Cameroon - Rural Development Projects

Document of The World Bank FOR OFFICIAL USE ONLY Report No- 8690 PROJECT PERFORMANCE AUDIT REPORT CAMEROON ZAPI INTEGRATED RURAL DEVELOPMENT PROJECT (CREDIT 776-CM) WESTERN HIGHLANDS RURAL DEVELOPMENT PROJECT (CREDIT 784-CM) NaRTHERN PROVINCE RURAL DEVELOPMENT PROJECT (CREDIT A I/LOAN 1919-CM) 10176- 4N' JUNE 1, 1990 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS ZAPI INTEGRATED RURAL DEVELOPMENT PROJECT US$1: At Initiation = CFAF 245 At Completion = CFAF 350 WESTERN HIGHLANDS RURAL DEVELOPMENT PROJECT US$1 = CFAF 245 CFAF 1 = US$0.0041 CFAF 1,000 = US$4.1 CFAF 1,000,000 = US$4,081.63 NORTHERN PROVINCE RURAL DEVELOPMENT PROJECT Fixed Parity: 50 CFAF = 1 French Franc EXCHANGE RATES Fiscal Year CFAF per Dollar 1983/84 409.5 1984/85 471.1 1985/86 386.6 1986/87 318.8 1987/88 291.7 1988/89 300 (projected) Fiscal Year: (Government and ZAPI): July 1 to June 30 Fiscal Year: (UCCAO) January 1 to December 31 Fiscal Yeer: (Northern Province) July 1 - June 30 FOR OFFCIAL USE ONLY THE WORLD BANK Washington. D.C 20433 U.SA Offce of Dwectcr-Cu~er Operatenns lvalate June 1, 1990 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Performance Audit Report on Cameroon Zapi Integrated Rural Development Project (Credit 776-CM) Western Highlands Rural Development Project (Credit 784-CM) Northern Province Rural Development Project (Credit 1078-CM/ Loan 1919-CM) Attached, for information, is a copy of a report entitled "Project Performance Audit Report on Cameroon Zapi Integrated Rural Development Project (Credit 776-CM) Western Highlands Rural Development Project (Credit 784-CM) Northern Province Rural Development Project (Credit 1078-CM/Loan 1919-C0" prepared by the Operations Evaluation Department. Attachment 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. FOR OFFICIAL USE ONLY GLOSSARY OF ABBREVIATIONS AGRRD - World Bank Rural Development Division BCU - Bottomlands Cultivation Unit (UCCAO) Caisse - Coffee Price Stabilization Board CCCE - Caisse Cntrale de Cooperation Economique (France) CFAF - Westafrican Franc Zone Currency Unit CFDT - Compagnie Franc9aise pour le Developpement des Fiber Textiles CNCE - Centre National du Commerce Exterieur CODES - Branch Committee CODEVI - Village Development Committee COGEST - Management Committee COOPMUT - Department of Cooperatives and Mutual Assistance CRF - Centre de Recherche Forestiere CTFT - Centre technique Forestier Tropical DEP - Planning Unit - Ministry of Agriculture DEP - Studies and Planning Department (MINAGRI) DPA - Provincial Delegation of Agriculture, Western Province EPL - Local Progress Enterprise ERAP - Inputs Supply Center FNFP - National Fund for Forestry and Fisheries FONADER - National Rural Credit Fund FONADER - National Fund for Rural Development FSAR - Fonds Special d'Actions Rurales FTDU - Field Trials and Demonstration Unit (UCCAO) Genie Rural - Rural World Department of Western Province IITA - International Institute for Tropical Agriculture IPD - Pan-African Institute for Development IRAF - Agricultural and Forestry Research Institute IRA-Nord - Institut de Recherches Agronomique, Maroua IRZ - Institute de Recherches Zootechniques MESRES - Ministere de 1'Enseignement Superieur, de l'Informatique et de la Recherche Scientifique MIDEVIV - Fooderop Development Agency (Ministry of Agriculture) MINAGRI - Ministry of Agriculture MINDIC - Ministere du Developpement Industriel et Commercial MINEL - Ministere de l'Elevage MINEQ - Ministere de 1'Equipement MINEP - Ministere du Plan MINEPIA - Ministere de 1'Elevage, de la Poche et des Industris Animales MINFIN - Ministere des Finances MINPAT - Ministere du Plan et de l'Amenagement du Terrioire NSMP - National Seed Multiplication Project OC - Office Cerealier OED - Operations Evaluation Department ONAREF - Office National de Regeneration des Forets ONAREST - National Scientific Research Organization ONAREST - National Research Organization ONCPB - Office National de Commercialisation des Produits de Base ONDAPS - Office National de Developpment Agricole et du Petit Betail OVP - Office Pharmaceutique Veterinaire This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents m.y not otherwise be disclosed without World Bank authorization. PCC - Project Coordinating Committee PCR - Project Completion Report PCU - Pest Contro Unit (UCCAO) PMB - Produce Marketing Board PMEU - Project Monitoring and Evaluation Unit (DPA) PPAM - Project Performance Audit Memorandum PPAR - Project Performance Audit Report PROVIV - Commercial Food Crop Marketing Organization RMWA - Regional Mission in Western Africa SEMRY - Societe d'Expansion et de Modernisation de la Riziculture a SODECAO - Cocoa Development Society SODECOTON - Societe de Developpement du Coton du Cameroun SPU - Seed Production Unit (UCCAO) UCCAO - Arabica Coffee Growers' Cooperative Union, Western Province USAID - US Agency for International Development ZAPI - Integrated Priority Action Areas PROJECT PERFORMANCE AUDIT REPORT CAMEROON ZAPI INTEGRATED RURAL DEVELOPMENT PROJECT (CREDIT 776-CM) WESTERN HIGHLANDS RURAL DEVELOPMENT PROJECT (CREDIT 784-CM) NORTHERN PROVINCE RURAL DEVELOPMENT PROJECT(CREDIT 1078-CM/LOAN 1919-CM) TABLE OF CONTENTS Page No. Preface ............................. .................. i Basic Data Sheets ............................................. il Evaluation Sumary .......................... v PROJECT PERPORMANCE AUDIT REPORT I. Background ................................................ 1 II. The Projects ........................... 2 A. ZAPI Integrated Rural Development Project (Credit 776)... 2 B. Western Highlands Rural Development Project (Credit 784). 4 C. Northern Province Rural Development Project.............. 6 (Credit 1678ILoan 1919) III. Main Issues ................................................ 9 A. Project Organizations .................................. 10 B. Extension ............................. ................. 15 C. Food Crop Production .................................. 17 D. Bottomlands Development ........ 19................... E. Cancellation Procedures ........... ..... .............. 20 MAPS IBRD Maps 12998R, 13072R, 13073R, 13821R CAMEROON - ZAPI INTEGRATED RURAL DEVELOPMENT PROJECT (CREDIT 776-CM) WESTERN HIGHLANDS RURAL DEVELOPMENT PROJECT (CREDIT 784-CM) NORTHERN PROVINCE RURAL DEVELOPMENT PROJECT(CREDIT 1078-CM/LOAN 1919-CM) PROJECT PERFORMANCE AUDIT REPORT PREFACE This is a Project Performance Audit Report (PPAR) of the ZAPI Integrated Rural Development Project (Credit 776-CM) involving a credit in the total amount of US$8.5 million; the Western Highlands Rural Development Project (Credit 784-CM) with a credit amounting to US$13.0 million; the Northern Province Rural Development Project (Credit 1078-CM/Loan 1919-CM) involving a credit of US$12.5 million and a loan of US$25.0 million. Credit 776-CM was approved in July 1978 and closed after cancellation of an undisbursed balance of US$0.7 million, on June 30, 1984. Credit 784-CM was approved in March 1978 and closed, fully disbursed, on June 30, 1984. Credit 1078-CM/Loan 1919-CM were approved in November 1980. The Credit was fully disbursed while an undisbursed amount of US$4.7 million of the Loan was cancelled and the Credit/Loan closed on December 31, 1987. The PPAR is based on three Project Completion Reports (PCRs) issued in 1987 and 1989 respectively by the Western Africa/Africa Regional Office, the Staff Appraisal Reports, the credit/loan documents, and a review of project files. An OED mission visited Cameroon in June 1989 and discussed the effectiveness of the Bank's assistance with staff of the Ministry of Agriculture and other ministries concerned as well as with staff and officials of the implementing project organizations. Field trips to meet participating farmers were undertaken. The kind cooperation and valuable assistance given by all persons met is gratefully acknowledged. The PCRs provide a good account and assessment of project experience, and discuss the performance of the Bank and project executing agencies. The PPAR, prepared several years after project completion, reevaluates project outcome and economic viability and elaborates on some issues which have, positively and negatively, affected project results: the selection of the implementing agencies, the extension services, aspects of foodcrop marketing, cancellation procedures and bottom lands development. Following standard OED procedures, copies of the draft PPAR were sent to the Government and to the implementing agencies. No comments were received. - il - 'Ap NTmrTD URAL /tVto NT PROJCT CRED17 776-06 PROJEC C?Mfat!M RUPet? LAtC DATA -NT * PROJtCt DATA Appvargal Actual or Actoal as 2 of testsat Esimate ACtus1 Appralete c*tt"etel/ Prolet Cota (US* aiton) 12.2 12.8 100 Credtt A~t•• (US$ attlimn) 8.3 7.8 92 (US*0.? acante ll*d Date Soard approal 3/07/78 - Dae Lft twees. 10/04/78 - Date thysteaI Caspen ta Camp~lteds 12/31/83 06/30/85 Ptpe s chein coepeted 1009 302 30 Clostag Date 06/30/84 12/31/4 tlatituatoast .na an Good Pet Agrømmele Yerfot a C?odPov *~ube of Dtract Unøftt&rl*s .0 E-tnese R~ of R~am (M) 25 Nee. CUMPIA=I DISBUIV~Vl M27 FMI MI1 F8 F8 YA F8 Appea~ rattaate (US* etion) 1.0 4.9 6.1 7.3 8.3 8.5 8.5 Aetm8l (Øs$ aillion~ 0.2 2.9 4.5 5.9 6.1 6.1 7.8 åsul &g 1 of ett~aa 20 59 73 81 14 72 92 Date o tnaL 0tabuaemao. 12/16/85 NSIM0 DATA 4/ Ne. Gt Nadaye Sp?eøtattons Pert. 3/ Type* of Nst Date fonge. ta ald ReprGaented 1/ R!ttg g/ roni probleoG (m7yr. ) Idfnt./Prøptrattoa 4/76 - - - - - fre.ØpprasL 8 11/74 - - - - • Pra-Øppratea4 II 2/77 - - - - - - Appreteel 3/77 8 25 - - - - pot-Apptel 9/7 3 18 f.t.C - - . Sup~ttel 1 1/79 4 13 a.e.ta 2 f Supeesten 2 6/79 l 18 2 2 fm stpeg~too 3 6/80 4 15 a.a.. 2 1 M Supettan 4 11/80 3 13 p.a.f 2 2 M 20pe~etatom 3 4/81 4 17 p.a.a.SoIG 2 1 m sapesviston 8 10/81 3 13 p.a.Dtv. Ch. 2 1 Supvs 7 5/82 2 18 p.a 2 3 fN? saevt"oa8 2/83 1 13 a 3 3 FNF SpeettIeU 9 10/83 2 13 a.e 3 2 Fm Supeuten10 6/84 3 9 a.a.f. 3 t 7T Søaoststen M 11/84 2 5 a.e 3 t 97 CeAette s/85 $ 4 t - - Co~pletten 3/86 3 e 6 - - - . . .3.& ? 8.9 ua-a a.* 36 8.6 4.4 t.* 1..9 a.* 76 12.6 ~.4 oTlm PROJCT DATA ~mGovernmet of Camao 2e10n Agefcy Uta, east vial t~ Julp 1 to June 30 w~ ot Curny C7A Cure ahbm*" Rate* Apprta~al Year Avea6a 245 CFA a USS1.0 taterwef Toars Auveage 329 CFA • UMI1.0 Camplaettea Tesr verage 380 FCA OSS81.0 Preaedtag Praltt No1 ro11u~. holo tt 9~m S a • g6~ t*uht s b o agaemutraL mmat3 c . fima=etal analyott @ « tem mt a * rmra~ «ngiaeers t • ag ~ t. 2/ 1 pmbl=teoere em ifflo probleah 2 = "Tdarat* probl~n and 3 0 ajtr prøbleas. 3/ 1 • taptostag; 2 • StatteayV Ba 3 det orattag. 1/ r * t~.maa; n * naage~$als T • teSta~1 p - pontical. - iii - wmuEN WtIIANPB tUAL SPV#03pNET cDUr? 7ea-CN BAMIC PATA Sppv7 RUT ~aaJEr lATA AppA~tee A r Aectul s 1 af Rim s ta tiu £ ~ tes Almretel snte project coeta (CA billjn) 6.1 8.3 1353 CTedit AMnt (089 MålitlMn 13.0 toa Dau Beer A~cowel 3M at* cffettU 12/08/78 Cloasn Date 06/30/84 EcoaoMl Rate ef ReerU (1) 20 17 Miehar of DSrcet Se tea (eaer 30.200 40,000 fem faltIs fata folSes 133 CUwUmT*T 9tsUussElls Apprtuel Estte (US MItt~> 0.3 2.5 5.8 0.3 13.0 - Actual (U u aulo) 0.0 1.1 4.3 7.3 10.1 12.3 13.0 Lutual as af aetaRae 0 4 74 79 78 93 l00 Date ofinal Dn§bans~t 10/12/84 PeSpeI Iapal d LhL 11h1~ oATA ud . f Nae9e Speetaltuettum Perf. TypTs Tf eOPM g fno, t M1,4 l Ss 1/ 1gt 41 v pm~ ldmttftieatia /73 • - Propeensten 4/74 - Appaisal 3/77 6. • SupersistaaI 71/78 1 6 d. 3 2 8 upervtisai 2 3/79 1 a.- 1 2 supurifion 10/79 2 10 a.C 1 2 N 8epvitat 4 3/60 2 10 e.d 1 2 8upitt 1 9/80 3 13 c.d.f 1 2 N alplsts6 3/81 3 12 .b.4 2 2 or 8sp sie6 7 10/81 1 9 b 2 1 11 Supevi~eS 8 6/82 2 6 h.e 2 1 r supesaises 9 11/83 1 12 b 2 2 8Up,ista 10 7/84 2 9 h.e 2 2 F Der.v Gewermef CGa&Mes use*ttag Agec Um.e Centrale de CooperatIVM Agicale de 10et (1~ I) Fissa TaG Juy 1 to Dua~ev 31 9e~ of dom~say (ahbreviette> CmA c e p mhne. atas AppreTsel Tes ~ avege 0ss1.00 a 245 Intseverv~g Tes Aveg. 0*1.00 * 329 c..platte TeI r AWerage 0ss1.00 • 440 Fll-up Projeets Iateffs 9eee uetr Prvaee 84ral Devele Proecet Lomn Iero ta 2406-Cm Lom dm~* f3921.3 maim Date 9eeyd Appesa 04/117/84 j/ s ~ egbeltuuietg b • eptcultural emamnes e0 fimemtal sns19eeg M etene~et tuel asttser f a enc~tat. / 1 O pble~fe öf Ditmr reles 2 w oderate problem and 3 * Oser probls. 3/ 1 I iupe~sg 2 ecctuffrvf nd 3 • deertraettag. ' • * finlesutal; lI • u ,erial: T • teesattclt P - politteal and 0 oetber. - iv - CAMEROON BASSC VATA SMEET CEY PRE.CT DATA Appraissi Actusl or Actual as 5 of Estimate Estimated Actual Appraisal EstImate Project Costs (CFAF billion) 13.9 Credit Amount (USS Millon) 87.5 32.8 87 Date Board N.A. 11/se Date effectiveness N.A. GOf91 Closing Date 12/88 12/67 Economic Rate of Return () 82.4 is Number of direct beneficiaries 168,996 116,ge 67 CUMULATIVE DISBURSMENTS FY02 FY98 FY64 FY65 FY88 FY67 FY88 Appraisal Estimate (USSNt II Ion) 5.0 18.2 19.8 28.6 88.0 37.S -- Actual (UStal I Ilon) 2.4 6.4 18.9 19.2 24.7 30.6 32.7 Actual as If Estlmt* 46 64 72 72 76 62 a7 Date of final disbursement 3/81/68 Principal repaid NIL MISSION DATA No. of Mandays Speciallzations Perf. Types of Weeli* r Dte In Field represented I/ R/tin Problem V/ (o.7r) Identifloation 8/77 - . . . . Preparation 11/77 - - . . - Appraisal 12/79 7 Supervision 1 19/81 1 1 a 1 2 Supervision 2 12/62 1 9 a 2 2 T,P Supervision 8 10/88 2 18 a 2 2 T Supervision 4 8/84 2 8 :,b 2 2 T Supervision 8 12/84 1 1 a 1 1 T Supervision 6 6/65 8 15 a,oe 1 2 T Supervision 7 12/18 2 N.A. N.A. 2 2 F,P Supervision 6 6/88 2 N.A. N.A. 2 8 F,P Supervision 9 1/67 1 .A. N.A. 8 3 F,P OTER PROJECT DATA Borrower Government of Cameroon Executing Agency Societe de D&veloppement du Cotes da Cameroun (SODECOTON) Fiscal year July 1 - June 89 Name of currency (abbreviatien) CFAF Currency Exchange Rate: Appraisal Year Average US08.9 a 216 CFAF Intervening Yeare Average US81.00 a 875 CFAP Completion Year Average US81.00 a 2S CFAF Fol low-up Projects None 1/asagriculturiftl bagricultural economist; eleanclal analyst; dneconeeit; esrural engIoeer; fasgronaeist. /Iaproblee-free or minor problem; teeaderate problem and Smajor problems. /lsaproving 2astationery; and audeteriorating. I/sfianclol; Mea agerial; Tateehlcal; Pupolitical and Onother. - v - CAMEROON ZAPI INTEGRATED RURAL DEVELOPMENT PROJECT (CREDIT 776-CM) WESTERN HIGHLANDS RURAL DEVELOPMENT PROJECT (CREDIT 784-CM) NORTHERN PROVINCE RURAL DEVELOPMENT PROJECT(CREDIT 1078-CM/LOAN 1919-CM) PROJECT PERFORMANCE AUDIT REPORT EVALUATION SUMMARY Introduction 1. The projects supported government efforts to improve agricultural production, the infrastructure, and agricultural as wall as social services. Agricultural services include the expansion of research, extension, credit to farmers, marketing and improved processing of agricultural commodities such as coffee, cocoa, cotton, and food crops. Social services provided for improved water supplies and access to primary health facilities. Investments in infrastructure focused on feeder road construction and development of marketing/cooperative service facilities. Two projects also aimed at bottom lands development and one aach of fishpond and forest establishment. Project Objectives 2. The projects' main objectives were intensify production of the individual region's major cash crops, i.e. cocoa and robusta coffee in the East, arabica coffee in the Western Highlands, and cotton in the North, together with the dominant food crops. Increased output of cash and food crops was expected to be paralleled by improvements in quality. To safeguard these efforts measures were to be undertaken to strengthen implementing agencies as well as associated ministriesiorganizations. Finally it was intended to substantially improve income levels and social/health conditions of participating smallholders. 3. It was assumed at appraisal that due to improved farming practices incremental seed cotton output would reach 43,600t at full development (Project year 10), robusta coffee 2,OOOt, arabica coffee 6,540t, and cocoa 1,500t. Incremental food production for all three projects were estimated to reach 271,500t at full development. Implementation Experience 4. Project implementation took about 6 1/2 years for each of the three projects, a time overrun of about 30%. Main reasons for the delays were the difficulties encountered in filling key Cameroonian staff positions of the Eastern and Western Highlands projects, while the Northern Region project experienced staffing problems only with components not directly under SODECOTON (Sociiti de Developpement du Coton du Cameroun) control. Slow project start-up was further accentuated by financial bottlenecks at ZAPI (Zones Agricoles de Production Intensive), cumbersome procurement procedures, and difficulties in locating the seed farm site for the Western Highlands project. - vi - Project Results 5. Project achievements of the three undertakings differ widely. Results of the ZAPI project were below expectations with coffee production targets reached but cocoa and foodcrop output remaining unchanged. The Western Highlands project was more successful, especially by substantially increasing food crop production and improving the quality of locally produced arabica coffee. The Northern Region project was most successful in promoting cotton production reaching dramatic yield increases, on average 1.5/ha on aoout 110,000 ha in project year 10. 6. All three projects did not reach the economic results estimated at appraisal. However, two projects, the Western Highlands and the Northern Region have achieved satisfactory ERR9, respectively 172 and 10%, while the Zapi project has a zero, or negative ERR. Main reasons for the lower han expected Economic Rates of Return were the time and cost overruns on the one hand and the serious price declines of the main cash crops coffee, cocoa and cotton on the other. Sustainability 7. Managerial, organizational, and financial difficulties encountered by ZAPI during project implementation has led to a complete loss of farmers' interest in ZAPI activities. Government has liquidated the organization and the only remnants reminding of ZAPI are a few shels and the coffee processing facility. 8. The Western Highlands project implemented by a strong cooperative entity, is successfully continuing its activities, supported by a follow-on project. Low coffee prices and signs of marketing difficulties for some food crops/vegetables, however, call for action to safeguard the project's sustainability. 9. Due to depressed world markets is the future of cotton production in jeopardy. However, latest commodity projections anticipate a price recovery and there are few if any alternatives to cotton as cash crops. Efforts undertaken by SODECOTON to reduce its administrativeloperating cost through delegating more responsibilities to village groups should also contribute to sustaining cotton production. Findings and Lessons 10. Project experience has brought some useful lessons for project design, institution building and Bank policy and procedures. Proiect Design a) in regions with relatively recent settlement/start-up of agricultural production (Eastern Cameroon) it is preferable to determine the agricultural production potential through pilot projects instead of full scale development schemes; b) low population densities and consequently poor road infra- structure render extension efforts costly and less effective; - vil - c) any planned increase in extension activities needs to take into account not only available research recommendations but also the financial implications of these recommendations. With climatic conditions prevailing in Cameroon and existing low world market prices, spraying of cocoa against black pod fungus is economically questionable; d) socio-economic considerations regarding division of labor between men and women has to be considered when planning extension activities; e) project components focusing on increased food crop production require more analysis in project appraisal, specifically in respect to marketing, on-farm and regional storage, transport costs, marketing intelligence, and availability of traders' working capital; f) bottomlands development in areas of low population densities and with adequate rainfall will not be successful especially when farmers will be confronted with health risks (Bilharzia); and g) rice is not the only crop for bottomlands Vegetable produc- tion is often more remunerative and helpful in overcoming local nutritional problems. Institution Buildini a) Although the integrated development approach was essential for the success of cotton production it demonstrated some of its intrinsic weaknesses: the fragility of fixed producer prices without consideration of the world price, the inability of price stabilization funds to cope with deep financial crisis and the insufficient cost consciousness of the project authority; b) There are differences of opinion within the Bank and between the Bank, cofinanciers and governments as to whether the existing institutional arrangements should be kept intact on fundamentally changed in favor of diversification of responsibilities or privatization of cotton companies. It is the opinion of this audit that the nature and status of executing agencies is not of primary importance as long as the vertical integration of cotton activities is retained. c) contrary to other cotton development schemes in Western Africa, the building of African management capacity has been slow in this project; d) strong and financially relatively independent agencies are capable to coordinate multifacetted development programs, but make difficult the taking over by line agencies at project completion; - viii - e) more emphasis needs to be given to improve commercial practices of project implementing agencies to ensure the viability of these organizations; f) secondment of goverment personnel to private sectorlproject implementing organizations can lead to over-bureaucratization and overlapping activities; g) rapid staff expansion can lead to diminished motivation of the original staff and hamper teamwork; h) organizations with roots in traditional hierarchical struc- tures can easily assure full cooperation of the local rural population; i) transferring some extension functions to village/farmer associations appears to be an interesting pilot undertaking by SODECOTON; j) following widespread adoption of basic extension recommenda- tions, a reduction of general extension staff densities and a parallel increase in the number of subject matter specialists is now warranted; Bank Policy and Procedures a) it is impossible to address key national policy issues in the context of regional projects; b) influx of substantial aid funds are likely to stifle selfhelp efforts. CAMEROON ZAPI INTEGRATED RURAL DEVELOPMENT PROJECT (CREDIT 776-CM) WESTERN HIGHLANDS RURAL DEVELOPMENT PROJECT (CREDIT 784-CM) NORTHERN PROVINCE RURAL DEVELOPMENT PROJECT(CREDIT 1078-CM/LOAN 1919-CM) PROJECT PERFORMANCE AUDIT REPORT I. BACKGROUND 1. In the late 70s, at the time the projects were identified and prepared, agriculture was the predominant sector in the Cameroonian eco- nomy, accounting for about 40% of the GNP, including 70% of the country's export earnings and about 35% of tax revenues. About three-quarters of the total labor force were employed in agriculture with smallholders predomi- nating. Nearly a million farm families, cultivating on average about 2 ha, contributed 90% of total agricultural production. 2. The three main export crops, coffee, cocoa, and cotton are grown principally by smallholders. Domestic food crops which supply about 96% of the country's needs are also produced almost entirely by small farmers. 3. Confronted with a slowed down annual growth in the production of export crops and a food crop production just ahead of population growth, Government development plans put emphasis on the development of the small- holder sector through a variety of means, such as integrated regional development and specific crop promotion. The Bank tried to support Govern- ment efforts to achieve balanced growth of rural development projects in traditional areas. 4. In its lending operations to agriculture the Bank had three main objectivest firstly to increase technical and managerial capabilities, strengthening institutions, and improving sector policies, mainly through training and technical assistance; secondly, raising foreign exchange earn- ings through expanding production and export of agricultural products; and thirdly, increasing the productivity of overpopulated and poor rural areas. The three rural development projects were designed to meet these objec- tives. - 2 - II. THE PROJECTS A. ZAPI Integrated Rural Development Project (Credit 776-CM) Project Design 5. The project was designed to support (a) institutional development through management development, including construction of offices, streng- thening extension, staff and farmer training, and monitoring and evalu- ation; (b) agricultural production through capsid control in cocoa, esta- blishment of coffee seedling nurseries, swamp rice development, establish- ment of fishponds, food crops research, and credit; (c) marketing and pro- cessing through Improvement and expansion of coffee processing, improved coffee and cocoa marketing, and provision of retail marketing facilities; (d) social infrastructure through improvement and provision of health ser- vices and domestic water supplies. 6. The project was expected to take 5 years to complete. Total pro- ject costs were estimated at appraisal to be US$12.23 million, of which IDA would finance US$8.5 million representing 1002 of the foreign exchange and 51% of the local currency costs. The economic rate of return was estimated at appraisal to be 25%. Project Implementation 7. Initial project start-up was encouraging with Government providing a higher capital increase to ZAPI and also earlier than stipulated by the Credit Agreement. Staff was transferred to ZAPI, the Administrative Coun- cil established on time and consultants were immediately hired to assist in the preparation of work programs and budgets. Shortly after the start, however, problems in credit processing resulted in procurement delays. Two fundamental difficulties became apparent at this early stage and persisted throughout the project implementation period. First the newly recruited staff lacked experience and showed little commitment and lacked motivation to perform in the expected manner. Second ZAPI experienced serious pro- blems with its cash-flow. Although Government had fulfilled its initial financial obligations, both, the capital increase and a revolving fund deposit of FCFA 150 million were used by ZAPI to pay-off outstanding debts. Consequently there were only limited funds available to meet operating expenses, a problem worsened during project implementation. 8. Bottomlands development started late and was soon abandoned with- out a single swamp reclaimed. Fishpond development suffered the same fate since ZAPI did not receive any support from the Ministry of Livestock which was responsible for inland fisheries development. Although ZAPI was ex- pected to take over pest and disease control in the project area, it was decided t leave this activity in the hands of the ineffective MINAGRI (Ministry of Agriculture) crop protection service. 9. The construction of the Belabo coffee processing plant and the modernization of other facilities was executed between 1980 and 1982. Vehicles and equipment were procured and technical assistance, staff train- ing as well as monitoring and evaluation activities were undertaken. - 3 - 10. Feeder roads financed under Loan 1494-CM did not materialize as expected and maintenance proved to be ineffective due to unsuitable equip- ment which turned out to be too light for conditions prevailing in the tropical forest area. Instead of 540 km envisaged at appraisal only 72km of feeder roads were improved and no new construction took place (280km planned at appraisal). 11. The project's contribution to improving the social infrastructure was also negligible. Instead of 23 consumer supply stores and marketing centers only 9 were constructed. The health center and the pre-pharma- ceutical construction program suffered a similar setback. Instead of the 19 planned buildings only 8 were constructed. 12. The total project costs estimated at completion were US$21.6 mil- lion. This cost included the Government's financing of ZAPI cumulative losses and capital increases. Without these amounts total costs amounted to US$15.0 million. (PCR paras. 3.16, 6.02).l/ Project Results 13. There is little information available on the project's impact on agricultural production. While coffee statistics for Eastern Cameroon show no significant change in coffees quantities marketed, ZAPI increased its purchases from 3,000t of robusta coffee in 1978/79 to 5,000t in 1984185. Cocoa production remained stagnant mainly the result of depressed cocoa prices and the questionable economy of black pod control. 14. Based on sales of seedlings it can be assumed that about 2,600 ha of cocoa and 830 ha of coffee were planted by 1984. These figures are, however, highly indicative since no baseline information on plantation acreage has ever become available and no indications exist how much of this acreage would have been planted without the project. Equally absent is any information on food crops production. 15. The project was unsuccessful in substantially improving coffee quality. This was mainly due to the decision to buy coffee cherries ins- tead of establishing decentralized hulling facilities. A particularly worrying aspect of the coffee marketing operations was the inability to improve the flow from collection through processing to exportation. As a result coffee marketing operations continued throughout the project to incur heavy losses. Of ZAPI's estimated debt of US$6.8 million in 1986, about 60% consisted of accumulated losses of marketing operations. 16. The project's research component was in a limited way successful by establishing a first adaptive research unit for the Eastern Region. The most successful project component, however, was the training of ZAPI staff. Although ZAPI no longer exists, trained staff make a valuable contribution to the work of the organizations/agencies to which they have been assigned. 17. The lack of data on production makes it impossible to calculate a rate of return. In the absence of supported data on coffee production 1/ OED Report No. 6808, dated May 29, 1987. - 4 - increases attributable to the project and the significant cost overrun, the audit estimates that the PCR's estimate of an ERR of 0% is still too opti- mistic. The 2,000t incremental robusta coffee represented a gross value of not more than US$4.0 million (cif New York), deducting operating and on- farm costs the ERR is more likely to be negative. B. Western Highlands Rural Development Project (Credit 784-CM) Project Design 18. The project was designed to assist agricultural development in the overpopulated Western Highlands. The major objective of the project was to provide for the development of institutions, procedures, and techniques to improve agricultural production and the standard of living of smallholders in the project area. The project supported (a) the strengthening of a new Directorate of Production Services within UCCAO (Union of Coffee Coopera- tives) responsible for extension, plant protection, seed production, train- ing and field trials/demonstration; (b) a training center for technical staff; (c) construction of 10 rural service centers; (d) a revolving credit fund for seasonal inputs and tools; (e) construction of 220 village water points; (f) development of about 400 ha bottomlands; (f) introduction of improved coffee processing equipment; (g) a monitoring and evaluation unit; (h) strengthening of the MINAGRI Planning Unit. 19. The project was expected to take four years to complete. Total project costs were estimated at appraisal to reach US$25.0 million, of which IDA would finance US$13.0 million, representing 100% of the estimated foreign exchange costs. The economic rate of return was estimated at ap- praisal to be 20%. Project Implementation 20. The project started slowly due to delays in hiring the Director of Production Services and the Chief of Extension as well as the ratification of the Project Agreement by UCCAO's Board of Directors. Difficulties were also encountered in locating the site for the proposed seed farm. Delays in the project construction program were caused by the lengthy government process of reviewing and awarding contracts. The difficulty in finding and deciding on a suitable site and the inexperience of local contractors exa- cerbated the delays. Due to the centralization of pcocurement in Yaounde and inadequate preparation, first tenders for vehicles, equipment and tech- nical assistance were delayed and caused vehicles and other equipment to arrive one year too late. 21. Training given to extension staff and the support provided by the project's technical assistance team was successful in building up an effi- cient agricultural extension service. By the end of the project, a regular schedule of extension worker visits had been established. The training program was highly successful with 2,500 man-months accomplished compared with 750 months envisaged at appraisal. Most of the training was for ex- tension agents (65%), the remainder for supervisory staff, farmer leaders, and cooperative officers. The intensified training program permitted the fielding of 784 extension agents and 515 supervisory staff versus appraisal assumptions of 378 and 295 respectively. - 5 - 22. The research component did not perform as well as expected mainly due to the research organization in charge followed traditional research patterns, i.e. pure crop stands oriented while mixed cropping is predo- minant in the Western Highlands. Field demonstrations did better with 36 plots established--more than five times the original target of seven. Although 460 ha of new forests were established (450 ha originally planned) ownership problems and lack of maintenance have led to a changed approach in new plantings which responds to the local farmers' aspirations. 23. Production of seeds from the project seed farm was lower than expected with 150 to 200t sold annually compared with a production poten- tial of 700t. All indications are that the farm was overdesigned and will never be run profitably since its theoretical output will always exceed demand. 24. Because of undersupply of pesticides, a responsibility of MINAGRI, the antestia control covered only 67,000 ha and anthracnose 34,000 ha, an implementation rate of 80% and 60% respectively in comparison with ap- praisal assumptions. 25. Construction and rehabilitation of cooperative service centers fell behind appraisal estimates. Belatedly it was recognized that assis- tance for management training was provided under the project. This short- coming has been corrected under the follow-on project. 26. Due to insufficient appraisal/analyses of credit needs the revolv- ing credit component did not perform well. The credit problem was further exacerbated by poor accounting practices and lacking linkages with exten- sion services. 27. A total of 235 village waterpoints were established compared with only 220 expected at appraisal. Equally successful was the reclamation of bottomlands. The draining of 400 ha were foreseen by the appraisal but by the enJ of the project 436 ha had been reclaimed and another 1,651 ha sur- veyed to be tackled under the follow-on project. Soil conservationlerosion control was limited to a single pilot project. 28. To improve the quality of coffee the project provided for the purchase of ten sorting machines, three machines to detect stinking beans, and 3,000 coffee pulpers. While the number of pulpers bought was in line with appraisal estimates, a larger number of coffee sorters and stinking beans detectors was procured. quality improvement during project implemen- tation did, however, not improve significantly. This changed In the last years and UCCAO management stated to the audit mission that in their view one of the project' major accomplishments was the improved coffee quality. 29. Total project costs increased by about 35% in FCFA terms. At appraisal total expenditures were estimated to reach FCFA 6.1 billion com- pared with actual expenditures of FCFA 8.3 billion. A comparison in dollar terms is not very meaningful in light of the declining value of the CFA franc vis-a-vis the US dollar. -6- Project Results 30. In light of limited survey data available, estimates of yield increases attributable to the project are difficult to make. Coffee pro- duction varied during project implementation between 15,200t in project year 0 (appraisal assumption) and 20,400t in project year 2. The fluctuat- ing production is due to several years of drought which had a severe impact on production. Since pest control covered a major part of the total coffee area and also fertilizer use increased considerably, a major impact on coffee production should be assumed. 31. Maize yields under mixed cropping increased above appraisal esti- mates. Average yields grew from 0.97 t/ha to 1.79 t/ha in 1983/84. Consi- derable production increases were achieved with groundnuts and vegetables especially potatoes and tomatoes. 32. Due to lower than expected coffee production the ERR was recal- culated at the time of appraising the follow-on project at 17% (20% at appraisal). Since UCCAO does not have sufficient data to support indica- tions which put incremental maize production alone at 10,000t annually, another recalculation was not attempted but It can be safely assumed that the ERR is between 17 and 20%. C. Northern Province Rural Development Project (Credit 1678-CM/Loan 1919-CM) Prolect Design 33. The project was to provide (a) an improved rural infrastructure, including roads, stores, and marketing facilities; (b) effective services, including input delivery, supervised credit, extension, marketing, adaptive research, and project monitoring, necessary for the achievement of a pro- gram of balanced development of agricultural production and incomes of some 150,000 farm families; (c) support to forestry development; (d) financial and technical assistance to SODECOTON required to discharge its broader responsibilities in becoming an organization capable of a wider approach to rural development; (e) assistance to facilitate the rationalization of remaining MINAGRI extension services in the area; (f) help existing admi- nistrative structures at provincial level to participate more effectively i the planning coordination and control of all development projects in the Northern Province within the framework of national development policies; (g) extend the agricultural project monitoring and evaluation network being formed by MINAGRI to the Northern Province; (h) help the agricultural re- search institute for the North to work more productively. 34. Project components included: a) strengthening SODECOTON agricultural extension services in the project area by increasing the number of extension sec- tors from 19 to 23, the number of extension zones from 75 to 150 and of extension workers from 560 to 600; b) establishment of an animal husbandry extension unit within SODECOTON, employing two veterinarians and 20 livestock - 7 - assistants, principally to look after the feeding and health care of draft animals; c) systematic training and retraining programs for SODECOTON and MINAGRI staff of all levels; two existing training centers of SODECOTON would be enlarged and additional training specia- lists provided; d) establishment of an agricultural experimentation unit within SODECOTON, to carry out field trials and demonstration at two new sites and on field plots in each of the 23 sectors in the area; e) staff and operating costs required by SODECOTON for the pro- vision of management services; f) construction of additional field storage facilities to handle farm inputs required for the project; g) improvement of about 800 km of feeder roads in the area in- cluding construction of fords and culverts, in order to faci- litate crop marketing; input delivery and extension work; h) improvement of rural market place facilities to ease market- ing of food crops within the area; i) financing of incremental costs of seasonal inputs used by farmers (revolving fund administration by SODECOTON); j) administration by FONADER (Fonds national de developpement Rural) and financing of an agricultural credit program for participating farmers requiring credit to buy draft oxen and equipment; and k) pilot tree plantations in forest reserves for fuel wood and poles and plantings of windbreaks in densely farmed areas to reduce soil erosion and help improve soil fertility; 35. Project components related to the Northern Province as a whole would be as follows: a) strengthening of the provincial planning unit at Garoua to improve planning and coordination of development projects; b) support for agricultural applied research work of IRA-North (Institut de recherche agricole), with a view to making better use of existing resources in particular for adaptive research on food crops; c) strengthening the savannah forest research unit of IRA-North; d) purchase of additional seed handling equipment required to furnish improved groundnut seed required for development projects in the province; - 8 - e) provision of new vehicles and improvei office facilities for MINAGRI, as well as staff retraining; and f) establishment of a monitoring and evaluation unit under MINAGRI to monitor and evaluate results of this and, later, of other agricultural projects in the north. 36. The project was expected to take five years to complete. Total project costs were estimated at appraisal to be US$74.7 million, of which the Bank Group would finance US$37.5 million, representing 96% of the fo- reign exchange costs. The economic rate of return was estimated at 34.2%. Proiect Implementation 37. Project start-up was relatively fast because of SODECOTON's expe- rience and management capacities. Activities not directly under its con- trol started more slowly. Livestock assistants to be provided by the Ministry of Livestock were not seconded until mid 1983. The Northern Seed Multiplication Project was unable to supply the desired quantities of im- proved groundnut seeds. FONADER did not meet the loan condition of the Bank to give proof that outstanding/overdue loans did not exceed 25% of the total until about two years into the project. 38. The project included over 20 separate activities and most were implemented satisfactorily and on schedule. The appraisal had anticipated an increase from 48,000 ha to 70,000 ha of intensive cotton production with a total output of about 82,300t of seed cotton by project year 5. Instead approximately 60,000t were produced on 52,500 ha, a 27% shortfall compared with appraisal. Average yields reached only 1,140kg/ha instead of the 1,200 kg foreseen at appraisal. This was mainly due also to extended years of low rainfall. In 1988 when rainfall was optimal, average per hectare yields exceeded 1,500 kg. 39. Efforts to improve groundnut production were handicapped by adverse weather conditions and also by differences between the Bank and SODECOTON about the approach. The Bank recommended a semi-intensive approach while SODECOTON promoted intensive cultivation practices. Total acreage under intensive groundnut cultivation amounted to 1,500 ha by pro- ject year 5. Sorghum, the most important cereal in the region, was to be intensified on 91,850 ha but only 25,000 ha were accomplished. One of the reasons was the bottoming out of sorghum prices following a record crop in 1984. Maize was found to be less suited to the low rainfall project area and the planned expansion of production was abandoned by SODECOTON. 40. The modernization of seven markets was not undertaken since it was found that marketing facilities were not the constraint in marketing food crops produced in the area. Construction costs of feeder roads were under- estimated. Consequently only 600 km were built in comparison with the 800 km estimated at appraisal. Lack of operating budgets has led to a near total absence of maintenance for the feeder roads since 1987. 41. FONADER was never able to disburse the budgeted US$3.2 million and only US$0.95 million was spent. Reasons for this shortfall are not clear and several explanations have been ventured. First, that credit demand was - 9 - overestimated; and second demand was further reduced when SODECOTON started to promote cheaper donkeys to replace the work oxen. Some 3,800 donkeys were distributed to farmers and trained. 42. Forestry research made good progress under the project but for lack of determined management forest plantations fell to over 50 percent below target. Village plantations and windbreaks were entirely dropped from the program. 43. The project was more successful in carrying out the agricultural research component. Research programs were executed at the Maroua station and nine substations. The experimental program involved 183 trials, inclu- ding 134 on food crops. By 1984, IRA had established 4,200 demonstration plots. The sustainability of the research effort is now seriously hampered by lack of government funds. 44. Actual project costs are not available except for those components which were fully or partly financed by the Bank Group. There are no records on actual amounts made available by Government to SODECOTON. Project Results 45. SODECOTON's financial position suffered from the decline of cotton prices in the world markets. During strong cotton prices SODECOTON showed a modest profit and increased revenues were largely passed on to the produ- cers who enjoyed the highest cotton prices paid in Africa. The fall in cotton prices in 1986 together with the Government's inability to provide agreed counterpart funding for rural development work, resulted in an acute financial crisis of SODECOTON. During the 1985 to 1988 seasons SODECOTON experiences losses of FCFA 32.4 billion. On the other hand financial returns of farmers remained satisfactory with net returns of FCFA 30-50,000 per average cotton grower. 46. The project executing agency SODECOTON performed well, other ins- titutions involved in the project were less satisfactory and the project's institutional development objectives were not achieved, because policy making and implementing structures in north Cameroon remain weak. 47. At appraisal the ERR was calculated to reach 34.2%. Recalculated the ERR stands now at 10%, due to lower international price of cotton and lower food crop production than estimated at appraisal. III. MAIN ISSUES Introduction 48. The three rural development projects were appraised almost simul- taneously and focused on the promotion of cash crops as well as food crops. The organizational set-up differed widely and since success or failure of these projects depended to a large extent on the strength/weakness of the implementing agency, results obtained vary. One of the projects can be termed successful in regard to production development (SODECOTON), in con- trast to it one turned out a complete failure (ZAPI), and the third (Wes- tern Highlands-UCCAO) a modest success. - 10 - 49. Valuable lessons can be drawn from the experience of the three projects, especially in respect to project organization, extension activi- ties, food crop marketing, and bottomlands development. A. Proiect Ornanizations 50. The three projects were typical of the integrated rural develop- ment approach supported by the Bank in the late seventies. Each project, however, had a different organizational pattern of the executing agencies: a new autonomous project entity in Zapi; an existing regional cooperative organization in UCCAO and a vertically integrated cotton parastatal in the Northern Province. 51. The ZAPI organization dated back from a farmer self-help scheme, originally supported by expatriate technical assistance and taken under the Government umbrella as a semi-autonomous development agency in 1972. The appraisal described the organization: ZAPI is the executing agency of one of the most innovative experiences in small farmer-oriented, participatory develop- ment programs in francophone Africa. The unique feature of ZAPI's philosophy is to induce and train farmers to partici- pate in, and eventually take charge of, local development actions, thereby creating an environment for sustained social and economic development. 52. In addition to guiding self-help, ZAPI also focused on developing smallholder directed cooperatives, which would be responsible for agricul- tural credit, input supplies, marketing of agricultural produce and con- sumer goods and communal village improvements (see PCR, paras. 1.06-1.09).!/ 53. From its start in the 60s until appraisal ZAPI management had gradually built-up a small organization of Cameroonians, highly motivated and who had succeeded in organizing farmers to follow self-help principles in improving their and their neighbors' lives. Most of the activities promoted by ZAPI did not require cash or Government funds. The absence of major Government development efforts is easily explained by the constraints facing Eastern Cameroon. Low population density was and is not conducive to productive extension work and makes input supplies and marketing operations costly. In addition natural conditions also pose serious problems to agriculture. While abundant rainfall seemingly promises good crops it is a handicap for cocoa, one of the regions major cashcrops. The incidence of black pod (Phytophtora palmivora) is extremely high and spraying against this fungus disease is not only costly but also requires large quantities of water which has to be carted or carried for long distances to the cocoa plantation. The difficult situation for cocoa was further exacerbated by depressed world market prices. 54. Enter the Bank to support rural development in this area. To start with the project provided for substantial staff increases. In the audit's -iw such decision did not take into account the following factors: (i) the country wide shortage of qualified agricultural staff; (ii) the 2/ OED Report No. 6808 of May 1987. - 11 - reluctance of staff to move to a region which, by Cameroonian standards, had an underdeveloped infrastructurel (iii) the inexperience of new staff not familiar with farmer attitudeslaspirations, and likely difficulties in integrating new staff members into a tightly knit organization. 55. This oversight led immediately to problems. Instead of working with farmers and continuing to encourage their self-help activities, new staff promised the advent of the "development miracle "by persuading far- mers that no more individual effort would be required, the World Bank and its money would take over, "time to sit back and relax". As a consequence farmers lost not only interest in contributing to development with their work input, but also stopped identifying themselves as members/owners of ZAPI. When most of the false promises did not materialize, farmers started to accuse ZAPI officials of having embezzled the expected World Bank funds. 56. Next to underestimating the "in-house" organizational diffi- culties, the Bank did not properly identify the "external" organizational difficulties ZAPI was confronted with. Having evolved from a private ini- tiative, having "semi-autonomous" status, ZAPI's standing vis-a-vis other ministries/development agencies was weak. By selecting ZAPI as the execut- ing agency, politically stronger entities felt challenged and reacted accordingly. A case in point is the failure of the fishpond component. The project provided for the establishment of 1,000 fishponds but none was ever constructed under the project. There was nobody on ZAPI staff who had any fisheries experience and the project had anticipated that experts from the Ministry of Livestock, responsible for inland fisheries development would be seconded. However, the Ministry had different development priorities and the politically weak ZAPI never succeeded to change the Ministry's mind. In contrast to this disappointing development in the ZAPI project, did the strong organizations SODECOTON and UCCAO relatively easily succeed in obtaining the services from other agencies or where the foot dragging became too strong, the established their own specialized branches (Livestock under SODECOTON for instance). 57. ZAPI's weakness was further compounded by its financial dif- ficulties which were ali not identified at appraisal. Not meant origi- nally as a marketing organization, ZAPI had taken on coffee exports in 1976, the year prior to appraisal. Although the appraisal report makes reference to likely marketing difficulties such as .."high moisture content of crops, low tonnage collected per km, bad roads and poor maintenance of vehicles, and inadequate and faulty storage (which) result in high market- ing costs and, frequently, a poor quality product" the likely consequences for ZAPI, as the project organization, did not receive adequate attention. 58. Due to the poor quality of ZAPI products destined for export, they remained in store in Douala for extended periods of time, burdening ZAPI with debts exceeding US$2.0 million just at the time of appraisal/project start-up. Government counterpart financing had to be used to pay ZAPI debts and during the whole implementation period, the project was beset by financial problems. Being financially unsound in addition to being also politically weak rendered the project organization ineffective which in turn caused the failure of the project. In the audit's view this could have been avoided by a more comprehensive appraisal and analysis of the proposed project organization. - 12 - 59. The UCCAOI executing agency for the Western Highlands project, was, and still is a successful coffee marketing organization, serving as the cooperative union of six coffee marketing cooperatives. One of the very few successful cooperative ventures in Africa, it derives it strength not only from excellent management but also by its entrenchment in the region's socio-economic hierarchy. 60. Contrary to the ZAPI appraisal a thorough analysis of UCCAO's apparent strength was made. The most important appraisal conclusion was that UCCAO leadership was firmly linked/identified with the region's traditional leadership. Farmers having confidence in their traditional leaders showed the same confidence in the cooperative leadership--and were not disappointed. Because of this strong leadership, UCCAO yields an important political clout, it found it not difficult to coordinate or enforce attention from project associated ministries/agencies, something ZAPI never succeeded in doing. 61. Also in getting farmers more interested in the project UCCAO suc- ceeded better than ZAPI. UCCAO recognized the importance of "project visi- bility". It decided on early implementation of the village water supplies and bottomlands project components. By giving priority to these develop- ments, the UCCAO succeeded in an early motivation of project farmers. This was in stark contrast to the ZAPI area where farmers sat back and waited for something to happen. The higher "project visibility" was of course much easier to accomplish in the Western Highlands with high population densities than in the sparsely populated ZAPI areas of the East, another factor militating against rural development efforts in sparsely settled zones. 62. In contrast to ZAPI staffing (plus 200), UCCAO was expected at appraisal to employ an extension staff of only 78. Although this number was exceeded during project implementation, it should be noted that the farmer: extension worker ratio, the Bank's preferred yardstick in planning for extension activities, needs to be more carefully scrutinized. Extension staffing has to take into consideration the density of the farming population and--to remain cost effective the likely production volume/value. While ZAPI accomplished a turnover of about US$1.4 million in 1984/85, at the end of the project, UCCAO showed already a pre-project turnover of US$100.0 million. It is obvious that a project induced annual increment of 1% of this turnover would provide a more acceptable financial justification for an enlarged extension effort in the case of UCCAO than for the ZAPI project. 63. SODECOTON, responsible for the Northern Region project imple- mentation, goes back to the former CFDT (Compagnie frangaise de dive- loppement des fibres textiles). Like UCCAO in Western Highlands, SODECOTON had a politically strong position in the North. SODECOTON is representa- tive of a development strategy--introduced in the 60s in about twelve fran- cophone countries in Sub-Sahara Africa--which resulted in rapid and consi- derable rise in cotton production. Although the development strategy started out with common components, following the Cotton Crisis of 1986 (para 66) different countries responded differently and it is instructive to compare project experience across these Francophone countries. - 13 - 64. The core of this development strategy is a vertical integration of (a) research, with good links with international product manufacturing, (b) extension, similar to the training-visit system but long before the Bank embarked on it, (c) provision of inputs to cotton growers under a credit scheme, (d) establishment of village groups for input distribution and primary marketing of cotton (para. 80), (e) rapid cash payment to farmers when the crop is collected and (f) processing and export of lint cotton. The success of such strategy resulted in a cotton production increase in francophone Africa from about 50,000 tons in the early 60s to about 350,000 tons in the mid-eighties, contrasting with the generally weak record of agriculture in Sub-Saharan Africa and the poor records of cotton production in other African countries. 65. Because of their success, cotton companies in most countries were given broader responsibility for rural development. They embarked on a large range of activities including food crop production, livestock and forestry development, and construction of roads and potable water supply systems. Cotton companies were to be financially autonomous and fully compensated by governments for their rural development activities. In all countries, cotton companies have been successful in their rural development activities, although the food crop objectives have generally not been achieved due to severe marketing constraints and low price of the incremental production (para. 84). For many years, cotton production financially benefitted farmers, cotton companies and governments. Despite project success, some questions remained controversial such as the possible competition between food crops and cotton where land is limited and labor force is constrained, and the potential conflict of interest between cotton revenues and broader rural development objectives of cotton companies. All OED evaluations have concluded, however, that far from competing, food crops have benefitted from cotton production, and cotton companies' performance in rural development has generally been better than that of other government agencies.3/ 66. The dramatic drop in world cotton price in 1986, together with the overvaluation of the CFAF resulted in heavy financial losses for cotton companies. SODECOTON was not an exception, with cumulative losses reaching the equivalent of US$200.0 million in 1989. The financial ramifications of the cotton crisis called into question the sustainability of the development strategy and illustrated some inherent weaknesses of the system. First, the inability of price stabilization funds to cope with a deep financial crisis. In most countries, monies received by stabilization funds when cotton prices were high have been heavily taxed and diverted from their original purpose to finance government investments. Second, the fragility of the fixed and guaranteed producer prices without consideration of the world price. In this resp -t, Cameroon is quite illustrative, with the highest producer price of all countries, despite the fall of world prices of cotton. Third, the government's failure to pay their share of rural development activities increased considerably the deficit of cotton companies. Fourth, some cotton companies have not been as cost conscious as they should have been during the good years and were taken by surprise at the time of the cotton crisis. In this regard, the building of African 1 See Impact Evaluation of Cotton Development Programs in Burkina Faso, Cate d'Ivoire and Togo (0ED Report 6878 dated June 1987). - 14 - management capacity and related reduction of expatriates in technical and managerial positions was almost fully achieved in some countries (Burkina Faso and Togo for example) but not in Cameroon where SODECOTON still employed 80 expatriates in 1984, gradually reduced to 24 in 1988-89. 67. Most countries have taken remelial measures such as reducing cot- ton cultivation in marginal areas, eliminating subsidies and introducing a better control of operating costs. SODECOTON is no exception, having re- duced its operating costs from CFAF 4.6 billion in 1985-86 to about 2.4 in 1988-89. During the same period, extension staff decreased from a high of 756 extension workers to 500. There are differences of opinion within the Bank and between the Bank, cofinanciers and governments as to whether the existing institutional arrangements should be kept intact or fundamentally changed in favor of redistribution of responsibilities among government agencies, disengagement of government from commercial activities and priva- tization of cotton companies. Diversification from cotton to other crops was also sought in all countries, but found to be extremely difficult on the basis of existing technologies. In addition, based on current yields and production costs, cotton is one of the few crops with comparative advantage in the West African context. The fact remains that cotton production is in jeopardy in Cameroon, as well as in all countries producing cotton essentially for export, due to a combination of depressed world prices and costly organizational structure. 68. The response to this crisis has varied greatly. In Burkina and Togo, the integrated strategy was retained, but the institutional responsibility was diversified: government line agencies are now fully responsible for support services to farmers. In other countries, provision of inputs is handled by the private sector in parallel with the public sector. Flexibility in producer price has also been introduced. Village groups have been given more responsibility for some activities previously carried out by cotton companies. Privatization of cotton companies, however, has been a failure in Zaire where revitalization of the cotton industry under an integrated organizational structure is now contemplated. Privatization of the national cotton company in Central African Republic, one Bank project objective, has not been successfully achieved. Promoting a privately owned cotton industry in parallel to the public system is also under consideration to submit cotton companies to market discipline and competition. For the time being, no satisfactory solution has been found to surmount the cotton crisis. Nevertheless, an important finding of OED evaluations of cotton projects is that the vertical integration inherent in the development strategy should be kept intact even if services are provided by different government or private agencies. The need for such integration is by no means limited to cotton but to a large number of industrial and export crops as shown by a number of OED evaluations.4/ 4/ On this subject, see Impact Evaluations of Morocco and Mexico Irrigation projects (OED Report No. 7876, dated June 15, 1989) and PPAR of fruit and vegetable production and marketing projects in EMENA Region (report under preparation). - 15 - B. Extension 69. Several aspects of the project's extension efforts merit atten- tion, especially the appraisal analysis required in determining extension densities and extension programs. 70. Main cash crops in the ZAPI project area were robusta coffee and cocoa. The appraisal identified some problems confronting coffee produc- tion, i.e. the need to regenerate old plantations, improving land clearing and planting technology as well as harvesting techniques. But somehow contradictory the appraisal report states that "growers consider rehabili- tation uneconomic at present prices". 71. In regard to cocoa the appraisal found that seed pods distribution was a major bottleneck confronting replanting programs. The SAR also found that a gradual decline in the general health and productivity of cocoa plantations is clearly evident in many parts of the region, the principal cause being a combination of capsid damage (Sahlbergella singularis and Distantiella theobromae) and black pod disease. Referring to black pod control the SAR states: "The treatment consists of 10-12 kg/ha of copper chloride mixed in about 1,000 liter of water, applied in 8 - 10 separate sprays. The high volume of water required poses a problem to most farmers particularly if plantations are situated at some distance from a water source. Because of the amount of labor involved in transport of water, spraying and weeding/cleaning, farmers will, at most, only practice partial control. In addition existing government-determined price margins between low and higher cocoa grades provide farmers little or no incentive to incur spraying expenses". 72. Intensifying extension efforts aiming at improved food crop pro- duction could be or should have been ruled out at appraisal because it was the project that was initiating an adaptive research program for food crops, and, as proven by actual developments, results could not be expected in a short time. Practically without any extension messages available or likely to be financially profitable, the project nevertheless provided for increases in the number of extension staff and allocation US$1.1 million to support extension. 73. Bank staff commenting on the project prior to Bank approval had serious and justifiable reservations. A memo dated December 2, 1985 noted that little knowledge on staple food techniques existed and that conse- quently a project "if not built around cash crops, they (extension ser- vices) might turn out to be a serious drain on the Government's budget". Another memo (October 5, 1977) casts doubts on extension success due to the region's low population density and queries the wisdom in respect to tech- nologies that "a number of problems have been identified for which solu- tions would be tested (only) during project implementation". Finally another memo (October 12, 1977) commenting on the project points out that" the report states on a number of occasions that no new proven technology exists". Despite these sound comments the project went ahead with its substantial investments in extension. 161 - 16 - 74. In the audit's view it would be essential to prepare detailed extension programs prior to and analyzed during appraisal. Extension pro- grams should be based on (a) available and proven research results; (b) financial viability at farm level; (c) economic justification; and (d) as far as possible on safeguards that extension staff would be product'vely active throughout the year and not only at limited periods. 75. The cost-effectiveness of extension services in regions of low population densities needs to receive special attention during project preparation/appraisal. The ZAPI project demonstrates that not the lack of extension services is the limiting factor, but in this case, the lack/poor maintenance of the region's roads. Low population density and poor transportation network are interdependent. Extension efforts to improve production will remain ineffective as long as transportation problems remain unresolved. 76. Extension was a new activity for UCCAO, it was, however, a suc- cessful undertaking as far as maize and vegetable/bottomlands development was concerned. Unsuccessful, on the other hand in ensuring major improve- ments required for higher coffee arabica yields, i.e. establishment of pure coffee stands. Two factors were militating against this goal, factors that were not adequately evaluated during appraisal. First the socio-economic aspects and second the income loss of coffee farmers during the required transition period. 77. Due to the traditional division of labor between men and women certain field activities such as land clearing and (coffee) tree planting are the men's responsibility--among other things--while the women's tasks include cultivation of food crops and up-keep of the coffee plants. Weeding operations--in the women's view--are facilitated by having food crops in the "vicinity" of the men's coffee. Extension efforts directed at persuading men to plant pure coffee stands will fail due to the women's opposition. Only farmers with substantially more land than the customary 1/2 ha will be able to employ hired labor and establish pure coffee stands. But even larger farms may be confronted with the problem of income losses during the time between uprooting of coffee trees and the coming into fruit bearing of newly planted bushes. For several years the male farmers would be deprived of their cash incomes. 78. General agreement was found that UCCAO extension activities were highly successful in teaching farmers the basics of general cultivation techniques, crop fertilization, and vegetable production. What is needed now is a gradual reduction of the general extension staff combined with an increase--in number and competence--of subject matter specialists. It should be noted that the reduction of extension personnel after an initial period would favorably affect the ERRs of agricultural development pro- jects. Extension intensity needs to be built up gradually, will level off after a certain period and can then be decreased as shown by the SODECOTON experience. 79. Average cotton yields stood at 1,500 kg/ha in 1988, about 75% of the northern region cotton farmers were making use of draft animal mechani- zation. In addition it can be safely assumed that practically all cotton growers are fully aware of cultivation techniques, i.e. seeding in line, - 17 - timing of seeding operations, proper weeding, etc; fertilization and pest control. It is therefore a sound SODECOTON decision to reduce general extension activities--without endangering the region's agricultural pro- duction. The reduction in SODECOTON's extension staff is helped by its promotion of AVPs and AVAs.51 80. The 350 AVPs represent a first stage in transferring activities from SODECOTON to the cotton farmers. Their main activity is cotton col- lection at village stores, grading and sale to the SODECOTON ginneries. As an incentive SODECOTON pays the AVPs a fee of FCFA 3/kg seed cotton. The amounts received are generally used to finance communal projects aiming at improved living standards of the villagers. 81. The second phase AVAs of which 219 existed by 1989, have chosen their owr, contact farmers, usually young farmers who have attended schools, who are expected to act as go-betweens, establishing lines of communication between extension/research facilities on the one hand and farmers on the other. In addition to having taken over some extension functions, the AVAs have been made for input sales next to the same marketing functions of the AVPs. C. Food Crop Production 82. All three projects had expected increased food crop production-- without the appraisal reports being specific about likely problems or issues associated with these crops. Although research results were recog- nized to be added, that extension advise would be given, etc. several other important questions remained unanswered by the appraisal. 83. First, a more detailed analysis of potential consumers is neces- sary prior to embarking on expansion of food crop production. In the case of the Northern Region (SODECOTON) project, the appraisal assumed that 15% of the Northern Region's population was non-agricultural, and had to be supplied with project generated foodstuffs. The population estimate was based exclusively on available statistics and did not take into considera- tion that most families in the northern townships still farm, either in the vicinity of the towns or in the villages of their origin. Local officials put the actual percentage of the region's non-agricultural population at maximal 5%. With the overestimate of the appraisal, actual market demands for staples amounted only to one third of the envisaged quantities. The project generated overproduction and led to price breakdowns of cereals in the region. 5/ AVP Association villageoise des producteurs--Village Association of Pro- ducers; AVA Association villageoise autogirie--Self-administered Village Association. - 18 - 84. Second, more attention needs to be paid to assessing food deficits of individual regions. It is not good enough to take national food deft- cita/import requirements as basis for food crop oriented projects/project components. Equally Important is the analysis if food deficits are only temporary or permanent. The Northern Region project serves again as an example. When reviewing production statistics it becomes apparent that only in years of drought, food deficits exist. As mentioned above project production led to market collapse in years of adequate rains. The conse- quences are indicated in a SODECOTON letter to the Bank dateds July 17, 1984. "...we have given up storing and marketing of food crops, especially sorghum. It is paradox to aim at improving output when, as a consequence prices will be depressed at the producers' level." The reason for the letter: sorghum prices had then fallen from FCFA 200/kg to FCFA 30/kg. 85. A similar observation was made in the Western Highlands project. In a letter to the Ministry of Agriculture, dated July 31, 1984, UCCAO pointed out that "..the adoption of hybrid maize by the farmers had led to surplus production and further to weak producer prices and marketing problems." 86. Third, consumer preferences play an important role and deserve more consideration during appraisal. Sorghum and millet are the main grain st*.ples in northern Cameroon while mainly maize is consumed in the center of the country and plantains/root crops are preferred over cereals in the rainforest area. Surplus production of basic carbohydrate staples in one area does not necessarily mean control over deficit situations in other parts of the country. In addition transport costs and accessibility to production areas play a significant part in the marketing prospects of food crops. Prevailing average transport costs of FCFA 30/ton/km limit the economically justifiable distances for food transports. 87. Marketing of food crops under the ZAPI project suffered from the lack or poor maintenance of feeder roads. Plantains and other easily perishable food crops rotted away in the villages or on the farms espe- cially during the extended periods of rain because no cars or trucks can reach the remote places of production. The Western Highlands project iden- tification report recognized the importance of access/distances to the marketing of staples: "..the project area is handicapped in this respect by its distance from, and poor comunication with Douala and Yaounde, the main consuming centers, which are already served by neighboring producing areas." 88. Fourth, lack of storage facilities are a major limiting factor to proper food crop marketing. Insufficient storage facilities can be found at two levels: (i) at regional level; and (ii) on the farms. The absence of grain storage at village/township/distriet/regional level is particu- larly felt in the northern parts of the country where in the Sahel zone, frequent droughts lead to substantial fluctuations in grain production. It is difficult to understand why the Northern Region (SODECOTON) project had an allocation for the improvement of village market places--later dropped-- but no provision for the construction of badly needed grainstorage. - 19 - 89. Proper analysis during appraisal would have shown that many farmers of the northern area are forced to sell the bulk of their grain crop immediately following the harvest due to the lack of on-farm storage facilities. In addition to the storage problem, some farmers are forced to sell because of cash needs, leading to the grotesque situation of some farmers having to repurchase grain at higher prices later during the year. In the context of on-farm storage of grain/other staples more research into protection of stores against insects/rodents would be recommendable. 90. Fifth, private traders fulfill a useful function--usually they are more efficient than state run food marketing organization, but these traders are also confronted with limitations. One critical constraint is their access to/availability of working capital, with commercial lines of credit usually unavailable for food crop trading by small traders. Another constraint limiting activities is their relatively small circle of opera- tion due to taking advantage of knowing individual producers and being able to personally supervising business transactions. Finally they depend on transportation capacity which is limited and usually consist of small pick- ups which, in contrast to heavy trucks, can negotiate poor roads during bad weather periods. 91. Sixth, the absence of adequate market intelligence. A case in point, the Western Highlands are ideally suited--by climate and soils-- for the production of a wide range of vegetables usually cultivated in moderate climates. The project contribution to substantial increases in the output of these vegetables and a surplus situation has developed in regard to potatoes and tomatoes. The area's export potential has not yet been fully * utilized due to insufficient information on market possibilities in neighboring African countries. It is paradox that urban markets of many West African countries rely on expensive, airlifted imports from Europe . while the same produce could be found in Cameroon. Bank support in study- ing possibilities of exporting vegetables to African markets would be of merit. D. Bottomlands Development 92. Two projects (Eastern Cameroon-ZAPI and Western Highlands) had provided funds for the development of "bottomlands", i.e. swampy areas in valleys, which were to be drained, protected against flooding and partly irrigated. The reclaimed/improved lands were expected by appraisal to be used for paddy production. While no bottomlands were developed by the ZAPI project, the same project component became a great success in the Western Highlands. 93. In the audit's view, the failure in Eastern Cameroon was predic- table. First, the low population density in the area combined with abun- dant land available, made farmers reluctant to invest their own labor into hard work required for reclaiming the swamps. Second, farmers were afraid to contact bilharzia in the stagnant waters of the swamps. Third, abundant rainfall and plentiful land assured them adequate yields from upland rice. Fourth, rice was not a traditional staple in the area and poor marketing/processing facilities were also not conducive to promote labor intensive rice cropping. - 20 - 94. Similar farmer considerations led to a changed approach in the Western Highlands. High population pressure made bottomlands development more interesting--even if health hazards existed. However, rice production was also found uninteresting, not only because of marketxng problems and eating habits, but because vegetable production, which had a long tradition in the project area, offered higher returns on land and labor. 95. A lesson learned is that vegetable production in the context of bottomlands development should be considered an interesting alternative, not only as high value crops where market outlets exist but also as contri- buting to improved farmer nutrition and health of the local population. E. Cancellation Procedures 96. After an extension of six months, a remaining total of US$4.7 million was cancelled from Loan 1919-CM, the Northern Region/SODECOTON project. The original closing date was December 31, 1986. By this date, the project had become a three-rated problem project and was rated to be deteriorating, due the Government's inability to provide counterpart funds. Extension of the closing date by the Bank was made contingent on submission by the Borrower (and SODECOTON) of a restructuring plan and a plan of action to rehabilitate the cotton sector. A response was promised to the Bank, and the closing date was extended to June 30, 1987 based on that promise. 97. By June 30, 1987 there was no further response from Government, but the Bank still did not formally close the loan, contrary to Bank procedures, in anticipation of reception of an adequate restructuring plan. A report by Price Waterhouse was given to a Bank supervision mission in October 1987, but turned out to be a financial audit and financial restructuring proposal, rather than a cotton sector development or policy plan. In the absence of a decision to extend the closing date, the loan was automatically closed on June 30, 1987, but even if an extension had been granted, no disbursements could have been made without revising the disbursement percentages in the Loan Agreement, since the Government counterpart funds were still not forthcoming. The loan account was kept open until February 1988 to honor project commitments made before June 30, 1987. In retrospect, the Bank should have sent the official confirmation of closing in July 1987 to avoid some subsequent confusion. MALI N I G E R C H A D SSUDAN * FA1 O *-4 N 1 ER e. 5 ~ EI -r NAFGRERCAAN C E .rE .Nd.,nn > PEPN' Z A R SC~AMEROONCAMEROON ZAPI~~~~ RUA DEEOMN PRJC P ECPTØ ZAARE GABON R OM UDR CAMEROON ZAPM RURAL DEVELOPMENT PROJECT PROJECT AREAS - - ZAP ZMNE ~OUARIES -- -P avedrods - 2r,G :q D !D t? 10 tEad ,Ralroads KILOMETERS International aipoS MILES Rivers - -- Provinda Boundaries -- - Iter-national Bundares 0v NIlG E R l A NO T HNk~e ) SO TH WEST SOUTHE CPUBLIC ES } }CENTRAL D NGUELENO irTToRAL °l• biN 4ATl A NTie OCEAN A G A B N PEOPLS REP. OF THE CONGO EQ GUINEA0 o - /~.... i IBRD 13072R b -30- T0*00 CAMEROON 011,0 KI~E ° _ _Western Highlands Rurol Development Project 0 2WESTERN PROVINCE 6•00 0os DELIMITATION OF PROJECT AREA Project area according to administrative boundaries Physical project orea Provincial boundaries Departmental boundaries - - District (Arrondissement) boundaries ---- Chief-dom (Chefferie) boundaries I Province headquarters g Prefecture headquarters o Subprefecture heodquarters 0 District headquarters Note: Chief-dom boundaries shown only in projeot reoa Magba 0 KILOMETERS L- j MILESO f. 0 S*W0 2 e WESTERN PROVINCE MAIN SOLS GROUPS r - Malantoven jj ffi--/ 0 Physical project area -Brown soils G boundary Black deposition soils - ----Provincial boundaries\ t] Province hendquarters Hydromorphic socls --- Rivers Red ferralitic solls r 7 Reworked ferralitic soils --•nternational L~~" *o boundories Dh.49 sa \h Massongom 9L same Bndio w NIGERIA CHAD JSontchoul *Kekem Bangn gat E RCRIA GUINCEA GAÄON A0 3 1 AP5*0198 걍·’ BIRD 13821 Magasins de copperatives existants CAMEROUN Amélicrés et transformés en centres de services PROJET DE DEVELOPPEMENT RURAL Nouveaux sites DES HAUTS PLATEAUX DE L!OUEST %ee Chef-lieu de province EID Chef-lieu de préfecture Routes de Desserte et Centres de Services Envisagés 0 Chef-lieu de sous-préfecture 0 Chef-lieu dcirrondissenient villages Routes principales Nkoupo Motopit Routes de desserre Cuurs d'eau 2? FOUMBAN la GALIMI] mites Kýpf*m[] Nkooie MBOUD a Ng-.,ýdm Nzindeng BATÎHAM Bofoudo FOUMBOT Ndo.be Fo.go T-go q PENKA MICHEL OUSSAM rý.go Ndeg G RANDJOUN 122 Fo Nffl. L Sontchu COMPANY BANGOU qJ L'IL, Bendenkop tu mww rý 0 z cc Z - ci z - Batchingou BAFANG Tm 0 A.1m KEKEM NGTE BANA OCTOBRE 1987