Cameroon - Northern Province Rural Development Project

Docwment of The World Bank FOR OmCIL USE ONLY Report No. 8420 PROJECT COMPLETION REPORT CAMEROON NORTHERN PROVINCE RURAL DEVELOPMENT PROJECT (CREDIT 1075-CM/LOAN 1919-CM) FEBRUARY 28, 1990 Agriculture Operations Division Occidental and Central Africa Department Africa Regional Office This doement has a resticted distribution and may be used by recipients only In the performance of ther oMcia duties. Its contents may not otherwise be disclosed without Worid Bank authorization.  CURRENCY DEFINITIONS Currency Unit: CFA Franc (CFAF) Fixed Parity: 50 CFAF - 1 French Franc WEIGHTS AND MEASURES Metric System FISCAL YEAR July 1 - June 30 GLOSSARY OF ACRONYMS CCCE Caisse Centrale de Cooperation Econoiique (France) CFDT Compagnie Fransaise pour le Ddveloppement des Fibres Textiles CNCE Centre National du Coanerce Ezt6rieur CRF Centre de Recherche Foresti&re CTFT Centre technique Forestier Tropical FNFP Ponds National Forestier et Piscicole FONADER Fonds National de Developpement Rural FSAR Fonds Special d*Actions Rurales IRA-Nord Institut de Recherches Agronomiques, Maroua IRZ Institut de Recherchps Zootechniques MESRES Ministere de l'Enseignement SupAricur, de l'Informatique et de la Recherche Scientifique MIDEVIV Mission de Ddveloppement des Semences et des Cultures Vivribres MINAGRI Ministire de l'Agriculture MINDIC Ministere du Developpement Industriel et Commercial MINEL Ministare de l'Elevage MINEQ Ministere de l'Equipement MINEP Ministere du Plan MINEPIA Ministere de l'Elevage, de la POche et des Industries Animales MINFIN Ministare des Finances NSMP National Seed Multiplication Project MINPAT Ministere du Plan et de l'Am6nagement du Territoire oC Office Cdrealier ONAREF Office National de Regeneration des Forets ONCPB Office National de Commercialisation des Produits de Base ONDAPB Office National de D6veloppment Agricole et du Petit Betail OPV Office Pharmaceutique Vetdrinaire PCC Project Coordinating Committee PMEU Project Monitoring and Evaluation Unit SEMRY Soci6td d'Expansion et de Modernisation de la Riziculture a Yagoua SODECOTON Societe de DAveloppement du Coton du Cameroun USAID United States Agency for International Development  FOR OMCUL usE ONLY rHE WORLD SANKt Washgton. D.C. 20433 U.S-A. OU1. al Ovt4r-G.wMa Opmtatom Ivahiasa February 28, 1990 HEMORANDUM TO THE EXECVTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report - Cameroon Northern Province Rural Development Proiect (Credit 1075-CM/Loan 1919-CM) Attached, for information, is a copy of a report entitled "Project Completion Report - Cameroon Northern Province Rural Development Project (Credit 1075-CM/Loan 1919-CM)", prepared by the Africa Regional Office. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment This document has a restficted 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 authofization.  FOR OMCIL USE ONLY PROJECT COMPLETION REPORT CAMEROON NORTHERN PROVINCE RURAL DEVELOPMENT PROJECT (CREDIT 1075-CM/LOAN 1919-CM TABLE OF CONTENTS Page PREFACE .. *#so ... . . ................. i BASIC DATA SHEET . ...............................................i EVALUATION SUMMARY ............................. ....................V I. INTRODUCTION . ........................................ .. . 1 II. PROJECT IDENTIFICATION AND PREPARATION ..................... 2 III. IMPLEMENTATION .................................. 6 IV. OPERATIONAL PE.FORMANCE AND PROJECT IPACT .................. 11 V. FINANCIAI. AND ECONOMIC PERFORMANCE . .................... *.. 19 VI. INSTITUTIONAL PERFORMANCE.. ........ ........ 22 VII. BANK PERFORMANCE ............................................ 23 VIII. CONCLUSION . ........................ ................. 25 ANNEXES 1. Disbursements by Project Component and Disbursement Category 2. Cameroon Northern Province Rural Development Projects Selected Annual Production Indicators 3. Seed Cotton Production, Acreage and Yields, Cameroon: 1970-1986 - Table 1 and Figure 1 4. Cameroon Northern Rural Development Project - Economic Rate of Return Worksheet - Base Case Analysis and Sensitivity Analysis HAP: IBRD No 21508 Cameroon Regions and Main Crops I 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.  - i - PROJECT COMPLETION REPORT CAMEROON NORTHERN PROVINCE RURAL DEVELOPMENT PROJECT (CREDIT 1075-CM/LOAN 1919-CM) PREFACE This is the Project Completion Report (PCR) for the Northe n Province Rural Development Project in Cameroon, for which Credit 1075-CM in the amount of US$12.5 million and Loan 1919-CM for US$25.0 million were approved on November 18, 1980. The credit and loan were closed on June 30, 1987, six months behind schedule. The credit was fully disbursed, and the amount of Us$4.72 million of the loan was cancelled. The last disbursement was on February 24, 1988. The PCR was prepared by the Agriculture Operations Division of the Occidental and Central Africa Department, Africa Regional Office, and is based, inter alia, on the Staff Appraisal Report; the Loan and Credit Agreements; supervision re'3orts; correspondence between the Bank and the Borrower; and internal Bank -.zeinoranda. The PCR was read by the Operations Evaluatic Department (OED). The draft report was sent to the Borrower on December 7, 1989 for comments, but none were received.  - ii - CAEOON NOlTMEN RUIL DEVELORMENT SOJECT KEY PROJECT DATA -Appisgal A eual or AcWal as S of Esti"mt Estimatd Acual Aceraisal Esltmate Pro.ect Costs (USm) 74.7 n.a. Credit and Loan Amount (US$m) 87.6 82.6 67 Date %ord - N.A. 11/U Date efetivenes N.A. 66/61 Closing &ate . 12/6S 06167 Economic Rate of R"urn (X) J2.4 t1 Numbr of diect benetiiearlos 163, 11119. 67 CWUILATIE DIh11MEN FY62 FP6 FVP64 FY6 FYN FY6T FP6 Appraisal Estimate (USImillten) 5.0 18.2 19.8 26.6 8.6 87.6 - Actual (UIStellIIOn) 2.4 0.4 16.9 19.2 24.7 W.0 32.s AcUal as* Of Estimate 40 64 72 72 76 *2 67 DatO Ot final dlsUrsq.pmt 21241S Principal repaid (USS$m) 4.6 MISSLON DATA No. Of Mandays Special liat in Per?. Typ Of Mission Pat& ( P - In Field m reet I/ Retina V/ Leg e / ProblI es Identification / - - - - - - Prepration 11/77 - - - - - - Appraisal 12/79 7 SUP.vieIO. I 1U/01 1 1 f 1 2 b Supervision 2 12/62 1 9 * 2 2 T,P Supervision J 1/"8 2 1S 2 2 T Supervision 4 8/84 2 6 ,b 2 2 T Supervisio S 12/64 1 19 a 1 1 T SupervisionO 5 0/5 8 1S s*,e, 1 2 T Supervision 7 12/65 2 N.A. N.A. 2 2 F,P SupervioI1n 6 S/O6 2 N.A. N.A. 2 8 F,P Supervision 9 1/67 1 N.A. N.A. 8 * F,P OTHER PROJECT DATA Borro r ove rn nt of Cameroon Executing Agency Societe de 10el Oppsm s O doa Cn du Cameroun (SODECOTON Flil year Jkly 1 - June SO Nam of currecy (abbreviation) CFWA Currency Excange Rat: Appraisl Year Averag UWIM * 219 CfAf nte rv ning Yeare Averag U0S1.11 a *75 CcAr Completion Year Avera US1.6 * 26 CPA? Follow-up Project: Non )/aagricvlterist; bagriculteral economist; tlinaU ial analys; I gsonemlot; eururaI enginer; #agronomist. /lmproblem-rze or omnor problems; 2mederat problems and Sneajor problems. 3/lnimproving; 2luttionary; an udsterloratlia. i/Winancalol Unesnagerial; Tutechnical; P1woliticl and Obotier.  - lli - BASIC DATA SHEET (Cont' d) STAFF INPUT 11 n4 15 lb U nI n9 o t 02 I U os b 61 UWIh wm~ _ OEIII_IIE RIIIIiilt v . . . . 10.0 9.2 5s 1.0 . . . . . . . . m.? FM,l21. . . . . . . .2 9.5 .3 . . . . . 9.0 IIIUIIS . . . . . . . . It.? . . . . . . . 89.1 O I, . . . . . . . . 5.1 IS1 20.7 It.3 I.2 .L IL9 2.9 I. 1O1n . . . . . .2 . .2 .1 . . . . . . . .5 '1TO NUMC XA 9 .0 .6 .0 .0 Q 99,4 5.1 202. U5. 15.1 0..2 26.6 l..9 2.9 332.9  - iv - PROJECT COMPLETION REPORT CAMEROON 'iORTHERN PROVINCE RURAL DEVELOPMENT PROJECT (CREDIT 1075-CM/LOAN 1919-CM) EVALUATION SOMMARY Obiectives 1. The project sought to provide the means for development of profitable cash crops other than cotton in areas of Northern Cameroon, notably groundnuts and maize, and to improve the productivity of staple food crops production (mainly sorghum), while maintaining the momentum on cotton intensification. 2. The specific objectives of the project were to (i) improve rural infrastructure, including roads, stores and marketing facilities; (ii) provide effective services, including input delivery, supervised credit, extension, marketing, adaptive research, and project monitoring and (iii) support forestry development in the North; (iv) provide financial and technical assistance to SODECOTON; and (vi) help existing administrative structures at the provincial level. Implementation Experience 3. The project was approved in Noveuber i8C and became effective in June 1981. After a five-year implementation period, the project was extended for a six-month period and closed on June 30, 1987, about six and a half years after Board Approval. 4. Several events occurred during project implementation that had a bearing on project implementation and impact: - the dollar appreciated sharply relative to the CFA franc from 1981 to 1985 and depreciated just as sharply in 1986-1987, with substantial repercussions for project finances. - the North Province was subdivided in 1984 into three provinces (Far North, North and Adamaoua). - the five growing seasons in project period were characterized by well below normal precipitation levels, - the cotton price peaked in 1980 at a little over two dollars a kilo and fell to just over a dollar in 1986. After 1985, further decline in world cotton prices combined with a large decrease of the value of the dollar in 1986-1987 led to a major financial crisis for SODECOTON. Results 5. The project included over twenty separate activities and most were implemented satisfactorily and on schedule. In particular, activities focusing on extension agent training, seasonal credit and internal SODECOTON management proceeded smoothly. Despite this, not all of the project objectives were achieved, or are today sustainable.  6. What was designed to be a rural development project with cotton as the most important crop turned into a cotton project with few substantial non-cottcn activities. The SAR overestimated the viability of expanding food crop production, as well as the incentives to farmers to do so. Maize cultivation turned out not to be practical under prevailing precipitation levels, while sorghum and groundnuts production growth was constrained by limited marketing possibilities. During implementation of the project little was done to promote agricultural extension in the non-cotton growing areas. The poorest area in the region, the Mandara Mountains. was excluded because cotton is not produced there. 7. The financial performance of SODECOTON, the main project implementing agency, was overshadowed by the fluctuations in the price of cotton. Financial returns to farmers throughout the project period were satisfactory. Economic performance of the project has been more favorable (10) than is suggested by the lack of financial profitability of SODECOTON. Sustainability 8. The Bank paid insufficient attention to the sustainability of project results. By 1986, it was clear that many of the project activities would end with the end of Bank funding, as continued counterpart funding was increasingly unlikely, and would undermine project results with sizeable recurrent expenditure implications. It was foreseeable that the project's end implied the slow decay of the feeder roads built or rehabilitated by the project for example. Findings and Lessons Learned 9. The success of this project was undermined by a series of exogenous events which could not have been predicted or controlled. Drought, the wide swings in the value of cotton and of the CFAF, and the growing government financial crisis undermined project implementation and the sustainability of its achievements. 10. The project aesign followed Bank and government policy in Cameroon at that time of implementing the project through the existing relatively efficient and experienced SODECOTON, rather than using the project for the longer run objective of increasing implementation capacity in the central administration. SODECOTON expanded into a number of non-cotton activities, such as extension, training, industrial training, village level infrastructure, food crop marketing and road maintenance, often at the behest of international donors like the Bank, because of the deficiencies of the central administration. Unfortunately, it is unlikely that much of the experience gained by SODECOTON in rural development with cotton growing farmers can be readily utilized by the other Government implementing agencies to benefit the majority of non-cotton growing farmers of the North and Extreme-North Provinces. Financial consider. tions dictate that SODECOTON will have to concentrate its efforts or, cotton and cotton farmers for the foreseeable future, and separate plans will have to be made for MINAGRI and associated Government institutions to assist the other two- thirds of the region's farmers.  I. INTRODUCTION 1.01 Despite the importance of oil in recent years, Cameroon's economy is still fundamentally based on agriculture. Over the period 1971- 75 prior to the advent of oil in late 1970's, agriculture contributed about 302 to GDP and agricultural exports accounted for 821 of the total. In 1984-85, agriculture only accounted for 202 of GDP while agricultural exports fell to 282 of total exports. However, the sector still provides employment to about three-quarters of the population, and contributes about 15 to 202 of state revenues. Agricultural exports have regained their importance in recent years however, and accounted for about half of total exports in 1987188. More importantly, it produces almost all of Cameroon's food and serves as the main source of raw materials for the agro-industrial subsector, the potential of which has not yet been sufficiently developed. 1.02 Although the shares of various components of the agricultural sector have varied over time, the food crop subsector has continued to provide the main agricultural contribution to GDP, which accounted for about 542 df value added in agriculture over the last 18 years, export crops about 212, livestock and fisheries about 162, and forestry about 92. 1.03 The Northern Province Rural Development Project (NPRD) fell within Govarnment's overall rural development strategy in the North, which comprised four main types of actions (a) relieving pressure on land in the Center North by providing infrastructure and services for settlement in the Benoue Valley; (b) use of irrigation of the main water courses in the region; (c) raising the productivity of rain-fed agriculture in conjunction with efforts to improve rural infrastructure (particularly water supplies) in the most densely populated areas and (d) exploiting the unpopulated Guinea savannah grasslands of the Ademous Plateau for livestock production. The NPRD was intended by Government to contribute an important part in both the implementation and the direction of the above development strategy. 1.04 The NPRD was the fourth Bank supported Rural Development Project in the country. The other three projects, Rural Development Fund (FY77), Plaine des M'Bo (FY78), and Western Highland (FY79) were still being implemented at the time of appraisal. To-date, the Bank has financed a total of 24 projects in agriculture amounting to USS 496.06 million over the past 20 years (1967-1987).  - 2 - II. PROJECT IDENTIFICATION AND PREPARATION A. Identification. Preiaration and Appraisal 2.01 The NPRD originated from needs identified in the Bank's agricultural sector survey in 1974. Following a formal Government request for assistance in November 1978, a Bank identification mission isited Cameroon in May-June 1977. The project was then prepared by FAO/CP which visited Cameroon twice, in October/November 1977, and in March 1978, when consultants from Compagnie Francaise pour le Developpement des Fibres Textiles (CFDT), financed under Technical Assistance Credit 673-CM, assisted in preparation of the Project. The Project was appraised by a seven-man mission in November-December 1979. The Project was approved by the Board in November 18, 1980. B. Proiect Area 2.02 The project area covered 19,700 square kilometers of the Center-North, a Sudano-Sahelian ecological zone of the then Northern Province, and comprised of the departments of Margui-Vandala, Diamare and Mayo-Danay (except the SEMRY area along the Logone River) 1/. In addition, the project would include the arrondissement of Guider in the Benoue Department. This area had an average population density of 61 persons per square kilometer, with a total population of 1.34 million people estimated by the 1980 Census. Excluded from the project were the mountainous areas of Mandara in Margui-Wandala department, in part because the USAID was then con:emplating a rural development project in that area and in part because SODECOTON was not active in the Mandara Hills where there is little or no cotton potential. Since that area was the most densely populated in the project zone, the project area had an estimated population density of 58 persons per square km. 2.03 The Bank identification mission originally suggested the project cover the entire Northern Province, but the Appraisal mission agreed with the Government's view that the project should be limited to the Central North region, in the middle of the Province. The focus on the Center-North was justified by the fact that it contained two thirds of the rural population of the then Northern Province, and that soil fertility was declining with farmers having a hard time maintaining yields. Thus, the Appraisal mission recognized the limited long term potential of the project area and its environmental problems even if it did not address the issue of declining precipitation levels. 1/ The Northern Province was subdivided into Far North, North and Adamaoua Provinces by presidential decree in 1984, in the middle of project execution. At the same time, the old Department of Margui-Wandala was sub- divided into Mayo-Sanaga and Mayo-Louti Departments, and also the old Diamare Department was divided into Diamare and Kaele Departments. Unless otherwise noted, this report will designate Northern Province as the new, smaller geographical area (see Map IBRD 21508).  3- C. Proiect Approach and Justification 2.04 The Appraisal saw the need for a twin approach to rural activities in the project area. Cotton intensification under the direction of SODECOTON in Northern Province since 1975 had resulted in rapid increases in cotton production and yields. Now the Government sought to diversify cautiously away from cotton. As the SAR explained, "the next phase in this strategy is, while maintaining the momentum on cotton intensification, to (a) provide the means for developing other profitable cash crops in areas where cotton is not suited, notably groundnuts and maize; and (b) to improve the productivity of staple food crops productions (notably of sorghum), in line with expected increases in demand' (SAR, p. 6). This strategy provided the justification for the rural development activities of the project. The decision to situate these activities in SODECOTON, and not in MINAGRI, was based on the view that SODECOTON was the most efficient development institution in the project area, on the one hand, and the then prevalent view, on the other hand, that parastatals outside the ministerial structures were the proper vehicles for development efforts. 2.05 The SAR noted technical and institutional deficiencies in the administrative structures in charge of rural development in the north and the second set of activities was designed to address these shortcomings, with assistance to strengthen extension, credit and agricultural research, as well as regional planning and project monitoring and evaluation capacity. As such, the project was strongly influenced by Bank integrated rural development projects being implemented elsewhere in Africa, including in Cameroon itself. D. Objectives and Description 2.06 The specific objectives of the project were: (a; to provide: (i) an improved rural infrastructure, including roads, stores and marketing facilities; (ii) effective services, including input delivery, supervised credit, extension, marketing, adaptive research, and project monitoring and (iii) to support forestry development in the North; (b) to provide financial and technical assistance to SODECOTON required to discharge its broadened responsibilities, including management of the proposed project. At the same time, assistance would be provided to facilitate a rationalization of agricultural extension in the project area by equipping the remaining HINAGRI  - 4 - structure with the means to perform its other tasks more effectively; and (c) to help existing administrative structure at the provincial level to participate more effectively in the planning, coordination and control of all development projects in the Northern Province, within the framework of national development policies; to extend the agricultural project monitoring and evaluation unit being formed by MINAGRI for the Northern Province; and to assist the Agricultural Research Institute for the North (IRA-North) enhance its research efforts on food crops. 2.07 The project was comprised of two sets of activities within what was then the Northern Provinces - rural development activities in the project area itself (Center- north), including infrastructural improvements, reforestation activities and food crop development. - at the provincial levels (a) to establish competent institutions for planning and coordination of rural development in the Province and for project monitoring and evaluation; (b) to establish a sound research base for future agriculture and forestry development in the North, including adaptive research for food crops. 2.08 The SAR outlined the following project components in the Center northt (a) strengtheni. g SODECOTON agricultural extension services in the project area by increasing the number of its extension sectors from 19 to 23, the number of its extension zones from 75 to 150 and of its monitors from 560 to 600; (b) establishing an animal husbandry extension unit within SODECOTON, employing two veterinarians and 20 livestock 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 training centers would be built and training specialists provided; (d) an agricultural experimentation unit within SODECOTON would be established to carry out field trials and demonstrations at two new field plotc in each of the 23 sectors in the area; (e) staff and operating costs required by SODECOTON for the provision of these project management services. (f) construction of additional field storage facilities to handle farm inputs required for the project;  - 5 - (g) improvement of about 800 km of feeder roads in the area; (h) improvement of rural market place facilities to ease marketing of food crops within the area; Ci) financing of incremental costs of seasonal inputs used by farmers and prcvided by SODECOTON; (j) financir@ cf an agricultural credit program for participating farmers .uquiring credit to buy draft oxen and equipment; (k) pilot tree plantations in forest reserves for fuelwoood and poles and plantings of windbreaks in densely farmed areas to reduce erosion and improve soil fertility. 2.09 The SAR described the following project components for the Northern Province as a wholes (a) strengthening the provincial planning unit at Garoua (NPED) to improve planning and coordination of development projects; (b) support for agricultural research at IRA-North, notably for adaptive research on food crops, and to strengthen the Savannah forest research unitt (c) purchasing additional seed handling equipment required by MIDEVIV (NSMP) to furnish improved groundnut seed required for development projects in the north; (d) providing new vehicles and improved office facilities for MINAGRI, as well as staff training; (e) establishing a monitoring unit under NINAGRI to monitor and evaluate results of this and later agricultural projects in the north; E. Proiect Organization and Management 2.10 The Project required the collaboration of eight agencies operating in the Center-North (SODECOTON, MINAGRI, FONADER, MINEL, MINEQ, local councils, FNFP and MIDEVIV), and two more (the Provincial Economic Division of MINEP and IRA-North) at the Provincial level. 2.11 SODECOTON would play the lead in assuming the execution and supervision of all components and would be responsible for the physical execution of the agricultural development program and construction of feeder roads in the project area. (At the national level, responsibility for overseeing the implementation of project activities would rest with the corresponding responsible Ministries (MINEP, MINAGRI, MINFIN) and other technical Ministries concerned such as MINEQ for roads and DGRST for research coordination at provincial level between the various executing agencies and other institutions involved in the project would be assured by  - 6 - a Project Coordinating Committee (PCC) to be established under the authority of the Governor of the Northern Province. F. Prolect Costs and Financing 2.12 Total project cost, including taxes, amounted to US$74.7 million of which the Bank would finance a Bank loan of US$ 25 million and an IDA Credit of US$ 12.5 million, making a total Bank group participation of US$ 37.5 million or about 50Z of total project costs. III. IMPLEMENTATION A. Effectiveness and Start-up 3.01 The project was approved in November 1980 and became effective in June 1981. After a five-year implementation period, the project was extended for a six-month period and closed on June 30, 1987, about six and a half years after Board Approval. Estimated targets and actual accomplishments for key project components are given in Annex 1, Table 1. 3.02 Start-up was relatively uneventful and rapid, not least because of SODECOTON's experience and management capacities. Several expatriates hired as experts by the project were already in Garoua working for SODECOTON, for example and were immediately operational. Activities not directly under its control started more slowly and the SAR proved unduly optimistic regarding implementation schedules. For example, livestock assistants to be provided to SODECOTON by MINEL were not recruited until mid 1983. The Northern Seed Multiplication Project (NSNP) proved incapable of supplying the desired quantities improved quality groundnut seeds and alternative sources were not sought before the 1985/86 growing season. Finally, FONADER did not meet the loan condition of submitting to the Bank a report on the results of loan recovery efforts in the north indicating a recovery rate of at least 75 percent. These issues are discussed at greater length below. 3.03 The Project Coordinating Committee (PCC) convened for the first time in May 1981 and again in August and appeared to play a satisfactory role in the early phases of the project. B. Progress and Revisions 3.04 Several events occurred during project implementation that could not have been predicted at Appraisal but that were to have a bearing on project implementation and impact: (1) the dollar appreciated sharply relative to the CFA franc from 1981 to 1985 and depreciated just as sharply in 1986-1987, with substantial repercussions for project finances. Appraisal had assumed a stable exchange rate of 210 CFAF per dollar throughout the life of the project.  In fact, exchange rates varied widely, climbing from a low of 206 CFAF per US dollar in the third quarter of 1980 to a high of 498 CPAF in the first quarter of 1985, and then back down to 260 CFAF in late 1987. (2) the North Province was subdivided in 1984 into three provinces (Far north, North and Adamaoua). 80 percent of the surface of the project was in the new Far North Province, while the planning and evaluation and monitoring units of the project continued to be based in Garoua, in the regional delegations for MINAGRI and MINPAT of the new North province. Further, technical assistance to MINAGRI extension was also based in the North Provincial Delegation at Garoua. The Bank mission of July 1985 (over a year after the subdivision) proposed that these activities all be moved to the Provincial delegations of the Far North Province in Maroua, but this was never done. (3) The five growing seasons under the life of the project were characterized by well below normal precipitation levels, perturbing the agricultural cycle and SODECOTON operations, and leading to production shortfalls, not only for cotton but also for food crops. (4) The Appraisal had predicted modest real increases in the world cotton price over the life of the project, from the level of 181 US Cents a kilo in 1980 (Mexican strict middling 1 1/16 inch, CIF Northern Europe). In fact, the cotton price peaked in 1980 at a little over two dollars a kilo and fell to just over a dollar in 1986. The sharp appreciation of the dollar vis a vis other major trading currencies between 1980 and early 1985 protected SODECOTON's cotton revenues, since cotton is traded in dollar terms. After 1985, the further decline in world cotton prices combined with a large decrease of the value of the dollar in 1986-1987 led to a major financial crisis for SODECOTON, particularly since the Government's budgetary difficulties led it to decrease its level of support for SODECOTON activities. C. Costs and Financing 3.05 Total project costs, including taxes, were estimated at appraisal at CFAF 15,690 million, or US$ 74.7 million with a foreign exchange component of 52 percent or CFAF 8,190 million (US $39.0 million), based on a exchange rate of 210 CFAF per US dollar. Actual project costs are not available except for those components of the project which were partly financed from the Bank loan and IDA credit, which are discussed below. The disbursements records show that almost all project exenditures were made in CFAF or in French Francs. It is therefore interesting to reflect that the loan generated considerably in excess of the estimated needs of CFAF 8 billion, as a result of the appreciation of the US dollar, even though $4.7 million was eventually cancelled. 3.06 The financing plan proposed in the SAR envisaged financial inputs from the Government, the Bank, SODECOTON, farmers and FONADER as follows:  -8 SAR ($0000) Government 27.5 Bank 37.5 Farmers 9.3 FONADER 0.4 SODECOTON 0.0 Total 74.7 3.07 There are no figures for the actual amount that was made available by the Government to SODECOTON, NMINGRI and the other agencies implementing the project, but it is clear that from 1985 onwards the Government's contribution to SODECOTON was reduced, so that SODECOTON itself had to contribute towards incremental operating costs of extension from its own resources in order for implementation of the project to continue. This lack of Government counterpart contribution was one of the most important factors affecting implementation of the project from 1985 onwards. However, the reduction in Government budgetary allocation may have been exascerbated by the provisions of the loan agreement that required a rapidly increasing Government contribution to incremental operating costs in the latter years of implementation. Thus, whereas initially the Government was only required to finance 10 percent of incremental staff salaries, the percentage increased to 30, 50 and 70 percent through the project period. Even without the national financial crisis, it might have been difficult for SODECOTON to obtain such rapid increases in Government recurrent budget allocations. 3.08 Although actual costs are not available, the Bank's contribution to each of the categories is available from the disbursement records. They are smaller than the total costs, since the Bank only financed a proportion of the costs of each component, but they can be used to estimate costs. Moreover, since separate disbursement categories were included in the loan and credit agreements for the different institutions responsible for implementing the project - SODECOTON, IRA, FNFP, FONADER, MINAGRI, MINEP and MIDEVIV, it is possible also to obtain some indication of the performance of the different agencies. 3.09 Annex 1, Table 1 presents the disbursements for each category and for each of the parts of- the project. The data, which is summarised in the table below, confirms that SODECOTON was able to implement the project more or less as envisaged in the SAR. Incremental staff salaries and other operating costs, as well as specialists' services were within 10 percent of appraisal extimates. There were some savings on vehicle costs. Costs of civil works, mainly the roads construction program, exceeded SAR estimates by 60 percent. IRA was also able to utilize the funds that were made available for research on food crop and trees. However, costs of the FONADER implemented medium term credit program for draft animals and  -9- equipment were only 30 percent of appraisal estimates, and equipment purchased for the MIDEVIV executed seed production program was only 40 percent of the amount estimated. Moreover, the costs of the MINAGRI and MINEQ programs were only 60 percent of appraisal estimates. Proiect Costs Financed by Cr.1075-CM and Loan 1919-CM Proiect Component SAR Actual Actual ($000) ($000) as 2 of SAR - Provision of extension services by SODECOTON - Operating Expenditures 9,552.1 10,384.7 109 - Civil works 4,486.8 7,312.2 163 - Specialist' services 6,056.3 5,472.0 90 - Vehicles and equipment 3,928.9 3,239.5 82 sub-total 24,024.1 26,408.4 110 - Strengthening of agricultural 2,707.9 2,806.6 104 and forestry research by IRA - Pilot tree plantings, and 1,386.8 1,145.6 83 windbreaks by FNFP. - Medium term credit program for 3,182.6 958.5 30 animal traction - Strengthening of MINAGRI and 1,755.3 1,036.1 65 Project Monitoring and Evaluation. - Development Planning 557.4 360.7 65 by MIMEP - Seed Production by MIDEVIV 195.8 71.2 36 Contingencies 3,690.0 0.0 - Total 37,500.0 32,787.2 87 D. Procurement and Disbursement 3.10 The project files indicate that there were no major procurement problems during implementation. 3.11 Disbursements of the Credit 1075-CM started towards the end of 1981 and were completed in early 1984, at which time disbursements from the Loan 1919-CM commenced. Disbursements lagged somewhat behind the appraisal estimates, largely as a result of the appreciating US dollar exchange rate. As early as June 1985, it was anticipated that appreciation of the dollar would result in underspending of project funds, perhaps by as much as US$ 17 million (Memo, June 5 1985, p. 2). In early 1985 SODECOTON proposed that it prepare a work plan of activities to undertake for two  - 10 - additional years with these funds. Horeover, the Government and SODECOTON proposed that these funds be used to finance a ginnery at Bibemi, and provided the Bank with detailed cost estimates for its construction. However, due to lack of availability of Government counterpart funds, the revised work plan was never fully implemented. Nevertheless, all but USS 4.7 million of the loan was disbursed by project closing. E. Reporting 3.12 According to the SAR, the project's three executing agencies, SODECOTON, MINAGRI and FONADER would each submit quarterly progress reports to the Government and to the Bank showing actual and budgeted expenditures, statements of progress achieved and objectives for the forthcoming quarter by each project component. SODECOTON's quarterly reports were to contain summary financial information indicating the levels of its short-term borrowing and of any outstanding receivables from the Government or other bodies with respect to budgetary allocations for the project and subsidies on production inputs. In addition, SODECOTON was to keep records for the project consistent with sound accounting practice to reflect not only its overall financial situation, but also the detailed costs of operations in the project components it was to execute. SODECOTON's existing cost accounting system was to be revamped to allow it more accurate cost allocation in the context of its new role as a more comprehensive development agency. MINAGRI's project accounts were to be kept by the Provincial Delegate in Garoua, in accordance with standard goverament audit procedures. 3.13 These procedures were followed satisfactorily. SODECOTON's annual and quarterly reports were regularly submitted to the Bank, as were detailed reports of expenditures. Similarly, after a slow start, MINAGRI's evaluation and monitoring unit provided regular reporting of its activities. SODECOTON's accounting system was adequately modernized, and now clearly distinguishes commercial from non commercial activities. The new approach has been quite useful to SODECOTON by providing precise cost estimates of SODECOTON's different activities, notably the rural development ones for which the agency receives no reimbursements from the Government. These estimates have been used by the Government, CCCE and the Bank in the recent debates over SODECOTON's rehabilitation. F. Proiect closing 3.14 Project activities were initially supposed to end June 1986 and the initial closing date for the project was December 31, 1986. SODECOTON requested a two year extension to the project, made possible by large unspent funds. The ostensible justification for prolongation was to continue and intensify project actions to alleviate the effects of drought. In addition, the onset of SODECOTON's financial difficulties made any additional funding highly desirable.  - 11 - 3.15 Bank missions initially agreed informally to SODECOTON proposals to extend the project with a new and revised work plan for 1987 and 1988. Such a work plan was designed by SODECOTON in early 1986. But it was soon overtaken by the deteriorating financial situation. With the Government's own financial situation worsening rapidly, counterpart funds for the project were increasingly being withheld. By mid 1986, the Bank continued to accept the principle of extension but now made it contingent upon agreement with the Government on a plan of action for the rehabilitation of the cotton subsector and agreement on counterpart funding. After some internal debate the Bank agreed to provide the project with a first six month extension of the closing date, to June 30 1987, to provide the Government enough time to complete an in depth study of SODECOTON's finances and rehabilitation. 3.16 In February 1987, the Bank sent the Government a telex urging the Government to alleviate SODECOTON's financial problems and disburse obligated funds to SODECOTON. The telex warned that if adequate measures were not taken, disbursements would be stopped, no second extension would be granted and the balance of funds under the loan would be cancelled. The Government responded shortly thereafter indicating that counterpart funding was now assured for SODECOTON and that a ministerial review of the parastatal was underway. In July 1987, the Bank informed the Government that it had not received any information concerning the ministerial review, and that consequently the loan closing date would be maintained at June 30, 1987. 3.17 Undisbursed Bank project funds still totalled US$ 5.6 million as of September 30, 1987. A Bank mission to Cameroon in October 1987 was given a study, by Price Waterhouse, which included a number of recommendations based on a comprehensive diagnostic of SODECOTON's crisis. The mission was informed that the Government accepted most of Price Waterhouse's recommendations and would begin implementing them. In view of this new information the Bank mission indicated it would reconsider extending the closing date. After some internal debate in the Bank, however, the closing date was maintained and no further extension granted. The Bank now argued that additional finance to SODECOTON be linked to the major policy dialogue underway in the context of proposed sectoral and structural adjustment lending. IV. OPERATIONAL PERFORMANCE AND PROJECT IMPACT A. Issues 4.01 An evaluation of project operations and their impact should take into account the environment in which the project activities were carried out. This is particularly true for the present project. First, it should be noted the project covered only 4 growing seasons. Though it was signed in January 1981, the project really only got started in June 1981,  - 12 - too late to have any effect on the 1981182 season. The 1982/83 season was the first season to benefit from project inputs, and thus at the original project completion date (June 1986), only 4 growing seasons (not including the 1986/87 season) had been concludled under the life of the project, or 5 years with the additional year of cperations, which did not include all project components, and was conducted in the shadows of the Government's financial difficulties. This is little time for technology diffusion in a rural environment, even when -- as in the case of cotton -- a tested technological package already existed at SODECOTON. For most of the food crops the project included a research and field test component to finalize the development of such packages because none existed at project inception. For these crops, it is therefore not surprising that project impact was limited in terms of production and productivity increases. The SAR contained unrealistic expectations in this regard. 4.02 In addition, the project was perturbed by a series of negative exogenous events, whose effect must be properly taken account of. First, precipitation levels were below average for 3 of the 5 growing seasons during the life of the project. The Appraisal argued that 700 m over the course of the June-September growing season represents a minimal requirement for successful cotton cultivation. From 1982 through 1984, the project zone averaged less than that, with such lows as 410 mm at Maroua in all of 1984 or 630 mm in Garoua in 1983. The drought had an effect not only on the agricultural potential of the project area, but also on farmer incentives. In particular, drought can be expected to lower the potential of all cultivations, albeit of some more than others. The lower level of precipitation apparently made maize cultivation all but impossible, for example, but must have had a much lesser effect on millet. More importantly, drought put a premium on food crops in farmer strategies, particularly when as is the case in northern Cameroon, food marketing networks are poorly developed. The drought could thus be expected to have had a disproportionate effect on cotton. The drought conditions prevailing during much of the project's existence undermined project results. Since 1985, however, the rainfall levels have been satisfactory and production levels have increased appreciably. Indeed, the 1988/89 season provided SODECOTON with its largest cotton crop ever, some 168,000 tons of seed cotton. 4.03 Secondly, in 1985, the world price of cotton underwent a dramatic and unexpected decrease. This combined with the appreciation of the dollar to make SODECOTON's cotton activities highly unprofitable, not least because of the generous producer price increases the governement granted cotton farmers between 1980-1985. SODECOTON's financial difficulties were compounded by the Government's own budgetary crisis which led it to reduce its support of SODECOTON, and other relevant institutions, notably of IRA. Additional projects, such as the Bank's National Agricultural Research Project were delayed due to lack of counterpart funding, which undermined continuity and eroded progress made under the project. 4.04 A final factor which perturbed project operations was the Government's decision to subdivide the Northern Province in 1984. This change was followed by indecision on the Government's part as to whether or not project activities taking place in the ministerial Provincial  - 13 - delegation in Garoua should be physically moved to Maroua or not. This hesitation can be expected to have perturbed planning and perhaps morale in the Delegation at Garoua. Once the decision was made to move most of these project activities to Maroua, more time and energy was lost moving equipment and personnel and starting new activities from scratch at Maroua. The low degree of impact of the project activities conducted in MINAGRI and MINEP were at least in part a result of this move. B. Proiect Components 4.05 The project included over twenty separate activities and most were implemented satisfactorily and on schedule. In particular, activities focusing on extension agent training, seasonal credit and internal SODECOTON management proceeded smoothly. Despite this, not all of the project objectives were achieved, or are today sustainable. The following sections focus on some of the key issues that were raised by project operations, on activities which proved problematic and on the reasons why some of the targets were not met. 4.06 Cotton activities: The appraisal had set out an objective of an increase from 48,000 to 70,000 hectares of intensive cotton cultivation producing 82,300 tons of cotton seed by PY5. Instead, approximately 60,000 tons of cotton were produced on 52.500 Ha during the 198511986 season, implying a 27 percent production shortfall compared with appraisal. Cotton production statistics are given in Table 1. In addition, yields averaged 1140 kilos per Ha, instead of the 1200 foreseen in the SAR. These totals suggest a fairly successful project. Moreover, with the return of higher rainfall levels in 1986-1989, the project production targets have been met. However, to put these totals in perspective, it should be remembered that the old CFDT had produced as much as 70,700 tons in more or less the project area as early as 196911970 and had extended its low intensity extension system over 70-80,000 hectares there through out the late 1960s and early 19708. The SODECOTON was not so much introducing as reintroducing cotton to most farmers in the area, and its inability to promote intensification more successfully suggest the presence of fundamental agro-ec-logical problems in the project area as discussed below. This is corroborated by SODECOTON's greater success in areas south of the project zone. Though production increased some 8.5 percent a year on average in the Far North Province between 1974 and 1986, it grew by 12.4 percent in the North Province and by 10.6 percent overall for SODECOTON. Between 1980 and 1986, the life of the project, these production growth rates were -1.22, 12.62 and 5.72 respectively (See Table 2) 2/. 4.07 Several factors explain this disparity between the North and Far North Provinces and seems to have been responsible for the project never reaching the production objectives laid out at appraisal. First, 2/ Comparisons are complicated by the fact that the Center-North project area is roughly 80 percent in the Far North Province and 20 percent in the Benoue Department of North Province. Data for the project area were nct available for the entire period. The Benoue Department, with greater precipitation levels, regularly outperformed the rest of the project area.  - 14 - drought hit the north of Cameroon in 1982-1984. It was particularly severe in the Far North, including the project area, whereas SODECOTON areas further south in the Benoue valley were less affected. With lower and more unpredictable precipitation levels, farmers have turned away from cotton and have focused on food crops, particularly in the Haroua region where food prices increased appreciably during the project period. In addition, there has been some out migration of farmers to the cotton areas further south in the Benoue Valley. Some observers were led by the drought years to argue that there is a secular decline in precipitation levels in the Far North that will continue to undermine the project area's production potential. With the returns of three years of adequate rains, a slightly more sanguine view may be in order. In any event, the sustainability of the cotton production increases achieved through the project will likely depend on the area's precipitation levels in the future. 4.08 Food croD activities; The food crop component of the pro4ect was no more successful at meeting project targets, albeit for different reasons. SODECOTON was slow to undertake the full program of activities for food crops listed in the appraisal, in part because of its lack cf experience with food crop extension. The early years of the project witnessed some groping. For example, it first focused on wholesale intensification of food crops on small perimeters, instead of the program of semi-intensification on large areas outlined in the SAR. Bank staff involved in supervision missions appear to have agreed with this departure from the SAR strategy though several Supervision missions did raise the issue with SODECOTON. While difficult to quantify precisely since SODECOTON did not provide the Bank with complete statistics for food crops in the project area, pro,ect results were well below the objectives listed in the SAR. Target shortfalls for output of groundnuts and sorghum, were well above 50 percent and the production objectives for maize were altogether scrapped when it became clear that maize was not a viable crop in the project area. Project results were hampered by the drought conditions. In general, it can be said that tested technological packages did not exist to promote intensification of production, at the project outset. For most crops, the SAR foresaw the need to begin by recommending simple improvements to farmers in a first stage while field trials tested new varieties and more sophisticated input use. SODECOTON's food crop intensification strategy was undermined by an initially poor understanding of farm level incentives and food crop/cash crop tradeoffs, as well as by the exiguity of food markets in the region, and the drought conditions preva.ling during the prject. Project results for food crops were thus slow to appear. Recent results have been much more encouraging, however, suggesting cause for a more positive assessment of the progress made under the project. The project experience with each crop is reviewed separately below. 4.09 Groundnuts: In the appraisal, it was widely felt groundnuts had the greatest pntential among food crops for the project area. SEMNORD had spoisored the intensification of groundnuts in the seventies before its abolition, and it was thought that farmers would be quick to accept extension recommendations. Local market prospects for groundnuts were better than for other food crops and simple technical innovations were thought to be sufficient to appreciably increase output. The seed multiplication project was  - 15 - supposed to be able to provide the project area with improved seeds. The appraisal promoted the twin approach of seeking wholesale intensification on only some 10,000 hectares by the end of project operations on the one hand and simple improvements on the another 29,500 hectares. The latter consisted of semi- intensification and included on farm seed selection and rudimentary greater planting density, for example, but no fertilizer use. The SAR estimated that these simple improvements could increase yields from 350 to 560-630 kilos of shelled nuts per ha. In fact, SODECOTON efforts in the early phases of project implementation focused entirely on intensification, leading to criticism by several bank supervision missions. By PY5, the project had promoted intensification on 1500 hectares, but had yet to begin the program of semi-intensification, despite Bank insistence. In part, this slow progress can be explained by the poor weather conditions for groundnuts and the absence of well adapted varieties for the project area. 4.10 Sorghum: Sorghum (wet and dry season or Mouskwari) represents the most important subsistence crop in the region. Under traditional systems of cultivation, mean yields were said to approximate 700- 750 kg of threshed grain per hectare for wet and dry season sorghum. The project focused exclusively on wet season sorghum, though SODECOTON estimated at 75,000 Ha the acreage devoted to this crop in the 1985/86 season (45,000 ha in 1984185). Acreage devoted to improved sorghum cultivation was supposed to increase from 12,500 ha in PYO to 91.850 ha by PY5; instead, SODECOTON reported increases to 25,000 ha., a 73 percent shortfall. By way of comparison, the 1984 Agricultural Census reports total sorghum and millets acreage at 292,000 ha in the Far North province. In part, marketing problems undermined production efforts. For example in 1986, the excellent sorghum harvest led to brutal fall in the market price paid to farmers, which went from 200 CFAF per kilo in August 1985 to 25-40 CFAF in December 1986. The good rains in 1986 and then again in 1988 seem to have led to a quantitative jump in semi-intensive sorghum cultivation in those years, however, suggesting that the project's impact on production is tributary to adequate rainfall levels. 4.11 Maize: SODECOTON had high hopes for maize cultivation in the project area, though it is not a traditional crop in the area and was grown mainly as a garden vegetable (culture de case). The project appraisal foresaw the experimental introduction of high yielding short season varieties of maize among the area's more progressive farmers, essentially as a cash crop. It noted that maize consumption in the area was practically nil and cautioned that current marketing outlets might be very thin. It established a target of some 5,000 Ha under advanced technology on 0.25 Ha plots. In fact, SODECOTON abandoned this project component when it became clear that intensive maize cultivation was not viable in the project area under the drought conditions prevailing in the project area. It should be noted that in the Benoue valley further south, SODECOTON's efforts to promote maize have been  - 16 - successful and production ihcreases have been rapid, often at the expense of sorghum cultivation. 4.12 Food marketing issues: The SAR had pointed to the low proportion of food production marketed as a key development constraint ir. the project area. As well as the feeder roads component, the project included a component to modernize seven rural markets as a way of improving marketing channels. These were not built, as SODECOTON appears to Lkve determined they would have little impact on marketing in the area. To justify this SODECOTON pointed to a number of farmer needs surveys that do not mention rural markets as a priority need. In any event, the decision to include this component in the project can be questioned in light of the fact that the major urban centers in the north do not have permanent market structures. Food marketing has been very active in certain parts of the project area, notably in the immediate radius of provincial towns like Maroua. 4.13 A consensus has now emerged among government agencies and donors that the biggest constraint to increasing food production in the project area is not the production possibility frontier under existing production technologies, but rather the exiguity of markets. With less than 20 percent of food production actually marketed, most farmers in the area are more or less self reliant. There is so little farming specialization that even SODECOTON's most progressive farmers continue to cultivate food crops for home consumption. Food demand comes mainly from the urban populations, which are small in the North and increasingly consuming non traditional food stuffs like bread and rice. SODECOTON efforts to promote food pro1action was not hampered by technological or ecological issues (though these were present, notably for maize in the Far North) so much as by the lack of market outlets for the resulting output. Farmers would not expand their food crop acreage and'use modern inputs like fertilizer unless they were assured of being able to market the surplus output. Until 1986 SODECOTON did guarantee a purchase price for some food stuffs, but supply soon out stripped demand and the operation became unprofitable. It was discontinued because of SODECOTON's mounting financial difficulties. 4.14 The Appraisal mission was aware that marketing channels were poorly developed, but it overestimated the demand for the surplus foodstuffs the project aimed to produce. In retrospect, greater attention should have been devoted to the immediate causes of food insecurity in the project area, the Osoudurew period every year and the failure of the rains that seems to occur one in three or four years. In particular, project efforts to improve farm and village storage capacitv should have been expanded to address this type of insecurity. On the research level, greater efforts should have been devoted to improving storage technology and increasing the shelf life of cereals. 4.15 Feeder roads: SAR cost estimates for feeder road construction proved overly optimistic and revised cost estimates in early 1982 indicated that the CFAF 3.2 billion (USS 10.7) earmarked this component in the SAR would cover only 80 percent of the original 800 kilometers, or some 650 km. Eventually, some 600 km were built or rehabilitated over the course of four  - 17 - seasons. SODECOTON focused on the eastern and southern regions of the project area, on the assumption that a later phase of the project could expand the road network north and west. By late 1986, over half of these roads had been classified by the HINEQ and thus was included in their maintenance schedules. The remaining roads were classified by 1988. After this component had been completed, some of the road building equipment was transferred to MINEQ and ONAREF. The lack of operating budgets at both of these institutions since 1987 has led to a near total absence of maintenance for the feeder roads. Under these circumstances, by 1990 some of these roads will in worse conesition than at project inception. 4.16 Medium Term Credit: The project set aside US$ 4.6 million (including 3.2 million from the Bank) to allow FONADER to extend medium term loans to farmers to allow them to purchase oxen from SODECOTON. Implementation of this component proved problematic. The project Loan Agreement (Schedule 1, para 4 (c) and (d)) stated as a condition of disbursement that FONADER had to provide the Bank with a report on the results of agricultural credit recovery efforts in the Northern Province showing that arrears had been reduced to no more than 25 percent of the aggregate principal amount of such credits which were due and payable. Such a report was provided to the Bank in April 1984, two and a half years after the beginning of operations, showing that arrears had been brought down close to 23 percent and the Bank auLhorized disbursements under this component. In fact repayment rates at FONADER appear to have continued to be low. 4.17 FONADER was never able to disburse the loans at anything close to the levels predicated by the appraisal. Of the US $ 3.2 million budgeted by the Bank in the SAR for this activity, only $0.95 million was spent by the end of operations. The reasons for this are not clear. There was some question as to the effective demand for such credit on the part of farmers. After 1985, SODECOTON successfully began promoting the use of donkey traction, and the medium term credit component was used to provide farmers with credit for donkeys rather than oxen. This appeared to fit farmer needs better, and the donkey's cheaper cost seems to have better adapted to farmers means. Some 3,800 donkeys were eventually distributed to farmers and trained. 4.18 Forestrys The appraisal had argued that the forestry component was important to addressing the ecological problems of northern Cameroon. Proposed targets in the SAR included 300 Ha of experimental tree plantations in IRA's forestry research station in Maroua, 700 Ha of pilot savannah plantations for fuel wood and 750 km of windbreak planting in selected parts of the project area. In addition, forestry research would be actively promoted. While good progress appears to have been made on the research front, the targets proved too ambitious. For lack of determined management and strict adherence to the work program, the savannah forest plantation lagged over 50 percent below target, with a meager 330 hectares planted, instead of the 700 planned. Technical problems led to a large attrition rate among young trees. The SAR had argued that the windbreak or village plantations component of the project was a promising way of protecting fields from wind erosion and would lead to increased yields. Eventually, it was entirely dropped from the project due to lack of village level interest. In retroapect, there was inadequate  - 18 - knowledge of technical issues to embark on a planting scheduleof the size foreseen by the SAR. 4.19 Agricultural Research: The project was more successful at enhancing the adaptation and diffusion of existing technologies than at developing new technologies, given the short life of the project and the exceptional drought conditions. The applied research program was executed at the Maroua research station and 9 substations located in the three northern provinces. The experimentation program was properly executed and involved 183 trials, including 134 on food crops, on the substations and about 400 trials performed throughout the cotton area. Food crops tested included sorghum, groundnuts, maize, paddy, cow pea, soybean and yams. Due to drought many of the trials failed. By 1984, the IRA had with the assistance of SODECOTON established 4,200 demonstration patches of a quarter hectare of Sorghum fertilized with urea to show its potential benefit to farmers. The impact of these research results on farming systems seems to depend on the precipitation levels. 4.20 Several factors seriously hampered the sustainability of this component's impact, however. The lack of counterpart funding for project activities was first compensated by SODECOTON's willingness to prefinance them, but it soon stopped doing this. In addition, some of the activities were designed to be continued by the Bank's Agricultural Research Project, initially slated to begin operations in late 1984. That project did not become effective in 1988. As a consequence, much of the research begun under NPRD was in effect discontinued. 4.21 Northern Province Economic Division; the NPED, based in the MINPAT was designed by the SAR to fill a gap in regional planning by coordinating regional interministerial forecasting activities and by coordinating the numerous interventions of multi and bilateral donors. It seems to have gotten off to a slow start but then to have played a useful role, in reporting regional trends and analyzing development constraints. Unfortunately, the NPED seems to have ended its activities with the departure of the expatriate expert. 4.22 Project Monitoring and Evaluation Units the PMEU established in the HINAGRI's regional delegation was less successful. It was meant by the SAR to monitor the project's impact at the farmer level and eventually monitor and evaluate all projects in the area using a consistent and rigorous set of techniques and criteria. In fact, it was never able to carry out a meaningful baseline survey, as it used an excessively complex questionnaire which it waa not able to completely master, and there were complaints that the Regional Delegate used the unit as little more than a agricultural statistics collection unit. After a laborious start, however, the unit did provide several useful reports on the project's impact on the farming population. Since the end of the project and departure of the expatriate expert, the unit has been closed. Its personr.el was integrated into the statistics division, project cars were taken away and the computer equipment was transferred to the DEP in Yaounde.  - 19 - V. FINANCIAL AND ECONOMIC PERFORMANCE A. Financial Performance 5.01 The financial performance of SODECOTON, the rain project implementing agency, was overshadowed by the fluctations in the price of cotton. During the early 1980s, while the price of cotton was strong, SODECOTON was able to show a modest profit, but increased revenues were largely passed on to producers in the form of increased prices for seed cotton, which increased from CFAF 80/kg in 1980-81 to CFAF 140/kg in 1985- 86. This does not include the CFAF 15/kg post-harvest bonus paid by ONCPB. Throughout this period, SODECOTON subsidized seeds and fertilisers and provided free insecticides. 5.02 In addition to the agricultural assistance to farmers for cotton, SODECOTON continued to expand its rural development activities during the 1980s. It addition to the crop development programs for sorghum, maize, and groundnuts and the roads construction included in the Bank financed project, SODECOTON was involved in groundnut processing and storage and adult literacy programs through the Bank financed FSAR I and II projects, as well as implementing a number of other rural developemnt activities. The total cost of rural development is estimated to have been CFAF 5.8 billion in 1985-86. 5.03 The fall in the price of cotton in 1986, combined with the failure of the Government to provide the agreed counterpart funding for the project related and rural development work, resulted in the acute financial crises of 1985-86 and 1986-87 when financial deficits of CFAF 20.3 billion and CFAF 12.1 billion were recorded. Furthermore a deficit of CFAF 11.8 was estimated for 1987-88. Financial audits prepared by Price-Waterhouse in 1986-87 and the CCCE in 1987-88, led to recommendations for restructuring SODECOTON, including reduction in the subsidies on cotton inputs from 56 percent in 1985-86 to 25 percent in 1986-87. In addition, a third grade of seed cotton was introduced in 1988, which reduced the average cost of seed cotton to CFAF 126/kg in 1987-88. Under the provisions of the Structural Adjustment Program, it is realistic to assume that the producer price could be reduced to about CFAF100/kg seed cotton, and other farmer input subsidies removed completely, which would permit SODECOTON to cover its variable costs. However, without some improvement in world cotton prices it is unlikely that would be much of a positive return on the capital employed. 5.04 Financial returns to farmers throughout the project period were satisfactory, with a net return of CFAF 60,000-90,000/ha equivalent to CFAF 30-50,000 per farmer. As noted above, even after the fall in world cotton prices, farmers were largely insulated from lower returns. In addition the project also assisted them to increase food crop production, some of which was marketed.  - 20 - B. Economic Performance 5.05 The economic performance of the project has been much more favorable than is suggested by the lack of financial profitability of SODECOTON. Over the longer term, the economic viability of cotton production in Northern Cameroon should be satisfactory, and considerably better than the financial figures suggest, due in part to the fact that the price of seed-cotton paid to farmers was not reduced when world cotton prices fell, and in part to low controlled prices for the sale cf cotton lint on the domestic market (nearly 20 percent of the total volume of lint), and cotton seed oil, all of which is sold on the domestic market at regulated prices well below border prices. Thus, the financial returns do not reflect the returns to the economy. Moreover the medium and longer term price projections for cotton indicate some recovery, which combined with some recovery of the US dollar relative to the CFAF, would lead to increased revenues from exports of cotton lint. 5.06 Throughout the 1980s farmers in the project a-ea have received a considerable economic rent from cotton sales. Even with a producer price below CFAF 100/kg seed cotton, farmers can still make a satisfactory net return to land and labor after deducting all input costs, particularly in the Northern Province, where conditions are more favorable and yields are higher. In the Extreme North, where yields are lower, farmers have still been able to make satisfactory returns from cotton and food crops, whereas without the project yields and the level of farm income would probably have fallen. 5.07 The project's economic rate of return has been estimated again to compare it with the estimates made in 1980 at the time of project appraisal. In the SAR some 55 percent of the benefits at full production in the 10th year, were projected to result from incremental seed-cotton produLtion of 37,400 tons, compared with the actual production of 51,000 tons in 1988-89, which was a good year. Taking into consideration the pattern of poor rainfall once every three years, incremental production is estimated at 32,400 tons of seed cotton annually, some 85 percent of that estimated at appraisal. The value of future cotton production for the remaining 12 years of the project is projected to be CFA 7.0 billion annually, largely mitigating the effects of low cotton prices and unfavorable exchange rates in some of the early years of the project. Cotton would account for 75 percent of the net value of project benefits, assuming a cotton fibre price of US$1.55/kg f.o.b. and a future exchange rate of CFA350s$1.00. 5.08 Incremental food crop benefits were projected in the SAR to account for 35 percent of total project benefits at full production, with an estimated annual value of CFA 3.6 billion. Figures are not available of the actual increased area or production, since SODECOTON keeps complete statistics for intensive food crops only, rather than the much larger area of semi-intensive crops. In the absence of detailed information concerning areas and yields of food crops, their incremental value at full production has been estimated at CFA 0.9 billion annually, some 25 percent of the SAR  - 21 - projection (para 4.08-4.11). This reduced estimate is based on the facts that it was not possible to introduce maize as projected in the SAR, due to lower than anticipated rainfall, and that SODECOTON concentrated on a small area of intensive food crop production in rotation with cotton, rather than a much larger area of semi-intensive food crop production. The actual area of cotton was smaller than estimated in the SAR (although yields increased as projected), therefore it is likely that the area of intensive food crop production was also smaller. On the other hand, the SAR did not take into account the benefits from niebe (cowpeas) cultivation, a crop which SODECOTON has had some success intensifying in recent years. Food crop benefits are assumed to account for 20 percent of total net benefits in the economic analysis in this report. 5.09 The remaining benefits assumed in the SAR were derived from roads and livestock benefits. The benefits from roads construction have been assumed to be 75 percent of the 3AR estimate, since only 600 kms of the 800 kms planned were constructed. No benefits were derived from the livestock activities to reflect the poor performance of the credit component. 5.10 On the costs side, the SAR did not include many of the long term fixed costs of SODECOTON. As discussed in Paragraphs 5.02-5.03, these have proved to be higher than anticipated. The Caisse Centrale estimates of actual and future SODECOTON fixed and variable costs have been used in the projected ERR. 5.11 Taking into account the actual performance from 1981-82 to 1988-89, assuming no further production increases would result from the project, and using current World Bank commodity forecasts, the revised ERR of the project would be about 10 percent over a 20 years project life. The project life has been increased from 15 years in the SAR to 20 years to compensate for the effects of the unanticipated drought (1983-85) in the early years of project execution (Annex 4, Table 1). 5.12 Another set of calculations has been made to determine how sensitive the ERR is to changes in production, in which the return of drought conditions over the project area has been assumed, such as late onset or partial failure of the rains, giving rise to higher food crop prices, and an increase in the area of food crops planted, to try to compensate for lower yields. Under these conditions, the value of incremental food crop production has been assumed to remain at CFA 0.9 billion/year, with incremental cotton production falling from 32,400 tons to 20,400 tons, resulting in an ERR over a twenty year period estimated at 7 percent (Annex 4, Table 2). If the value of food crop benefits was also to fall by 25 or 50 percent, the ERR would fall to 5 or 2 percent respectively, indicating that the project is rather sensitive to changes in the value of food crop benefits. 5.13 The implication of these calculations is that project performance was adequate, albeit not as strong as foreseen by the SAR, and that future economic performance of cotton will be highly dependent on weather patterns in the North. Though project benefits were not weighted according to social criteria, it should be clear that project results would be even more valuable to the region, and to its social welfare, if drought  - 22 - conditions do prevail in the near future, since the project did make peasants less vulnerable to the effects of drought. Had the project output been valuated taking this into account, its rate of return would have been higher. Finally, it is important to note that the economic assessment of the project is not directly related to that of SODECOTON, whose large fixed overheads are by definition not considered in the incremental cost and benefit estimates used in the project economic calculations. VI. INSTITUTIONAL PERFORMANCE 6.01 The institutional problems in northern Cameroon identified by the SAR do not appear to have been relieved by the project. Though the physical goals for cotton production were largely met at high financial cost by SODECOTON, other institutions involved in project implementation were less satisfactory and the project's institutional development objectives were not achieved. As a result, policy making and implementing structures in north Cameroon remain largely deficient. The lack of progress during the project has now been compounded by the Government's inability to continue funding operating expenditures. 6.02 SODECOTON was and remains an effective cotton development agency. During the course of the project it demonstrated this effectiveness by meeting many of the physical targets of the project with minimal delays. The infrastructural components of the project (road building, rural markets etc) were all executed on time. Training of extension agents was similarly satisfactory. The shortfall in meeting cotton production outputs can be blamed on the drought rather than on SODECOTON. 6.03 On the other hand, the project helped SODECOTON improve the scope of its non cotton activities, notably in the area of food crops, even if they remain secondary to its cotton activities. SODECOTON was not transformed into a rural development agency, but it did begin to integrate rural development activities in its promotion of cotton based farming system. During the project implementation period, SODECOTON expanded its integrated rural development activities, and acquired new or enhanced experience in such areas as credit distribution, road building and food crops research and extension. Much of this institutional growth has continued since the end of the project, though it is jeopardized by the cotton sector crisis as well as its and the Government's fiscal difficulties. 6.04 The performance of the rest of the institutional apparatus in the north does not appear to have undergone sustainable improvements during the life of the project. The project decision to decrease duplication between HINAGRI and SODECOTON extension efforts by placing all extension activities under the competence of the latter did not improve MINAGRI efforts in other areas (moreover, MINAGRI extension staff levels for the project area were not decreased correspondingly as agreed to at negotiations). It may also have undermined MINAGRI's extension efforts in areas adjacent to, but not covered by the project area, notably the Mandara mountains. Monitoring and evaluation capacity in the provincial delegation of MINAGRI was not noticeably improved beyond the life of the project, nor  - 23 - was provincial planning and coordination efforts by the MINEP's provincial offices, despite a promising start. If anything, with the onset of budgetary cuts after 1986, the provincial ministerial structures at MINAGRI and MINPAT have suffered institutional decay. 6.05 Some progress in institutional development was made, notably in the area of agricultural research where the project trained local personnel, improved coordination efforts between projects and undertook useful and time consuming field trials for a number of crops. Cooperation between IRA and SODECOTON was enhanced during the project and has been sustained. Valuable experience was gained during the course of four growing seasons (1982/83 - 1985/1986). This progress was then greatly undermined by the termination of those activities, the endless delays in beginning implementation of the Agricultural Research Project and the major cuts in government funding for the IRA after 1986 as these have fallen disproportionately on its field offices. VII. BANK PERFORMANCE A. Prolect Justification and Obiectives 7.01 The Government's development priorities for the north were not clearly or precisely outlined at the time of appraisal and combined inconsistent social and commercial objectives. The project was justified by the perceived need to enhance development efforts in the Par North province, by far the poorest region of the country, to reach the poorest farmers and to promote the country's third most important agricultural export, cotton. In addition, the project area's ecological deterioration needed to be reversed. The SAR properly identified these issues, but the resulting project design effectively focused more on the commercial and environmental issues than on the social ones. The project for example excluded the poorest area in the region, the Mandara Mountains, because cotton is not produced there. More generally, by focusing on the cotton growing areas, the project excluded the poorest farmers even in the cotton region, who typically do not cultivate cotton. 7.02 The criteria behind the delineation of the project area are not clear. By Appraisal, it was clear that the Center-North was not a high potential area for cotton, and that SODECOTON's future growth lay in the higher potential Benoue Valley. The Appraisal noted this point but argued that Bank assistance could reverse or at least slow down the region's decline, and that social welfare arguments justified intervention. However, the project area included Guider, an arrondissement in the Benoue Department with higher precipitation and income levels which consistently outperformed the other project zones, and excluded the poorest areas of the Far North Province. In retrospect the Bank seemed ambivalent about emphasizing commercial criteria, by focusing exclusively on cotton in the higher potential areas further south, or concentrating on social objectives by pursuing integrated rural development objectives all over the Far North. SODECOTON's geographical area of activities seems to have influenced the  - 24 - choice of the project area; the Center-North was the only major sector of its activities where SODECOTON did not already have a foreign funded project. 7 *'3 The choice of SODECOTON as the lead implementing agency was strongly supported by the Government throughout negotiations. It did reflect the Bank's preference for short term advantages in project execution capability to longer term institution building. At the same time, it was justified by the fact that SODECOTON was the most if not the only effective rural development institution in Northern Cameroon. The SAR did nonetheless attempt to balance the short and long run considerations by entrusting the bulk of project implementation to SODECOTON, while also including several project components that would in the long run serve institution building objectives in the central administration. In retrospect, the project design did not devote nearly enough resources to ensure the success of these non cotton objectives. Moreover, these objectives were slowly forgotten during implementation and SODECOTON's components became the focal point of most supervision missions, particularly after the onset of the cotton crisis. Indeed, successive supervisions progressively lost sight of the project's rural development objectives for the Center North area and concentrated on SODECOTON's cotton activities in both northern provinces. The last two supervision mission reports confuse overall SODECOTON output goals with those of the project for example. By the end of execution, it is fair to say that the Bank considered the project a cotton project based in SODECOTON. 7.04 The Bank actively sought the increase of cotton producer prices during appraisal and negotiations, in keeping with its view that swings in world commodity prices should be transmitted to farmers. In Cameroon, this led to a rapid increase in the producer price, which went from 80 CFAP per kilo in 1980/81 to 140 in 1985186 (excluding the additional 15 CFAP provided by the ONCPB). The Government imposed these price 1-creases on a recalcitrant SODECOTON, while maintaining the subsidies on fertilizers and pesticides. The Government refused to decrease the producer price after 1986, however, to reflect the decrease in world prices. The high prices are a major reason for SODECOTON's financial difficulties since then. The lesson for the Bank is that producer prices are likely to be 'sticky, downwards when the marketing parastatal does not have price setting authority. Political pressure and the slow speed of government decision making can have as a consequence the rapid accumulation of deficits for the parastatal, which does not control the parameters of its own profitability. B. Bank Supervision 7.05 There were nine Bank supervision missions. The first supervisions focused on start up issues and were effective in prodding the different actors to implement agreed to measures. The Bank was quick to remedy minor appraisal weaknesses and agree to revised targets when these seemed more reasonable. This was notably the case with the feeder roads component, where the target was brought down from 800 to 600 km of roads, when it became clear the former was too ambitious. After 1985, supervision missions were increasingly focused on SODECOTON's financial  - 25 - difficulties, the absence of counterpart fundinA and on the project's extension with undisbursed funds. These missions were much more lax in monitoring project activities not controlled by SODECOTON and paid little or no attention to the output targets established in the SAR. Thus, it is impossible to determine from supervision mission reports the number of loans made by FONADER, the quantity of groundnuts and sorghum produced in the project area in PY5 or the area and yield of cotton in the Center-North Project Area. 7.06 The Bank paid insufficient attention to the sustainability of project results. By 1986, it was clear that many of the project activities would end with the end of Bank funding, as continued counterpart funding was increasingly unlikely, and would undermine project results with sizeable recurrent expenditure implications. It was foreseable that the project's end implied the slow decay of the feeder roads built or rehabilitated by the project for example. VIII. CONCLUSION 8.01 The success of this project was undermined by a series of exogenous events which could not have been predicted or controlled. Drought, the wide swings in the value of cotton and of the CFAF, and the growing government financial crisis undermined project implementation and the sustainability of its achievements. 8.02 Project design was consistent with the Bank's policy for rural development projects at that time, though it did not pay enough attention to institution building and it overestimated the project area's long term cotton potential. More successful institution building would have required greater attention to the long term needs of the central administration. instead of emphasizing the strengthening of SODECOTON. Project components of regional planning, project monitoring and evaluation based in MINAGRI and MINPAT were inadequate in this respect. The meager results of these project components did not last beyond the end of project operations. The project design followed Bank and government policy in Cameroon at that time of implementing the project through the existing relatively efficient and experienced SODECOTON, rather than using the project for the longer run objective of increasing implementation capacity in the central administration. While it would have been counter productive not to use SODECOTON's existing capabilities for rural development in the region, in retrospect the project's emphasis was excessively tipped against the needs of the central administration, and institution building in North Cameroon was delayed. This policy of relying on project authorities for rural development has since been reversed by the Government and the Bank, which has begun more substantial institution building efforts in the MINAGRI, notably through the National Agricultural Extension and Training project, and the earlier Technical Assistance project. 8.03 In addition, what was designed to be a rural development project with cotton as the most important crop turned into a cotton project with few substantial non cotton activities. The SAR overestimated the  - 26 - viability of expanding food crop production, as well as the incentives to farmers to do so. It overestimated SODECOTON's ability to diversify its activities away from a narrow concern with cotton to the other crops -- sorghum, groundnuts and cowpeas -- associated in the crop rotation. Maize cultivation turned out not to be practical under prevailing precipitation levels while sorghum and groundnuts production growth was constrained by limited marketing possibilities. SODECOTON needed to develop experience and technical expertise before it could support these new activities effectively. During implementation of the project little was done to promote agricultural extension in the non-cotton growing areas. Although personnel were not removed from MINAGRI's extension service as agreed to during negotiations, its agents never implemented any extension program in the regions left out of SODECOTON's program. By 1986, the lack of a recurrent expenditure budget in MINAGRI excluded the possibility of any meaningful action on its part. 8.04 The financial calculations at appraisal proved to be excessively sanguine regarding the financial viability of SODECOTON. It is unlikely that SODECOTON can become a profitable parastatal in strictly financial terms, in the near future. 8.05 The Appraisal was too optimistic about the likely evolution of the world cotton price. More serious it did not predict the exchange rate volatility that has characterized the last decade. The exchange rate used at appraisal of 210 CFAF per US dollar turned out to be just about the highest level for the CPA franc ever, and actually underestimated the SODECOTON's profitability once the dollar began to appreciate. SODECOTON enjoyed its best years in 1983 and 1984 as a consequence, but its windfall profits were passed on to the farmers through generous producer price increases and the maintenance of input subsidies. These were imposed by the Government despite SODECOTON's opposition. SODECOTON also financed an increasingly expensive administrative structure and took on a series of unprofitable rural development activities. When the dollar peaked and began to decline after 1985 and world cotton prices tumbled simultaneously, SODECOTON accumulated losses. A recent Caisse Centrale report on SODECOTON suggests that the parastatal would continue to lose money even after major internal reforms and a decrease of the producer price granted farmers to around 100 CFA/kg. 8.06 SODECOTON is generally considered one of the most effective public institutions in Cameroon. It expanded into a number of non cotton activities, often at the behest of international donors like the Bank, because of the deficiencies of the central administration. Thus, it financed expensive activities such as extension, training, industrial training, village level infrastructure, food crop marketing and road maintenance. No other public institution was providing these critical public goods for sustained rural development, that were in any event preconditions for SODECOTON's program of cotton intensification. As a consequence, the major questions posed by this project for the cotton growing areas of the North and Extreme North Provinces are whether SODECOTON is the institution best suited to provide these public goods, and if so, does this justify subsidizing SODECOTON's rural development activities for the foreseable future?  - 27 - 8.07 Tt is difficult to attach an economic value to SODECOTON's accumulated e..perience in these areas, but one of basic indicators of the success of a project relates to its replicability to the other two-thirds of non-cotton producing farmers in the region. Unfortunately, it is not likely that much of the experience gained by SODECOTON in rural development with cotton growing farmers can be readily utilized by the other Government implementing agencies to benefit the majority of non-cotton growing farmers of the North and Extreme-North Provinces. Financial considerations dictate that SODECOTON will have to concentrate its efforts on cotton and cotton farmers for the forseeable future, and separate plans will have to be made for MINAGRI and associated Government institutions to assist the other two- thirds of the regicn's farmers.  - 28 - Annex 1 Table 1 Page 1 of 3 Disbursements by Project ComDonenet and Disbursement Category Loan 1919-CM and Credit 1075-CM CateRorY Allocated Disbursed Proiect Conoonent ($000) ($000) Part A (ii) of the Proiect - Provision of extension services... 3 Ln. 6,460.0 6,132.2 Operating Expenditure of SODECOTON Cr. 3,092.1 4,252.5 and of PCC for Project Tot. 9,552.1 10,384.7 Administration under Parts A (ii) and B of the Project (including local staff salaries). 1 Ln. 3,000.0 5,349.4 Civil Works under Part A (ii) of Cr. 1,486.8 1,962.8 the Project Tot. 4,486.8 7,312.2 15 La. 4,030.0 3,548.1 Specialist's Services Cr. 2,026.3 1,923.9 Tot. 6,056.3 5,472.0 2 Ln. 2,600.0 955.2 Vehicles and Equipment under Part Cr. 1,328.9 2,284.3 A (ii) of the Project Tot. 3,928.9 3,239.5 Sub-total Part A (ii) La. 16,090.0 15,984.9 Cr. 7,934.1 10,423.5 Tot. 24,024.1 26,408.4 Part D of the Proiect - Strengthening and support of agricultural research by IlA, and Part E (i} - Strengthening of the Savannnah forest research unit of IRA, 8 La. 1,200.0 1,031.0 Operating Expenditures including Cr. 605.3 337.5 local salaries uncer Parts D and E Tot. 1,805.3 1,368.5 (i) of the Project. 7 Ln. 600.0 937.5 Civil Works, vehicles and Cr. 302.6 500.6 equipment for Parts D and E (i) of Tot. 902.6 1,438.1 the Project. Sub-total Parts D and E (i) Ln. 1,800.0 1,968.5 Cr. 907.9 838.1 Tot. 2,707.9 2,806.6  - 29 - Annex 1 Table 1 Page 2 of 3 Parts E (ii) and E Uiii) of the Proiect - Pilot tree plantings and windbreaks to be implemented by FNFP. 10 Ln. 400.0 403.4 Operating expenditures including Cr. 210.5 266.0 staff salaries under Part E (ii) Tot 610.5 669.4 9 Ln. 500.0 80.9 Civil Works, vehicles and Cr. 276.3 395.3 equipment for Parts E (ii) and Tot. 776.3 476.2 (iii) Sub-total Parts D and E La. 900.0 484.3 Cr. 486.8 661.3 Tot. 1,386.8 1,145.6 Part A (iii) of the Proiect - A medium term credit program to finance the provision of draft animals and agricultural equipment to farmers. 5 Ln. 930.0 948.7 Animals for Part A (iii) Cr. 460.5 0.0 Tot. 1,390.5 948.7 4 Lu. 1,200.0 9.8 Agricultural equipment - Part A Cr. 592.1 0.0 (iii) Tot. 1,793.1 9.8 Sub-total Part A iii) Ln. 2,130.0 958.5 Cr. 1,052.6 0.0 Tot. 3,182.6 958.5 Part F of the Proiect - Strengthening of MINAGRI and Project Monitoring and Evaluation. 12 Ln. 570.0 460.8 Operating E*penditures, including Cr. 289.5 249.4 local staff salaries, under Part F Tot} *859.5 710.2 of the Project. 11 Ln. 580.0 142.2 Civil Works, vehicles and Cr. 315.8 183.7 equipment under Part F Tot. 895.8 325.9 Sub-total Part F Ln. 1,150.0 603.0 Cr. 605.3 433.1 Tot. 1,755.3 1,036.1  - 30 - Annex 1 Table 1 Page 3 of 3 Part G of the Project . Development Planning 14 Ln. 160.0 186.4 Operating Expenditures, Cr. 92.1 52.2 including salaries under Tot. 252.1 238.6 Part G 13 Ln. 200.0 96.0 Civil works, vehicles and Cr. 105.4 26.1 equipment Tot. 305.4 122.1 Sub-total Part G Ln. 360.0 282.4 Cr. 197.4 78.3 Tot. 5S7.4 360.7 Part C of the Proiect _ Seed Production 6 Ln. 130.0 5.5 Equipment for Part C Cr. 65.8 65.7 Tot. 195.8 71.2 Continaencies Ln. 2,440.0. 0.0 Cr. 1,250.0 0.0 Tot. 3,690.0 0.0 TOTAL DISBURSEMENTS Ln. 25,000.0 20,287.2 Cr. 12,500.0 12,500.0 Tot. 37,500.0 32,787.2 - For Credit 1075-CM, the same rate of conversion of SDR 1.00 - $1.3158 has beei' uded for the purposes of comparing appraisal estimates with actual disbursements. Due to strengthening of the US dollar against the SDR over the disbursement period, a US dollar equivalent of US$10.1 million was disbursed, but since most expenditures were in CFAF or French Francs, the CFAF equivalent was similar to the CFAF 2.6 billion estimated in the SAR. 2/ - Annex 1, Table 1 presents the disbursement information for the Credit and the Loan separately, since the detailed data is not readily available elsewhere because the disbursements file for Credit 1075-CM has been misplaced.  Camarsa Northern Proviso Rural ivel_pset Project: Selete Annual Prodution Indicators 61^/ 12/01 0/64 64/86 "5/ P11 P12 PS P14 PC App. Lob. App. Acb. App. Ach. Ap. Lch. App. Ach. I. Pasr pased l.t rjet ('lo) 1/ 16 181 148 arealdast 1i U 8 as Naom,.uwthCott" so 77 n2 7 92 Oa-dras withCoett" 16 17.5 19 2.6 22 tot settee 95 94 114.6 so U1 IIJ 142.5 i6 16S 1i U1. Area Unr Citivatleo ('M lb) a. seteeo traditisal 21 8.6 6.1 2.6 7.1 2.0 5.6 2.6 5.6 2.0 4.7 Cott" to4al Iteasivo U6.6 89.6 66.6 26.6 09.6 41.1 64.0 40.1 66.0 47.6 Total Cetten 54.2 45.0 68.6 85.4 62.6 46.9 00.6 46.6 76.O 62.6 b. ied orm (et iee.) Ia Sote loe wIth Cetdt 22.2 16.4 * .6 27.5 86.7 16.9 46.6 24.6 U2.7 In letetlee wlith Croeedmt 2.2 5/ 7.6 12. 190.3 2.8 TOtal Sorgbh 24.4 16.4 57.1 27.5 11.5 16.9 66.5 24.6 79.0 25.0 c. lbenro" eved Orawa e 2.2 19.6 27.5 23.6 29.5 Isteaslfifed Groomig" 1, 6.4 2 6.6 8 6.6 7.4 1.8 1i 1.5 ToteS Or osw st 8.2 0.4 21.1 6.6 66.6 6.6 3 1.5 59.5 d. Improwe 1Iba 0.7 6.4 1.5 4/ 2.6 8.3 w III. iProdiactios '6 Seed Cott". 65.9 62.6 65.1 4J.9 76.8 66.6 76.6 54.4 S2.8 6.6 ixe, 1.8 0.4 2.6 0.6 4.2 4/ 0.4 6.6 IV. Asime_ Trestle ('_ Tote 1xistIag 16 18.9 17.5 14.4 10 18.4 2U.S 15.5? 22 16 PaIrs Of trOe *xeA - - - 6.05 - 6.214 - 6.488 - .609 Pairs of traleed fow.. total 16 18.9 17.5 14.4 10 18.6 26.6 15.6 22 16.69 total wO"Is deo"ps 1.5 2.6 1.6 2.? 1.9 2.06 2.2 X.7? 2.5 3.97 There are umepleiloe slight differemoes to totals betsweena SODECOTOanI reports cnd Supervision reports atotals lis* I Supvleol. reports. Rne used figures from latter, ecept for 1886/1 yer 1/. Accordng to ses as crap. 21. Supsviis report. ree to 'nprove tdi)tioal'. 5/. PIgmaros net reported by SCDECOTO PU ws supposed to report thm but nevr did. 4/. ProJect act deold x  See Cotton Prodwtise, Acreg and Yields, Caesroon: 107s-1m Tl 862C016N : Extreme North Pronce a Nort ProvIn am Cse.r-Wlrt& year Sfla Prodctlo Yields: Surface Produ)om Yields: Surf* ProoducION Yields: Surtae Production Yields: ,(In lb.) (tee) No/be :(t He.) (tees) o/he s:(n lb.) (toes) Kg/ha :(In sl.) (ton) Kg/ae 1190/26 1581Sr64U 9188 4 *44f WM 7sm 27 a 27584 238 749 : 90441 625 In8 1076/171 16666 U8o*So4 70s t74544 s6m 8 * 27511 1101t 48U a *9142 sU8K2 87M 1071/1072 46sow 48197 48 6 71917 21918 891 s 27 @ 1ism4 Us : MOI5 87857 416 : U7108/11 P 9079 4523 51? 02t69 2854 458 a 26 107m6 SW9 t 7416 877 495 108/127t4 01S70 17687 466 a 8SO21 1168i1 S 2 n86s 15376 748 * 56327 70 84 : 1274/1273 04432 4002 022 a 42012 21156 472 no m 127 9QC 58191 2740 ' s64 107Lv10 mn78170 48461 070 : 2877 25074 464 t 2GM1 24146 1160 : 01044 8a96m 5o l 12o970077 _908s 477 727 : 42816 25916 612 t 17615 216s 1241 : 4915 82684 664 : 1277/3078 48486 4686 9867 a 88865 36i2 64 t 14J1 1776 1217 : 823 2721 096 108/1909 4718U 56427 1202 s 81789 8566 116 a 1u 28O44 18 : 85688 41156 1158 108/106 i@5694 mm8 1419 8 $6846 46789 1219: 19749 880m 17M1 : 42085 : 1966/191 6646U" 844o8 1298 s 42287 45467 175 a 2865 8*960 1092 : 1061/1162o 08848 70619 1168 a 89867 42466 170w 2970 7158 1C55 : 4636 520M 1146: 1961/1268 540 726 15 195 822 16 24764 4J111 16 864K 4891 1165: 1808/1984 73992 4566 1886 a7 9 8o5 41907 164 a SI6 SM 18 406 60 118: 16/1 7 073* 186 8M n4 1 1 876 0171 1640 48 6_ 1111 196soo966 66282 116544 16 42411 4179 114 s 401 67420 1441: 621 _9 1148 a 1@06/3187 4469 1 U_11011 18a 45110 4 J I76 I 4Y58 s 466U 6 10 s 65100 540 SW s sobEelm ~ ~ ~ ~ a * a $00root SOKC91011 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~o w  Cameoon Seed Cotton soz - 10 - uvwo 9 73474 ^ Se x o3 Totd+ Fw Nort Not  CAMERM #01FITHE-N RIPAL De NW WFRO eT - RR ECIIC RTE OF 1E83 W - OM6 CASE *4LYS.S increemntol Production table aedecoton ttl ic r* nrpjt poet fixi pr ctopion soo sltareo * e tf *itm *rJ *ilmt pr withow1bf r; *Il rhoj sifA, project -ith projet project isar. _as _" r"to conr-nrAb total orodi f;S cam :0 1*14 (tam g prod.o (WMe ~naduct4t9 prod. (twon CP4bi Iion ton twoCMbllo 1".0/01 wt " e&;))t tl-t < C ren 'i;