVA surveys which were then extrapolated and applied countrywide on the following assumptions: (a) linear gro-th in agricultuiral yipeds as a fuinction of in- puts; (b) the increase in the mechanization rate resulted in intensification 1/ The Region argues that "experience has shown both through the credit projects ana on-go.ng LuLi.L UevsuCen. project atj it. woul nt hav been possible then nor is it possible now to grant credit to individuals given the nature of official agricultural credit, marketing and coopera- tive organization policy." (increased yields) and not in extensification (expansion of farmlands); and (c) the supplementary implements were distributed in the most efficient manner, i.e. supplied to those farmers who were able to make the best use of them. On the above assumptions, the PCR estimates the projects EARn at about 30%. These three hypotheses however do not necessarily reflect the true situation. 30. A different approach, consisting of measuring the project's economic impact against the inputs it has helped finance, would lead to a somewhat different conclusion since the Bank loan aimed at financing agricultural equipment and draft animals only and not fertilizers. SODEVA made an attempt to isolate the benefits generated by the different input usea Dy farmers, but the various monitoring surveys for the SODEVA agricultural program arrive at contradictory findings on the effects of some equipment. The only clear- cut conclusion from these surveys is that among the agricultural equipment (seeders, hoes, lifters, toolbars, attachments, ploughs and carts) sold to farmers, only the combined use of hoes and seeders has a clear and measurable effect on production. By using the conclusions of the SODEVA study, the proj- ect's ERR would fall to about 12%. There is substantial evidence, however, that equipment other than hoes and seeders might have some beneficial impact on production but the methodology used by SODEVA and its consultants does not permit such impact to be measured. 31. To arrive at a more realistic assessment of economic effects, additional statistical information is required: the farm typology carried out by SODEVA in the Sine Saloum region could serve as the basis for a study of the situation without project; a survey on the distribution of equipment purchased during the project by different types of farm, according to their degree of initial equipment, land area and labor situation, would also make it possible to gauge the impact of the equipment on yields or on the expansion of cultivated areas and hence to evaluate its true economic effects. 32. In the absence of these surveys, the only possible conclusion is that the equipment financed by the Bank has elicited a demand from the farmers. This means that farmers consider this equipment really useful. Good weather conditions and the increase in producer prices can offset input prices, however, and are major factors that can disguise the true economic effects of the equipment, even if a suitable method exists for measuring the benefits. If it is assumed that the economic effects of an agricultural credit project should be measured against the inputs it has helped finance, the nroject's economic rate of return has been over-estimated both in the completion report and the appraisal report. On the other hand, the effects of the hnp, and qepders, as measured by SODEVA, do not accurately reflect the benefit which can be expected from a complete set of equipment sold to farmers. Based on the ansumntinn that- the whole set of eauinment delivered under the project had some beneficial impact on production, the project's ERR - 11 - The Issue of Disincentive Producer Prices 33. Numerous surveys were carried out by SODEVA to measure the impact of different inputs on agricultural production. But it SeemS tat the project did not pay sufficient attention to the key factor, i.e. the producer price, which still represents the greatest incenive. 1A. LLUUrLl ULM FUL 1LOO F groundnuts has remained stable in current money terms since 1973, it has deteriorated considerably in constant terms. it should be pointed out that (a) according to the Bank's inflation index, the producer price fell in real terms by 48% between 1973 and 1980 and Dy LU/o between 17/3 anu 17/1 duuing Lue project period; (b) in comparison with the price of fertilizer the producer price of groundnuts fell by 38% between 1975 and 1970, according to SODEVA and (c) the existence of a parallel market, on which groundnuts are now being sold (in 1981) at CFAF 100 per kg, against an official price of GFAF 46, serves as additional proof that the producer price is unremunerative. 34. A more thorough analysis should be able to determine the facts with precision, but an initial approach leads to the conclusion that the farmer reacts to the deterioration of prices in real terms by reducing his cultivated areas and limiting his demand for inputs, resulting in a drop in production. The withholding of production and its sale on the parallel market are also consequences of an official producer price that is not high enough. 35. While it is established that the producer price is inadequate to stimulate the farmer, this does not mean that an immediate increase in that price would be desirable without first measuring the net value added of groundnuts at the national level, by reference to world prices. The groundnut marketing system is unwieldy and complex. At the State level, it brings in revenues, but it also involves costs. The revenues comprise (a) levies by the Equalizing Fund when world prices are holding well and (b) the guarantee and investment funds set up by the cooperatives at BNDS. Costs comprise: (i) input, equipment and fertilizer subsidies (until recently); (ii) operating costs of the extension and marketing agencies and (iii) cancellation of the farmers' debts in case of drought. It should be possible to draw up a balance sheet of receipts and costs at the national level over a fairly long period, making different assumptions about trends in world groundnut prices and in production and marketing, in order to reach a valid conclusion. At present, it can only be said that the farmgate price is inadequate, without being able to determine whether groundnut production is profitable for the country as a whole. It is unfortunate that yearly discussions between the Government and the Bank on prices and subsidies did not contribute to the setting of a fair price for groundnuts at the producer level. - 12 - ATTACHMENT SENEGAL SECOND AGRICULTURAL CREDIT PROJECT k%Lreit '+V-+ On) C^ -An -m- 1n,rn r ----- LVrLr~LUA' LU%1A I. INTRODUCTION 1. In 1966, the Government of Senegal asked IDA to help with the 14 nan4 ng 044-. "Poram A-4COlp fDNI -.44,1, 4a ,4.A 4 4o o4. A.~~~ .t& LL5.L, tA .LL A. % -. which isJI aimed increasing L f agricultural production and productivity through the supply, on credit, of sion of extension services to farmers to improve their cultivation methods. TT'%A ~ - - A A 4-14-- f-.-A4*- 4-.-* fr,- 1A _Q V/T. 584-SE) covering the groundnut basin from 1969/70 to 1972/73. 2. Under this project, funds were provided through BNDS, Senegal's agricultural Development Bank to finauce cuperative purchases of farm inputs and technical assistance was given to improve the performance of the key institutions involved in the agricutural sector. Te institutions benefitting from this assistance were BNDS, the National Office for Coopera- Lion and Development (uNaA) and the extension agency involved in the groundnut basin, SODEVA. The project was limited to the basin which accounts for over t33 of the groundnut production and over 67% of millet production in Senegal. 3. At first, project performance in terms of expecred input supply and crop production was not satisfactory. The poor performance was attributed to the successive droughts, which occurred during the project period, as well as Government price policies making it unattractive for farmers to invest in extra inputs. Thus a loan portion attached to the credit was cancelled in 1971. During the last two years of the project however, input demand began to increase with improved weather conditions and tavorable price policy. It was also generally agreed that without the inputs distri- buted during the project, production would have been much lower than actually occurred and hence a strong case was established for a follow-up project. 4. The institutional reform component, especially for ONCAD, was also regarded as a failure. Substantial improvements however, were made in the performance of BNDS and SODEVA, and both Government and IDA agreed that the little improvement that had been made in ONCAD performance was to be consolidated. The performance of the project has been discussed in the PCR and PPAR issued after completion. 5. Thus in view of the rising input demand and the need to consoli- date the gains made in institution performance, IDA agreed to a Government request to finance a Second Agricultural Project. The project was prepared by SODEVA with support from the Bank's Regional Mission in West Africa in May 1972 and appraised in November/December 1972. 1/ The Project Completion Report dated March, 1980 was prepared by the West Africa Regional Office and based on a country visit in September 1979. - 13 - THE PROUJECT A. FURMULATION Objectives 6. The Second Agricultural Credit project was designed to be a follow up of the first project with two exceptions. The project area was extended to cover the whole of Senegal and credit for fertilizer was not included under the project since BNDS was expected to be able to finance, any additional requirements. 7. The basic objectives were to provide: (a) medium term credits for the purchase of animal drawn implements, carts and draft animals; (b) technical assistance to SODEVA and ONCAD; (c) equipment and vehicles for SODEVA and the Agricultural Research Institute, IRAT (now ISRA); and (d) various studies to provide information on seed storage and fertilizer response at the field level and to do a feasibility study of the Sine Saloum Agricultural Development Project (now CR. 549/Ln. 1113). 8. The project sought to distribute a total of 87,000 light implements - seeders, hoes and lifters - over the three year project period, 1973/74-1975/76. 75% of the implements were expected to be used for replacement and 25% acquired by farmers not owning any implements. In addition, a total of 7,000 heavy ox-drawn multi-purpose equipment plus 26,000 carts were to distributed. And provision was made under the project, for the purchase of 5,600 Pairs of oxen and 500 horses on credit. 9. During appraisal, it was not expected that there would be any significant increase in the total area under cultivation, but that the agricultural implements to be supplied under the project would benefit about 55,000 farmers (about 15% of farming population) and lead to an incremental production of 21,250 tons of groundnuts and 3,950 tons of millet by year 3. when the project would have distributed all medium term credits. Farmers production however was exnected to keen increasing as they apply fertilizer to larger areas of their plot and total incremental production was expected to reach 35.575 tons for groundnuts and 7,000 tons for millet in years 7-10. Moderate increases in production were also expected to be registered for sorghum. rice and cotton. - 14 - 10. To ensure improved farmer services, the technical assistance started under the first project was to be continued. Assistance was to be provided for ONCAD in the fields of accounting and data processing with emphasis on counterpart training. Assistance was also to be given to SODEVA to reinforce its study unit as well as to handle the seed stockoile. In addition, provision was made under the project to purchase vehicles, extension equipment and office materials and to out u three additional buildings at the training centre for SODEVA. These were aimed at equipping SODEVA to handle its workload effiriently nnd to nrennre it for the Sinp S.lnitm nrnipc- the feasibility study for which was done under the project. IRAT (now ISRA) was to hP Tirnwi Hatii t+ n -nn1 1 nrmniirit nf I nl%-rt-r%ru an ii1i ari- t-n~ nnn,4nii rhizobium research started under the first projec't. B. TMPTMENTATTIO 11w The credit- was A sTndo une 25, 1973 and the project became effective on October 30, 1973 as envisaged. There was only one condition between Government and BNDS for on-lending of US$6.3 million of the IDA effective on the envisaged date. But the date was far into the 1973/74 credits given by BNDS for that season and hence start-up of the project was 11U L aLCUe RevisLOUs *LV a6ittU LU a L ut:4U L UY UUVq[- LC £H ±CULUGLJ LLIL LU F-Vy the technical assistance for the seed storage study to ONCAD instead of SODEVA. Thus a reallocation was made in the financing plan to LicLease Lhe cost of the technical assistance component for ONCAD and reduce it by the same amount for SODEVA. The consultant group, which had an extended contract with ONCAD for general reorganization, was given the contract for the seed study. However, the general consultant contract was cancelled by Government in early 1975 and IDA financing for the technical assistance to ONCAD therefore ceased. General reallocations were made in the funds allocated to SODEVA beginning in early 1975 and after IDA financing for ONCAD ceased, the SODEVA technical assistance was again increased. Also IRAT requested in January 1974, and IDA agreed, that the funds allocated to it for equipment be used to cover operating costs of the rhizobium research instead of equipment since it had enough equipment for the research. 13. The reallocations in funds did not affect project performance badly except for the technical assistance to ONCAD which did not improve the office's performance. Experience has shown clearly that the problems of ONCAD are structural and it is certain that performance would not have improved even if all the funds allocated for the assistance had been spent. Government continued financing of some consultants for the ONCAD - 15 - reorganization even after the IDA financinp had stonned hut no qnnrPriqhlP improvement in ONCAD efficiency has been observed. Physical Progress 14. 2he medium-term credits were supplied c.., schedule by BNDS and the c-reJit nnrtioln of the projecat closed by- thei end oF 197ke Since the credit became effective in October 1973, after BNDS had disbursed farmer credits for the 1971/74 season, IDA allowed retroactive financing to cover this first year of operation. 15. The contract of the consultants, engaged under the first project j~J.L~ L~tiI±L.L L~IL LU SJ1EI~iV, Wa: tf2LLC1LUtU U JU1Itku. d1lU dppluveu by IDA in March 1974. The extended contract for accounting improvement ended in early 5 after the two consultants aea-ung with accounting an data processing left Senegal at the end of 1974. Since the contract was Uue to expire in a few months, Government decided, with IDA agreement, to cancel the remaining portion of the contract. The seed storage specialist however remained till the end of 1975, and was financed in part under the remaining funds of the first project. 16. Orders for vehicles for SODEVA were placed towards the end of 1974 and these were delivered in 1975. Extension equipment was ordered in the 1975/76 season for use in the Sine Saloum project. The delay in ordering equipment was partly due to the time taken to approve the fund allocations within the SODEVA category. SODEVA with the help of RMWA, completed the feasibility study for the Sine Saloum project in th first half of 1974 for appraisal in July 1974. 17. The monitoring and evaluation, and the fertilizer studies started during the 1975/76 season and were carried out over three years. The late start of the studies was due in part to delays in appointing consultants and personnel and in training local investigators. These delays were minor and did not affect project performance. The only regrettable part was that the 1973/74 and 1974/75 seasons could not be evaluated as had been hoped under the Monitoring Component. 18. In general then the project proceeded on schedule except for the studies. The credit was finally closed in August 1979 after final disburse- ments had been made to cover the consultant services for the fertilizer study. Reporting 19. General reporting was quite good. The consultants provided semi-annual reports on technical assistance to ONCAD, and SODEVA also gave reports on its activities especially those concerning the monitoring and evaluation of the Programme Agricole as well as the fertilizer study. - 16 - ONCAD renorts on diRtribution of aarimilturnl innurs were nrnuitit nnvinin missions, which were quite frequent. ISRA did not provide periodic reports on ifQ ndv hut aiihmit-Pei a finial ronnrr n t-hp h vb4nhinm raoivrh t-n TnA in November 1975. Procurement 20. Under the project, an exception was made to international competi- idding with respec toLL famr LL * &metsexp ca rts .C. .. LMLementsii were to be purchased under negotiated contract between ONCAD and SISCOMA, the J.. J. t.L~ LAL C L 0 .LLLL-C JLL LILC p~ L L I U.Ut% LICIU ULI.1Y UCCLL Lt=Lt=Vt:U LLU011 SISCOMA even under procedures of ICB. These contract procedures were Unuu anAeA Laifactorily. LLoULemeCL UL LC LEL U) ULJJ LUo1UWeU LcsL^LHA tional competitive bidding, the results of which were approved by IDA in June i. um 0. veLL s y IVJ-vA ro".owea the procedures of International Competitive Bidding but in view of the limited time between the call for bids and the opening of bids and the limited size or the purchases, the bids were only advertised locally. IDA also agreed to waive the ICB condition on the purchase of office equipment, furniture and extension equipment for SODEVA. This followed requests from SODEVA for the waiver, arguing that the range or local suppliers in Senegal was large and represen- tative enough of the international market to ensure price competitiveness. 22. It should be noted that the procedures worked well and the use of local competitive bidding did not affect the quality of materials or equipment purchased by SODEVA. Project Costs 23. The appraisal report estimated total project costs at CFAF 2,644 million (US$11.5 million) including price contingencies of CFAF 131 million (US$570,000). Though more units of agricultural implements were distributed than estimated at appraisal, most of them were light equipment hence explaining the lower costs. Actual costs compare with appraisal estimates as follows: - 17 - Appraisal Actual Estimates Actual Appraisal ------CFAF million----- % Animal Drawn implements & Carts 1,935.0 1,668.3 86.2 Draft Animals 200.0 312.0 156 Equipment and Buildings 119.0 116.1 97.6 Technical Assistance ONCAD 73.0 41.3 56.6 SODEVA 69.0 183.5 266 Studies 117.0 114.3 97.7 Subtotal 2,513.0 2,435.5 96.9 Contingencies 131.0 - - Total 2,644.0 2,435.5 92.1 24. Total project cost was slightly less than estimated owing to the lower unit costs of agricultural equipment distributed and the consequent lower costs to BNDS. and also owing to the cancellation of the ONCAD contract earlier then planned. Financing and Disbursements 25. The initial financing plan was adhered to with IDA providing 75% of financing for sh-loans made by RNDS, R7% of the exnpnditures on eauipment and buildings and 100% of technical assistance costs, except for the technical to the consultants to cover the local costs. This was in accordance with the table below: PROJECT FINANCING IDA RNDS GOVERNMENT TOTAL CFAF '000 US$ CFAF '000 US$ CFAF '000 US$ CFAF '000 US$ million miLlion million million Animal Drawn Implements Carts and Draft Animals 1,485.2 6,293.7 495.1 2,098 - - 1;980.3 8,391.7 Equipment and Buildings 101.1 418.3 - - 15.0 62.7 116.1 481.0 Technical Assistance 1/ ONCAD 31.9 145.9- - - 9.4 43.1 41.3 189..0 SODEVA 183.5 806.5 - - - - 183.5 806.5 F Studies 114.3 509.6 - - - - 114.3 509.6 00 Subtotal 1,916 8,174.0 495.1 2,098 24.4 105.8 2,435.5 10,318.3 Cancellation 42.4 42.4 Total 8,216.4- 495.1 2,098 24.2 105.8 2,435.5 10,420.2 1/ US$16,400 relating to the seed storage study was paid under the remaining funds of the First Agricultural Credit Project (Cr. 140-SEN) in October, 1975. - 19 - 96. Crdit- diqhirqemnt did not start until Auast 1974. This was due to delays in sending reimbursement requests from the agencies to IDA. Tt- was alsn narl-v hpraiAp disbtirnaments wprp qtill hPina mad under thp first credit. When actual disbursement started, the pace quickened and by the ani nf 1Q7A Ahnllt Rr nF thp irpdit had hoon diahursCd. The rammining 15% was however disbursed over 2-1/2 years, since the studies components weare t c o41 -ntnungn and weoreo n17, C-1mr.i f amA m4l...1070, 'P0 mljglf- delays in disbursement did not affect the project. Covenants 27. Compliance with covenants was generally good. One covenant not compLJLu .L wi iru-Vul-VU L-ue yealty auudiing u. VAkU tLLLcVUnLL0. 1LuLM w LL: result of the poor state of accounting at ONCAD. Following an audit of the ONCIO 1971/2 accounts by the CVCCEP and because of the strong criticisms voiced by this Government audit organization of both ONCAD and its consultant group, new measures were inititated to strengthen accounting procedures. However balance sheets have been considerably late in coming out and therefore annual audits were not possible. Under constant prodding from IDA, Government awarded a contract in 1977 to some consultants to reestablish ONCAD accounts since the i9i5tio season. The new consultant group is still preparing the accounts. 28. Another covenant not fully complied with involved on-lending rate of interest. BNDS was to make medium-term loans at a rate of 2 percentage points above the discount rate of BCEAO. This rate had been raised from 3.5 to 5.5 per cent in January 1973. However Government asked BNDS not to increase its rate of 5.5 percent to farmers and promised to pay the difference to BNDS. Thus the lower rate was applied in 1973/74 and 1974/75 contrary to one of the covenants of the credit agreement. IDA did not request Government to allow BNDS to raise the rate and Government still owes BNDS CFAF 84.7 million for the interest charges lost. The situation was resolved in 1975/76 when Government allowed the interest to farmers on medium term loans to be raised by two percent. IDA should remind Government to settle the interest charges lost by BNDS for 1973/74 and 1974/75. C. AGRICULTURAL IMPACT Credit Component 29. There is considerable difficulty in estimating, ex-post, the benefits from projects of this nature. This difficulty was especially apparent in the ex-post evaluation of the First Agricultural Credit project in which there was no monitoring and evaluation to give an indica- tion of even input-output relationships. The benefits from this project then can only be evaluated by: (1) looking at the benefits from a micro- economic point of view using the work of the monitoring and evaluation unit as a basis, (2) the increase in productive capacity through the increased purchases of implements and fertilizer during the project period and (3) the success of the fertilizer study which could lead to a more rational fertilizer policy for the country. - 20 - In ThP annraiAal renort. using two tvoes of farm budgets. estimated an incremental production of 21,250 tons of groundnuts by year 3 increasing o 3r5,75 tnns h yaAr 10 Tn addifon millet and sorghum oroduction was to increase by 7,800 tons by year 10, plus increases in rice and cotton procion. nnEyA the mnitnrina anency did not maRure actual produc- tion increases resulting from the project. However, the monitoring studies sh_oWed certain - relatatshimel hot-won Hp rAtp nf Pnuiinpnt used and the yield per hectare. Thus for example, the studies show that for a 1% 4nr...e in the rate of use of hno yields increased by 3.57 kg/ha in the south. Using the results of the monitoring studies plus other work done h inte groundnut basi wehveetim-atedt an inc-remental0 nrir,rtinn of 29,000 tons for groundnuts by year 3 and 11,000 tons for millet/sorghum kse Z'Unex -+, LOUL=~ a..J --rL th s a cl ti n. they indicate, in a rough order of magnitude, the impact of the project. 31. However, the impact of the project is also underlined by the total production of all crops registered in Senegal from 1973/74 (Annx 1). Total groundnut production increased substantially from 1973/74, when the project started, and reached a record 1.4 milion tons in 1975/76, the lat year of the credit component. For the first time since 1965/66, the average groundnut yield for the country reached over 1 ton/hectare for that year. Millet production also picked up in 1973/74 increasing to over //U,UUU tOnS 1n 19714/J. OLgnilicant prouuutiOL -u----- w^_ -Jo- registered for all the other crops - maize, rice, cotton and cowpeas. Although production increases cannot wholly be attributeU to the project, since all the three project years were good rainfall years it should be noted that the production potential has nevertheless substantially, increaseu since the predraught years. Also the increase in the producer prices of groundnuts and millet relative to those or equipment and fertilizers helped to increase farmers' efforts. And as explained in paras 29 and 30, the impact was measured on a micro-basis and extrapolated to the project area. 32. There was an upward trend in the distribution of implements throughout the project period. Agricultural implements delivered in 1973/74 and 1975/76 exceeded appraisal estimates for all categories except for carts and heavy equipment. A slight shortfall in deliveries occurred in 1974/75 compared to appraisal, but the shortfall was due more to the inability of ONCAD to supply the amounts requested rather than a lack of demand for these inputs. Thus under the project medium term credits totalling CFAF 1,980.2 million 2/ (US$8.4 million) were supplied to farmers in the three years. A total of 150.9 thousand units of both light and heavy implements were distributed compared to the expected 120 thousand at appraisal. In addition 6,700 pairs of oxen were supplied and about 345 horses were bought on credit. 1/ The rate of equipment use was defined as the actual area cultivated by the equipment divided by the potential area it can cultivate as defined by norms set by ISRA. 2/ This figure excludes interest and government subsidies. - 21 - 33. The impact of the project is also revealed by comparing a calcula- tion of the "without project equipment stock" with the "with project stock" for the three years 1/. STOCK OF EQUIPMENT ('000 units) 1972/73 1973/74 1974/75 1975/76 Total Stock without with without with without with Seeders 137.7 111.8 123.1 92.5 112.0 75.9 112.8 Hoes 159.9 145.6 164.5 138.2 172.1 129.2 190.0 Lifters 31.6 29.9 34.9 29.0 38.8 27.2 42.6 Carts 55.8 54.3 62.0 51.8 64.7 48.3 65.4 Ploughs 19.7 18.2 20.9 17.5 23.0 15.8 25.5 Source: mission estimates See also Annex 4 Table la. 34. The table shows that the project increased the rate of hoe use, for example, by about 47% and the use of seeders by 49%. These increases in equipment rate were used to estimate additional production from the project (Annex 4). 35. Under the project, short term credits were to be supplied for fertilizer by BNDS. The appraisal report estimated that demand for fer- tilizer was not expected to increase beyond the average levels reached by 1972/73. Deliveries of fertilizer in the first two project years were not much higher than the appraisal expectation. There was a substantial increase in demand and delivery. however, in 1975/76 and the demand level has been maintained since then. Annex 1, tables 5 and 6 show the deliveries of arlnitiral imnlmPntR and fprtili7er over the credit neriod and the years following. 36. Though the calculation of incremental production only approximates the actual nrnim-ntion increaqPz_ the increases made in the stock of eauipment and fertilizer demand reveal that the credit component of the project had a 1/ The stock for 1972/73 was calculated by adding the equipment supplied ,rnA~,- ~Anw4,.ra fn? annh yanr w.ith t-hp nAqnTmni)n that equipment is replaced every ten years. The "with project" situation s calculated from the stock for anch year hv addine the deliveries under the project and subtracting the equipment placed 10 years earlier. Thus 1973/7". 11-hpjet "hoe eafils hinp Qtnrc 1Q79/71 plus deliveries in 1973/74 less hoes supplied in 1963/64. It is assumed for the "withoutt project situation that therewould have been no deliveries without the project and the stock would therefore be reduced by placements 10 years earlier. - 22 - favorable impact on production. It helped improve mechanization, which studies have shown indispensable in advancing seeding dates and improving timely weeding, factors that have significant impact on yields and thus incomes. The actual distribution of the farm implements per beneficiary is not known. But using similar assumptions on distribution of new and replacement equipment as per the appraisal report, an estimated 78,300 farm families benefitted from the medium term credits supplied under the project 1/ compared with 55,000 estimated at appraisal. D. INSTITUTION BUILDING COMPONENT 37. The objectives of the two agricultural credit projects included building up the various institutions involved with the PA to perform more efficiently and effectively. Thus under the first project, BNDS, ONCAD and SODEVA benefitted from technical assistance. The latter two institutions benefitted from additional assistance during the second project. One major objective in institution building was also the help which the two projects were to offer to BNDS to build up a significant revolving fund for financing PA. Credit Recovery and BNDS 38. Under the two projects a total of US$9.4 million (CFAF 2.7 billion) of the IDA funds was onlent to BNDS over 1969/70 to 1975/76 by Government, at the BCEAO discount rate with four years of grace and a twenty year repayment period in each case. BNDS in turn was to lend these funds to cooperatives at 2 percent above the discount rate for a period of 5 years in the case of medium term credits, and for one year for short-term credits. It was expected that BNDS would have built up a large revolving fund by the end of the two projects to be able to finance the PA without outside assistance, assuming reasonable credit repayments by cooperatives, and with only minor assistance from the Central Bank. 39. Since 1973/74 total credits by BNDS to cooperatives have increased considerably and have been running higher than repayments by cooperatives. The increase in credits is the result of a rising demand for both fertilizers and agricultural implements. Credits repayment however, which was improving un to 1975/76, has not been satisfactory for the past few years and was very poor in 1978/79 despite a good harvest. The table below gives an idea of thp crpdit Ritiirinn nnd rpnavments by cooneratives for 1973/74 - 1978/79. Annex 6 Table 1 gives a breakdown of both seed and BNDS credits from 1960/70 to 1978/79. The assmpton gie 9*6 Sfarer buyix.ng hoes an~d seeprlers for th P f ~t time; 27,300 farmers who owned seeders and hoes replaced them and 23,Z.'0 farmers who owned seeders bought hoes; in addition 18,200 fabrmers ars.1,qt various other implements including ploughs and toolbars. - 23 - BNDS CREDITS FOR PA IN SENEGAL 1/ 1973/74 1974/75 1975/76 1976/77 1977/78 1978/79 Credits to Cooperatives Short Term 710.8 461.8 1,020.4 1,743.5 1,355.7 2,156.1 M___ tnn , r I _nn~l nfl. - rlqeu-Luu lekrm DJ) I,UJo.o 1,725._J1 Z,52.21ZO-Z Total Credits 1,536.1 1,112.1 2,059.2 3,468.6 3,927.9 4,978.8 Credit Repayments Mature debt 1,304.6 1,441.8 2,138.0 2,5. 3,162.8 4,453.1 Repayments 944.6 1,106.7 1,934.9 2,483.6 240.12/ 2,405.1 Repayment Rate (7) 72.4 76.7 90.5 84.0 - 54.0 Unpaid Balance 360.0 335.1 203.1 471.6 2,922.7 2,047.5 Balance not yet due 1,333.3 1,434.8 1,690.9 2,407.3 3,631.9 5,782.6 1/ The situation for each year is as of 31st December, i.e., 1973/74 figures refer to the situation on 31st December 1973. 2/ Government cancelled CFAF 1,296.8 million of the mature debt on 1977/78 and carried forward CFAF 1,625.9 million to be repaid over three years due to the drought of that year. The cancelled debt was to be repaid by the stabilization fund (CPSP). From the debt due, the cancelled debt, and the amount carried forward, the mission estimates that only CFAF 240.1 million was repaid in 1977/78. 40. Through the technical assistance of the first project, the performance of BNDS has improved substantially. The two credits have also expanded the capabilities of BNDS to give cooperatives credit. However, owing to the large increase in the proportion of medium-term debts and the increasing defaulting of the cooperatives, BNDS has not been able to set up any large revolving fund. And if steps are not taken to improve credit repayment, continuous funding of the Programme Agricole could be in jeopardy. It should be noted here that Government policy of cancelling debts has - 24 - created the impression among farmers that any time the weather is slightly below normal the Government would either cancel their debts or postpone them. This policy has not helped the repayment rate and IDA should urge Government to intervene in debt repayment only under extreme and closely defined circumstances or set up an insurance system which would automatically determine debts to be cancelled under different weather circumstances (see Annex 6 - STABEX). Some measures beina considered by both Government and IDA, which can be taken to reduce the increasing reliance of cooperatives nn rrpdit and also to imorove the rate of repayment, are discussed in Annex 6. OMNCAn 41. The"VOffiea ntirmnl rip r.nnnprni-inn Pt e'A.q.1-tanep ponur le Develop- pement" (ONCAD) was established in 1966 to surpervise cooperatives, maintain their accounts and train their leadersa w th timp- nwrAT) wAs giupn a mnopnnny in purchasing groundnuts and millet from farmers and of importing and dis- t.4kU*4-ngr 4e 1 I T- n a srn t-ha ranncnniMlit-v fnr nrnotirpmpnt nnei LtLLuLI L _L1~ -.t -.--- -- - --- -------- - distribution of fertilizers, farm implements, draft animals and seeds. In the perforanc oUfLL thes actiitie rIXrA U-V I,-- -1A- 'km~.~' J cient. 42. During the first agricultural credit project, financed by IDA, the U1JI~~.~J emphasisA of 4. -4 -4 __ Ap-1 - -f-l - t-n _q1Qi- inc _q rpr%an i zation of ONCAD. Prior to the First project, Government had asked a consult- ing firm to study the Office and delineate itS main inefficie"ncs Bnaedo the recognized deficiencies of ONCAD therefore, one third of the proceeds of the first project was devoted to technical assistan-e to reorganiz nMrAn ndA eliminate the inefficiencies. 43. The Government with IDA agreement, selected a consultant group to do the reorganization and make recommendations for its implementation. ONCAD Component Under Second Project 44. The consultants' contract was exteuueu at tne end ouu 17i I mid-1975, to continue the reorganization of ONCAD's accounting department and to provide a seed storage specialist who would study seed turage problems and propose ways to improve groundnut seed distribution and handling. 1/ In November 1979, Government removed ONCAD from marketing of millet and rice. Rice imports and distribution is now entrusted to the CPSP and millet marketing is to be done through licensed wholesalers. - 25 - 4a. Tn early 107A, CVCCEP, the Governemnt audit naonni7ntion iqqiipd its report on an audit of ONCAD accounts and the report criticized financial prcdrsadmngment~ 4,n OCCAD. The,~ PCR ,a the PPAPR of the., first rr.c also judged the ONCAD component as a failure in as much as the reorganization assistance to ONCAD under the first project. These included: (a) a better organizatLion oI. grundutI mareng,LI ini particular, bte aca oto of marketing activities and prompt payment of farmers; (b) better transporta- tion of grounanuts - though not inputs - and (c) improvements i Uuperative account keeping. These achievements, however, have now been lost and the overall UNUAD situation has deteriorated since 19706/77, and in addition LU Le problems that were not resolved under the first project: (i) overall financial control; (ii) inventory control; (iii) excess staffing; (iv) handling of the seed stockpile; and (v) input transportation and distribution; the overall ONCAD picture has become gloomy. 47. Cooperative accounting during the last few years has deteriorated and analysis being made by SODEVA in the Sine Saloum region have shown large discrepancies between accounts kept by the weighers tor cooperatives, ONCAD accounts kept at the regional level and the debts as recognized by individual farmers. Groundnut marketing has also deteriorated over the last few years and ONCAD is reported to have lost a substantial amount on the groundnut account in 1977/78. Thus in overall accounting and tinancial matters and other ONCAD activities, the situation is, at best, similar to when the first project started in 1969/70. 48. At the end of 1974, the two financial experts involved in the extended contract and who were helping in reorganizing ONCAD's accounting department left Senegal, Government with IDA agreement cancelled the consultants' contract and thus the technical assistance to ONCAD component of the project was ended. Some payments were still made to the seed storage specialist from the proceeds of the first project, since the study fell under the category of the general reorganization envisaged by the first project. 49. The major failure of both projects in terms of the ONCAD component involved the role assigned to the consultants in the ONCAD reorganization, Government's reaction to their recommendations and the terms of reference approved by IDA for the consultants. The consultants had made a number of recommendations, which if implemented, could possibly have led to an improve- ment in accounting and financial management of ONCAD. The consultants, however, had no mandate to ensure implementation of their recommendations, and while some changes were made in accordance with these recommendations not all of them were followed. Thus the consultants, for example, recommended com- puterization of accounting and data processing. ONCAD therefore abolished the positions of certain data processors in the regions but failed to obtain - 26 - and institute a computerized system to take their place. Hence information flow and data processing were hampered and this has accounted to a large extent for the gaps and errors in ONCAD accounts. ONCAD had to rely on the CEP for computer services which were not always assured. Thus it has not been able to put out a balance sheet since 1976/77. 50. An important lesson from the ONCAD technical assistance is the role to be assigned to consultants in projects of this kind. It is absolutely essential for the success of reorganizations of this nature to obtain a comprehensive terms of reference for consultants, with Government commitment to take required action immediately, after recommendations have been discussed with all involved parties, in this particular case, IDA and the consultants. Though it is difficult for IDA to supervise the work of consultants, it may be necessary to set up mid-term progress reviews, in projects of this nature, and to have independent experts visit organizations benfiting from the technical assistance during project implementation. SODEVA 51. Under the Droiect. IDA agreed to finance the cost of extension equipment, vehicles and office materials for SODEVA headquarters. Provision was also made to finance three buildings at the SODEVA training school, and technical assistance to reinforce its study unit. 52. Financing of extension equipment helped SODEVA to re-orient its ncivities towards a more farmer oriented extension service. New activities including the establishment of demonstration plots were undertaken and films on extension nractices were shown to farmers. The impact of these activities however in terms of actual production or yield increases has so far hoon nealipihlp thouch it is expected that with the re-appraisal of the Sine Saloum project (Cr. 549/Ln. 1113), the equipment supplied for SODEVA wniAl hp used more Pfficientlv. The buildines. Dut up for the SODEVA training school, one finished in April 1977 and the other two finished in 1978/79 shouA h sefu1 in Hho rp-trninine of extension aents in new systems of extension to be introduced in the re-appraised Sine Saloum project. 53. Technical assistance to SODEVA during the project period has considerably strengthened the Monitonrincy nnt R%Ynliintinn Unit cof thep agCency. It was through the work of the unit that most of the problems facing the Sine Saou proec were-+ -- ----A- Tb uni hia provNdAnn fairly good data base for evaluating the impact of extension activities and agricultural nputs in tILe grudu 1-4-U 4L L ----- I -- 4 - +- C1- - - -. OnTNUTT J e4. lUUgh sVeLd LUU leml L1UC1.L1 WLIH LL= j J. LUL ILL U JVy, the general impact of the project in reinforcing SODEVA has been positive. it is expected that with the assistance being provided under the Sine Saloum project, especially in training of extension agents, the agency would be able to provide good service to the farmers in the groundnut basin. - 27 - E. THE STUDIES COMPONENT 55. Five studies were envisaged to be done under the project (i) the feasibility study of the Sine Saloum nroiect- (ii) a seed storage studv ii) a monitoring and evaluation of the programme agricole; (iv) a fertilizer study; and (v) rhizobium research by ISRA. 56. The feasibilitv study was finished qiieressfuli1v hv mid-1Q74_ hv SODEVA with the help of the Bank's regional mission in West Africa. This study formed the hasis of the Sine Snlnim Acrirultirnl lP-,rA1onmPnt Prnoet (Cr. 549/Ln. 1113). 57. Government decided after the project had started that ONCAD should be rnnoni'hl for the ced storace study. An ornprt wn therefore contracted to do the study. However it was done in the general context of thea nCrAD) -ranranni7n4nn otiel nn "nt 4nc"abrto withi the Mini Ctr17 of Agriculture, SODEVA and the Research Institutes as envisaged, and the imact- hase therefore beean limitala Du,r4ry the sztudy thea Euirnoenn Development Fund (FED) also financed a similar study and the consultant wrkedl,a 4in colabo,,ra,tin w4ith the- Wnf-J-4,~1 CQ00A Va\T4rn 4m,,lonont-jncn the VPQ~T project. The emphasis was more on storage of all groundnuts rather than se.Tefnlrepot o--f .the consua nt ws issued- in A--41 IQ#; For seed management, it suggests a program of unifying the two levels of seed u.,1*4 4-14ca"on n,ow,~.,44, 4xi,i C Senegal CO 9. 4:4 A'JU LtWLQ t~L 5lJ J5. The most part of the seed sLuUy Was HLo Li.Linucu unut %L ne project as envisaged, owing to the confusion arising out of Government canceiation ut some part of the consultants contract. A tnerefCue uiu not have much influence on the orientation of the study and hence the project objectives of fiUaing ways to transter respousibiLy Lor seeU storage from ONCAD to the cooperatives, and of providing a seed storage policy in Senegal, were not achieved. 59. The monitoring and evaluation of the PA was successfully done over three years beginning in 1975/76. The study provided useful socio-economic information for the grounutU Dasin aU maue a 'air assessment uL LUe relatLion- ships between agricultural inputs and outputs in the basin. It is unfortunate that the study could not be started with the project and also did not measure actual production increases in the groundnut basin as a result of inputs distributed under the project. However the lessons learned from the first year monitoring enabled SODEVA to carry out a more useful analysis in the following two years and the experience gained by this component has strengthened the Monitoring and Evalution unit of SODEVA, which has been doing useful evalua- tion work in connection with the Sine Saloum Agricultural Development Project. 60. The fertilizer study was done over the years 1975/76 to 1977/78 by the study unit of SODEVA with the help of IFDC, a consultant firm appointed tor the purposes. Thirty-two tertilizer trials were carried out each year by SODEVA on millet and groundnuts on farmers' fields and the results analyzed by IFDC. The analysis also included all fertilizer research done by IRAT, ISRA and IRHO since the early 1950s. IFDC presented - 28 - a working draft of the results in mid-1979, a second draft in September 1979 and the final report is expected in April 1980. 61. The preliminary results of the IFDC analysis indicate that there should be a change in fertilizer grades and application rates, and the recommendations of the study, if accepted, could have considerable impact on fertilizer usage in the country and help to reduce costs to both govern- ment and farmers. It also gives a mechanism for ferlitizer price setting that can be considerably useful for the country. The Government has indicated its interest in the results, and plans, after receiving the final IFDC report, to organize a meeting between the extension agencies, research institutes and eventually the fertilizer industry to evolve a new strategy for fertilizer use in the country. 62. This component of the project has been successful in terms of impact. It has generated a lively debate on the appropriateness of compound fertilizer use in Senegal as well as questioned the economic or even social justification for uniform fertilizer prices and subsidies in Senegal. A full analysis of the results and reactions to the recommendations is provided in Annex 5. F. ECONOMIC REEVALUATION Economic Returns 63. For the project as a whole costs were slightly lower than estimated at appraisal both in economic and financial terms. Economic farmgate prices were also higher in real terms than appraisal estimates due mainly to the higher C.I.F. prices for groundnuts than forecast at appraisal. Thus, the economic rate of return calculated over 12 yearF -' "ompletion 1/ was higher than estimated at appraisal (over 13 years). Appraisal Actual Forecast 1979 prices Appraisal Prices Rate of Return 22% 29.7% 18% 64. The economic rate of return for the proiect has been estimated at 29.7% as compared with 22% at appraisal. Using the prices forecast at anraisal and actual costs in 1973 terms the rate of return is estimated at 1/ At appraisal it was assumed that the useful life of equipment supplied under the project would be 10 years-excludinc thu ver nf irfh s. Tho PCR includes the year of purchase in the 10 year useful life of each equ4 p-.nt - 29 - 18%. The difference, is the result of the relatively lower output prices, projected at appraisal, than actually occured. 65. Assumptions underlying the calculation of the rate of return are in Annex 4. G. INSTITUTIONAL PERFORMANCE 66. Project implementation was effectively carried out by the executing agencies. SODEVA and BNDS in particular did their work well and helped in the success of the project. Despite the organizational and financial problems of ONCAD (paras. 47-51) input distribution was done satisfactorily and in general, amounts of implements and fertilizer successfully distributed exceeded apprai- sal expectations. 67. SODEVA, helped by the technical assistance component, improved its monitoring and evaluation unit and the unit has been doing creditable work in the groundnut basin. BNDS also efficiently handled project accounts and finances for the credit component. 68. Cooperative development was not achieved under the project. Though this was not an explicit objective of the project, it was hoped that the continuation of ONCAD technical assistance could help accelerate the development. Agency coordination was lacking and thus services to farmers were not integrated. H. BANK PERFORMANCE 69. The Bank performed satisfactorily overall during project preparation, appraisal and supervision. 70. Project preparation and appraisal reasonably took into account the increasing demand for agricultural implements by farmers. The project supported the Government's Programme Agricole which constitutes the country's maior priority in terms of the agriculture sector. The project lacked a component for fertilizer credit though fertilizer distribution expanded durine the last year of the project. The Bank however had anticipated correctly that BNDS credits, which without the project, would have gone for medium-term loans for imnlements could be used to finance any incremental fertilizer demand. 71. Despite the problems encountered by ONCAD especially in organiza- tinnl and finanrial mattprs innut distribution did not suffer during the project period. Cooperative organization and development, however did not iinrr daQnita th affn-rt-Q msra uinder the Fir.st AgrrJ r1ilt11rq1 rrid it Project. On hindsight, the Bank should probably have urged Government to Onvolve agencies other than ONCAD is topatre dexopment ansi Mnaement ONCAD's poor cooperative management is to a large extent responsible for the critical debt situation that has d-eveloped in the countrys Disreancieso - 30 - in accounts exist at all levels, from BNDS to ONCAD to cooperative and to the farmer, such that a complete reassessment of cooperative accounts will have to be made to ascertain the real debt situation. 72. Supervision was frequent and adequate. Frequent discussions were held with SODEVA and that accounts for the success of the SODEVA components of the project. The lack of adequate data from ONCAD and its confused organizational and financial situations made supervision of the ONCAD components difficult. This coupled with the reluctance of Government to take any concrete action, to reorganize ONCAD, during the project period, makes it difficult to see what the Bank could have done to improve the situation. Several Bank discussions with Government on the ONCAD situation led to the establishment, in mid-1979, of working groups to study ways of improving the situation in and several positive results are beginning to emerge from the group's work. The Bank Group is still continuing its dialogue with Government and it is expected that through these discussions ONCAD can finally be put back on the right course. I. CONCLUSION 73. In terms of appraisal obiectives, the prolect successfully contributed to improving agricultural production in Senegal. The production estimates made at completion may be partial owine to the lack of comnlete data on actual yield or production increases. Thus, although production estimates have been made on a micro-hasis and extranolated to the nroiPrt area, the impact of the project may be underlined by the development of tntal nrnduiitinn of All rron in Rpnpon1 And the Jnreed demand and distribution of fertilizer and equipment registered during-the three years nf thp Prpdit ronnnc-t. Ricynifirnnt- n-ndiintinn int-ran.--p were regiszteared for all crops during those years and while all these cannot be attributed 5), the project made a significant contribution to the increases. 74. The project conformed with the Government's priorities in agricultur- al deeopet Throug the montoin and Ui hepe to highlight the failings of the input distribution and credit systems that nLave e tou a new assesment u Lue overli organiiLon Ut Lue rural sector. This is a highly positive achievement of the project. The fertilizer study financed under the project has also opened up debate on the appropriate- ness and social and economic profitability of present fertilizer grades and subsidies. Inrougn the study, a new national fertilizer policy is expected to evolve that can lead to savings for both the state and individual farmer. 75. The only significant failing of the project was its inability to help in the ONCAD reorganization and improve its management. Here, it is pertinent to note the conclusion of the completion report of the - 31 - First Agricultural Credit Project that an extensive project of this type is not the best medium for effecting institutional improvements in agencies (like ONCAD) for which the project constitutes only a small part of their responsibilities. 76. Some indirect benefits have come from the proiect. Government is now seriously tackling the problem of participation of rural people in develonment as a result of the failines revealed by the Monitorine and Evaluation. Potentially, also, BNDS can achieve financial independence that it rpndrpq to dpvplnn Pffertivp lpndina nrnarnmn- throuah thp PrPditq nn- lent to it by Government from the proceeds of the project and its predeces- nor. Th- finnnrinl nutinomv nf Re hNp ir hdvAndAnr nhwnindirnt P hz u fl it can be in recuperating debts owed to it by Cooperatves and ONCAD. SENEGAL SECOND AGRICULTURAL CREDIT PROJECT Groundnut Cultivation in the Whole Country Area Production and Yields Area (ha) Yields_(kg/ha) Production ('000 tons) Groundnut Total Groundnut Total Groundnut Total Basin Senegal Basin Senegal Basin Senegal 1969/70 793.0 953.1 840 827 666 788.8 1970/71 877.0 982.7 513 593 450 583.0 1971/72 865.0 1,060.3 940 932 813 988.5 1972/73 922.0 1,071.4 465 532 428 570.0 1973/74 903.3 1,026.2 526 657 475.0 674.9 1974/75 89:2.3 1,152.1 840 862 750.0 993.0 1975/76 1,017.8 1,203.0 1,105 1,174 1,125.0 1,412.0 1976/77 1,03:3.1 1,347.0 868 897 896.9 1,208.0 1977/78 968.2 1,113.0 472 466 456.8 519.0 1978/79 957.7 1,156.9 751 910 719.0 1,053.0 1979/80 - 1,035.9 - 712 - 737.2 Sources: DGPA for Total Senegal figures. SODEVA for Groundnut Basin figures 1973/74 - 1979/80. m and PCR 1st Agricultural Credit Project 1969/70 - 1972/73. S INEGAL SECOND AGRICULTURAL CREDIT PROJECT Groundnut Marketed by ONCAD Controlled Marketable MarkEted 3/ Production Seed Quantity Oil Mills Exportt- Other- Lost-- ('000 tons) ('000 tons) ('000 tons) ('000 t ons) ('000 ton, 1969/70 789.0 87.5 601.0 595.0 - - - 1970/71 583 88.9 454.0 401.0 - - 1971/72 988.5 101.4 764.0 755.0 - - - 1972/73 570.0 101.5 466.0 433.0 - - - 1973/74 570.1 126.5 443.6 429.0 b.4 6.2 - 1974/75 882.9 125.3 756.2 728.7 12.1 14.0 1.4 1975/76 1,302.2 149.8 1,152.4 911.8 185.3 34.9 20.4 1976/77 1,077.5 123.9 953.6 826.0 70.5 30.4 26.7 1977/78 459.0 149.0 316.1 273.8 18.0 24.3 - 1978/79 901.6 123.7 778.0 456.7 -- 279.4 41.8 Source: ONCAD except for Exports and Others (see below). 1/ - means not available. 2/ Exports of shelled groundnuts converted to unshelled by dividing by 0.7. Source -, Direction des Douanes. F. 3/ Quantities unaccounted for. 4/ Figures provided by ONCAD on losses. ANNEX 1 - 34 - Table 3 SENEGAL SECOND AGRICULTURAL CREDIT PROJECT Millet and Sorghum Cultivation Area, Production, Yields and Production Marketed Hectares Production Yields Production Marketed ('000) ('000 tons) kg/ha through ONCAD(tons) 1969/70 1,037.3 634.8 612 10,472 1970/71 972.2 400.9 412 346 1971/72 974.6 582.7 597 2,866 1972/73 936.3 322.9 344 21 1973/74 1,093.5 510.8 467 29,725 1974/75 1,155.1 777.0 673 35,969 1975/76 963.2 621.0 673 12,123 1976/77 952.4 553.8 581 4,993 1977/78 898.4 416.7 464 1 '4G 1978/79 1,034.6 802.9 776 1(< .00 1979/80 954.8 796.1 834 Sources: DGPA and ONCAD (for Marketing figures) ANNEX 1 'ale A - 35 - SECOND AGRICULTURAL CREDIT PROJECT Confectionery Groundnuts Area, Production, Yields Area Production Yields 1970/71 6,080 5,636 927 1971/72 8,530 8,650 1,014 1972/73 11,300 11,200 991 1973/74 14,100 10,200 723 1974/75 17.000 13.000 765 1075/176 24000 1900 90 /0J 4 - LJ /14 L'~ .LA ,~II :,P. 14 1978/79 23,633 12,253 518 1979/80 21,191 12,720 600 Source : DGPA SENEGAL SECOND AGRICULTURAL CREDIT PROJECT Credit Purchases of Implements and Draft Animals 1973'74 1S74/75 1 576 Total Appraisal Ackual Appraisal Actual Appraisal Actual Appraisal Actual Estimates ________ Estimates E________jstimates ________ Estimates _______ Item Number Value Number Value Number Value Number alue Number Value Number - r^-L (CFAF __ (CFAF (CFAF __ _ (CFAF m (CFAF (CFAF (CFAF Seeders 9.000 112.5 L . 11 141.8 9,000 112.5 8,224 102.2 9,000 112.5 17, R2 215.5 27000 337.5 36,880 459.5 Hoes 16,000 124.8 18,01 148.2 16 000 124.8 14,965 118.9 16,000 124.8 26.885 211.7 48,000 374.4 60,110 475.0 Litters 3,700 18.5 4,976 19.1 4,000 20.C 4,772 18.9 4,300 21.5 5,607 21.7 12,000 60.o 15,355 59.7 Toolbars(with attachments) 2,300 96.6 264 11.4 2,300 96.6 278 11.8 2,400 100.8 205 10.6 7,000 294.0 747 33.8 Tollbars(without attachments) - - 3,107 17.8 - - 2,134 12.2 - - 3,852 22.1 - - 9,095 52.1 Attachments - - 382 1.8 - - 655 4.6 - - 513 3-3 - - 3,410 17.9 Ploughs - - 2.734 28.7 - - 2.T3 26.7 - - 4.173 40.0 8,421 91.0 Carts 8,000 267.2 7,742 215.L 8.500 3 5,191 147.9 9,500 Il 4,189 116.1 26,000 86 17,122 479.7 Subtotal 619.6 584.5 637.A 443.2 676.9 641.0 1,934.3 1,668.7 Drart Animals Horses 100 5.0 - - 200 6.0o 128 3.9 200 6.0 217 8.7 500 15.0 345 12.6 Heifer - - - - - - - - - -- 254 17.8 - - 254 17.8 Oxen (Pairs) 1,600 52.8 - - 2,000 66.0 3,086 . 2,000 66 _ 3,637 _191.3 5,600 184.8 6,734 281.6 Subtotal 55.8 - 72.0 94.2 72.0 217.8 199.8 312.9 Total Value (m. CFAF) 65435 09 57.4 8 888 1 ;ources: BNDS and Appraisal Report (n5 - 37 - ANNEX 1 Table 6 SENEGAL SECOND AGRICULTURAL CREDIT PROJECT COMPLETION REPORT Total Implements Distributed in Senegal (Units) Hoes Seeders Lifters Ploughs Carts Multipurpose Others 1/ Frames 1974/75 14,965 8,224 4,772 2,773 5,191 278 2,797 1975/76 26,885 17,323 5,607 4,173 4,189 205 4,321 1976/77 27,836 27,946 9,626 4,718 15,291 93 6,070 1977/78 23,035 18,563 11,944 3,010 11,786 40 6,448 1978/79 17,592 15,341 8,526 4,664 18,811 543 4,787 SOURCES: BNDS (1973/74 - 1975/76) ONCAD (1976/77 - 1978/79) 1/ Mainly toolbars without attachments and implement parts. ANNEX 1 Table 7 - 38 - SENEGAL SECOND AGRICULTURAL CREDIT PROJECT Fertilizer ConsumDtion (tons) 1/ 2/ Groundnut Basin Senegal Total Fertilizer Grounn M411le Gr1mnnu M411et Tr4114=ieA/ Gt r ou "IM, t-o~,0I.t I 19f % 70/71"7 5,82 16,99A0 6,49 6,200 14,828) 19/4// 21,527 10,884 2,639 ZO,514 43,9 9 1975/76 37,457 26,703 46,520 2,6'-a 1976/77 38,370 28,094 63,545 32,197 115,318 1977/78 25,330 19,808 40,637 19,844 74,574 1978/79 27,673 25,504 48,614 35,733 99,093 1/ Source SODEVA Journies d'Etudes (includes fertilizer used by SODEVA). 2/ Compiled from various sources, but mostly from DGPA figures. 3/ Rmrnp - VMfR- 1q7R/79 fiaiirp frnm nGPA. The total fertilizer distributed reflects the quantities on SENEGAL SECOND AGRICULTURAL CREDIT PROJECT 1/ Estimates of Stock of Farm Equipment in Senegal with and without Project ('000 Units) 1971/72 1972/73 1973/74 1974/75 1975/76 Without With Without With Without With Project Project Project Project Pro ject Project Seeders 137.3 137.7 111.8 123.1 92.5 112.0 7i.9 112.7 Hoes 145.3 159.9 145.6 164.5 138.2 172.1 129.2 190.0 Lifters 25.8 31.6 28.2 33.2 27.3 37.1. 25.5 40.9 Carts 49.7 55.8 54.3 62.0 51.8 64.7 43.3 65.4 Ploughs 15.13 22.5 21.0 23.6 20.3 25.2 13.6 26.9 Source: Mission Estimates 1/ Calculated on the basis of deliveries from 1960-61 under Programme Agricole with assuIMptions that each equipment is replaced every ten years, and w.ithoumt project no equipment would have b.!en supplied. OAD - 40 - Annex 2 11VV'e AT SECONDr4 AGICIMM.UM~LUAL CAIT4.4L ROuJECT UWLETLUN KEPORT ONCAD Financial Situation The actual financial situation of ONCAD is difficult to establish since it has not been able to put out a balance-sheet from 1976/77. It is however generally accepted both by Senegal and by donors, that the ONCAD financial situation is alarming. As of the end of August 1979, ONCAD's short-term aggregate over- drafts amounted to a reported CFAF 71 billion (US$322 million). It is estimat- ed that of the CFAF 71 billion overdrafts, about CFAF 8 billion was lost during the 1977/78 marketing season and the last audit of ONCAD estimates that CFAF 10 billion in assets should be written off. This latter figure could even be higher since (a) about three-quarters of the millet purchased during the 1978/79 season is still not sold and its condition at this time is doubtful, and (b) most of the fertilizer and equipment purchased by ONCAD for 1979/80 was not distributed owing to Government hard line policy on credit defaulters. Stora2e for both the millet and the inputs. especiallv fertilizer, is not adequate and hence considerable losses could be registered on these two. By the end of August 1979 ONCAD had exceeded its authorized overdraft nielina for the vpar. of CPAF 41.7 hillion at the central hank hv abnit CFAV 1.2 billion and was operating with BNDS' own resources. The basic problems the lack of internal budgetary control in ONCAD itself. The Government has recognized the alarming state of ONCAD finances d si nfe mi-la$4 sev.. 0 o I A ea Wo rkling groupsi tryng o, stud-r,A, waysa to reorganize ONCAD. The initial indications have been the reduction in the aL. L±V A. LJ.VO UL ~~~~~J_ J0%L1Z%.jj A.LL.L~ LLU~ . IQ LL . L. t~~j . ..J QLA~ J .~iL ..L its staff to streamline activities, or measures to eliminate fraud, which has been reported, in the offi.L.U What is needed is a concise redefinition of vUN A Lu L u Lural sector with a study of how best to achieve this in terms of organization, staffing and financial management. ONCAD losses, now amounting to close to CFAF 20 billion may have to be written off. It is not certain though that either the stabilization fund or the treasury have enough money to absorb the debt. However it is imperative that for a new ONCAD to succeed some means may have to be found to absorb the deficits so as not to saddle the new organiza- tion with outstanding debts that would hamper its operations. A X'r' I Table 1 - 41 - SENEGAL SECOND AGRICULTURAL CREDIT PROJECT ONCAD Operating Accounts (million CFAF) 1973/74 1974/75 1975/76 e0it Initial Stock less Provisions 751.2 5,737.2 4,274.9 Purchases 39,407.5 55,640.4 74,086.3 Personnel Costs 1,604.4 2,326.0 2,806.1 Taxes and Duties 375.2 184.9 614.9 Outside Services 2,046.3 2,251.2 2,901.9 Transportation 2,110.6 2,881.2 5,486.4 Miscellaneous Management Cost 84.4 109.7 112.9 Financial Charges 1, /. 3,674.5 4, 626.9 Depreciation 198.0 295.4 276.2 Refund to CPSP 15,290.0 301.1 367.7 ODerating Profit 481.0 614.7 8.251.1 63,762.7 74,016.3 103,804.9 redit Final Stock 5,765.1 4,274.9 4,589.7 q49q 1 t9 4Al 71 - Q9 ,A5A Repayments and Operating 2,133.9 6.348.2 3,537.1 Cook.4 A4 ae wu- b- a if dI e - Rebates and Commissions 46.2 - Aman-----~~J%J a.ad a 1. 4 O 0 £I..GDDULy L'WWWAyba JLAUe*7 .Jguuved IVJA* Financial Receipts 705.9 33.4 ONA Transportation 56. -- Refund CPSP (for Rice) 5,301.4 - - 63,762.7 74,016.3 103,804.9 Souce OCA SENEGAL SECOND AGRICULTURAL CREDIT PROJECT Groundnut Marketinj& Conparison of Bareme and Actual Operating Cost 1973/74 1974/75 1975/76 Bareme Actual Bareme Actual Bareme Actual Placing of Funds 67 - 46 44 17 9 Wholesaler's Commission 500 977 500 950 846 216 1 Guarantee Retention - 488 - X Financial Costs 721 425 721 2,540 721 2,063 Insurance 30 44 30 62 :30 121 Premia 50 214 50 72 50 131 Transport and Handling 3,000 3,320 3,000 2,984 3,000 3,216 Taxes - - - - - 111 Loss in Transit 64 - 64 64 - Information Collection 71 50 49 28 21 Economic Control 289 427 200 519 174 366 Storage Charges 450 441 450 205 450 569 Coop Management 444 - 307 264 155 158 Shelling - 91 - 684 - 104 General Expenses 2,640 3,760 1,828 2,547 1,650 2,890 Seed Service - - -- 331 335 Total Cost/ton 8,326 9,749 7,245 11,359 7,156 10,310 ('000 tons) 450 433,5 650 677.4 1,150 1,152.4 Source: ONCAD and Mission Estimates - 43 - ANNEX 2 Table 3 SENEGAL ONCAD Balance Sheet as at October 31, 1976 (million CTAF) ( not certified) ASSETS LIABILITIES Fixed Assats 3,502.5 Equity and Long Term Long Depreciation (2,274.9) Capital and Reserves 3,589.3 St 1 LL/. ACeR D/-Ug70 .7W%G\IO Net 1276Account Brought Forward (778.7) Suspense Account (1,092.2) Other Fixed Assets (Net) 30.8 Net 1718.4 inenores4,182 Provision for lossesa exene inventorles V.Lue 0LVau KU J.nc m. A.pau.cAUU Less Provisions (627.9) Long and Medium Term Debts 53.0 Net 3,990.3 Subtotal Equity and Lon Te--. Deblts Current Assets Current Liabilities Clients and Receivables 23,967.8 Suppliers and Accounts T.oa Prn ions C 931.6) Pawhla (Mae) RA A Net 23.026.2 nth. ar'Ature Carriers 3,299.7 FCI-FMDR 113.3 Less Provisions (2,810.9) Miscellaneous 393.3 Equalization Account 1,042.1. 488.8 Bank Overdrafts 46,738.1 Other Debtors Subtotal Current Liabilities 49,l74.2, Rural People 434.2 Seed Capital 7,962.9 Programe Agricole 3,105.2 Stabilization Fund 8,544.1 Miscellaneous 2,344.7 Less Provisions (3,624.2) Net 18,766.9 Equalization Account 28.6 Cash in Hand and Bank 3,036.5 TOTAL ASSETS 50,945.6 TOTAL LIABILITIES 50,945.6 SENEGAL SECOND AGRICULTURAL CREDIT PROJECT ONCAD's Staff Total Consultant's Administrative Unit Cooperation Administration (April 1979) 1/ Total 1974 Estimate (1974) Head Office General Management - 21 21 16 16 General Inspection and Legal - 33 33 16 18 Personnel Management - 28 28 25 18 Planning and Control - 72 72 61 52 Financial Management - 73 73 58 24 and Accounting Operational Management - 50 50 - - General Services 62 64 126 102 74 Technical Services - 150 150 94 81 Marketing Services 75 75 50 32 Procurement (AMR) - 4 24 24 -__ Sub-total 62 590 652 446 315 Regional Offices Cap Vert 23 43 66 30 15 Casamance 69 302 371 347 257 Diourbel 49 280 329 359 306 Fleuve 47 136 183 98 t3 Eastern Senegal 80 246 326 167 136 Sine Saloun 78 430 508 444 367 Thies 40 194 234 206 160 Louga 49 246 295 - - Sub-total 435 1,877 2,312 1,651 1,294 Total 497 267 2.914 2097 1.609 1/ Excludes seasonal staff and includes 969 temporaries. Note: As at the end of August 1979, the total staff of ONCAD stood at: Permanent staff 2,075 Temporary staff 2 969 Seasonal staff 1,300 4,344 Source: ONCAD OprIng_Account (nillion CFAF) 1973/74 1974/75 1975/76 1976/77 1977/78 Incomue Interent and Commisslone on Operations - AgrLcultural guaranteed 156.2 147.6 207.4 283.8 214.4 non-guaranteed 921.1 2,186.9 3,791.7 3,009.3 2,283.8 - Cottage Industries 19.0 22.0 86.3 145.2 258.6 - Commercial 42.8 58.8 379.7 531.3 649.1 - Real Estate 52.5 90.5 124.0 160.3 180.9 - Small Equip. and Autos 3.3 6.6 62.3 72.8 140.6 - Industril 167.1 251.8 599.8 746.0 990.3 - Coumunal 10.9 19.9 37.2 23.7 10.9 Sub-total 1¯372.9 2,784.1 5,288.4 4,972. 4,728.6 Financial Operations 291.5 607.1 197.6 168./ 326.3 Management Operations 27.6 87.7 103.0 495.6 451.4 Loss - - - TOTAL. 1692.0 478. 9 5.589.0 5636/ 5,506.3 Expenses Office Expenlsea 16.0 27.3 34.6 47.0 55.9 Pertonnel 218.6 331.5 411.7 492./ 533.5 Taxes 147.8 294.2 484 .4 553.1 668.7 Service and Supplies 83.7 69.1 78.1 79.5 107.6 Transportk t l on 11.3 15.1 24.3 25.1 32.9 Misc. Management Expenses 12.8 15.7 29.3 39.8 48.9 FIinancial Charges 571.9 1,600.5 2,414.4 2,394.1 2,226.7 Iepreciation 15.9 37.4 46.8 56.9 43.4 l'rovisions 194.5 93.8 1,109.4 420.:s 289.9 lrofI l _ 419.5 994.3 956.0 1 526. / 1,498.8 TOTA1. 1692.0 I,478.9 5,589.0 5,636.1 5,506.3 SENEGAL BNDS Profit and Loss Accounts (million CFAF) 1973/74 1974/75 1975/76 1976/77 1977/78 Losses Operating Loss Losses oi Past FY 84.8 33.2 74.0 105.7 112.1 Write-Offs 6.4 48.4 - 79.1 0.4 Exceptional Provisions 375.3 1,013.0 858.1 996.4 946.7 EndoT,nent - 27.4 Taxes 15.2 20.9 42.0 167.8 256.5 Profilt 30.3 41.8 83.9 335.7 513.0 TOTAL 512.0 1,157.3 1,058.0 1,712.-1 1,828.7 0N Profits Operating Profits 419.5 994.3 956.0 1,526.7 1,498.8 Profits Past FY 45.6 38.4 5.0 17.4 123.7 Recoveries on Write-Offs 3.7 0.6 6.1 6.0 2.0 Profits on Sale of Assets 6.6 4.4 0.5 - - Cancellation of Provisions 25.6 119.6 89.6 158.2 204.1 Exceptional ]Profits 11.0 .8 3.9 .1 Loss .... ~ - TOTAL 512.0 .1,157.3 1,058.0 1712.2 l828.7 <D.~ - 47 - SENEGAL ANNE 3 BNDS BALANCE SHEET (in million CFAF) .kss1973/74 i1974f-17 i 1q7ý/7A, 1 Q1971 !=! ASSETS Tangible. Assets (Net) 194,8 253,- ZO2,3 170,4 153,5 Other Fixed Assets: - Deposit and collateral 365,8 365,7 365,7 416,- 416,- - Investments 642,- 852,7 9&9,3 1 290,8 1 507 - - National Loan 12,6 12,6 12,5 12,6 12,5 1 Use of French Loan 2 544,1 2 379,3 2 203,2 2 09,4 1 395,2 - Miscellaneous - 5,7 ToaFixed Assets 1 3 6 3 3 919,2 4 j1D,c Lcans: -Medium Term 5 167,9 7 8351,5 10 515,3 17 49 2 - Short Term o o 1R o ,0 'R - Doubtful 723,9 1 977,9 !Total Loans 20 462,- 142 038,3 60 843.5 47 666.9 58 807.4 State Operations 5 094,9 7 030,6 12 509,4 16 462,- 17 436,9 Accounts Receivable 851,2 795.3 796.3 817,2 306,2 Cash 706,2 1 263,6 3 103,7 3 227,6 3 926,2 ISuspense Account 1 613,9 805,6 673 817,' 1 131,1 Losses- - TOTAL ASSETS .................... 32 487,5 55 796,7 al 699,- 72 910,3 86 117,3 - 48 - ANNEX 3 SENEGAL Table 3 b BNDS BALANCE SHEET (in million FCFAF) 1973/74 1974/75 1975/76 1976/77 19 77/78 LIABILITIES 1973/74 Capital .1. 70, - 1.750, 1.750,- 1.750,- 1.750, Tied up Capital 700,- 650,- 650,- 650,- 650,- ubidies.C 29, 3 295, 295 ,3 2975,3 295,3 Reserves 77,3 78,8 80,9 99- 169,5 Rough Of.orwardA 10,1 39 78,7 150 , 435,3 Provisions 1.428,1 2.391,6 4.348,7 5.800,4 6.976,i Guarane ,1 20Fun 41,1 56,6 Proprietor-sh n 4.224,3 5.225,5 7*234,6 8.785,8 10.333,4 Guarantee Deposits -331,8 402,4 411 9 418.- 430,8 Loans (NT & Ltj 3.030,6 3.700,5 3.291,- 4.283,4 10.044,6 Debenture Loans 2.544,1 2.379,3 1 2.203,2 2.015,1 1.895,2 Medium and Long Term Debts 5.906o5 6.482,7 5.906,1 6.716,5 V 1 0 Permanent Liabilities 10.1308, 11.707,7 13.140,7 15.502,3 22.704,0 Short-term Liabilities 4.940,3 27.750,6 41.008,8 24.961,1 26.754,1 Deposits- 4.,3 I6. .6U7,9 8.409I3 10.228,1 12.144,5 Government Advances 4.915,6 6.624,- 11.499,- 12.024,4 14.073,4 Accounts Pa-able 5.122,6 3, .9,8 I.AM 50, Miscellaneous Creditors 0,7 122,- 59,8 97,- 135,8 Suspense Accounts 2.602,6 1.429,6 4.012,1 5z. 15 C. I,A 5.946 1 Banks 16,3 1.087,- 805,6 3.149,3 3.286,6 Current Liabilities 22.326,4 44.047,2 68.474,4 57.072,4 62.900,8 Profit 30,3 41.8 83,9 335.6 513,0 Total Liabilities....... 32.487,5 55.796,7 81,699,- 72.910,3 56.117,3 - 49 - Annex 4 Page 1 SENEGAL SECOND AGRICULTURAL CREDIT PROJECT Completion Report Economic Rate of Return The Monitoring and Evaluation studies financed under the prniect did not provide data on the actual areas or production due to the project. It did however estimate the effects of aericultural inputs on yields, and based on these we have estimated the incremental production due to the project. We have divided the country into two areas: NORTH and SOUTH, for niivnnoa nf aQti mtin h n thp imnnt nf the nrniert- Rine- the innnt-oiutntlt relationships were different for the two zones. The NORTH comprises the reaonna nf Can Vart. b nivrhkl Fleuve T.nno7 and Thip.q and the SOUTH- Sine Saloum, Eastern Senegal and Casamance. Equipment Based on the annual supply of various implements under the Govern- ments rogram Ag r4cole westi ae t-he stck-i of eac'h iTnlPmpnt in, 1Q9/71- Assuming then that the useful life of each equipment is 10 years and that nout thOU Ie pro jec no L-l [--p 1-n- a--ljLor[l I-tw a 1(717 and 1975/76, the stock for each year with and without project was estimated usinIg Lte forula:id S.x - ST (x-1) -Sup.x -Sup fx-10' WLLere oLat -x L Stoc in ea xan Sup .x - Supply in year x. Thus tChe without project situation is where Sup.x is set at zero for the years 1973/74-1975/76, Annex 4 Table 1 gives the resulting stocks for Hoes, seeders, lifters, carts and ploughs for (PY 1 to PY 12) (1973/74 - 1984/85) for the North and the South. From 197/08 (PYa 7 - PYl 12 it wa assumed that- 4the supply.t of new. eqi-kmot would be equal to the average of the last 6 years. Area Cultivated Since there were no estimates of actual area cultivated, it was assumed that the area cultivated by each implement would be equaL tU te norms of ISRA. Using these norms, the total area cultivated by hoes and seeders supplied were estimated. For example, each Jlght hee wdb :Zum; t cultivate 3.0 ha and heavy hoe 3.5 ha. Annex 4 tables 2 and 3 give the - 50 - Annex 4 corresponding areas for hoes and seeders in the North and South. The cropping the groundnut basin. Incremental Yields From the SODEVA monitoring studies of 1975/76 and 1976/77, input- output relationships were cacuaeu ror mus. u ne 1_1mp-m eIther the relationships did not exist or calculations were not made. The significant relationships were only for hoes in 1976/77 and seeders in 1975/76. Our incremental yields are therefore based on these two implements plus an aggregate incremental yield attributable to all the other inputs. Following the SODEVA studies, for incremental yield from noes, I/ it was assumed that a 1% increase in the rate of hoe use would lead to a 3.57 kg/ha in yields in the South and a 1.72 kg/ha increase in the North, for groundnuts, and 1.88 kg/ha increase in millet yields in the South and 1.33 kg/ha in the North. For seeders a negative relationship was shown between yields and seeders/ha of groundnuts. It was assumed that a 1% increase in the ratio of seeders per groundnut area would lead to an incremental yield of 1.53 kg/ha in the South and 1.46 kg/ha in the North. Table lb and Table 2 of Annex 4 give the details of the calculation. For the other inputs it was assumed that they enabled farmers to semi-intensify 40 percent of the area cultivated by the hoes supplied under the project. Very conservative incremental yield assumptions were made about the area semi-intensified 1/ as follows: North: 80 kg/ha for groundnuts and 120 kg/ha for millet. South: 160 kg/ha for groundnuts and 130 kg/ha for millet. These low incremental yield estimates were made because fertilizer use on semi-intensified areas is not much different from non-intensified areas and hence not much yield increase can be attributed to more fertilizer use as assumed at appraisal. Production Total incremental production was calculated from the incremental yield figures and area cultivated figures (Tables 2 and 3) and for the semi- intensified by taking 40% of the area cultivated by hoes. 1/ The figures were those obtained by the Monitoring and Evaluation unit of SODEVA in the South and North aroundnut basin. 2/ Semi-intensified fields are defined here as those on which heavier ox-drawn equipment were used. Annex 4 Page 3 Prices~ TnfPrnation1 prices at f.o.b. or c.i.f. levels including freight and insurance were used to arrive at the economic farmgate prices. The prices -ere-0 theT c"7rPei intn 1Q79 termq iuina the Bank's index of international inflation (with 1979) = 100 Tables. Details are in Tables 6 - 8 of the A nnex. it was assumed that the area cultivated from the hoes and seeders supplied would remain constant between years 3 to 10, equivalent to the total number of tLie implement multiplied by the TSRA normse The incrpmental yield per ha on semi-intensified fields was also assumed to be 12 years compared with. 13 years at appra 4sa!6 This assumpt-Jon nalres intn account that the useful life of an implement is 10 years including the year of purchase. Farm Production Costs Incremental farm production costs are based on the assumptions that; ki) the-----------------4o- equipment i 10 oV -f t-he eonomic price (mairr'hasp price + subsidy) from years 2-10 and 5% in year 1, the year of purchase; (ii) only a marginal increase was made in the use of fertilizer becuase of the project, equivalent to 15 kg/ha on the semi-intensified fields and valued at CFAF 67/kg, the ex-factory pri plus trnptbation for 1970 4i-Ad of the, subsidized price of CFAF 25/kg. Extra family labor requirements on the semi-intensified fields was assumeu to De 1u hours per ha (bau Xu studies) and the wage rate put at CFAF 150/hour. Implements Costs The economic cost of implements and draft animals distributed under the project was taken to be the total cost paid by the farmers plus the subsidy for each year paid by Government including transportation. Extension Costs and Cooperative Service Since it is estimated that about 30% of all farmers benefitted from the project, 30% of SODEVA operating costs, outside of specific projects (like the Sine Saloum project), was imputed to the project. For cooperative services 30% of the bareme paid ONCAD for cooperative development has been imputed to the project. Other Costs Actual costs for studies, equipment and buildings and technical assistance have been included in the project cost estimates. - 52 - Annex 4 Page 4 Project Cost in 1979 Terms Using the Bank's index of international inflation all the costs were converted to 1979 terms for purposes of the economic analysis. Rate of Return Under the above assumptions, the estimated rate of return of the projects at completion is 29.7%. If, however appraisal forecast prices for outputs and inputs are valued in 1973 terms the rate of return falls to 18% which is slightly lower than that estimated at appraisal. This shows the importance of price effects in the estimated completion rate of return. SENEGAL SE(X)ND AGRICULTURAL CREDIT PROJECT COMPLETIC REPORT Stockof EqMent (1000 units) with and without Project Pyl PY2 PY3 PY4 PY5 PY6 PY7 Pr8 PY9 PILO Pyll FT12 HOES NORTH With Project 72.9 7n.6 77.6 78.2 77.7 73.9 71.2 75.6 75.8 76.1 75.9 .78.4 ithout Project 65.1 60.7 56.0 56.6 56.1 52.3 49.6 54.0 54.2 54.5 62.1 69.7 Percent Increase 12.0 21.25 38.6 38.2 38.5 41.3 43.5 4o.o 39.8 39.6 22.2 112.5 SOUTH With Project 91.6 98.5 112.4 118.3 114.7 116.8 125.1 136.7 142.9 143.5 147.1 151.9 Without Project 80.5 77.5 73.2 79.1 75.5 77.6 85.9 97.5 103.7 104.3 119.0 133.7 Percent Increase 13.8 27.1 53.5 49.6 51.9 50.5 45.6 40.2 37.8 37.6 25.6 13.6 SEEERS NORTH W91h Project 47.7 40.6 39.7 47.9 49.6 49.1 50.4 55.2 58.3 60.1 61.4 65.1 Without Project 43.0 33.6 27.2 35.4 37.1 36.6 37.9 42.7 415.8 47.6 52.' 58.' .53 Percent Increase 10.9 20.8 45.95 35.3 33.7 34.15 33.0 29.3 27.3 26.3 17.4 11.7 SOUTH . . With Project 75.4 71.4 73.1 79.0 78.5 81.4 89.0 98.0 102.3 104.7 108.6 113.2 ithout Project 68.8 58.9 48.7 54.6 54.1 57.0 64.6 73.6 77.9 80.3 90.8 101.3 Percent Increase 9.6 21.2 50.1 44.7 45.1 42.8 37.8 33.15 31.3 30.4 19.6 11.7 LInERS SENEGAL %ith Project 4.9 38.8 42.6 44.9 47.4 51.4 56.2 62.3 67.4 67.0 68.9 71.0 Without Project 29.9 29.0 27.2 29.5 32.0 36.0 40.8 46.9 52.0 51.6 58.5 65.k Percent CARTS With Project 62.0 64.7 6514 73.8 77.1 88.4 93.1 100.0 102.9 105.0 107.8 112.8 Without Project 54.3 51.8 h8.3 56.7 60.0 71.3 76.0 82.9 85.8 87.9 98.4 108.9 PLOUGHS SENEGAL 4. Ith Proiect 20.9 23.0 25.5 29.2 30.2 72.7 34.9 37.4 38.2 38.9 4:.4 41.8 ' ithout Project 18.2 17.5 15.8 19.5 20.5 23.0 25.2 27.7 25.8 29.2 33.4 57.6 1/ See text for method of estietion SENEGAL SECOND AGRICULTURAL CREDIT PROJECT Completion Report Seedera 1973/74 1974/75 1975/7~ 1976/7f 1977/78 1978/79 19790 1980/81 1981/82 1982/83 1983/84 1948 GROUNDNUT AREA ('000 ha) 1/ North 412.2 458.8 550.4 576.2 444.0 485.3 434.5 480.2 ~~ 480.2 480.2 4,80.2 480.2 1/ South 614.0 593. 3 732.2 770.3 669.0 671.6 601.4 664.5 - 664.5 664.5 664.5 664.5 AREA/SEEDER - ha I North With project 8.64 11. 13.86 12.03 8.95 9.88 8.62 8.7 8.24 7.99 7.82 7.38 Without project 9.6 13.65 20.2 16.28 11.97 13.26 11.46 11.25 10.48 10.09 9.18 8.24 Decrease 10.0 17.2 31.4 26.1 25.2 25.5 24.8 22.7 21.4 20.8 14.8 10.4 South With project 8.14 8.31 10.c2 9.75 8.52 8.25 6.76 6.78 6.5 6.35 6.12 5.87 With project 8.92 10.CT 15.03 14.11 12.7 11.78 9.31 9.03 8.53 8.28 7.32 6.56 % Decrease 8.74 17.5 33.3 30.9 31.1 30.0 27.4 24.9 23.8 23.3 16.4 10.5 1/ Assuned equivalent to the average of the last seven years from 1980/81 - 1984/85. -> 0 X ANNEX 4 Table 2 -55 - SECONDLj AA CREDIT~ PRJECT - jr=LE.LION REPORT (Area Cultivated by Roes Supplied (ha) ±] PYl PY2 PY3-10 PYll PY12 Groundnuts 13,750 22,890 38,360 24, 610 15,480 Millet/Sorghum 9,690 16,130 27,030 17,340 10,910 a50 vL0 100 70 Rice - - - 4 7 Sub 66,250 42,500 26,730 SOUTH Groundnuts 22,840 43,890 81,090 58,240 37,200 Y.illet/Sorghurn 15,520 29,830 55,120 29,590 25,280 Maize 270 530 970 690 440 Rice 12n01 22n 1 300 n9 Other Crops 160 300 550 400 250 Subtotal 38,910 74,770 138,140 99,220 63,360 TOTAL 62,660 114,300 204,390 141,720 90,090 1/ Derived from hoes supplied under project and norma of ISRA. cropping pattern according to those provided by SODEVA for the North and South groundnut basin. Alr"IV A pm&41&wL F - 56 Table 3 SENEGAL SECOND AGRICULTURAL CREDIT PROJECT - COMPLETION REPORT Area Cultivated by Seeders Supplied COOO ha) PY1 PY2 PY3 - 10 FY11 PY12 RTH 3roundnuts 10.86 16.41 29.22 18.13 12.68 Millet/Sorghum 7.605 1.620.6 12.95 9.05 Maize - ice 0.40 0.5 v.O 0 U.V4 0.2 18.99 28.47 50.5 31.5 22.02 -,;TH ;roundnuts 17.31 33.02 62.21 44.89 29.19 Aillet/Sorghum 11.77 22.66 42.98 30.91 20.15 Aaize - - - - - Rice 0.52 0.57 0.79 0.60 0.39 29.5 56.25 105.98 76.4 49.73 SECOND AGRICULTURAL CREDIT PROJECT Incremental Yields From ui ient u kg/ha PYI PY2 PY3 PY4 PY5 PY6 PY? PY8 PY9 PY10 PYul PY12 HOES NORTH Groundnuts 21 37 66 66 66 71 75 69 68 68 38 22 Hillet 16 28 51 51 51 55 58 53 53 53 30 17 EOIUTH Groundnuts 49 97 191 177 185 180 163 144 135 134 84 49 Millet/Sorghum 26 51 100 93 98 95 86 76 71 71 44 26 SEEDIRS NORTH Groundnuts 15 26 48 4o '9 i9 38 35 33 32 16 SOUTH Groundnuts 13 25 4q 45 45 44 40 36 35 34 24 15 NOTE: Incremental yields on hoes and seeders was calculated using the increase in equipment rate as a result of the project. 1/ 1% increase in the hoe rate gives 3.57 kg/ha more of groundnuts in the South and 1.72 kg/ha in the North. It also gives L.33 kg/ha more of millet in the North and 1.88 kg/ha in the South. z-- SENEGAL SECOND AGRICULTURAL CREDIT PROJECT Completion Report Incremental Production Tons) PY I PY 2 PY 3 PY 4 PY5 PY 6 PY 7 PY 8 py 9 PY IO PY 11 PY 12 FROM HOES SUPPLIED North Groundnuts 289 847 2,532 2,532 2,532 2,724 2,877 2,647 2,608 2,608 95 341 Millet 155 452 1,379 1,379 1,379 1,487 1,568 1,433 1,433 1,433 520 185 South Groundnuts 1,119 4,257 15,488 14,353 15,002 14,596 13,218 11,677 10,947 10,866 4,892 1,823 Millet 404 1,521 5,512 5,126 5,402 5,236 4,740 4,189 3,914 3,914 1,742 657 FROM SEEDERS SUPPLIED North Groundnuts 163 427 1,403 1,169 1,140 1,140 1,110 1,023 964 955 417 203 South Ln Groundnuts 225 826 3.048 2,799 2,799 2 737 2,488 2,240 2,177 2,115 1,493 933 0D FROM SEMI -INTENSIFICATION + FERTILIZER North Groundnuts 440 732 1,228 1,228 1,228 1,228 1.228 1,228 1,228 1,228 788 495 Millet 465 774 1,297 1,297 1,2 1,297 1,297 1,297 1,297 1,297 832 524 Maize 25 40 65 65 65 65 65 65 65 65 50 5 South Groundnuts 1.462 2,809 5,190 5,190 5,190 !5.190 5,190 5,190 5,1.90 5,190 3,727 2,380 Millet 807 1.557 2,866 2,866 2,866 2.866 2,866 2,866 2,866 2,866 2,059 1,315 Maize 155 265 485 485 485 485 485 485 485 485 345 220 TOTAL INCREMENTAL PRODUCTION FROM PROJECT Groundnuts 3,698 9.898 28,889 27,271 27,891 27,615 26,111 24,005 23,114 22,942 12.252 6,175 Millet 13851 4.304 11,054 10.668 10,944 10,886 10,471 9,785 9,510 9,510 5,153 2.681 Maize 160 305 550 550 550 550 550 550 550 550 395 255 en pI'l - 59 - ANNEX 4 SENEGAL Table 6 OcVArT ArDTrTMTAT rDVT1r 3DnT ECT COMPLETION REPORT Economic Analysis Economic Farmgate Price - Groundnuts 1. Breakdown Appraisal Completion 1973 1979 UST1 $ /ton". Reference Price CIF European Ports 309.8 564.5 Equivalent Unshelled Basis (70%) 216.9 395.2 Ocean Freight / Insurance 23.5 42.6 Exnr Price FOB Dakar UnAhelled 193.4 ____ ... CFAF /ton ... FOB Dakar 44,015 77,572 Port Handline Charges 750 6-500 =. Transport / Handling Costs 4,330 10,455 4 Primary Marketing Cost 1,120 3,895 Economic Farmgate Price 37,815 56,722 2. Trend Completion Farmgate Prices Appraisal and1 Inlaio Economic Farmgate Price CIF zl' Ireight 6n/ FOB Ludexin Faragate Price used (1973 Terms) 1979 Terms CFAF / kg ......... US $ / ton ........ 1979 = 100 CFAF / kg 1973 37.82 275.1 23.5 251.6 49.7 90.52 1975 33.45 316.4 32.3 284.1 71.7 66.32 1976 30.68 296.8 33.9 262.9 73.0 58.38 1977 28.01 385.8 36.2 349.6 80.0 75.29 1978 25.29 432.3 38.0 394.3 92.1 73.34 1979 22.92 395.2 40.5 339.6 100.0 56.72 1980 + 20.90 410.9 41.1 369.8 106.0 55.90 1981 20.90 441.7 44.2 397.5 112.4 56.95 1982 + 20.90 457.8 45.8 412.0 119.4 55.06 1/ Appraisal report cites CIF price in 1973 as E125 / long ton equivalent to CFAF 72,585 / ton. At the exchange rate of CFAF 230.6 = US $ 1 used in the report, the 1973 price would be US $ 314.78 / long ton, i.e. US $ 309.8 / metric ton. The actual 1973 C.I.F. price (European ports) was US $ 393 / ton. 2/ Annarent exchanae rate used is US I 1 =.CVAF 227.6.at anoraisal. Comoletion rate is CFAF 220 -= $ 1. 3/ Handling cost for 1979 was CFAF 3,500 / ton for grains with bagging at C-AF 3,000 / ton. The same rate is used. 4/ Derived from the margin ("bareme") paid ONCAD for groundnut marketing. 5/ CIF is on unshelled basis and actual figures from 1973 - 1974, for 1980 and after estimates. 6/ The actual 1978 rate was about US $ 38.0 / ton from Europe to West Africa, the others deriVed using thne Bank's ocean freight indices From 1980l the rates are assumed to be 10% of CIF. 7/ From commodity price forecast put out by the Bank's Economic Analysis and Projections Department, May 1979. - 60 - ANNRT A SENEGAL Table 7 SECOND AGRICULTURAL CREDIT PROJECT COMPLETION REPORT Economic Analysis Economic Farmgate Price - Millet / Sorghum 1. Breakdown Appraisal Completion 1973 1979 Reference Price FOB Gulf 48.0 111.0 Ocean Freight / Insurance 23.5 40.5 CIF Dakar 71.5 150.5 ... CA I.i~AU.L - 2/ LO JU33,±.LU Port Handling 750 6,500 Less Trnprtto I )Inntc ut 3/ Economic Farmgate Price 17,780 A~' A t 2. Trend Completion Farmgate Prices Ann- 4 .l Farmgate Price 4/ Freight and Inflation Farmgate Price FOB - c CIF CFAF / kg 1973 17.78 930 29.5 11 4597 53.07 1974 18.11 121.0 28.5 149.5 62.2 54.38 1975 18.46 111.9 31.1 143.0 71-7 4S-IR 1976 18.72 105.2 32.7 137.9 73.0 43.06 1977 18.85 88.4 35.0 123.4 80.0 35.44 1978 18.96 93.8 36.7 130.5 92.1 32.67 1979 19.10 108.1 40.5 148.6 100.0 34.19 1980 19.19 140.5 40.5 181.0 106.0 39.07 1981 19.19 154.5 40.5 195.0 112.4 39.67 1982 19.18 166.9 40.5 207.4 119.4 39.71 1/ Exchange rate used at appraisal was US $ 1 = CFAF 255.0; completion rate used was US $ 1 = CFAF 220. 2/ 1979 rates were CFAF 3,500 / ton for grains plus CFAF '30007ton fo'r bagging. 3/ Assumed to be 15% of CIF Dakar value. 4/ Actual prices through 1979, estimates from 1980. 5/ Actual 1977 rate was US $ 35.00 / ton, others derived using index of freight rates. -61- ANNEX 4 Table 8 SENEGAL SECOND AGRICULTURAL CREDIT PROJECT COMPLETION REPORT Economic Analysis Economic Farmgate Price - Maize 1. Breakdown 197 3 1/ 1979 TT.q -C ' o Reference Price20. Ocean Freight / Insurance 40.5 CIF Dakar 160.5 CFAF / ton CIF Dakar 35,310 Port Handling 6,500 Less Transport / Distribution 51000 36,810 2. Trend Completion Farmgate Prices /FOB 2 Freight / CIF Inflation Farmgate Price Insurance - Index 1979 Terms 1973 98.0 23.5 121.0 49.7 55.060 1974 132.0 28.5 160.5 62.2 58.27 1976 112.4 32.7 145.1 73.0 45.23 1978 100.7 36.7 137.4 92.1 34.32 1980 159.0 40.5 199.5 106.0 42.91 ,i ^0 - 1/^ %I I M III 1 1)' /. LnI no 1982 + 190.7 40.5 231.2 119.4 44.10 Ei Lae farmgate price was not cstimated at appraial. 2/ Actual rates through 1978 estimates thereafter. Acteu. rate fOr 1977 , others derived u indce o Jf&.L6LL6 rates. SENEGAL SECOND AGRICULTURAL CREDIT PROJECT Lcmpetion Report Economic Costs and Benefits 1973/74 1974/75 1975/76 1976/77 1977/78 1978/79 1979/80 1980/8L 1981/82 1982/83 1963/84 1984/85 1985/86 Project Costs Implement Purchases 584.5 537.4 858.4 - - - Subsidy on Implements 192.9 123.5 465.1 Equipment and Buildings (SODEVA and IIAT) - 2.1 19.2 33.9 9.7 51.2 Technical Assistance and studies (ONCAD and SODEVA) 13.0 69.9 42.6 63.1 68.0 82.4 Extension Costs (SODEVA) 78.0 83 3 121.4 137.3 135.4 150.0 177.6 177.6 177.6 177.6 127.6 85.1 ONCAD Cooperative Services 59.0 58.0 .6 42.0 42.0 42.0 42.0 42.0 42.0 42.0 30.2 20.1 Sub-Total 927.4 874.2 1.560.3 276.3 255.1 325.6 219.6 219.6 219.6 219.6 157.8 105.2 Inflation Index 49J 62 2 71- 73.0 80.0 92.1 100.0 Sub-Total in 1979 terms 1,866.0 1,405.5 2.176.2 578.5 318.9 353.5 219.6 219.6 219.6 219.6 157.8 105.2 On-Farm Costs 1979 terms Implement Maintenance 78.2 209.5 355.0 447.3 447.3 447.3 447.3 447.3 447.3 447.3 290.8 184.6 Fertilizer 24.8 45.3 81.0 81.0 81.0 81.0 61.0 81.0 81.0 81.0 56.2 35.7 Labor 37.0 67.6 121.0 121.0 121.0 121.0 121.0 121.0 121.0 121.0 83.9 53.5 Total Project Cost, 1979 terms 2,006.0 1.727.9 2,733 2 1,Ce7.8 968.2 1,002.8 868.9 868.9 868.9 868.9 588.7 378.8 Total Project Cost, 1973 terms 997.5 859.2 1,359.1 511.1 481.5 498.7 432.1 432.1 432.1 432.1 292.7 188.4 Project Benefits In 1979 terms 440.7 1.429.4 2,443.8 2,076.4 2,508.3 2,399.8 1,858.7 1,747.8 1,717.2 1,665.1 910.9 456.2 At Appraisal prices 1975 terms 219.0 710.4 1,214.6 1,0132.0 1,246.7 1,192.7 923.8 868.7 366.5 822.6 452.7 226.7 Original Appraisal Estimate 242.9 582.6 785.4 1,029.8 1,105.4 1,179.1 1,154.2 1,081.6 1,081.6 1.081.6 1.081.6 721.1 560.5 Net Benefits In 179 terms (1,565.i) (298.5) (289.4) 1,048.6 1,540.1 1,397.0 989.8 878.9 848.3 796.2 322.2 77.4 At Appraisal Prices, 1975 terms (778.5) (148.8) (144.5) 520.9 7651.2 694.0 491.7 436.6 421.4 390.5 160 38.3 Original Appraisal Estimate (952.1) (675.6) (457.9) 630.0 709.3 786.5 785.6 715.7 715.7 715.7 715.7 477.2 238.5 Rate of Return Completion 30% Appraisal 22% - 63 - Annex 5 Page 1 STUDTF.S Five studies were envisaged to be done under the project; N)~ fPq.qhJ1i-v _Qtijdv nf tha 1Rna Rn1mim projec t; (441see ntrg study (iii) a rhizobium research to be done by IRAT, now ISRA; (iv) a monitoring and Pvn1iintinn -iidv nf tha imnnntt nf the nrroram aicon-le; and f-l a ntuy On the response of groundnuts and millet to fertilizer in various ecological zones. The seed storage study, entrusted to ONCAD by Government, and the rhizobium research though done by TOD_Ad no y4-eld any- usfu -rsutshc can be applied immediately. The other three studies-all under SODEVA were LtJLtCV aLit. LLOV UOCLUJ. Li.LtJ.UfLtUJ.I cOycL_td .Ly ULL LILC groundnut basin. Perhaps the most successful to the studies has been the ACULAA41 14 zerreeach which uas upeuCU u1uc U L--C -L - UL the present fertilizer recommendations. The rest of this annex summarizes THE FERTILIZER STUDY The study was started in early 1975 when SODEVA signed a contract with the Tennessee Valley Authority (TVA) for a consultant to monitor the study. Field trials were startea in i DuL aid not yield any usexu results since the program was implemented late, and also SODEVA staff had yet to be trained. Effectively therefore the study started in 1976/77 when 32 trials, each for groundnuts and millet, spread throughout the groundnut basin, were undertaken. Farmers for the first time showed a keen interest in trials being made on their fields. In early 1977 the consultancy contract was transferred from TVA to the International Fertilizer Development Center (IDC) which is an out- growth from TVA's international division. IFDC supervised the trials for the 1977/78 season. The transfer however did not affect, in any adverse way, either the quality or timing of the study. The analysis which follows is of the preliminary reports and the results and recommendation are not yet definitive. Final conclusions can therefore not be drawn yet. Page 2 RTTT.TS A THW TTTnY IFDC submitted a preliminary draft of its findings to SODEVA, the Gveyrment and DA .or comments 4 TuL -17. "Lek LrecelVCg Comments, It submitted a second draft in November 1979 and discussion on the draft were SSEPC I/ and Government representatives. The final draft is expected at the ,.,,... ~ ~ ~ ~ ~ aton are~ not... expecte toI. differQLUI~~LLLLUULL much from those made in the second draft. Findings The main findings have been discussed at length in the Bank and LnLIateL unI±y a vLL nummary is gJVULn netLe. Ine main conclusion of the study is that the present fertilizer recommendations' at least for ground- nuts, are too nign in some nutrient application rates and that they are not feasible for all farmers to optimize returns to fertilizer. The findings Lnuicate also that the response of Millet and groundnuts to various nutrients is at variance with the present recommendations since, (a) millet responds well to N and P but not to K and (b) groundnut responds well to P, inconsis- tently to N and only marginally to K. Present fertilizer recommendations however contain relatively heavy doses of all those elements. The millet response functions were in general agreement with previous research work but grounduut response to N and K disagrees with past research. Also the response of the two crops to the nutrients varied from one ecological zone to the other. IFDC then made an analysis of the economics of fertilizer use and proposed "optimum" nutrient application rates for the four regions of the groundnut basin. Tie "optimum" rates were calculated using; (1) response functions derived from the mean values of variables and observed rainfall (2) fertilizer prices both subsidized and unsubsidized and producer prices, and (b) the condition that the optimum application rate is where the marginal product from the use of a nutrient is two times its marginal cost (three times marginal cost for millet in Louga). After discussions with the Bank and interested parties in Senegal, IFDC is recalculating the optimum rates and would propose new optimum rates where marginal product equals costs but also show the rates at which marginal product is twice marginal cost, it would also take account of farmer's risk by using the lower quartile of rainfall observed over the past 10 years in calculating the nutrient rates. The findings of the fertilizer study have shown that savings can be made both by the country and the individual farmer by changing the present fertilizer formulae. It does not tackle questions concerning fertilizer subsidy. Though it was not expected to address the subsidy question, the 1/ SSEPC: Societe Senegalaise des Engrais et Produits Chimiques, is a wholesaler of fertilizer in Seneeal. Annex 5 -65 - Page 3 report's recommendations will need to be discussed taking into account Government's subsidy policies. The final report is to be discussed in early April 1980 in Senegal together with the general question of output prices and input subsidies. It is hoped that the discussion would lead to an adoption of a realistic national fertilizer policy and resolve the subsidy question. RESPONSE TO THE STUDY The general response of Government and SODEVA to the IFDC study was favorable. Government is at the moment looking for ways of improving fertilizer use and thereby reducing the cost to both the state and the farmer. Thus the recommendations of the study proposing an optimization of nutrient application meets Government criteria for an economic allocation of resources - where nutrients are used only when they give reasonable returns to the farmer and the country in terms of additional production. Government would however be reluctant to regionalize fertilizer prices based on nutrient content. Thus even with the new formulae some farmers, especially those in the north, would be Daving more in terms of the real benefits in terms of yields per nutrient that they get from fertilizer application. When the final formulae are decided upon, Government may favor an increase in the groundnut rebate to farmers receiving lower returns per applied fertilizer which would mainly be farmers in the North. The initial reaction of SODEVA was favorable. However both SODEVA and DGPA 1/ would prefer to test the recommended formulae in farmers fiP1P for at least a year. The field test, they argue. would make it easier for Government to propose a switch from the present recommendations to tbp npw np qRinr some onnosition to the new formula has alreadv started in ISRA, the main agricultural research institute. IFDC also recognises th-at, ho- reOmmenaations may need to be modified as data are examined in greater detail especially for site parameters and other variables. QonVA alsn eaets durina the tria1s to seek ways to define acceptable means of localized placement of fertilizer, which should lead to savings in fertilizer use and reduce weed growth connared with the present system of fertilizer broadcasting. ISRA has critized the new IFDC recommendations on the grounds that two years is too short a npriod to make final recommendations and argues for the continuation of the present formulae. ISRA's main argument s bsed on te fact that ach nutrient used un by the plant should be replaced. However, as the IFDC report has proposed, continuous monitoring OC -,-1 Ariency 4n nu*,r4--*. nc-an4n1v nntn,qii1m_ Rhould ensure that LL U L.L.L~* ~ -*Ie -- -- j ---, _ soil deterioration does not occur as feared by ISRA. Such monitoring plus some of e n et-44r. reosech "rnnncn1Q made by TFDC would be taken up in the framework of the Production Systems Research component of the recenly apraL~ised A- - +- Ine M~UC report alopooe friieJitibtoUo1rv Discussions with SSEPC indicate that such a study has been done and SSEPC, which is the main supplier o ifetlier to AMPA, would be id getting more involved in fertilizer distribution even down to the cooperative 1/ Direction Generale de la Production Agricole. Annex 5 Page 4 level. SSEPC would also want to get involved in fertilizer extension work, though the details are yet to be worked out. It would however be useful if SSEPC and SIES, the fertilizer factory, could be involved, at least, with training of extension agents with respect to the proper use of fertilizer. Other Recommendations The draft report recommends further research as a follow-up to the study. These would involve: (a) Confirmation and explanation of the strong indication (in the trials) that the response of millet to U and P declines with increasing rainfall; (b) A test on supplying Phosphate through the use of phosphate rock, partially acidulated phosphate rock, and soluble superphosphate instead of through compounds; (c) A study on the need of sulfur in fertilizers for millet and sorghum; (d) Monitoring the occurence of Potassium deficiencies to be expected from more intensive production of millet; and (e) Examination of the possible benefits from localized placement. As mentioned above, the recently appraised Agricultural research Droiect would take up most of thesc studies and SODEVA would also he3p in promoting localized placement and monitoring potassium deficiencies. -7- Annex 6 Page 1 SENEGAL SECOND AGRICULTURAL CREDIT PROJECT COMPLETION REPORT Cooperatives The Government of Senegal set up cooperatives in 1960 in an attempt to have more control over groundnut production and marketinR. ReRional farmers' organizations, the Centres Regionaux d'Assistance au Developpement (CRAD) were responsible for providing assistance to these cooperatives when the movement was first set up. In 1967/68 responsibility for cooperative assistance and development was transferred to ONCAD, created in 1966. The number of cooperatives has kept on increasing and by mid 1979, there were 1.870 cooperatives in Senegal with 1.072 in the aroundnut basin. ONCAD has remained the main vehicle for cooperative development, since 1967/68, charged with: (i) sunervision of cooneratives. maintenance of their accounts and training of their leaders; (ii) procurement and delivery of farm inputs for the cooperatives; (iii) marketing of groundnuts and other crops purchased from CoonorntJ7AQ-nt ~(fv, - i4a n ^f a nannnl nrA mg-,i fg-rm rrltil-r n ,rn-irc and recovery from them. ONCAD activities in all these spheres have been judged to be unsatis- factory most of the tt4me.* But because of the control of ONCAD on ----er-ti4-s their development has been gagged and the most important ultimate objective set ULy Go1,vernmtic i nL es a b ih n -FU .L- LO4FLL - -CLU iaI.'t U L U.LXaLL5fllt i ta people in economic development--has not been attained. During the First Agricultural Credit Project (Cr. 140-SE), it wa xpce tha te- tec~h4ncal assistaance for t-he gnal reorgniz *ation of ONCAD, financed under the project, would lead to more independence for coop- ratiVes MU CHOUULOge LuECL. ueveAvpmote Lucoc UvJec-ves were however not achieved. Several recommendations were made by the consultants and these are LCpULLCU in L I u o LUL L fis pr1je pLUJCL (AnnxCAJ) * LUm these LcLmc dations a pilot operation was set up to merge cooperatives into larger units capaole or marketing their own prouuce anu uperating a sccu supply sceme. The experiment was not as successful as envisaged because the ONCAD control still remained. During the past few years management of cooperatives has been deteriorating. The cooperatives have just become appendages of ONCAD and most farmers do not believe that the system represents their interests. Thus, credit repayment by farmers has been decreasing (Annex 6, Table 1) and each year the debts due from cooperatives increase. What is worse, large - 68 - Annex 6 Page 2 discrepancies exist between ONCAD regional accounts, individual cooperative accounts and debts as recognized by the farmers themselves. In comprehensive village surveys being conducted by SODEVA, it is becoming clear that actual reimbursement of Rroundnuts by farmers may be much higher than declared by the weighers and ONCAD and the amount of seed declared as supplied by ONCAD may be much lower than actually distributed. It would be necessary therefore to establish a clear situation of the level of indebtedness of each cooperative and to reconcile the differences in accounts where nossible. In some cases the differences may be due only to calculation errors. If after exhaustive search. the differences cannot be exnlained. and. where fraud cannot he eRtah- lished, the debts could be absorbed using the cooperative guarantee funds at 'RMnR. The eed distrihuted hv ONrAn ham Alan hen aptprinrAtino inianality owing to the poor storage and it may be unfair to ask farmers to pay the huge Qooe delht which has Aciimiated (Anno A- Tnh1a 1). Tha r-n-ucornmPnt- ham herne an~rp nf thp I.Qrkl nf doeuolnnint- nf cooperatives and the increasing complaints of farmers. In 1979 therefore several workiny nrou- uwre et n- to redefine the -nlas of the rural develop- ment agencies and especially ONCAD and find ways to reactivate the cooperative mmno Fin-- decisi4o4ns have not yet- beeo. takeon but 4in December 1070 Government decided to remove the Department of Cooperatives from under ONCAD control.,l and. reattac 4*- 4-a +.- U4-.4-4--.r ^a D,-1.c fo.1ano,-- The staff from the department are to be selectively reassigned as needed to the development. A decision on restructuring of cooperatives has not yet been taken UUL GUVeLreL LVUL s LLL UL mAerLgie UUpthaL.VCD LU sx-Lsti L-Rm --I (minimum of 1,500 tons of marketed groundnuts). However there are demands for restructuring cUperati-VCS tlrougul LLn ULCtLiV UL mULUC. rUcLtLL 6LUUp (sections villageoises) within each cooperative unit. It should be noted that these two solutions are not mutually exclusive and larger cooperatives can be formed with the guarantee groups constituting their basic units. Several proposals have been made for improving the debt situation and involving cooperatives in rural development. These include: (1) establishing clearly the debt situation of individual cooperatives and its members; (2) creating small voluntary units of 20-30 farmers within cooperatives mutually responsible for credits and repayment; - 69 - Annex 6 Page 3 (3) reducing and eventually stopping seed credits to cooDeratives and encouraging mutual guarantee groups to keep their own seeds; (4) keeping a regional security stock of seed to meet natural calamities where the groups are unable to constitute their own seed stock; (5) asking farmers to contribute cash towards input purchases and to encouraging farmers, who can afford it, to buy their inputs with cash. This can be done by increasing the interest on credits or reducing subsidies on credit purchases; (6) putting into practice an automatic insurance system along the lines of STABEX 1/ where a given nercentage of debts for a vear (sv 2n can be cancelled for a cooperative or district if the production for that vear due to the weather iR lower than a civen nPrPnteP (ay 20%) of the average of the last three years; (7) having the restructured cooperatives multifunctional, to eventually prrvie all serPYvircs for themselvesP. (m.Qrketina producerP, scurlring ndA repaying credit, transportation, and procurement of inputs); (8) providing a mass education system for all farmers especially in (9) givin cooeraive an-uta -gaantee groups the financial means to properly execute their new function, first by releasing billion--Annex 6 Table 2) and by increasing the rebate (ristourne) democratization of each cooperative and having the ability to employ LheiL UW peimanetL LaL. Most of these suggestions are now being considered by Government and the Bank is also asking Government to introduce them in the reappraised SinC Sdlum prLUCL 1t Lfthy are intLUUULeU, CAperiene galneU woulu enable their easy application in the whole country. 1/ STABEX is the mechanism whereby the EEC pays certain African, Caribbean and Pacific countries a given amount when exports for a year fall below a certain percentage and where the crop supplies at least 7.5% of export revenue. SENEGAL SECOND AGRICULTURAL CREDIT PROJECT COMPLETION REPORT Coopertive Debt Situation (Seed and _mplements Mature Debts Seed Credits Credit R ments Due BNDS TOTAL Seed Credits BNDS TOTAL Producer Quantity Value Credits Due Due % of Value Credits Repayments Repayment Price '000 Tons Million Million Million Seed Ouantity illon Million Million Rate Groundnuts CFAIF CFAF CFAF Credits *000 Tons CFAF CFAF CFAF CFAF/kg O 1969/70 158.1 2,554.8 832.8 5.587.6 75.4 105.2 1,945.3 716.2 2,661.5 78.6 18.5 170/71 166.7 5,084.6 625.7 3,710.3 83.1 53-1 985.1 519.1 1,502.2 55.1 18.5 1971/72 128.5 2,965.6 752.3 5,715.9 79.7 123.1 2,842.9 745.5 3.588.2 96.6 25.1 1972/75 145.9 35,25.5 1,014.0 4,537.5 76.6 63.8 1,472.6 482.4 1,955.0 45.1 25.1 1975/74 148.8 4,383.5 1,304.6 5,695.1 77.1 88.5 2,611.6 944.6 5,556.2 62.5 29.5 1974/75 157.1 6,519.5 1,441.8 7,961-3 81.9 135.6 5,628.8 1,106.7 6,733.5 84.6 41.5 1975/76 180.1 7,474.3 2.138.0 9,612.3 77.8 149.8 6,216.0 1,954.9 8,15o.9 84.8 41.5 1976/77 179.6 7,455.1 2,955.2 10,410.5 71.6 150.8 5,426.5 2,483.6 7,909.9 76.0 41.5 1977/78 202.7 8.412.8 3,1.-2.8 11,575.6 72.7 31.9 1,322.1 240.1 1,562.2 13.5 411.5 1978/79 218.4 9,064.4 4,453.1 13517.5 67.1 15..7 5,629.6 2,405.1 8,034.7 59.4 41.5 SOURCES BNDS and ONCAD> ANNEX 6 SENEGAL Table 2 SEuUU AULLULTURAL CLL PRUJEC COMPLETION REPORT Cooerates - Financial Po4tion at RNc (May "A 1071 A. COOPERATIVE FUNDS B. COOPERATIVE DEBTS Under PA. Million CFAF I. SOCIAL CAPITAL I. Amount Due on PA 1978/79 2,720,661 Million CFAF I.1 Statutary 165,527 II. Outstanding from previous .1.2 FMDlR guarantee account 20,646 PAS (1974/75-1977/78) 1,190,772 III. Due from 1977/78 postponed Subtotal 186,173 debts (1st. Year) 541,668 II. INVESTMENT FUND 1,019,203 TOTAL DUE 4,453,101 III, DEVELOPMENT FUND 167,241 IV. Repayment 2,392,055 TOTAL COOPERATIVE FUNDS 1,372,616 Unpaid Balance 2.061,046 V. Rebate on Groundnut Primary Marketing (Est.) 300,000 NOTES to Table 2 1. OCALCAITA. heto a. amunt o ocal capitLal pJaid by copraie sU.Vo far is CAI 275,8J79Lh±L~L but only 186,173 was at BNDS as of April 30, 1979 and the rest transferred to FMDR for credit guarantee purposes. 2. THE INVESTMENT FUND (If)is constituted by blocking 50% of the rebate (ristourne) paid on groundnut marktin and AddA rn the lpaAl reperveA nf the -noo-tva (1 9) to ar-tea cooperative credit. When the IF reaches 5 times the social capital of a cooperatives it ceases to be obligatory and the balance is transferred to the DEVELOPMENT FUND (DF) The DF can be used to finance cooperative investments. As of May 31, 1979 various cooperatives had used CFAF 436.8 million of the DF, distributed as follows. Million CFAF Louga 28.4 Diourbel 85.4 Casamance 83.4 Senegal Oriental 21.6 434.8