Cameroon - Second Livestock Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 10068 PROJECT COMPLETION REPORT REPUBLIC OF CAMEROON SECOND LIVESTOCK DEVELOPMENT PROJECT (CREDIT 1010.CM) NOVEMBER 15, 1991 Agriculture Operations Division Occidental and Central Africa Department Africa Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency Units - FCFA AVERAGE ANNUAL EXCHANGE RATES 1980 226 FCFA = US$1.00 1981 287 FCFA - US$1.00 1922 33E FCFA - US$1.00 1983 417 FCFA - US$1.00 1984 480 FCFA = US$1.00 1985 378 FCFA = US$1.00 1986 323 FCFA US$1.00 1987 267 FCFA - US$1.00 1988 303 FCFA - US$1.00 1989 289 FCFA = US$1.00 ABBREVIATIONS AND ACRONYMS ERR = Economic Rate of Return FONADER Fonds National de D6veloppement Rural IDA International Development Association IMF International Monetary Fund KFW Federal Republic of Germany MINEL Ministere de I'Elevage et des Industries Animates MINEPIA Minist6re de L'Elevage, des P&ches et des Industries Animales PCR Project Completion Report PCU Project Coordination Unit SAR - Staff Appraisal Report SEDA Soci6t6 D'Etudes pour le Deveoppement de L'Afrique SODEPA Soci6t6 de Developpement et d'Exploitation Animale FISCAL YEAR OF TH_ BORROWER July 1 - June 30  FOR OMCIAL USE ONLY THf W ORlD SANK Washington, DC. 20433 U.S.A. O0laC Of DfuF,KtcW4GnV&I O9w.aX hlJtK November 15, 1991 MEMORAND_M TO THE EXECUTTVE nl"iECTORS AND THE PRESIDENT SUBJECT: Project Completion Report - Republic of Cameroon Second Livestock Development Proiect (Credit 1010-CM) Attached, for your information, is a copy of a report entitled "Project Completion Report on Republic of Cameroon - Second Livestock Development Project (Credit 1010-CM)", prepared by Africa Regional Office with Part II of the report contributed by the Borrower. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT REPUBLIC OF CAMEROON SECOND LIVESTOCK DEVELOPMENT PROJECT (CREDIT 1010-CM) TABLE OF CONTENTS Fage No. PREFACE . . . . ... . . . .. 1 EVALUATIONSUMMARY . ..... . . . . . . . . . *0.* * . . .ii PART I: PROJECT REVIEW FROM BANK'S PERSPECTIVE . . . . 1 A. Project Identity . . . . . . . . . . . .1 B. Background . . . . . . . . . . . . .1 C. Project Objectives and Description . . . . . 2 D. Project Design and Organization . . . . . . 3 E. Project Implementation . . . I . . . . . . 4 P. Project Results . . . . . . . . . . . . . . 5 G. Project Sustainability . . . . . . . . . . 11 H. Bank's Performance . . . . . . . . . . . . 12 I. Borrower's Performance . . . . . . . . . . . 12 J. Consulting Services . . . . . . . . . . . . 13 K. Project Relation . . . . . . . . . . . . . . 13 L. Project Documentation and Data . . . . . . . 13 PART II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE . . . 14 PART III: STATISTICAL INFORMATION . . . . . . . . . . . . 22 ANNEX: Response to the Borrower's Perspective to the Project Completion Report . . . . . . . . . . . . . . . . 35 MAP. IBRD 14521R This document has a restricted distribution and may be Lused by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  PROJECT COMPLETION REPORT REPUBLIC OF CAMEROON SECOND LIVESTOCK DEVELOPMENT PROJECT (CREDIT 1010-CM) PREFACE This is the Project Completion Report (PCR) for the Second Livestock Development Project in Cameroon, for which Credit 1010-CM in the amount of US$16 million equivalent was approv6d on April 22, 1980. The Credit was closed on December 31, 1988, four years behind schedule. It was fully disbursed and the last disbursement was in February, 1989. The Federal Republic of Germany (KFW) cofinanced the Project with a grant of US$1.0 million and a credit of US$8.0 million on terms similar to IDA. Parts I and III of the PCR were prepared by the Agricultural Division of the Technical Department of the Africa Region and Part II was provided by the Borrower. The KFW was requested to comment on the PCR and we have not received any comments. The PCR is based inter alia on the Staff Appraisal Report, credit and project agreements, supervision reports, Bank correspondence files and interview with staff involved in project.  - iii - PROJECT COMPLETION REPORT REPUBLIC 0. CAMEROOF SECOND LIVESTOCK DEVELOPMENT PROJECT (CREDIT 1010-CM) EVALUATION SUMMARY Obiectives 1. Despite its la:ge livestock herd, Cameroon was in the 1960s importing about 10,000 metric tons of meat annually. This was about 13% of its total consumption, but import of meat was projected to increase to 25,000 tons by 1985. The main reasons for the meat shortage were the low productivity of the national herd and constraints i.n the marketing system. In the early 19708, Government had prepared a long term strategy, the "Plan Viande", for the development of the livestock sub-sector. The main objective of this project was to continue the implementation of the "Meat Plan" begun under the First Livestock Development Project financed by the Bank. Specifically, the Project aimed at: (a) increasing meat production and the income of livestock producers; (b) freeing more land from tsetse infestation; and (c) strengthening the implementation capacity of government agencies promoting livestock development and (d) completing construction of the Yaound6 and Douala slaughterhouses begun under the first project. Emphasis would be placed on promoting smallholder production of livestock, espeL>:l1y pig and poultry. (paras 1,4). Implementation 2. Start-up of the Project was delayed by 9 months due to problems ins (a) preparing the subsidiary loan agreement between Government and the credit agency and submitting the required legal opinion to the Bank; and (b) recruitment of technical assistance. Effectiveness of the cofinancing credit from KFW was delayed for more than three years, which constrained the quality of credit administration and supervision. Lack of counterpart funds for the veterinary and extension components further delayed project implementation. Consequently, the Closing Date for the Project was postponed four times and the Project closed in December 1988 (paras 11-12). Results 3. The Project achieved most of its objectives. However, it failed in making significant progress in increasing meat production in state owned and private ranches on a sustainable basis and in strengthening the institutional capacity of government agencies in promoting livestock development in the country. Four of the seven project components were successfully completed. These were: (a) strengthening of SODEPA; (b) the  - iv - tsetse eradication campaign, (c) training staff; and (d) preparing a successor project (para 17). 4. The tsetse eradication campaign of this and the previous project freed 1.3 million ha. as intended. It was responsive to the environmental impact of the insecticides used. Government commissioned consultants to study the environmental impact and discontinued the use of Ensodil and Dieldrin on their advice. It started experimenting with alternative low cost and less damaging methods of tsetse control using traps and attractants (para 37). 5. SODEPA managed to complete the two slaughterhouses started in Yaound6 and Douala under the first project. The number of cattle slaughtered per d.ay is 260 and 180 for Yaound6 and Douala respectively. This surpasses the expectationa of the SAR by 73 and 20 percent, respectively. Cameroonian staff were trained in sustainably operating them without financial loss. On the other hand it did not establish the 12 improved private butcheries in Yaound6 and Douala. However, it opened two butcheries, which it had to close as they were making eubstantial financial losses. Unacceptably low technical and financial results continued to plague the three parastatal ranches operated by SODEPA until the fifth year of the project's seven year life due to lack of proper management. SODEPA failed to effectively carry out a task delegated to it, i.e. rendering extension services to cattle ranchers in the areas where it operated cattle ranches (paras 30-35). 6. The credit agency, FONADER, gave loans to 1079 producers (83% of that foreseen in the SAR). The monetary value of the loans surpassed that allocated in the SAR by 23%. However, this component did not succeed in achieving the expected increase in meat production and producers income as the institutional capacity to supervise and support smallholders with extension, veterinary and supervised credit services were not sufficiently developed. The overall recovery rate of the credit that was due for repayment (about US$ 2.3 million) as at June 30, 1986 was 70%. However, this varied from province to province. The lowest credit recovery was in Adamoua (61%) followed by West (70Z) and North-west provinces (75%). The global recovery rate may deteriorate substantially as FONADER was dissolved and replaced by an agricultural development bank after a two year hiatus (paras 20-29). 7. Strengthening of MINEL's veterinary field services to ensure adequate disease prevention and control was not effectively carried out. The Project Coordination Unit (PCU) was established. However, it failed to coordinate project implementation. It did not set up any management information system that could enable MINEL to effectively monitor and evaluate the diverse activities of the Project and it did not develop institutional capacity to undertake such coordination for the successor project. It did not prepare the Project Completion Report as stated in the SAR ard the Credit Agreement (para 38). 8. Calculating the ERR was not attempted as there is no sufficient data on the benefits and costs of livestock production by the private sector project beneficiaries. The ERR estimated in the SAR was 18%. Considering the poor performance in the credit an d livestock services components which were reflected in marginal impact on productivity, the continued unviability of the  -v - parastatal ranches, and the uncertainty with respect to use of the tsetse cleared areas the ERR is likely to be less than 10%. Sustainabilitv 9. Although 10% of the tsetse cleared area (about 150,000 ha.) was reinfested and had to be resprayed, the tsetse eradication program was sustainable, especially with the utilization of less costly and less environmentally hazardous methods of tsetse control introduced under the Project. Sufficient Camerooniane wero trained to continue the progrsn on a sound footing. The two slaughterhouses can be sustainable, especially if slaughter fees are charged in accordance with prevailing economic circumstances. The experience with the management of the three ranches and the two slaughterhouseb by SODEPA showed that they could not be financially viable under paras.atal management. Efforts are being made under the successor project to privatize them. The livestock ministry did not develop its capacity (a) to effectively extend veterinary and extension services; and (b) to supervise and monitor development projecti under its wing, a problem still plaguing the successor project (paras 39-41). Performance of IDA. the Consultants, and Government Agencies 10. Bank advice contributed significantly to improvements in project design and implementation. Close supervision was maintained throughout project implementation. Bank staff were instrumental in discontinuing the use of chemicals toxic to non target fauna in the tsetse eradication campaign. The low performance of SODEPA ranches, the poor supervision of the credit program and the inadequate support of extension and veterinary services to project beneficiaries were brought to the attention of Government. Bank staff could have expedited the RFW credit by putting pressure on both Governments. The objection of Moslem butchers to stunning animals with pistols prior to slaughter could have been handled with better diplomacy (paras. 42-44). 11. The performance of Government agencies involved in the execution of the projects was variable. The Tsetse Eradication Unit executed its program extremely well. It trained its staff properly to take over the functions of the technical assistance. Although the SODEPA management was trying its best to implement the Project, it was constrained by its public nature and structure. While its physical achievements improved substantially towards the end of the Project, it could not be financially viable due to its high overhead costs and overstaffing in its headquarters. FONADER was unable to institute a carefully scrutinized and adequately supervised credit program. Consequently, a substantial part of the credit for ranching operations were utilized for other purposes and credit recovery became a difficult task. MINEL failed to support its own veterinary and extension program and to supervise the Project Coordinating Unit in discharging its functions with adverse effects on the overall efficiency of project implementation. In general, Government was supportive of the Project despite delays in releasing counterpart funds and processing procurement of goods and services (paras 46- 47).  - vi - 12, The performance of most of the technical assistance consultants was satisfactory (para 49). Overall Assessment 13. On balance, the project must be rated as unsatisfactory. Findings and Lessons 14. The Project demonstrated that a large area of tsetse infested land can be freed and maintained at reasonable cost. It also showed that the lack of participation of the beneficiaries in the planning and implementation of tsetse eradication became a constraint in safeguarding tsetse freed areas from reinfestation. The Project showed that a good potential exists for the development of small livestock in Cameroon and that credit could accelerate this development. Unfortunately, FONADER was dissolved and the momentum generated under the Project was dissipated. The Project amply demonstrated once again the difficulty of SODEPA, a public enterprise encumbered with inflexible financial and personnel procedures, to successfully manage a commercial operation. Assigning to SODEPA extension functions, not only duplicated the responsibility of the livestock ministry, but also detracted SODEPA management and resources from its commercial operations. 15. The non-synchronization of the credit effectiveness of the cofinancing agency constrained the efficiency of FONADER and prevented the release of IDA funds that were tied to disbursements of the cofinancier.  PROJECT COMPLETION REPORT REPUBLIC OF CAMEROON SECOND LIVESTOCK DEVELOPMENT PROJECT (CREDIT 1010-CM) PART I: PROJECT REVIEW FROM BN_K'S PERSPECTIVE A. Prolect Identity Project Iuume: Second Livestock Development Project Credit No.: 1010-CM RVP Unit: Africa Country: Cameroon Sector : Agriculture Subsector: Livestock B. Back&round 1.1 Despite its large livestock herd (2.3 million cattle, 2.9 million sheep and goats) compared to a population of less than 6 million people, Cameroon was in the 1960s importing about 10,000 metric tons of meat annually. This was about 13% of the total consurption, but import of meat was projected to increase to 25,000 tons bz 1985. The main reasons for the meat shortage were the low productivity of the nati-nal herd and constraints in the marketing system. In the early i970s, Government prepared a long term strategy, the "Plan Viande", for the development of the livestock sub-sector. IDA's involvement in tha subsector started with a credit of US $11.6 million equivalent to finance the first phase of the "Plan Viande" under the First Livestock Development Project (1975-80). 1.2 The first project aimed to: (a) free 800,000 ha. of highly productive pasture from tsetse infestation; (b) establish three 20,000 ha. state cattle ranches; (c) develop 150 private cattle ranches and farms; (d) construct and operate two slaughter plants; (e) modernize 12 butcheries; and (f) provide technical assistance for (i) a nascent livestock extension service; (ii) training of slaughterhouse and credit managers; and (iii) the establishment of the Government parastatal, Soci6t6 de D6veloppement et d'Exploitation Animale (SODEPA), to implement the state ranch and slaughterhouse components of the project. 1.3 The first project met most of its physical targets except the slaughterhouses, which were to be completed under the second project. However, the project's impact on national meat production fell far short of appraisal expectations (2000 versus 4,300 tons p.a). The economic and financial performance of the parastatal ranches was disappointing. By contrast the performance of the private producers that participated was  -2- beyond expectations. The tsetse eradication had the greatest impact as an estlmated 60,000 head of cattle moved into the area and produced an incremental 1,000 tons of meat p.a. The economic rate of return of the comp;leted project was calculated at 42 compared to 13% at appraisal. C. roiect Obiectives and Description 1.4 The objective of the Second Livestock Development Project, planned for implementation during 1980-84, was to complete some of the components not fully completed during the first project and to implement the second phase of the "Plan Viande4' aimed at increasing meat production and the income of livestock producers. The necond project would complete construction of the two slaughterhouses, expand the tsetse free area and implement a revised trainin&: proeram began under the first project. Emphasis would be placed on promoting swallholder production of meat, especially pig and poultry and stre_gth-ari.ng the implementation capacity of government agencies promoting livestoc3: development; i.e., the MinistAre de I'Elevage et des Industries Animates (MINEL), SODEl'A and the agricultural credit agency, Fonds National de D6veloppement Rural (FONADER). 1.5 Pro1ect Descriptions The Second Livestock Development Project would provide for: (a) i.nvestment credit and working capital through FONADER: for pig breeding and fattening (200 loans), broiler production (100 loans), eggs and cockerel production (100 loans) and cattle production on small mixed farms (500 loans), private cattle breeding and fattening ranches (140 loans) and small-scale cattle production (800 producers in 40 group loans) in the tsetse cleared areas and small ranches in the North-West province (30 loans); (b) strengthening SODEPA through provision of needed civil works, equipment and vehicles and technical assistance to: (i) complete the construction of and manage the two slaughterhouses; (ii) improve its financial control capacity through a financial controller; and (iii) improve its mangement capacity through a ranch management specialist; (c) completing the program of eradication of tsetse flies in the remaining 276,000 ha. of the target area of 800,000 ha. under the first project, further expanding the tsetse cleared area by an additional 163,000 ha. and preventing the infestation of some 900,000 ha. of area jeopardized in the southern slopes of the Adamoua Plateau; (d) strengthening MINEL's veterinary field services in the Adamoua Plateau to ensure adequate disease prevention and control;  3- (e) improving the program of formal and in-service training for: (i) MINEL veterinary field dervice staff; (ii) SODEPA agents in slaughterhouse management and operation; (iii) FONADER agents in credit supervision and monitoring; and (iv) MINEL staff assigned to the Special Mission in tsetse eradication, survey and perimeter maintenance; (f) establishing a Project Coordination Unit (PCU) under MINEL with overall responsibilities for project coordination, headed by a Project Coordinator; and (g) providing funds for the 'paration of a possible follow-up project. D. Project-Design and Organization 1.6 While the project design was not innovative, its scope and scale of operations were appropriate. The Project was well prepared and its design took full account of the experience gained under the first livestock project. Its objectives conformed with `overnment's strategy of increasing livestock production for domestic consumption. The emphasis of production investment markedly shifted away from state operations to private activities. Parastatal investment in the second project was planned for only $ 4.6 million or 15% of base cost. Moreover, the tsetse eradication and improved animal health and extension services components (55% of total base costs) was to support mostly private sector production. 1.7 In retrospect, assigning SODEPA a combined function of commercial activity and a development activity not related to its commercial interests, i.e assisting FONADER in its ranch loans and providing extension services to pzivate ranchers, was a mistake. In the event, SODEPA proved incapable of discharging this responsibility, which should have been borne by the regular extension services of MINEL. 1.8 The first project was totally financed by the Government and project beneficiaries ($ 4 million, or 26Z) and IDA ($ 11.6 million or 74%). The second project was expected to cost US$ 36 million net of taxes, out of which Government and livestock producers would finance $11 million (30%). The Federal Republic of Germany would finance $ 9 million (25%) and IDA the remaining $ 16 million (45%). 1.9 The Project would mainly be executed within the institutional arrangements developed during the first project. The operational capability of MINEL, SODEPA and FONADER was to be strengthened through establishing direct lines of authority and clear definitions of responsibilities and by providing all three institutions with sufficient resources and support to carry out the Project. 1.10 FONADER would administer the credit component of the program promoting increased livestock production by the private jector. SODEPA would  4- oversee the completion and operation of the two slaughterhousee and the operation and management of the three parastatal ranches as w"ill as provide extension services to private ranchers in the tsetse cleared areas. MINEL would be responsible for veterinary and animal production extension services and the tsetse eradication program. The PCU established under MINEL would be responsible for overall coordination, monitoring, evaluating and advising the agencies executing the various components of the Project. E. Project Implementation 1.11 Credit Effectiveness and Prolect Start-up: The Credit was approved on April 22, 1980 and signed on June 20, 1980. It became effective on May 12, 1981. Credit effectiveness was postponed twice due to (a) delays in getting the subsidiary loan agreement between Government and FONADER prepared and signed and the necessary legal opinion on it submitted to IDA; and (b) delays in recruiting the technical assistance specialists. 1.12 Implementation Schedule: The Project was originally scheduled to be completed by December 31, 1984, but most of the project components were completed three and half years later in December 1987. The Closing Date was extended four times to December 31, 1985; December 31, 1986; December 31, 1987 and December 31, 1988. The final extension was made to allow an orderly completion of the physical expansion of the Yaound6 slaughterhouse. The main reasons for the delay were: (a) the late start-up of the Project; (b) late effectiveness of the KfW credit in 1984, three years after project start-up; (c) lack of counterpart funding, especially for the veterinary and livestock extension component; (d) ineffectiveness of the first project coordinator, a technical assistant; and (e) lack of interest by MINEL ia the implementation of the Project during 1981-84. 1.13 Procurement: Procurement for the execution of the components under SODEPA and the Tsetse Eradication Unit were by and large done on schedule and smoothly, while those for the veterinary and livestock extension were inordinately delayed and caused problems of reimbursement due to lack of compliance with Bank procurement procedures and guidelines. 1.14 Project Costs. The estimated cost of the Project at appraisal was about US$ 39.9 million. The final cost was about xx as shown in Table 5. 1.15 Disbursement. The estimated and actual disbursement of the Credit is given in Table 3. Disbursement was delayed due to the late start-up of the Project. Government was slow in allocating counterpart funds co the project, especially to the infrastructural development of MINEL's veterinary services. Government funding for this component was released only in the middle of 1983, two years after credit effectiveness. The KfW credit was sig..ed in April 1983 and became effective about 9 months later. This prevented the release of IDA funds for reimbursement of FONADER expenditures. This contributed to the slow rate of disbursement in the first three years of  -5- the Project. Disbursement was made till February 27, 1989 when the total Credit amount was fully disbursed. 1.16 Credit Allocation. The original, revised and actual allocations of the Credit are shown in Table 4. The original allocation was revised on February 28, 1986 to reflect changes in cost estimates. The allocation to SODEPA was increased by nearly 75% to cater for the expansion of the Yaound6 slaughterhouse and purchasing of more cattle to enhance the viability of the ranches. The allocation to FONADER for private sector credit was significantly reduced (by 31%) because FONADER had used its own resources in the early life of the Project when it was ineligible for IDA disbursement due to delays in the effectiveness of the RFW credit. The allocation to *MINEL for veterinary and extension services and to the PCU also reduced by 25% and 33%, respectively, due to slowness in implementing those components. F. Project Results 1.17 Project Objectives. The Project achieved some of its objectives. However, it failed in the major objective of increasing meat production in state-owned and private ranches on a sustainable basis and in strengthening the institutional capacity of government agencies in promoting livestock development in the country. Four of the seven project components were eventually completed successfully. These were: (a) strengthening of SODEPA; (b) the tsetse eradication campaign; (c) staff training; and (d) preparing a successor project. The credit component managed to disburse more funds than originally allocated but the program was not implemented as foreseen in the project design (See para. 22-24). Strengthening of MINEL's veterinary field setvices to ensure adequate disease prevention and control was not effectively carried out. The Project Coordination Unit (PCU) was established. However, it failed to coordinate project implementation. It did not set up any management information system that could enable MINEL to effectively monitor and evaluate the diverse activities of the Project and it did not develop institutional capacity to undertake such coordination for the successor project. 1.18 SODEPA managed to complete the two slaughterhouses started in Yaound6 and Douala under the first project. Cameroonian staff were trained in sustainably operating them without financial loss. On the other hand, unacceptably low technical and economic performance continued to plague the three parastatal ranches operated by SODEPA due to lack of proper management. 1.19 The tsetse eradication component was successfully implemented and Cameroonian staff obtalned sufficient training and experience to sustain the operation on their own after project completion. Smallholder production of meat was promoted but meat production was not substantially increased because the institutional capacity to support smallholders with extension, veterinary and supervised credit services were not sufficiently developed and sustained after project completion. FONADER has been liquidated under Government's  -6- banking reform program with no bank yet adequately assuming its function of making credit available for omallholders. Physical Results 1.20 The credit component, with an allocation of $9.1 million (30% of base costs) succeeded in achieving 96% of the number of loans targetad in the SAR. The monetary value of the Loans surpassed that allocated in the SAR as shown in Table 4. However, this component did not succeed in achieving the expected increase in meat production and producers' income as the institutional capacity to support smallholders with extension, veterinary and supervised credit services were not sufficiently developed. 1.21 The technical assistance for the credit component wa3 delayed so much that most of the credit had already been granted and disbursed with little technical plannirng, support and supervision. Upon arrival of the technical assistants, much of their time was spent in crisis management of how to effectively increase loan recovery than in training FONADER staff the arts of responsible agricultural portfolio management. 1.22 The Project enabled FONADER to expand its livestock credit operations from one province, Adamoua, to two others, the North-West and the West. Although the total target of loans was largely met, there was a significant variation in reaching the targets set for the different types of producers. Whereas the number of loans for egg production and cattle ranches were overachieved by 14 and 101 percent, respectively, those for pig breeding and cattle production in mixed farms were respectively 27 and 88 percent of the SAR targets. No loans were given to two categories specified in the SAR; i.e., for broiler production and cattle production for groups of farmers to be settled in tsetse cleared areas of the Adamoua Plateau (See Table 4). 1.23 The variation was much more accented in terms of the value of the loans. According to the SAR, credit for cattle ranching was only 39% against the actual which was 60% of the total credit. Total credit given for poultry production was 23%, which is only 3% more than that allocated in the SAR for broiler and egg production. While the value of credit allocated for pig production in the SAR was 25%, that realized by the Project was only 6% of the total credit given under the Project. 1.24 There are several reeaons for this significanit variation. First, loans for pig production were suspended following outbreaks of the African Swine Fever disease, a major killer for which there is no vaccine nor an effective cure. Second, the importation of cheap frozen poultry and the lack of an organized marketing system that could purchase, dress and transport broilers from the North-West and West provinces to urban centers in Douala and Yaound& were major disincentives to investment in broiler production. Producers opted for egg production and sold cockerels and old layers for meat instead of producing broilers. Third, the 40 group loans to 800 farmers never materialized as settlement in the tsetse cleared areas became difficult  -7- due to problems of land tenure and difficulties in organizing farmers into groups. 1.25 The overall recovery rate of the credit that was due for repayment as at June 30, 1986 was 69.8 2%./ However, this varied from province to province. The lowest credit recovery was in Adamoua (61%) followed by Waest (70%) and North-west (75%). The global recovery rate may deteriorate substantially as FONADER has been dissolved and replaced only by the Cr6dit Agricole du Cameroun, an agricultural bank after a two year hiatus. Impact of Credit 1.26 Although one can discern that the Credit had a positive impact on improving somewhat the technology of livestock production and thus on increasing meat production, it is difficult to estimate the incremental amount produced because neither MINEL nor FONADER kept production records. Moreover, the production plans proposed in the credit applications were not followed properl. In the case of cattle production loans, which were 60% of the loan portfolio, only 38% of the borrowers actually invested the loan in cattle production. Even then only half of these followed the proposed plan in their credit application. The others implemented the approved proposal at various degrees below the targets in their applications. This situation would make estimating the incremental production on the basis of loan amounts given highly conjectural. 1.27 Several factors contributed to the non-realization of the plans. These were: (a) inexperience of the borrowers with livestock rearing; (b) shortage of breeding and fattening cattle; (e) weakness of MINEL's extension and veterinary services; and (d) inadequate supervision, lack of discipline and lack of tachnical skills in livestock production on the part of FONADER field staff. 1.28 According to the SEDA study 2/, a majority of the borrowers for cattle ranches (62x) had no previous experience with cattle production. They had other interests like trading, transport, the civil service, etc. They became interested in cattle production only because of the opportunity the Project created for acquiring land and cattle via the tsetse eradication program and the credit facility. These town dwellers used their influence in obtaining what was communal land from the traditional chiefs. In some cases the fenced ranches blocked access of local communities to water points thus *X/ The credit due for payment was FCFA 752.7 million (USs 2.3 million), which was about 40% of the total credit disbursed by FONADER under the Project. 2/ Soci6t6 D'Etudes pour le Developpement de L'Afrique (SEDA). 1987. Etude sur le Cr6dit aux Eleveurs. Programme CMR 85/004. Ministere de L'Elvage, des Pgchee et des Industries Animales, Yaound6.  -8- disrupting herd movements and hampering the traditional management of grazing resources. This often created a conflict between the ranchers and the local communities. The few producers that tried to raise cattle had gross difficulty due to lack of experience in cattle raising, unavailability of breeding and fattening stock and lack of supervision and assistance from MINEL and FONADER. The SEDA study estimates that only 27% of the borrowers were authentic producers that were serious and could implement a ranching development plan. Moreover, there was a serious general drought during 1983-84 and a major outbreak of Rinderpest. This not only increased mortality of the cattle that the borrowers had at the time, but also made it difficult to restock in subsequent years. 1.29 In contrast to Adamoua, most of the borrowers in the two other provinces attempted to seriously implement the proposed plans in their credit application for poultry and pig production. About 67% of the borrowers for pig production were farmers in contrast to only 14% of borrowers for poultry production. 1.30 SODEPA Results. slaughterhouses: The slaughterhouses at Yaound6 and Douala, whose construction was started during the first project were completed in 1984 and 1985, respectively, with technical assistance from a third consulting firm and with a total delay of 7 years. The slaughterhouse offering to butchers custom slaughter service in good hygienic conditions are used to full capacity. The Yaound6 slaughterhouse which was expanded in 1987 is processing about 260 head of cattle per day with a throughput of about 11,000 tons of meat per year. The Douala plant is smaller, processing about 180 head of cattle per day with a throughput of almost 8,700 tons of meat per year. This throughput represents about 60-65 percent of the weat consumed in the two cities. In addition the two plants are producing annually about 150 tons of blood, meat and bone meal. 1.31 The slaughterhouses would have been commercially viable were it not hor the high overhead costs of SODEPA, which are about 30% of the operating costs of the ranches and abattoirs. Operating results of the two slaughterhouses for the financial year 1986-87 showed a gross profit of FCFA 119 million (about US$ 446,000) before depreciation. The net profit reduced to FCFA 2 million ($7,500) after accounting for depreciation. Profit was dampened by Government control of slaughter fees, which had not increased since the opening of the slaughterhouses. 1.32 SODEPA did not achieve its target of modernizing 12 butcheries in Yaounda and Douala as foreseen in the SAR due to lack of interest of butchers. However, it set up and operated two butcheries in Yaound& to demonstrate how modern butcheries operate. The butcheries generated high demand, particularly because their prices were about 20% lower than those charged by private butcheries. Consequently, the operation could not be sustained as it lost about FCFA 6.8 million ($25,500) during its first year of operation and the butcheries were closed dcwn.  -9- 1.33 SODEPA Ranches. Unacceptably low technical and economic performance continued to plague the four ranches operated by SODEPA until the fifth year of the project's seven year life. Calving rates were below 652 and mortality rates were high, particularly for adult cattle at about 7% p.a. (see Table 7b). In addition, there were too many cases of cattle disappearance due to theft and going astray. Emergency slaughterings, to allegedly save sick animals from dying and being wasted, were too frequent. This low performance was a result of the poor management provided by inexperienced ranch managers and indiscipline of ranch staff. Moreover, supervision from SODEPA head quarters in Yaounds was not frequent and rigorous. 1.34 A change in the top management of SODEPA in 1984 instituted improved management techniques and discipline in the work routine of the ranch staff. By 1987, calving rate increased to 74% and adult mortality rates declined to 3%. However, the rates quickly deteriorated in 1988 to 65% and 4.3%, respectively. The total ranch population of cattle increased from 26,880 in 1984 to 34,261 in 1987 and sales from 3,400 to 4,445 head. The ranches became self supporting, though with very low returns of 4% p.a. 1.35 SODEPA Extension Service. In 1985, SODEPA established three extension units in the Northwest and Adamoua Provinces in an effort to disseminate modern ranching techniques and improved breeding stock to cattle producers in the areas where its ranches are located. Although some livestock and range management techniques were imparted to few producers, this task proved to be beyond the capacity of the management of the ranches at that time. It only served to detract the attention of SODEPA's management from the commercial objectives of the company. Moreover, it introduced an undesirable duplication of livestock extensior channels in the country. This function was appropriately transferred to the livestock ministry under the successor project. 1.36 SODEPA Financial Results. The cumulative loss sustained by SODEPA at the end of the 1986/87 financial year was FCFA 625.66 million ($1.9 million) as shown in Table 7c. This loss may be as high as FCFA 900 million if depreciation of the infrastructure is rigorously accounted for. The net worth of SODEPA as at June 1986 was calculated as FCFA 5.7 billion vis a vie a cumulative capital investment of FCFA 6.6 billion. High overhead costs of the central headquarters of SODEPA, incompetent management of the ranches in the early years, and inflexible personnel policies and administrative procedures ill suited to a commercial operation contributed to its poor financial performance. 1.37 Tsetse Eradication. This component of the Project was successfully implemented. A total of 1.55 million ha of land was sprayed, out of which 548,800 ha represented areas, which were sprayed eiuring the first project but were reinfected by tsetse flies. This favorably compares to the target of 960,000 ha of spraying in the SAR, but did not increase the total area intended to be freed by the campaigns under the two projects which was 1.3 ha. The Project used external consultants to monitor the environmental  - 10 - effect of the insecticides used on the fauna of the region. The use of Dieldrin was discontinued and concentration rates of Decamethrin adjusted to avoid environmental damage. Towards the end of the Project, a safer method of tsetse control using traps and attractants was introduced with encouraging results. A chronic problem of the program was the lack of regulated land use in the tsetse eradicated area and effective means of controlling the movement of cattle from tsetse uncleared to cleared areas, making maintenance of tsetse eradicated areas difficult. The major reason for this problem was the failure of the Project to involve the local community in its tsetse eradication program through educational campaigns and participatory planning and execution of the eradication campaign. The cost of tsetse eradication was estimated at US$ 15 per ha with a favorable ERR of 202 by Cuisance et al (1987), assuming the tsetse cleared area was fully stocked.3/. 1.38 Strengthening of Veterinary Services in the Adamoua Plateau. This component of the Project showed little progress due to non provision of counterpart funds and lack of interest by MINEL in its implementation. In 1983, Rinderpest broke out in 5 Departments of the Northern Province and MINEL refused to launch a vaccination campaign in time despite the advice of Bank staff. The bids for building the veterinary infrastructure and for procuring equipment and vehicles were delayed substantially due to lack of government funds. The contract for building veterinary clinics, vaccination parks, offices and staff housing was signed in the fourth quarter of 1985, more than 4 years after effectiveness of the Credit. Five cars purchased for this component were diverted elsewhere. About three fourths of the veterinary infrastructure built was completed at the end of 1986 and the remaining at the end of 1987. While the vaccination parks were used regularly, the offices space provided was in excess of what was required. In 1990, only half of these offices were being utilized. 1.39 Studies. Four important studies were concluded during project implementation: (i) A socio-economic study of traditicnal livestock production systems was conducted to shed light on the major constraints of these systems and possible solutions. The results of the stud) were used in designing the Third Livestock Development Project, which emphasizes forming producers' associations and strengthening extension and training services to these producers. (ii) A study of the impact of the tsetse eradication program was conducted to find out the environmental and economic impact with the purpose of making recommendations on how the tsetse freed areas could be maintained and how the campaign would further be expanded under the successor project. The study influenced the design of 3/ Cuisance, D. et al (1987). Lutte Contre les Glossines: Bilan et Perspectives. HiniBtare de L'Elevage. des P6ches et des Industries Animales. CMR85/004. Yaound6.  - 11 - the campaign for the Third Livestock Development Project. It recommended that the beneficiaries be fully involved in the planning and execution of the campaign to improve the maintenance of the tsetse freed areas and to facilitate cost recovery for this maintenance. It also recommended a less costl method of tsetse eradication. (iii) An evaluation was made on the investment and operations of SODEPA's ranches and slaughterhouses and how it should be restructured to make it financially viable. The evaluation properly analyses the ills of the SODEPA; but its recommendation was a complex proposal to involve SODEPA in establishing more slaughterhouses in the provinces and taking over an existing fattening ranch and establishing new ones in association with village organizations (10% share), private sector (252) and SODEPA (65Z), which made it unattractive to incorporate in the design of the successor ptoject. (iv) A study of the impact of the credit program on livestock producers was carried out. While it characterized the shortcomings of the program and how they could have been avoided, it failed to quantify the benefits and costs realized by the credit beneficiaries. G. Proiect Sustainability 1.40 Although 15% of the tsetse cleared area (about 150,000 ha.) was reinfected and had to be resprayed, the tsetse eradication program was successful and is sustainable, especially with the utilization of less costly and environmentally hazardous methods of tsetse control introduced under the Project. Sufficient Cameroonians were trained to continue the program on a sound footing. It was shown that the participation of the beneficiaries in the planning and implementation of tsetse eradication is important for safeguarding tsetse freed areas from reinfestation. The Project showed that there exists a good potential for the development of small livestock in Cameroon and that credit could accelerate this development. Unfortunately, FONADER was dissolved and the momentum generated under the Project was dissipated. 1.41 The two slaughterhouses are sustainable, if slaughter fees are charged in accordance with prevailing economic situations. The experience with the management of the three ranches and the two slaughter houses by SODEPA showed that they could not be financially viable under parastatal management. Efforts are being made under the successor project to privatize them. 1.42 Under the Project, the livestock ministry did not develop its capacity: (a) to effectively extend veterinary and extension services; and (b) to supervise and monitor development projects under its wing, a problem still plaguing the successor project.  - 12 - H. Bank's Performance 1.43 Despite delays in implementation, Bank assistance and advice contributed to improvement in project design and in implementation, particularly implementation with respect to the tsetse eradication, strengthening SODEPA and the provision of technical assistance components. Close supervision was maintained on the environmental effects of tsetse eradication and Bank staff were instrumental in banning spraying with Ensodil and Dieldrin from the program. Bank supervisors repeatedly pointed out the unacceptably low technical and economic performance of the SODEPA ranch operations, which together with the change in the management of SODEPA led to some temporary (see table 7b) improvement in the performance of the ranches. Bank staff gave valuable counsel in the arrangements for the completion of the Yaound6 and Douala slaughterhouses. 1.44 On the other hand Bank staff could not effectively influence the credit operations of FONADER and the veterinary and extension services of MINEL. The Bank could have expedited project implementation by recognizing the ineffectiveness of the Project Coordinator and arranging for his removal much earlier than October, 1984. The Bank could also have pressurized GTZ, KFW and the Government to expedite the preparation, signing and effectiveness of the KFW credit. 1.45 Bank staff could have been more diplomatic and avoid a confrontation over their insistence that cattle be stunned with gun shots for slaughtering at the Yaound6 slaughterhouse. Although they were right on grounds of efficiency and the practice in other Islamic countries, Bank staff could have been more sensitive to the request of the Moslem butchers who wanted the traditional gorging of the animal without stunning. 1.46 The Bank adequately supervised the Project fielding 12 supervision missions in the 7 year life of the Project. The missions were technically strong and there was good continuity between missions. I. Borrower's Performance: 1.47 The Borrower's performance in executing the project components was variable. The tsetse eradication program was smoothly executed with excellent dialogue between the Borrower and IDA. Although the SODEPA management was trying its best to implement the Project, its public nature and structure did not augur well for efficient commercial management of its enterprises. The opening of the new Yaound6 slaughterhouse was delayed because Government was very slow in closing the old unhygienic abattoir. MINEL did not effectively supervise the Project Coordination Unit, which failed to carry out its monitoring and supervision functions properly. Moreover, it failed to prepare a Project Completion Report, which is clearly stated in the SAR, and is an obligation of Government in the Credit Agreement. MINEL did not give adequate suspport to strengthening its own  13 - extension and veterinary services. FONADER granted and disbursed loans without adequate project preparation and it failed to properly supervise the implementation of the investment. 1.48 The Bank had to resort to notifying Government on December 8, 1983 that it would suspend disbursements by February 15, 1984 unless it would: (a) second nine technicians and recruit a small livestock specialist for the credit component; (b) recruit a training specialist to strengthen veterinary services; (c) close the old Yaound6 municipal slaughterhouse; and (d) present a strategy on how to improve the management of the three SODEPA ranches. These conditions were met by March 15, 1984. The 'laughterhouse was closed by Presidential decree. IDA did not accept the strategy for the financial viability of the three ranches and asked for their privatization or closure. The notice of suspension drew the attention of Government to the problems of the Project a9-A aroused interest in accelerating its implementation. J. Consulting Services 1.49 A number of consulting services were provided by various institutions and individual consultants. The performance of most of the consultants was satisfactory. The exceptions were the first two technical assistants appointed as Project Coordinator and a consulting firm for supervising the construction of the Douala slaughterhouse. K. Project Relation 1.50 Despite delays in project implementation and the notice of disbursements suspension, a cordial re'lationship prevailed between IDA and Borrower staff throughout the implementation of the Project. L. Prolect Documentation and Data 1.51 The Credit and Project Agreements and the SAR provided adequate and appropriate guidance for project implementation. No amendment was effected to the Credit Agreement. A management information system was not instituted by the PCU. Consequently, data was not available on the benefits and costs of private sector beneficiaries of the Project nor data that enable accurate estimation of overall project benefits. FONADER, SODEPA and the Tsetse Eradication Unit kept well organized records. No information was available on the activities of the veterinary and extension service. Consequently, no attempt was made to calculate the ERR of the Project. Considering the poor performance in the credit and livestock service components which were reflected in a marginal impact on productivity, the continued unviability of the parastal ranches, and the uncertainty regarding the extent to which the investment in tse tse eradication is being utilized, the ERR is likely to be less than 102.  - 14 - PROJECT COMPLETION REPORT REPUBLIC OF CAMEROON SECOND LIVESTOCK DEVELOPMENT PROJECT (CREDIT 1010-CM) PART II. PROJECT REVIEW FROM BORROWER'S PERSPECTIVE 1/ 2.1 While the evaluation of those components regarded as successful (Part I, para. 17), the findings of the others, such as sustainable increase in meat production and the strengthening of the capacity of governmental organizations to develop livestock farming, are more nuanced. As regards the increase in meat production, there was undeniable growth in the poultry farming sector, thanks to the loans distributed by FONlADER. The constraints cited: such as competition from imports or the lack of marketing channels, are temporary and, in the case of the second, always a factor in the development of increased production. The increases in cattle and hog production were constrained by the occurrence of two epidemic diseases which the project did little to counter. Consequently, it is difficult to assess whether an increase in meat supplies brought about by the Project was not temporarily affected by these two epidemics, which, as witnessed in other countries are difficult to control. 2.2 The average results obtained in extension services are closely associated with the fact that this component of the Project had been entrusted to agencies for which it was not their primary activity. The efforts of SODEPA suffered from a lack of necessary funding, brought about by the financial and economic crisis, and the IMF and World Bank recommendations with respect to a halting of subsidies to public and parapublic entities. FONADER was unable to devote all the attention required to extension activities with stock farmers because of their other different tasks such as credit and monitoring. Closer attention to these aspects during the appraisal stage of the Project might have made it possible to propose other alternatives to implementing extension services. 2.3 The strengthening of veterinary services in Adamaoua was delayed not by budgetary problems, but because of constraints assoctated with the construction of the requisite infrastructures: selection of locations, awarding of contracts, shortcomings of some contractors. As a result, the buildings could not be completed until after successive extensions of the Project. After the close of the Project, budgetary constraints arising from the economic and financial crisis made it I1/ See Annex: Response to the Borrower's Perspective to the Project Completion Report.  - 15 - impossible to make use of these facilities which explains the low rate of utilization. FONADER 2.4 The credit component experienced difficulties. These were partially overcome, and were ultimately reflected in greater flexibility in the distribution of subloans, as compared with the projections of the appraisal report. Despite the Bank's support on this matter, the agreement between FONADER and KWF could not be signed until 1984, the date initially planned for the end of the Project. In the meantime, at the Bank's insistence, MINEPIA had seconded technicians to FONADER who were unable to obtain maximum benefit since they were hired rather late. The involvement of two cofinancers, together with the existence of different implementing strategies (FONADER and MINEPIA) who did not share the same degree of interest in the distribution of subloans, led to intra-institutional constraints which resulted in considerable delays. 2.5 The occurrence of Rinderpest and African swine fever had varying repercussions on the distribution of loans. Swine fever appeared in the provinces of Western Cameroon, where the project had made no plans (as in the case of Adamaoua), for strengthening veterinary services to combat the epidemic. However, Rinderpest was not the only reason for the lack of purchase of young breeding stock or cattle for fattening. The difficulties encountered were rather structural, as shown by the experience of SODEPA, which ran into the same constraints. The Rinderpest epidemic only accentuated the other problems for a period of time. 2.6 There were two reasons why the extension agents made available to MINEPIA by FONADER were not as effective as expected in monitoring and extension. On the one hand, the late hiring of the two technical assistant experts caused by intra-institutional difficulties, meant that the extension agents could not take full advantage of their experience. Moreover, the stress placed during supervision missions on the objectives of recovering loans certainly prompted the managers concerned to devote greater attention to this aspect of their activities than to extension services. On the positive side, the recovery rates distributed under the Project are among the best by far ever obtained by FONADER. 2.7 Above all, it bears noting that the credit did not take into consideration the operating costs of the extension structure. IBRD's disbursements were limited to the subloans, to the purchase of vehicles and equipment, and to training and technical assistance. From the outset, the very design of the financing plan entailed the risk that the necessary funds would not be available. The Third Livestock Project deals with this aspect more realistically, on the one hand by entrusting the provision of extension services to stock farmers to a single extension agency, and on the other, by providing funds to cover the operating costs of that agency.  - 16 SODEPA 2.8 Slaughterhouses, The technical and financial performance of these slaughterhouses are surely cne of the Project's major successes. The slaughtering fees were set when the slaughterhouses first went into operation at a level considerably higher than the Bank had recommended (CFAF 4,000 instead of CPAF 2,500). Subsequently the directors, who are the parties responsible for setting these rates, not the Government, sought constantly, as did supervision missions, to improve the profitability of the slaughterhouses by improving productivity rather than by taking the easy path of increasing fees; the latter solution also risked inciting users to turn their backs on the slaughterhouses and use clandestine facilities. Accordingly, productivity in terms of tonnage of meat slaughtered per worker-month rose from 5 metric tons to over 11 metric tons in Yaound6, and from 3 metrie tons to nearly 14 metric tons in Douala. Concomitantly, gross profits before amortization amounted to CFAF 137 million 1987/88, or 70 percent higher than the results for 1986/87. 2.9 The installation of modern butcher shops, as might have been expected, met the obstacle of the deep-seated conservatism of this powerful guild. Similarly, the buying habits of virtually all consumers show that more importance is attached to the price of meat than to the modernity of the store where it is purchased and the technical skills of the vendor. Nevertheless, SODEPA did a good job of providing professional training needed to 12 traditional butchers in both Yaound6 and Douala. Despite the closing of the credit on 12/31/86 as far as Category 1 is concerned, FONADER and SODEPA signed a protocol of agreement on June 3, 1987 on the modalities for their collaboration with a view to the distribution of credits to modern butchers. Because PONADER was dissolved shortly thereafter, continued collaboration in this area was not possible. 2.10 Ranches. Factors other than those mentioned explain the average performance of the ranches: (i) the quantitative problems of gathering sufficient numbers of animals made it impossible to stock these units at the planned pace. The good results obtained since 1989 reflect the reality of the operation; (ii) the animals available on the market are those that are put up for sale because they are deemed unsatisfactory. This problem and the preceding one, as also indicated by the recipients of FONADER credits, show that this was a structural problem, since private operators acting directly in their own interest found it just as difficult as Ministry officials to find a solution. (Iii) specialists are of one mind in acknowledging that mixing animals from different sources cannot but lead to high  - 17 - mortality rates; this view, which is well known in poultry farming and swine raising as reflected in the adage of "all in, all out" is also valid for ruminants. In the case of the stocking process excessive mortality was observed until the normal rate of operations was achieved; (iv) the sometimes contradictory changes requested by supervisory missions in the general approach (breeding ranch, fattening ranch, mixed) or in the short-term management (ratio of heads of cattle to ranch employees) did not permit SODEPA to develop and implement a long-term policy; moreover, when the first Project Coordinator, with the tacit approval of the Bank, played a key role in the redefinition of certain objectives, he ended up not devoting as much intsrest as might have been desirable in carrying out the mandate for which he had been hired. 2.11 SODEPA's extension operations were carried out by qualified personnel, who were specially seconded by -INEPIA and had participated in training and research missions devoted to the essential features of the Project. Thus there was no intervention by the ranch management personnel, who were not directly involved in these operations. However, the dual aims of the activities proposed for SODEPA came to light quite rapidly, when the company was asked at one and the same time to provide this extension work and to make all its activities profitable, to conduct its business with an eye on the bottom line. Thus, after having used its own funds for two years, without any subsidy to finance the operation of the extension activities, SODEPA virtually halted all activities in this area. At roughly the same time, the Third Livestock Project appraisal suggested that this activity should be withdrawn from SODEPA, which also influenced its short-term decision. STRENGTHENING OF VETERINARY SERVICES 2.12 The delays noted in this component do not stem specifically form the causes cited. For example, the lack of counterpart funds cannot be advanced, since, quite the contrary, the Government had paid its entire share (40 percent) in the financing of the construction by disbursing it for the startup of the enterprises; this met with the Bank's approval, moreover, as a result of which credit was subsequently to provide 100 percent of the costs up to the initial ceiling. The delays actually were caused by the choice of sites for buildings, the Bank's refusal to finance renovations only, the complexity of the contracting procedures, and, finally, the difficulties experienced by certain enterprises in completing their construction work. 2.13 The sites to be selected for developing infrastructures were not clearly identified during the Project preparation and identification phases. It was therefore necessary to carry out thia lengthy preliminary work before issuing calls for bids, in order that bidders would have the maximum possible amount of information not only about the type of structures desired, but their location in relation to their  - 18 - decision-making center or supply points. The first Coordinator played a significant role in carrying out this preliminary work, as well as in producing the documentation materials for the call for bids. 2.14 Once the sites were identified, it appeared that some existing infrastructures, which were already rather old, could be restored to complete functionality if they were renovated and rehabilitated. The Bank rejected this proposal, for which the drafting of a construction contract would have been quite simple from the standpoint of design, execution, and technical and financial control. These exchanges of views and the decision-making process not only added to the delays, but also to an unfortunate duplication of infrastructures in the field. 2.15 The signing of contracts for the construction involved the work of two separate administrations: MINEPIA, a technical department, and the Ministry of Government Procurement, the only authority under national regulations which is empowered to award contra 's in such large amounts. This structure raised problems, caused slowdowns, and led to delays, which the Bank repeatedly stressed in its Project reviews and which negatively affected this component. In July 1986, however, a Decree was issued which, inter alia, adjusted the thresholds required for review by Government Procurement from CFAF 10 million to CFAF 50 million, with the technical departments retaining financial authority for lesser amounts. In November 1988, a second Decree established the DGTC, whose responsibilities include the procedure for announcing and awarding government contracts. These two measures, in particular the first of them, have made it possible to simplify and speed up the contracting process. 2.16 The procedure of calling for open bids resulted in the awarding of contracts to enterprises which certainly offered the best guarantees from the national standpoint, but which were probably not the best equipped to carry out construction Projects spread over an entire province. In hindsight, they underestimated the conditions under which they were involved withir the very framework of their bids. Moreover, the distances to and scattering of the more than 20 construction sites from their head offices significantly impeded the execution of construction, and in the case of one of the bidders led to considerable delays. SODEPA experienced the same type of problems for the construction of the extension services building in Adamaoua. In future, it would be preferable to limit the calls for bids to enterprises that are located nearby in the same province, and even, by splitting up the bid lots, within the same department. 2.17 The buildings were finally delivered at the planned closing date for the Project. However, the Project had failed to provide funding againsi the Credit for equipping the facilities, and because of the economic and financial crisis, it was not possible to equip them completely using domestic funds. The equipment is being obtained gradually, when budgetary constraints permit, and should make it possible to use the offices constructed in their entirety.  - 19 - 2.18 The Rinderpest epidemic broke out at a time when the infrastructures were unfinished which would have been the case regardless of the procedural rigidity of MINEPIA. Moreover, the veterinary health authorities are constantly on a watch to minimize vaccinations in order to establish animal populations which are definitely identifiable as serologically immune, so as to maintain export potential. The EEC has just forbidden vaccination against foot- and-mouth disease within Community for the same reason. The combination of these two facts thus explains the "delays" in carrying out the vaccinations. However, the Government undertook, following this epidemic and when budgetary appropriations so permitted, to establish mobile prophylaxis teams equipped with vehicles and ample equipment of a suitable quality to enable them to intervene as rapidly as possible whenever there were new outbreaks of epidemic diseases. The use of the term "strengthening of veterinary services" by MINEPIA. The latter interpreted the expression broadly, i.e., strengthening throughout the entire national territory, whereas the Bank obviously stuck to the terms of the Project documents, which limited the purchase of goods (vehicles and equipment) to the province of Adamaoua; after the Bank made these observations, the vehicles were returned or replaced in Adamaoua. PROJECT COORDINATION UNIT 2.19 As the project records bear out, the first Coordinator, immediately upon taking up his duties, performed the important task of translating the Project documents into understandable technical, financial, and budget ry programs. The subsequent delays noted in the implementation of mob. components of the Project, for the various reasons indicated above (especially in paragraphs 4 and 12), prompted him to embark on various studies and activities (ranch reorganization, the extension component of SODEPA, the establishment of veterinary infrastructures in Adamaoua) which were beneficial to the Project in general terms, but probably harmful to the Project Coordinator in particular. It must be noted, however, that the Bank never reported this change in the Coordinator's activities, which obviously did nothing to correct the situation. The second Coordinator was responsible for the creation of a concertation arrangement involving all the players in the Project and set up a system of follow-up indicators. As he left Cameroon after only six months for professional reasons, he was unable to implement this system. The third Coordinator took up his duties at a time when the major concern of the Government and the Bank was shifting to preparation of the Third Livestock Project; he therefore played an active role in preparing and pushing forward all ;hese studies, including, on the Bank's recommendation, the drafting of the proposals relating to the development of the raising of small ruminants in the feasibility study for the following Project. 2.20 The introduction of a coherent monitoring and evaluation system was thus compromised by the fact that three different technical assistants held tha Coordinator position, as well as by the nature of the most pressing tasks confronting them. It should be added that the design for project coordination was probably not the most appropriate.  - 20 Even more thani the project direction itself, coordination depends as much if not more on the human relations the Coordinator is able to establish than it does on his technical capacities. These relations also depend on those with whom he is dealing, persons who may also change over time. Thus, for a project with multiple components and calling for the involvement of various different persons, and sometimes inivolving different supervisory structures (as was the case with FONADER in the Project), simple coordination was probably not enough. 2.21 Similarly, the task assigned to the project coordinator of drafting the Project Completion Report (PCR) six months after the final disbursement, was inadequate, not to mention inconsistent with the terms of the Loan Agreement, there was no way to finance this work after the Project wae shut down. Moreover, despite repeated verbal requests to this effect, the Bank never communicated the Terms of reference or drafting plan of the PCR. The Third Livestock Project provides for the establishment of a Project Monitoring and Evaluation Unit, which will speed up the drafting of the PCR; it should be noted, however, that the provisions of Section 7 of the general guidelines on the preparation of PCRs were not formally agreed jointly. BANK PERFORMANCE 2.22 The Bank's recommendations and observations both during the supervisory missions and from its headquarters, basically allowed for the implementation anid improvement of the technical aspects of the Project. Tthe successive changes in Project leadership were nevertheless harmful because of proposals, in particular as regards ranch reorganization or management, that were contradictory, with each new person responsible for supervision having his own notion of how things should go. The institutional aspects, whether internal or external, even if clearly perceived, did not result in any apparent improvements following interventions by the Bank. The fact that recommendations were issued on speeding up the awarding of contracts, the need for which MINEPIA was already convinced of, did nothing to bring about improvement at the level of the other departments concerned. In such a case, more direct intervention predicated on the evidence of problems experienced by a number of projects would have been more convincing. Internally, the Bank did not stress when it should have the extent to which the Project Coordinator had departed from the activities set forth in his mandate; likewise, it did nor communicate the guidelines on preparing the project completion report in a likely manner. 2..3 The Bank made a great contribution to the secondment of the MINEPIA specialists to FONADER; however, subsequently insisting on the loan recovery rates to be achieved. This prompted officials to devote more energy to that objective than to organizational i8sues and extension services. On the other hand, however, the recovery rates for the Project are among the best recorded by FONADER. 2.24 The Bank's officials made themselves readily available for helping to resolve even the most unforeseeable problems standing in the  - 21 - way of the project's proper functioning. This great flexibility made it possible, for example, for SODEPA to finance trips by religious leaders into muslim areas, but with the animals slaughtered after they were first stunned. The acceptance of this procedure by the islamic community of Douala following this trip resulted in better productivity at that city's slaughterhouse than at the one in Yaound6 (14 metric tons per employee month as compared to 11 metric tons). 2.25 The approach adopted by the Bank to provide SODEPA with the resources necessary for extension activities was inconsistent with the foreseeable approach for the Third Livestock Project, i.e., returning the extension function to MINEPIA staff. The buildings constructed by the project are oversized and inappropriate for this new approach, and will require the managers of the Third Livestock Project to make the necessary adjustments in the first stages of project implementation. PERFORMANCE OF BORROWER 2.26 Finally, the performance of each component roughly reflected the degree of autonomy enjoyed by each implementer in carrying out his part of the project. The components placed under single responsibility, and which also had the benefit of experience accumulated in the preceding project, such as the tse tse eradication, are unsatisfactory. Those depending on a cofinancer (FONADER), the awarding of contracts by another agency (strengthening of veterinary services), or decisions from a higher level in the hierarchy (closing of the municipal slaughterhouse) suffered from the delays inherent in such intra- institutional relationships. In order to avoid these delays, the Third Livestock Project should have provided for single and simple channels for decision making. BANK-BORROWER RELATIONS 2.27 Generally speaking, the relations between Bank officials and the supervisors of the Project were marked by the common concern with finding a consensus view on the most appropriate ways and means of advancing the project. As indicated above (paragraph 26), this objective was more easily attained when it depended on just a single institution in the borrowing country. The single notification of the possible suspension of disbursem6nts, which occurred in December 1983, was triggered mostly by cases in which several players were involved. In the event, it helped speed up intra-institutional procedures and decision making.  - 22 - PROJEC CDNPLE REPO EPUBLIC OF-CA-NEROOM (CR98 010-CO12.sM) SEON LIVESTOCK DEVELOPMENY PROJECT PART III; STATISTICAL INFORMATIO TABLE 1. RELATED BANK LOANS AND/OR CREDTIT Title Approvat Purpose status 1. Caowroam Livestock 1974 To assist Goverrment fn the impLementatIon of completed covelopmmt the first phase of its "lmeat plan", which was Project (L*oan 983-CM) designed to increase meat produtifon in order to reduce substantiatLy the level of meat Imports of 10,000 metric tons p.a. projected to increase to 25,000 mewtric tons p.a. by 1985. This objective was to be achieved mainly through (a) clearing 800,000 ha. of tsetse infested area, Cb) establishing 3 state and 150 private ranches and Cc) providing technical assistance for livestock extension, training of officers of the credit agency and SOCEPA, the parastatal involved in managing the cattLe ranches and the two abattoirs at Yaounde and Doumla to be btilLt under the project. 2. Carieroon Livestock 1989 The project follows-up on its two predecessor Ongoing Sector Oeveotnent and alms to increase meat ard miltk production, Project CLoan 3014-CM) raise producers' income, in particular for poor herders, and reduce the financial burden of the sector on govermtent budget, through Ca) policy reforms aiming at privatization of input distribution and veterinary services and the optimization of public services; and (b) investments in CO) institution building In extension services and herder organizations; Cif) range improvement; and Clii) credit for private producers and heatth service agets. Table 2. Proieet TiS_met_ablea Item ~~Planned Revised Actual - identifLcatton IdentLfied as a follow-up of a predeesgor projoct - PreparatLon 02-05/78 03-06/78 - Appraisal Mission 02/79 02/79 - Loan Negotiations 10/79 12/79 04/80 - Board approval 12/79 01/80 04/22/80 - Loan signature 06/20/80 - Loan Effectiveneus 09/22/80 12/31/80 05/12/81 -Project Completion 06/30/84 12/31/88 - Loan Closing 12/31/84 12/31/85 12/31/88 12/31/86 12/31/87 -------------------- -