Cameroon - Livestock Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 4556 PROJECT PERFORMANCE AUDIT REPORT CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) June 16, 1983 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. WEIGHTS AND MEASURES 1 Hectare (ha) = 2.47 acres 1 Kilogram (kg) 2.2 lbs. 1 Metric Ton 2,204.6 lbs. ABBREVIATIONS FONADER - Fonds National de Developpement Rural FRG - Federal Republic of Germany MINEL - Ministere de l'Elevage SODEPA - Societe de Developpement et d'Exploitation des Productions Animales FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) TABLE OF CONTENTS Page No. Preface .. . * * * * * * * * * * * * ** * * * * * * * * * * * * Basic Data Sheet .. . .0 . . . .0..........**** **** **** **** **** **** **** .. ii Highlights ....********* *************************************** iii PROJECT PERFORMANCE AUDIT MEMORANDUM I. SUIMMARY ... ....** *** *** ** *** *** *** *** *** ** 1 II. MAIN ISSUES ................ *********************.. 3 A. General ............. 0............................ 3 B. Intensification of Livestock Production ......... 3 C. Tsetse Eradication .................. ..... ..*... 5 D. The Slaughterhouses ... .......................... 6 PROJECT COMPLETION REPORT II. Project Formulation and Appraisal 13 II. Project Implementat ion .. .. .. .. .. .. .. . ... .. . ... . ...... 17 IV. Project Impact ... .. . .. .. . ... .. ... . .. . .......... ..... 32 V.* Rates of Return ................ . .*............... 33 VI. Institutional Performance ..o......... 0...........*** 35 VII. Consultants Performance . ... .. .. .. . ............ .36 VIII. Bank Performance * .- ............... .......... 37 BaCocgr on .............****************************.. 138 Annexes I-IV Map This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  PROJECT PERFORMANCE AUDIT REPORT CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) PREFACE This is a performance audit of the Livestock Development Project in Cameroon, for which Loan 983-CM was approved in March 1974 in the sum of US$11.6 million. The final disbursement was made on June 2, 1981, and an undisbursed amount of US$66,000 was cancelled on May 31, 1981. The audit report consists of an audit memorandum prepared by the Operations Evaluation Department and a Project Completion Report dated October 16, 1981. The PCR was prepared by the Western Africa Regional Office on the basis of a country visit in March 1981. The audit memorandum is based on a review of the Appraisal Report (No. 295-CM) dated March 6, 1974, the Presi- dent's Report (No. P-1367-CAM) of March 13, 1974, the Loan and Project Agree- ments dated May 14, 1974 and the PCR. Correspondence with the Borrower and internal Bank memoranda on project issues as contained in Bank files have also been consulted, and Bank staff associated with the project have been interviewed. An OED mission visited Cameroon in November 1982. The mission held discussions with officials of the Ministry of Livestock, the Rural Development Fund (FONADER) and the Livestock Development Agency (SODEPA). A field trip to the tsetse control area and ranches of participating cattle owners was under- taken. The information obtained during that mission was used to test the validity of the conclusions of the PCR and permitted discussion of the ranch and slaughterhouse aspects. A copy of the draft was sent to Government on March 15, 1983 for comments but none were received. The audit finds that the PCR covers adequately the project's salient features, its accomplishments and shortcomings, and the PPAM generally agrees with the conclusions. The issues discussed have been selected because of their importance for this as well as other Bank-assisted livestock projects. The valuable assistance provided by the Government, the agencies involved, and their staff, as well as the farmers met during the preparation of this report is gratefully acknowledged.  - It - PROJECT PERFORMANCE AUDIT REPORT CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) BASIC DATA SHEET KEY PROJECT DATA Appraisal Actual or Actual as Z of Item Estimate Estimated Actual P;,raisal Estimate Total Project Costs (US$ million) 14.6 19.6 134 Loan Amount (US$ million) 11.6 11.54 *99.5 Date Board Approval 03/19/74 03/26/74 Date Effectiveness 08/16/74 09/16/74 Date Physical Components Completed 12/79 not yet completed Proportion then completed (2) 90 Closing Date 06/30/80 05/31/81 Economic Rate of Return (2) 13 4 31 Financial Rate of Return (Z) - 6 Agronomic Performance good to poor (parastatal ranches) Number of Direct Beneficiaries (year 82) CUMULATIVE DISBURSEMENTS FY75 FY76 FY77 FY78 FY79 FY80 FY81 Appraisal estimate (US$ million) 1.9 4.3 6.7 8.7 10.3 11.3 11.6 Actual (US$ million) 0.2 1.5 4.2 7.0 10.2 11.0 11.5 Actual as 2 of estimate 11 35 63 88 99 97 99 Date of final disbursement Principal repaid to (mo./day/yr.) (US$ million) MISSION DATA Date No. of Mandays Specializat)ons Performace Types of Mission (mo./Yr.) Persons in Field RepresentedLl Trend Prble Identification 08/71 2 80 Preparation 10/72 3 25 Appraisal 05/73 5 140 Subtotal 245 Supervision 1 11/74 1 T b 2 2 F Supervision 2 05/75 1 6 b 2 2 F Supervision 3 12/75 1 12 d 2 1 F,T,P Supervision 4 06/76 2 40 b.d 1 1 T,M Supervision 5 08/76 2 24 d,b 1 1 T,M Supervision 6 02/77 1 20 d 2 1 T,F,M Supervision 7 07/77 2 40 b,d 1 1 F,M Supervision 8 03/78 2 40 b,d 1 1 M,F Supervision 9 06/78 1 12 e no rating given Supervision 10 12/78 3 40 b,c,d 1 2 F,M Supervision 11 03/79 5 35 2xb,c,d,e 2 2 F,M,T Supervision 12 11/79 2 70 b,d 2 2 F,M,T,0 Supervision 13 03/80 3 40 d,e,f 2 2 F,M,T,0 Supervision 14 05/80 1 6 d 2 2 F,M,T,O Supervision 15 11/80 2 24 d,c 2 2 F,M,T,0 Subtotal Completion 03/81 I Total OTHER PROJECT DATA Borrower United Republic of Cameroon Executing Agency Societe de Developpement et d'Exploitation des Productions Animales (SODEPA); Fonds National de Developpement Rural (FONADER) Fiscal Year July 1-June 30 Name of Currency (abbreviation) Franc de la Communaute Financiere Africaine (CFAF) Currency Exchange Rate: Appraisal Year Average US$ 1.00 - 222.70 Intervening Years Average US$ 1.00 - 227.21 Completion Year Average US$ 1.00 - 211.30 Follow-on Project: Name Second Livestock Development Project Loan Number 1010-CM Loan Amount (US$ million) 16.0 Date Board Approval 04/22/80 /a a - agriculturist; b - agricultural economist; c - financial analyst; d - livestock specialist; e - teetase specialist; f - slaughterhouse specialist. /b 1 - problem-free or minor problems and 2 - moderate problems. Te 1 - improving and 2 - stationary. T_ F - financial; M - managerial; T - technical; P - political; and 0 - other.  - iii - PROJECT PERFORMANCE AUDIT REPORT CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) HIGHLIGHTS The project was to assist the Government of Cameroon in the imple- mentation of the first phase of its livestock development program. It would comprise establishing and operating three 20,000 ha state ranches, two slaugh- ter plants, modernizing 12 butcher shops, developing livestock production on 150 private ranches and farms, and eradicating tsetse-flies from 800,000 ha of good pasture land. Direct benefits from the project would be the incremental annual production of about 4,300 t of meat, 2,000 breeding heifers, 1,500 improved breeding bulls and 1,500 feeder steers. The project accomplished most physical objectives, except for the slaughterhouses which are nearing completion in the follow-on project. The three Government ranches have been established but their financial viability is in doubt due to low production coefficients. About 800,000 ha grazing lands have been cleared from tsetse. Introduction of mixed farming and small private ranches has been successful. The number of loans, especially for small farms, exceeded appraisal estimates by 70%. Overall project impact on national meat production was less than expected. Incremental meat production is likely to reach only about 2,000 t per annum at full development. The economic rate of return has been recalculated at 4% compared with the apprai- sal estimate of 13%. Other points of interest are: - parastatal ranching proved to be unsuccessful because of over- staffing and lack of proper management control (PPAM, para. 13; PCR, paras. 3.30-3.40); - mixed farming and small private ranches were highly successful and could serve as models for livestock development in other African countries with similar ecology (PPAM, paras. 15-19; PCR, paras. 3.02 and 4.06); - the tsetse eradication program will be only partly successful because the delayed preparation of a land use plan has led to uncontrolled influx of herdsmen. Establishment of successful small ranches will be difficult if not impossible due to the immigrants- claims to the land (PPAM, paras. 20-21); - considerable cost overruns and design deficiencies were encoun- tered in the implementation of the slaughterhouse component due to poor appraisal and supervision (PPAM, paras. 23-32). - iv - Major lessons learned from project experience are: - to firm up cost estimates of agro-industries, it would be advisable to invite bids immediately after appraisal and prior to Loan/Credit negotiations (PPAM, para. 33); - specialized consultant supervision for agro-industry projects is called for since there is a lack of "in-house" expertise (PPAM, para. 34); and - there is a need for careful prequalification of consultants involved in the design of agro-industries (PPAM, para. 35). - 1 - PROJECT PERFORMANCE AUDIT MEMORANDUM CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) I. SUMMARY!! 1. At the end of the 1960s, Cameroon was facing a meat deficit of approximately 10,000 t annually. The main reasons for this shortage were the low F:Gductivity of the national herd and constraints in the marketing system. As a basis for a long-term strategy the Government prepared the "Plan Viande". The Bank evaluated this plan and, with certain modifications, accepted it as the basis of a livestock project. 2. On the production side, project objectives were to assist livestock owners in intensifying their grazing methods and to safeguard grazing lands through tsetse fly eradication. On the marketing side the project was to raise meat hygiene and to minimize waste by improving slaughter and retail facilities. The philosophy underlying the main project components, tsetse eradication and ranching, was to use experience gained in other African countries (in Nigeria for tsetse control program and in Kenya for ranching). The credit component was kept small because only limited experience was available in this field at the time. 3. The project was appraised in 1973, and the loan (US$11.6 million) became effective in 1974. The project consisted of (i) providing credit facilities to small, mixed farmers and settled graziers through the Fonds National de Developpement Rural (FONADER); (ii) exterminating tsetse flies on about 800,000 ha in the Adamoua; (iii) establishing three parastatal ranches and constructing two slaughterhouses which would be managed by the Societe de Developpement et d'Exploitation des Productions Animales (SODEPA); (iv) training personnel; and (v) preparing a second livestock project. 4. At appraisal it was acknowledged that the project would involve a considerable risk. The ranching component was therefore reduced from the requested seven to three ranches, the tsetse fly eradication campaign was superseded by a survey, and the credit component was kept small. Project implementation was entrusted to an experienced livestock consultant and was supplemented through technical and financial bilateral assistance. 5. The credit scheme was the most successful project component. It was well adapted to actual needs of livestock owners and provided, through land allocation and credit, the basis for a sedentarized and improved livestock 1/ Adapted from the PCR. - 2 - production. Implementation of this component was, however, affected by problems of lack of extension services, high credit administration costs and repayment difficulties. 6. The tsetse extermination campaign achieved its objective and cleared 800,000 ha of Adamoua grazing lands. The use of insecticides was found to have no lasting impact on the environment. The main problem of the tsetse component was its high cost per ha, due to a more severe than expected degree of infestation. The long-term success of the program, however, would depend on the total area to be controlled in order to prevent reinfestation (which could prove larger than expected) and on the capacity of the Government to provide the necessary resources to the Tsetse Control Division. 7. The three SODEPA ranches were established but did not achieve their objectives. Their financial viability has not yet been assured due to high cattle mortalities, an extended fattening period, unavailability of female stock and high administrative overheads. It has been agreed with Government to review the future development of the ranches on the basis of objective technical criteria. 8. Extreme difficulties were encountered during construction of the two slaughterhouses. The first contractor declared bankruptcy before actual works even began. Construction of the Yaounde plant was almost completed (95%) while that of the Douala plant had reached only 60% when the second contractor also declared himself financially unable to continue. Under the follow-on project funds have been provided to complete both plants. 9. The overall impact of the project on the national meat production was less than expected. The few technical innovations which became available could only be implemented on a limited scale due to the lack of an effective extension service. Within the project, the tsetse eradication campaign had the strongest impact on the national meat production. Approximately 60,000 head of cattle have re-entered the tsetse freed area, and the resulting incremental production is estimated at 1,500 t of meat annually. 10. The traditional sector credit scheme has demonstrated that land allocation, access to credit and technical advice can have a significant impact on the productivity of the traditional pastoralists. Given these encouraging results, this component was expanded under the second project. The economic rate of return for the project, recalculated at 4%, fell short of the 13% appraisal estimate. This drop resulted mainly from the difficulties encountered in the ranching and slaughterhouse components. In terms of institutional development, the creation of a separate Tsetse Control Division located at Ngaoundere represents a significant achievement. On the other hand, as indicated above, the project has not succeeded in turning the three ranches into financially viable enterprises. 11. The consultant's performance suffered from frequent changes in key personnel, and the evaluation of the first project also proved deficient as it failed to highlight the ranch problems, which remained undetected for some time. - 3 - II. MAIN ISSUES A. General 12. The project was a mixed success. While the introduction of modestly sized private ranches should be considered as a model for future livestock development in Africa, especially in areas of nomadic or semi-nomadic pastur- ing of herds, performance of the parastatal ranches was disappointing. A substantial area of the Adamoua region was successfully cleared of tsetse fly infestation, but proper land use is not safeguarded and insufficient control of herd movements may lead to an early reinfestation. The least successful project component was the slaughterhouse construction which was and still is beset with a host of problems. In the following these issues are discussed in detail. B. Intensification of Livestock Production 13. The project was prepared and appraised when the Bank had only limi- ted experience in livestock development in Africa in general and the little that was available was mainly restricted to ranching in East Africa. Based on the commercially successful ranches in Kenya and Tanzania with whom the Bank was dealing and also encouraged by the performance of private ranches like the Compagnie Pastorale in Cameroon and others in West Africa, the emphasis of livestock development was to be put on large-scale ranching. Since con- cessionary ranch development by private expatriate companies was no longer acceptable, the concept of parastatal ranch development gained in importance. The experience with parastatal ranching in this project proved to be dis- appointing, with the same negative results as encountered in East Africa../ 14. The same reasons for the poor performance can be found for all these projects. Due to prevailing un- or under-employment in the countries con- cerned, political pressures led to considerable overstaffing of the ranches, thereby hampering their financial viability. The financial and economic viability of the enterprises is further curtailed by "unaccounted" calf and cattle losses. It is obvious that theft by staff occurs frequently when comparing the reported excessively high losses due to predators or snake bite with the much more favorable coefficients on neighboring private ranches. Introduction of stricter controls, for example surrender and condemnation of cattle perished due to predators/snake bite would certainly reduce these "losses" rather rapidly. Equally important would be regular controls at relatively short intervals to determine the number of cows in calf. Calf losses should also be verified by surrendering dead calves for inspection and condemnation. 1/ See Tanzania Second Livestock Project, OED Report No. 4241, dated December 27, 1982 and Madagascar Beef Cattle Project, OED Report No. 1559, dated April 11, 1977. 15. In contrast to the poor results obtained on SODEPA ranches (see also PCR paras. 3.29 ff), development of the private small ranch subcomponent succeeded beyond most optimistic expectations, thanks to the excellent work of project managers supplied under German aid, assisted by motivated Cameroonian staff. 16. The appraisal expected to fund 150 small-scale cattle fattening and larger breeding/fattening operations. Smallholder fattening introduced the concept of mixed agriculture:livestock farming systems to smallholders in the arabica coffee-growing Northwest Province. The average farm size of 4 ha would permit introduction of grass-legume fodder crops. A total of 35 farms was expected to participate in this category. However, demand proved much stronger, and 172 farmers participated (see PCR para. 3.04). 17. The strong demand for development loans can be easily explained by the high profitability as shown by a 40% financial return on the investment. Furthermore, farmers have recognized the beneficial effects of manure produced by the animals. By substituting manure for costly fertilizers in their coffee plantations, savings as well as higher coffee yields are obtained. The scheme is similar to the operations introduced successfully in an East African project, / and should be considered as a breakthrough in introducing modern mixed farming systems to a society of traditional agriculturists without any previous experience in animal husbandry. 18. Equally successful and impressive was the development of the private ranches. Two different types (250 ha and 750 ha) were to be financed. Intro- duction of permanent, fenced grazing in an area where previously nomadic or semi-nomadic herdsmen dominated was at least as revolutionary as introducing the agriculturists to mixed farming. The possibility of obtaining long-term leases, which could serve as collaterals for obtaining loans, proved to be an important innovation, contributing to the success of the scheme. Comple- mentary feeding of cottonseed cake during the dry season and first efforts to make hay also for improving cattle feeding during this period are important steps in modernizing the traditional sector. As a result production coeffi- cients2/ have reached levels unparalleled in most developing countries. 19. The relatively rapid spreading of the ranching idea is, to a certain extent, also attributable to project management efforts to make best use of the successful ranches- demonstration effect. However, ranch visits were limited by transportation bottlenecks since only a few herdsmen could be 1/ Malawi, Lilongwe I-III, OED Report Nos. 751, 1597 and 3414. 2/ Calving rates between 55-65%, calf mortality 2% and less, adult mortality negligible. - 5 - brought to the ranches with the relatively small vehicles available to the project. In the audit's view it would be appropriate to consider financing small buses in the framework of supporting extension activities in a given region. Since demand of a single Bank-assisted project might be insufficient to justify the purchase of a bus, efforts should be made to analyze demands of other projects, not only agriculture but also for instance education, so that one vehicle could serve a number of different schemes. C. Tsetse Eradication 20. One of the project's subcomponents assisted a tsetse eradication scheme which freed almost 800,000 ha from the fly, opening up a vast area of good grazing potential. The spraying campaign was successful but development of the area has been hampered by the late preparation and issuance of a sound land-use plan. The project provided for 30 man-months of consultant services to assist Government in preparing detailed plans for the effective use of areas cleared of tsetse flies. This study got off to a late start and despite an agreement that disbursements against expenditures on tsetse fly extermina- tion would be made only if this land-use study was undertaken. However, such a detailed study did not become available until 1981 (PCR para. 3.18). 21. The audit questions the decision to go ahead with the eradication campaign without proper land-use plans. Since preparation of these plans cannot be considered too time consuming - the appraisal had anticipated a total manpower input of 30 months, i.e. completion of the report would be feasible in about one year - the question should be asked why this study was not undertaken during project preparation or at least immediately following appraisal. The delay in preparing the study has already had negative conse- quences. There are problems with properly controlling the influx of tsetse infested cattle, overstocking and overgrazing. In addition, possibilities of expanding the above-mentioned small-scale ranching systems, with higher productivity, appear limited, and, even if they are further pursued, uncon- trolled land use will lead to litigations with traditional pastoralists claiming "traditional" rights to certain areas. 22. There is another aspect to the tsetse eradication campaign. The PCR (para 3.27) reproduces the results of a study which investigated the effects of the insecticides applied on the environment. The conclusion of this study is that "no significant adverse impact on the environment was detected". Cameroonian officials complained to the audit mission that, despite the favorable findings of the study, the Bank had informed the authorities of its refusal to continue financing the same insecticides in the follow-on project.-! The recommended substitutes are not only much costlier, but during 1/ The Region states that during late 1981, the Bank received a new report which concluded that: 'The application of Dieldrin has to be considered the last resort, to be used only in case all other alternatives are ineffective, not only because it has a high immediate impact on non- target organisms, but also because the lasting impact on warm blooded animals is very strong.- These comments led the Bank to recommend to Government to discontinue the use of Dieldrin. In search for an en- vironmentally less harmful insecticide, the Bank repeated an earlier recommendation that Government should begin trials with Decamethrin. - 6 - the first applications severe damage to bird and fish life was noticed.l In the audit's view, it would be worthwhile to investigate the implications to the Bank of prohibiting application of insecticides instead of restricting itself to recommending a different chemical. The question should be asked if it were not more appropriate for the Bank to advise on the pros and cons of available insecticides and leave the decision which one should be applied to the'Government. D. The Slaughterhouse Experience 23. The project provided for the construction of two slaughterhouses, one in Yaounde and one in Douala. Total cost of these facilities was esti- mated at appraisal to reach US$1.822 million. The lowest bid received under ICB shortly after project start-up amounted to US$2.8 million and the second lowest to US$3.1 million. However, both of these bidders were not fully qualified, as events demonstrated, and the two plants were never completed during project implementation. A salvage operation had to be started under the Second Livestock project, and, by late 1982, only the Yaounde abattoir had been completed, while the Douala slaughterhouse is only 50% completed. Total construction costs are now estimated to exceed US$4 million, a cost overrun of 220%. In the audit's view, the Bank has to take substantial blame for the disappointing outcome of this project component because mistakes were made during appraisal and supervision. 24. The audit did not succeed in obtaining a copy of the consultant-s preparation report and was, therefore, unable to determine how many changes, if any, had been introduced during appraisal. 25. The appraisal was based on a rather general concept of municipal slaughtering facilities, where SODEPA would slaughter its cattle from the project-financed ranches and where local butchers would bring their slaughter animals for the same purpose. Capacity was to be based on the likely growth in population and consumption. For cattle the appraisal estimated a peak slaughtering throughput of 25 heads per hour or 200/day. In addition sheep/ goats and pigs were to be slaughtered. Emphasis was put on utilizing by- products such as sheep intestines for the production of casings destined for export, as well as blood- and bone meal. Due to religious customs of a large population segment, ritual slaughter was anticipated, and there was strict separation of the hog slaughtering facilities. 26. The audit found the appraisal flawed in several aspects. First, the anticipated peak capacity for cattle could never be reached if ritual slaughter was to be observed. Under the religious laws slaughtered cattle have to face in a certain direction and have to be completely bled until dead, 1/ The Region states that this resulted from the fact that the expert's recommended dosage was well in excess of an environmentally safe appli- cation and the Bank did not have the in-house capability of cross- checking this recommendation in advance and that this experience shows that such schemes are very risky. - 7 - i.e. at least for five minutes. Since only one cattle race, one stunning area and one bleeding pit were foreseen, actual peak capacity is only 12 head/hour. Second, although available statistics, reproduced in Annex 8, Tables 3 and 4 of the SAR, indicate that the maximum small stock slaughter throughput per annum reached only 1,400 in Yaounde and 5,300 in Douala, designs were requested to bring annual capacity to 30,000. What was completely overlooked is the fact that most small stock have been and still are slaughtered at home or in the butchers' back yards. Third, the anticipated utilization of small stock intestines was also treated inconsistently. While ;:he appraisal report, correctly, stated that traditionally all intestines are used for human con- sumption, it nevertheless provided for the purchase and installation of machinery to prepare the same intestines for export as casings. Irrespective of the traditional consumer habits, it was overlooked that local breeds are of such small size that the intestines are unsuitable for casings. 27. Following the appraisal, consultants were to be employed by SODEPA to prepare detailed slaughterhouse designs, specifications and tender docu- ments. SODEPA evaluated the proposals of 5 firms and sought Bank approval for starting negotiations with one firm. From documents on file it is evident that the cost factor was determining the choice. However, it can be stated that the first consulting firm had a fine reputation and experience in slaughterhouse construction. A fee of US$110,000 was determined, but then negotiations bogged down because SODEPA, on the Bank's advice, refused to agree on advance payments and/or issuing an irrevocable letter of credit to cover the consultant's risks. In the audit's view, the Bank not only gave the wrong advice to SODEPA but also got involved in an issue which was beyond its responsibilities. In discussions with other departments, it became apparent that the position taken by the Region regarding advance payments or guarantees to consultants is not shared by them. A common approach for all Bank depart- ments seems warranted. 28. The second choice for the consultant assignment was less qualified and experienced, and some of the problems arising during construction can be traced to his work. While some major shortcomings in his design could be corrected by the Bank during its review prior to tendering, others remained undetected (for example, the slaughtering/processing hall which is at least 1 m too high; insufficient roof overhang at loading ramp, etc.). 29. When bids were opened, the lowest was 56% above appraisal estimate (see para. 23 above). Confronted with this unexpected cost overrun, the Bank got - in the audit's view - too much involved in making sure that the lowest bidder was also a "qualified" bidder. Learning about the bidder's financial difficulties, a mission was dispatched to Sweden to investigate the firm's banking connections, plant facilities, etc. and concluded that everything was all right. As things turned out, the qualification and the usefulness of this mission can be questioned because only a few months later the firm went bank- rupt. By undertaking this kind of investigation and checking the Bank assumed a responsibility which goes beyond its tasks and which could be wrongly inter- preted by borrowers and, in their eyes, render the Bank liable for resulting losses. 30. Additional losses occurred in this case because the Bank, contrary to its own rules and regulations and contrary to the advice of the supervising consultant, approved an additional 10% advance to the same firm, although this - 8 - amount was not covered by any bank guarantee. Eventually the US$280,000 of this transaction were lost by the Borrower due to the firm's bankruptcy. 31. Further problems with the abattoirs went unnoticed when a new contract was approved with the second lowest bidder. In his contract it is stated that the building would be constructed by using his brand-name panels, sandwiching a concrete filling. There is no indication that anybody ever investigated the proposed technology. As it turned out the panels are based on a wood shaving-cement mix and are now subject to termite infestation. 32. Other problems include the approval of blood drying equipment with a 7000 1 capacity, although only 1600 1 could be obtained under the unrealis- tically assumed peak capacity of 200 head of cattle/day. Overcapacities were also tolerated for generators and chilling equipment, procurement of "luxury" equipment, such as a cloth ironing machine or an electronic microscope or a mechanical hoist for 15 kg small stock, to name only a few, were approved. Unsuitable equipment for processing of intestines or dehairing of pigs was accepted. Some of the shortcomings have been corrected by the plant manager - at cost - but others cannot be rectified. Major problems can be expected with the Douala plant, not yet completed, due to a change in site. The high water table at this site will require costly installation of drainage and provision of an affluent pumping system. In regard to the change of site the only Bank reaction found by the audit was in a supervision report (see PCR para. 3.49), and no action was ever taken to persuade SODEPA to use a more appropriate site. The same mistakes made in Yaounde in regard to equipment will be repeated in Douala since everything was ordered prior to the comple- tion of construction. 33. There are several lessons to be learned from project experience. First, more realistic cost estimates have to be introduced for agro-industrial components. One way would be, and this is already practised by one of the Bank's regions, to invite bids for an agro-industrial undertaking in the months after appraisal but prior to negotiations. This procedure would not only firm up cost estimates and avoid sizeable cost overruns for which financing is not assured, but would also contribute to immediate startup of construction, reduction of the disbursement period and elimination of likely time overruns. 34. Second, there remains a weakness in Bank expertise on agro-indus- tries. Involvement of specialized consultants in supervision is necessary and must not be eliminated on grounds of "budgetary constraints". 35. Third, there is a need for careful prequalification of consultants involved in slaughterhouse design. The remark found in the files when select- ing the design consultant for this project that he "reputedly" had experience with small abattoirs is not sufficient. 36. A postscript. The Yaounde abattoir has been ready since August 1982 but was not operating at the time of the audit mission's visit in November. It was explained that religious leaders had reservations about the method to stun cattle prior to slaughter and also that slaughter cattle would not face -9- in the prescribed direction. Section 3.06 of the Loan Agreement stipulates that the Borrower shall close the existing municipal abattoirs "as soon as the slaughterhouses become operational". It remains to be seen if the Bank will succeed in enforcing this covenant.21 1/ The Bank has been notified that the abbatoir started operating in Decem- ber 1982 but the old municipal facility also continues to function.  - 11 - PROJECT COMPLETION REPORT CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) October 16, 1981 Western Africa Regional Office  - 13 - I. BACKGROUND 1.01 At the end of the 1960s, the Government of Cameroon became aware that the country had insufficient meat to feed its own people. The productivity of local herds was low, the existing marketing system did not function properly and, as a result, the country had to import increasing amounts of meat. Something had to be done. 1.02 The low productivity of the national herd was the main constraint. Despite its size -- 2.3 million cattle and 2.9 million sheep and goats, compared to a human population of only 5.7 million people -- the meat supply remained low. The average per capita consumption was estimated at 12.5 kg of meat in 1969. In order to reach this level, the country had to import annually 10,000 t of carcasses or 13% of the total meat consumed. By 1985, the annual meat deficit was expected to reach 26,000 t. 1.03 Marketing was identified as the second important constraint. Cameroon's best grazing areas are in the Adamaoua, the North, and in the North West of the country. These are also the regions where the main cattle, sheep and goat populations are held. The main consumption centers, however, are the cities in the south and southwest, with a large urban population (29% of the total) and an annual growth rate of 7%. 1.04 To overcome this growing meat deficit, the Ministry of Livestock prepared the "Plan Viande", consisting of proposed measures to increase meat production and improve meat supply to the fast growing urban consumption centers. The main components of the Plan were: - Rehabilitation of existing and creation of new research facilities to assist producers through applied research and training (using agro-industrial byproducts); - establishment of feedlots; - improvement of existing marketing infrastructure, (markets, cattle routes, holding grounds and abattoirs); and - improvement of marketing practices. II. PROJECT FORMULATION AND APPRAISAL 2.01 Between 1969 and early 1971, several Bank missions visited Cameroon to investigate the possibilities of Bank participation in the development of the Livestock Sector. In February 1971, the Bank reviewed the "Plan Viande" and concluded, because four years out of the five-year program were suggested to be used for studies and planning, that the proposals were too time consuming. Instead, the Bank recommended to move into project operation more swiftly. 2.02 During September 1971, an identification mission visited the country. The mission found very favorable conditions for the development - 14 - of animal production, and a Government with a strong interest to improve the livestock sector. The mission reported Government's intention to implement the "Plan Viande" and that the Government was already actively looking for external aid financing to implement the proposals. 2.03 The mission confirmed the Bank's general interest in assisting the livestock sector and reiterated the Bank's preference for more operation-oriented project components. The mission therefore proposed: (i) 7 State ranches of at least 50,000 ha each in the East; (ii) 3 State Ranches of at least 20,000 ha each in the North West; (iii) transport facilities for cattle and meat, and stock routes; (iv) equipment for cattle markets and holding grounds; (v) construction and renovation of slaughterhouses; and (vi) pasture improvement trials and scholarships. Total project cost was estimated at US$11.5 million and it was assumed that, as a result of the project, an additional 10,000 t of meat could be produced, which was equal to the national meat deficit at the time. Government responded very favorably to the Bank's proposals. 2.04 In the following months, Government fully endorsed the operation-oriented project objectives. It refused free bilateral assistance for project preparation because the proposed terms of reference favored a more research-oriented approach, preferring instead to hire the best consultants available who were financed out of its own resources. 2.05 In the meantime, two follow-up missions from RMWA expressed concern that the tsetse infestation in the Adamaoua might be more severe than previously expected, thereby endangering the proposed ranch development scheme. The Bank therefore proposed to modify the terms of reference for project preparation, to include an assessment of the incidence of tsetse flies, and in case it was found to be necessary, the cost and methods of tsetse eradication. 2.06 The project was prepared by an experienced consulting firm in the relatively short time of only six months. The final report, submitted in April 1973, contained as major new elements, proposals for: (i) a credit scheme to finance private ranches and butcher shops; (ii) a tsetse eradication campaign; and (iii) the creation of the Societe de Developpement et d'Exploitation des Productions Animales (SODEPA) which would become responsible for the operation of ranches and slaughterhouses. - 15 - 2.07 The Bank accepted the preparation report as a working basis and sent an appraisal mission to the field. The mission recommended a project which included a credit scheme, a tsetse fly eradication and the SODEPA activities. The philosophy underlying the main project components was that practical field experience which had been gained in other African countries could be used: the tsetse eradication campaign was partly based on the ongoing efforts in Nigeria where, already for more than ten years, tsetse control measures were being implemented; and the ranching component on the example of successful private ranches in East Africa and of the "Compagnie Pastorale" of Ngaoundere/Cameroon. It was therefore assumed that these two components would not face any special risks. Only the credit scheme was considered innovative since, at the time, no experience was available from other West African countries. To minimize its potential risks, the credit component was kept relatively small and flexible through the introduction of different types of subloans. 2.08 Negotiations were held in February 1974. The Cameroonian delegation agreed to the proposed project and accepted all conditions but one: the immediate elimination of meat price controls for the 12 butcher shops to be financed under the credit scheme. The Bank agreed to drop this condition. 2.09 A Bank loan of US$11.6 million for 20 years, including a grace period of 5 years was approved in March 1974. Loan signature was in May 1974 and the project became effective in September 1974. The Project 2.10 The "Livestock Development Project" was considered as the first phase of a Government program to modernize the livestock sector and was to be implemented over a six-year period, 1974-79. It comprised: (a) providing credit facilities and technical assistance to about 35 mixed farmers to carry out steer fattening, and to 115 settled graziers to develop breeding/fattening ranches; (b) exterminating tsetse flies on about 800,000 ha in the Adamoua to provide additional grazing areas and contain the southeastward advance of tsetse flies toward the fly-free grazing lands of Central Cameroon; (c) establishing and operating three state-owned 20,000 ha ranches; (d) constructing two slaughter plants, each with an annual capacity of 12,000 t and modernizing 12 butcher shops in Douala and Yaounde; (e) training ranch and slaughter plant managers, animal production extension officers, credit officers, and tsetse fly extermination and survey teams; and - 16 - (f) preparing a second livestock project that vould include measures to ensure the effective use of areas cleared of tsetse flies, the expansion of credit facilities to the traditional sector, and the continuation of tsetse fly extermination. Targets and Goals 2.11 The project's main objective was to help satisfy Cameroon's beef requirements through increased domestic production. The project was to assist livestock producers in two important ways: firstly by helping them to intensify their grazing methods; and secondly by safeguarding grazing lands for livestock through tsetse fly extermination. Recognizing that changes in the production system would involve risks and that innovations needed to be tested, the project was to involve both state and private sector ranch development. As a secondary objective the project was to raise meat hygiene and minimize waste by improving slaughter and retail butcher shops in Yaounde and Douala. 2.12 The key role for the development of the livestock sector, was assigned to the Societe de Developpement et d'Exploitation des Productions Animales (SODEPA) which was established by decree of April 24, 1974. Its main responsibilities were: (a) to develop and manage livestock enterprises (ranches, slaughterhouses); (b) to provide technical assistance to livestock owners and assist in the implementation of the FONADER credit scheme; and (c) to sublease to private individuals approximately 140,000 ha range lands for which SODEPA had received a 99- year leasehold title from Government. 2.13 The credit scheme was designed to be implemented by the Fonds National de Developpement Rural (FONADER) which was created in 1973 with the dual role of (a) providing credit to farmers or groups of farmers; and (b) appraising, financing, and supervising a wide variety of rural development projects. Since FONADER had no qualified field staff, SODEPA was to establish two livestock extension teams. These teams, based in Ngaoundere and Bamenda, were to identify potential borrowers, carry out technical evaluations of loan applications, and subsequently give technical assistance and supervise the borrowers. These teams were to be supported by field staff of the Ministries of Agriculture and Livestock. Revisions 2.14 By 1978, an amendment of the Loan Agreement became necessary for two reasons: (i) disbursements in categories I(a) Consultant Services; I(b) Slaughterhouse Construction; and II(e) Ranch Equipment had already exceeded the allocated amounts; and (ii) in view of the bilateral technical assistance for tsetse eradication and FONADER credit components, it was possible to reallocated Bank funds. In an exchange of letters (March 3, 1978), Schedule 1 to the Loan Agreement was amended (Annex II, Table 1). The revised disbursement percentages remained unchanged until project completion. - 17 - III. PROJECT IMPLEMENTATION A. Traditional Sector Development. Objectives 3.01 Under the project, mixed farmers/settled graziers were to be provided with credit facilities to be provided by FONADER and technical assistance by SODEPA, with a view to developing small-scale cattle fattening and larger breeding/fattening operations. The credit program was to be the first for cattle production and was to involve, besides on- farm and ranch investments, the introduction of improved grazing systems and changes in the land tenure arrangements. Three different types of credit were conceived: Model I: 35 mixed farms of about 4 ha each in the coffee-growing areas of the Northwest were to fatten 3 steers annually. Project-financed investments included materials for fencing, cattle shelter, equipment, seeds, fertilizer and feeder steers. Model II: Private ranches of about 250 ha each in the Northwest and Adamaoua were to be developed. Project-financed investments included breeding stock, feeder steers, fencing, watering and stock handling facilities, dips, spray races, and small equipment. Model III: 25 private ranches of about 750 ha each were to be developed. All credit arrangements were the same as for the Model II ranches. Actual Implementation 3.02 The traditional sector credit scheme was the most successful project component. It was successful because (i) the possibility to receive a 20-year leasehold title made the scheme attractive for participants; (ii) the need for credit as a constraint to develop livestock production was correctly identified; (iii) FONADER realised the importance of small mixed farms and expanded the program for Model I loans; and (iv) farmers decided to invest mainly in cattle and fencing. However, its implementation was affected by: (i) lack of technical supervision; (ii) unavailability of farm inputs; (iii) slow repayment rate; and (iv) high credit administration cost. 3.03 Out of 986 applications received, FONADER approved 317 loans, of which only 255 loans were financed from project funds (Annex III, Table 1). To satisfy additional credit demands, FONADER financed another 96 loans with approximately CFAF 250 million from Government funds. These loans have been excluded from this evaluation. 3.04 In the first year of the credit program (PY 2), the number of loans disbursed was lower than projected, but beginning with PY3, the - 18 - number of loans annually disbursed exceeded appraisal estimates. At the end of PY6 FONADER had disbursed 255 loans, or 70% more than the SAR had planned. The strongest increase occurred in the Model I scheme for mixed farms with 172 loans (i.e. 137 above the SAR estimate), (Annex III, Table 2), whereas the number of loans to larger ranches was reduced. 3.05 Emphasis on mixed farms: This was the result of three developments: (i) During project implementation an increasing number of disputes arose between agriculturists and pastoralists. The usual causes were that agriculturists claimed rights to land allocated to ranches and had begun to plant crops inside ranches, which were in turn damaged by the herds. Many disputes were referred to the courts for decision. These difficulties discouraged the authorities to make full use of the Credit funds earmarked for private cattle ranches; (ii) the area of Dibi in the Adamoua was over-populated by humans and cattle to such an extent that it became impossible for the authorities to allocate ranch lands to individuals as this would have been at the expense of the community as a whole; and (iii) it was discovered that the area of Tourningal was infested with tsetse flies (para 3.23). Since it was technically not feasible to eradicate the tsetse flies from this isolated area, FONADER could not justify financing investments in cattle. At the same time it was also noted that Model I loans were usually implemented without difficulties and that the impact of introducing livestock into crop farming was very positive. As a result of these developments the Bank and FONADER agreed to increase the number of loans to mixed farms (approximately 400%). 3.06 To finance the total number of 255 loans, FONADER disbursed CFAF 247 million. Out of this amount the Bank reimbursed CFAF 235 million (US$1,017,356) or 92% of the reallocated amount of US$1.1 million (para 2.14). FONADER financed the difference of CFAF 12 million from Government funds. Loan Utilization 3.07 For Models I and II, the average amount disbursed per loan exceeded the appraisal estimate by 65% and 50% respectively; and for Model III it was 20% less than projected (Annex III, Table 3). Average Loan Amount (in CFAF'000) Actual as % Appraisal Actual of Appraisal Model I 200 330 165 Model II 1,400 2,101 150 Model III 3,400 2,717 80 - 19 - 3.08 The main reasons for these cost overruns were (i) that the number of cattle purchased per farm was about 100% higher than projected (Annex III, Table 4); and (ii) that the average cost per head of animal was 20% higher than estimated in the SAR. While investing more in cattle, other investments (mainly for buildings) were reduced from approximately 60% to 35% of the total loan amount (Annex III, Tables 5). Investments in % of total loan Model I Models II and III Appraisal Actual Appraisal Appraisal Actual Model II Model III Models II & III Livestock 50.7 74.2 29.9 31.2 56.9 Others 49.3 25.8 70.1 68.8 43.1 3.09 These three changes approved during project implementation -- i.e. the increase in the number of Model I loans from 35 at appraisal to 172 actual, the increase in the average loan amount from CFAF 200,000 to CFAF 330,000 and the emphasis on cattle purchases instead of physical investments were proper adjustments to the borrowers' needs and proved correct in terms of the returns for the farmer. Credit Supervision 3.10 The traditional sector credit scheme suffered from the lack of technical advice during credit implementation. It was foreseen at appraisal that SODEPA would set up field teams in Ngaoundere and Bamenda, to be responsible for these tasks (para 2.12). However due to SODEPA's technical problems in its own operations (ranches and slaughterhouses) and a shortage of qualified personnel, these teams were never established. SODEPA's actual participation in the credit scheme was therefore reduced to three functions: (i) approval of final loan applications. This formality was dropped during the second half of the project; (ii) supply of land to farmers on the basis of a 20-year leasehold title; and (iii) sale of approximately 100 heifers to FONADER-financed ranches during 1979. 3.11 The FONADER field offices, realizing the need for technical advice tried unsuccessfully to obtain this assistance from the Ministry of Livestock (MINEL). Since no other help was available, the FONADER field staff assisted as much as possible in the implementation of the investment proposals but concentrated mainly on the evaluation of loan applications and on credit recovery. - 20 - Farm input supply 3.12 The implementation of the credit scheme also suffered from the unavailability of farm inputs i.e. barbed wire, fence post, nails, hand sprayers, supplementary feeds, seeds, veterinary drugs etc. To overcome this shortage, the two FONADER field offices began to supply on their own some of the needed farm inputs to the farmers. Although this a-civity was in violation of FONADER's general policy, it enabled to meet an urgent need and helped at least some of the loan recipients. Credit Repayment 3.13 Payments for interest and principal were received in time only as long as the two field offices were responsible for their collection. In 1979, the default rate was only 11%. However, later in the same year FONADER decided to centralize the system which meant that credit recovery was handled directly from headquarters. As a result of this change, the repayment rate dropped considerably. Thus, for an example, the situation for the Bamenda office as of June 30, 1980 was as follows: Amount CFAF '000 % Interest & principal due June 1980 24,604 100 Received by FONADER in time 16,051 65 Received by September 30, 1980 1,691 7 Received by December 31, 1980 836 3 Not received by March 15, 1981 6,026 25 This level of repayments not collected within nine months of the due date, shows deficiencies in FONADER's credit administration (para 6.06) and could lead to future cash flow problems. However, in comparison with similar credit schemes elsewhere in West Africa, the level of late payments is not exceptionally high. Costs 3.14 The implementation of the credit scheme was expensive. A comparison between the total loan administration costs of CFAF 268 million and the value of all loans disbursed of CFAF 247 million, shows that for each CFAF 1,000 of loans disbursed, the administrative costs were CFAF 1,085. Even excluding the cost of technical assistance, this ratio is still CFAF 508 of expenses to each CFAF 1,000 of loans disbursed (Annex III, Table 6). This high cost of loan administration can be justified only when the benefits of institution building are taken into account (para 6.05). 3.15 Detailed analysis reveals that the largest cost overrun was for vehicles and equipment with CFAF 40 million above appraisal estimates. Whereas it was assumed at appraisal that the program could be implemented with two vehicles, FONADER actually purchased 6 four-wheel drive vehicles, 2 light vehicles, 2 tractors, trailers and agricultural equipment, 16 motor cycles and 2 radio transmitters. The large vehicle park caused an increase in vehicle operating expenses of CFAF 16 million above appraisal estimates. - 21 - B. Tsetse Fly Extermination 3.16 The objective of the tsetse fly extermination campaign was to eradicate flies from the Adamoua and to stop the southeastward advance of fly infestation. Basic Survey 3.17. During project preparation insufficient time was available to determine the distribution and behavior of the tsetse fly in the Adamoua. It was therefore agreed that the actual extermination campaign would be preceded by an in-depth survey. The survey was conducted by a specialist in tsetse fly eradication methods. During his field work, the consultant complained about lack of support and about a lack of surveyors to determine the exact extent of tsetse infestation. He submitted his report in August 1976 and made specific recommendations for the eradication program. During project execution continued fly discoveries showed that the survey had been inadequate. It had not correctly identified the extent of infestation nor had it determined the specific behavioral patterns of the fly population in the Adamoua (para 3.21). It is difficult to judge whether the consultant would have been able to determine the actual extent of infestation had he been assisted by more surveyors. 3.18 The project documents specified that funds for the tsetse eradication would only be disbursed after approval of the Eradication Plan and preparation of a Land-Use Plan (Schedule 1, 3(b)). While the first condition was met, the second condition was not enforced. Although the Upper Faro Valley Development Plan was ready in 1978, the detailed land-use plan became available only in February 1981. In the meantime, parts of the tsetse-freed area had already been reoccupied with cattle since 1977 on an uncontrolled basis and in a manner not foreseen in the land-use plan. Revisions. 3.19 A major amendment of the Loan Agreement was introduced when in 1976 bilateral aid became available to finance equipment, helicopter flying hours, operating expenses, two tsetse specialists, consultants and studies. This change was reflected in the Amendment of Schedule 1 to the Loan Agreement (para 2.14). 3.20. Another revision became necessary when project management found that no suitable contractor was available to construct the tsetse barriers. It was therefore agreed that SODEPA would purchase equipment and construct barriers under force account (para 3.27). Physical Progress 3.21 The extermination campaign began in 1976/77 and continued during four dry seasons, one season longer than planned. A major unforeseen fact was that instead of the expected 10% of the total area, 21% of the habitat - 22 - was infested with flies (para 3.17). As a result, an area larger than expected had to be sprayed. This was accomplished by using additional helicopter flying hours, which were actually 74% above appraisal estimates. During the four campaigns, the project freed 794,500 ha of tsetse flies, i.e. approximately what had been projected at appraisal (Annex III, Table 7). Reclaimed Area in ha 1976/77 1977/78 1978/79 1979/80 TOTAL 90,300 203,700 277,600 222,900 794,500 3.22 One significant technical difference compared to the appraisal estimates was that approximately three times more insecticide was consumed per ha of reclaimed area than foreseen. The reasons for this increase were: (i) the helicopters applied 4.5 - 5.5 liters of insecticides per ha instead of 4.0 liters; (ii) instead of 10%, 21% of the total area had to be sprayed by helicopters; and (iii) approximately 13% of the area had to be resprayed instead of 10%. Issues 3.23 Infested Area. In the initial study (para 3.17) the consultant had been unable to determine the exact extent of tsetse infestation. Successive surveys carried out between 1977 and 1980 found other infested areas in the northwest, southeast, northeast and east of the area originally thought to be infested. The tsetse infestation in the northeast, covering most of the Dibi-Tourningal area, was the reason for a partial stop of credit disbursements in that region (para 3.05). Due to limited survey capacity the actual extent of the infestation discovered in the east could not be determined. As a consequence and in order to consolidate the sprayed area, the Bank and Government agreed in 1979 to limit the extermination campaign to the Ngaoundere-Ngaoundal railway line. In summary, information available at the end of 1980 indicated that the actual infested area in the Adamoua was probably two to three times larger than originally expected. 3.24 Barriers. Based on the knowledge available at the time, it was assumed at appraisal that the tsetse-freed areas would be protected by natural barriers, such as the mountain chain to the west of the project area and through man-made tsetse barriers, such as 2.5 km wide deforested strips of land. However reinfestation continued despite these barriers which indicated that tsetse flies were able to pass mountain ranges of up to 1,800 m elevation and to cross deforested strips of 2.5 km wide. In - 23 - response to these discoveries, more effective barriers were established by applying a heavy dosage of residual insecticide. In general, the discovered ability of tsetse flies to cross previously believed to be natural barriers, will increase future annual maintenance costs considerably. 3.25 Survey Capability: A freed area can only be maintained free of tsetse flies as long as regular surveys are conducted and, in case reinfestation is discovered, an extermination team is despatched immediately. However, despite continuous requests by the Bank, Government did not supply the Tsetse Division with sufficient funds to establish the necessary number of survey teams. It may eventually be discovered that reinfestation has already occurred but has remained undetected. Costs 3.26 Given that the infested area was 21% rather than 10% of the total area, the cost overrun of 103% above the appraisal estimate is not surprising. The highest cost overruns were recorded for vehicles and equipment expenditures, mainly because land clearing was carried out under force account (para 3.20); and for surveys and consultant services, because two studies not foreseen at appraisal (Ecological Impact and Second Land-Use Plan) were found necessary. The higher average eradication costs of CFAF 2,177 per ha 199% of appraisal estimate reflects mainly the higher than expected degree of infestation (Annex III, Table 8). Ecology 3.27 The spraying over four years, of 800 t of Thiodan and Ensodil insecticides which have been partially banned in the United States and Europe, raised strong objections from environmental groups. A study financed by the bilateral cofinancier was therefore carried out to determine the impact of insecticide application on non-target species and the accumulation of residues in the human food chain but no significant adverse impact on the environment was detected. The study revealed that: (i) One year after treatment there was still a significant reduction in the number of insects, spiders, scorpions, crabs; (ii) no family of insects, spiders, crabs etc. was completely eradicated; (iii) unlike Thiodan, Dieldrin (Ensodil) had no influence on fish; (iv) no harmful pesticide levels were found in human foods; (v) in vertebrates no acute mortality could be found after spraying; and (vi) spraying did not reveal any structural changes of the gallery forest eco-system as a whole. - 24 - C. SODEPA 3.28 Objectives. Within the project framework, SODEPA's objectives were: (i) to establish three ranches; (ii) to build and o- arate two slaughterhouses; and (iii) to assist FONADER in the implementation of the traditional sector credit scheme (para 3.12). Ranch Development 3.29 Start up was relatively fast. During appraisal it had been agreed that the Ndokayo ranch should only be developed in case the site was found to be free of tsetse flies. In August 1974, a consultant confirmed that a survey had been conducted at Ndokayo and that no tsetse flies had been discovered. Therefore preparations began immediately on two sites, Dumbo and Ndokayo. The expatriate ranch managers arrived in November 1974 (Dumbo) and February 1975 (Ndokayo). After some basic infrastructure development, the first cattle was purchased in mid 1975. The Faro ranch was opened later (June 1978), after the area had been freed of tsetse flies. Problems 3.30 During the first four years of the project, both Government and the Bank considered the ranch component as relatively successful. At the end of 1978, however, some difficulties were noted and in March 1979, major problems became obvious: (i) The technical assumptions, used as a basis for the ranch models, had been too optimistic for Cameroonian conditions: the fattening period was longer, the mortalities were higher, and the calving rate was lower than projected; (ii) the ranches were overstaffed, had high overheads, lacked incentives to improve productivity, and paid high purchase prices and received low sales prices for cattle; and (iii) as the ranch sites had previously been used by private graziers, pastoralists' cattle kept trespassing into the ranch sites, causing health problems to ranch herds. Technical Assumptions 3.31 The fattening period for steers, expected to be between 18 and 24 months, actually proved to be 36 months (Annex III, Table 9). The reasons for the extended fattening period were that only young, (18 months old) and poor stock was available and could be purchased on local cattle markets. The poor health condition of the purchased stock is evident in the high tracking losses that occurred between the cattle market and their arrival on the ranches: - 25 - Tracking Losses of Purchased Stock 1975/76 1976/77 1977/78 1978/79 3.4% 13.0% 10.5% 9.3% 3.32 The actual mortalities in all stock classes were two to three times higher than projected (Annex III, Table 10). Whereas the appraisal mission had assumed an average mortality rate of 3.5% for the total herd, including calves, the actual averages were the following: Average Herd Mortality Rates on SODEPA Ranches 1975 1976 1977 1978 1979 1980 Dumbo 8.5 3.1 6.6 6.7 6.2 6.7 Ndokayo 3.9 5.2 8.4 9.7 5.5 1.7 Faro - - - 13.2 4.1 3.1 The main reasons for these high mortilities were: (i) insufficient use of dips and spray races; (ii) lack of prophylactic and curative drugs; and (iii) infections received from trespassing herds. In addition to these actual death causes, it can be assumed that a number of animals were stolen but erroneously recorded as mortalities. 3.33 The calving rate, estimated to reach 73% in PY4, actually remained as low as 47% in Dumbo, and 38% in Ndokayo. The main reasons for this low calving rate were: (i) females available on cattle markets were not very fertile (para 3.34); (ii) sanitary problems reduced the conception rate (brucellosis); and (iii) nutritional and mineral deficiencies limited fertility. However, as the actual number of breeding cows kept on the ranches was much smaller than expected, the low calving rate had only a minor impact on the overall ranch productivity. 3.34 The supply of breeding heifers was more difficult than expected. At appraisal, it had been assumed that Dumbo and Faro would become breeding ranches, while Ndokayo would be a fattening operation. In 1977, project management found that feeder steers were in short supply. It was therefore agreed between Government and the Bank that Ndokayo was also to become a breeding ranch, producing its own fattening stock. This meant that over six years, 18,000 heifers would have to be purchased. Actually, however, the cattle markets did not regularly offer breeding females or heifers for sale. Experience showed that at the most 10% females could be purchased together with a lot of feeder steers. But these females were usually old, in poor health and often sterile. Due to this unavailability of females, the ranches remained mixed operations with the emphasis on steer fattening. 3.35 This shortage of breeding females completely changed the expected herd structure. Instead of showing a ratio of about 40% breeding females and 25 to 30% calves, the actual herds mostly comprised feeder steers. - 26 - Herd Structure (in percentage) Cows Bulls Calves Heifers Steers Total Appraisal 42.0 1.6 27.3 16.9 12.2 100.0 Actual 18.9 1.1 12.2 14.5 53.3 100.0 Due to this lack of breeding females, herd productivity on the ranches will be low and herd growth will depend for some years on the number of animals available for purchase. Project Management 3.36 Personnel: The ranches employed more staff and had a higher proportion of administrative staff than normally found in a private enterprise. Staff numbers rose to a maximum of 403 persons in PY4 as against an appraisal estimate of 108 persons (Annex III, Table 11). Upon recommendation of the Bank, staff was reduced to 312 persons at the end of PY6, which was still 86% higher than projected. Productivity of staff, expressed in persons employed per 100 head of cattle, remained about 40% lower than in other comparable ranches (Annex III, Table 11). 3.37 Similarly, the managerial staff was 108% above appraisal estimates (Annex III, Table 12). For instance, instead of three veterinarians as foreseen at appraisal, the ranches employed 14 veterinary assistants. The large number of managerial staff also increased the overhead expenses, mainly for motorized transport facilities improved housing. 3.38 Productivity. Despite Bank recommendations, SODEPA management gave no incentive to improve ranch productivity. As the herdsmen were paid on a daily basis, there was no incentive for them to increase daily gains or offtake. Instead of leaving for the grazing grounds at daybreak, as private herdsmen do, SODEPA's herdsmen adhered to the official working hours. As a result the actual available grazing time for the animals was short and coincided with the hottest hours of the day when cattle prefer to rest in the shade. It is also noteworthy that ranch management spent twice as much money for personnel transport than for animal health purposes (Annex III, Table 13). 3.39 Prices. As a result of inadequate supervision and lack of personal incentives, SODEPA staff paid on average 15% to 20% higher prices for purchased stock and received 15% to 20% lower prices for animals sold, than comparable private enterprises. Previous Land Utilization 3.40 At appraisal Government had confirmed that the proposed ranch lands were not settled and that no traditonal pasture rights would be violated. During implementation, however, and particularly when the - 27 - ranches were expanded beyond the originally planned size (30,000 ha instead of 20,000 ha), the Dumbo and Ndokayo ranches experienced problems. Traditional pastoralists claimed pasture rights inside the established ranch boundaries or demanded compensation. In Dumbo, a financial compensation was discussed for a while, but the issue was later dropped in court. In Ndokayo a fence was built along parts of the eastern boundary and guards were assigned to prohibit herds from entering into the ranch. However, in both cases, neighboring cattle continued to enter the ranches and their poor sanitary status was one of the causes of general animal health problems and occasional outbreaks of foot and mouth disease. The problem needs to be redressed in the future. Investment and Operating Costs 3.41 Total actual investment and operating costs were about 11% lower than the appraisal estimates (Annex III, Table 13). However, while total expenditures remained in line with projected costs, certain categories showed significant differences from the appraisal estimates. Expenditures for buildings were approximately 55% higher than projected, due to the higher number of houses that had to be built for the increased staff. The allocated amount for vehicles and equipment was overspent by 132% because approximately twice as many vehicles and motorcycles were purchased. These larger number of vehicles caused an increase in vehicle operating costs of 31%, and in replacement costs of 56% above appraisal estimates. The actual expenditures for wages and salaries remained 9% below the appraisal estimate, despite the increase in staff size, mainly because the considerable hardship allowance proposed at appraisal to attract highly qualified personnel to the remotely located ranches, was actually never paid. The low level of expenditures for animal health, which was only 15% of appraisal estimates, was certainly a major reason for the low herd productivity and high mortality rates (para 3.32). Income 3.42 The actual income was far below expectations. Instead of a total income of CFAF 1,167 million over the six-year period the actual income on the three ranches was only CFAF 340 million, or 29% of the appraisal estimate. The reasons for this shortfall were: (i) The actual fattening period was 50 to 100% longer than projected (para 3.31), which resulted in a slow turnover. (ii) SODEPA's management purchased less animals than projected. At the end of PY6 only 26,288 cattle had been purchased instead of the projected 39,330 head. (iii) High tracking losses (para 3.31) and high mortalities (para 3.32) reduced the number of animals available for sale. (iv) The actual purchase price was 13% higher while the sales price was 9% lower than projected. (Annex III, Table 14). - 28 - Net Benefits 3.43 Excluding investment costs and livestock purchases, the ranches accumulated a net operating deficit of CFAF 228 million over six years of operation, due mainly to SODEPA's high overheads and low productivity. For both types of operation, fattening and breeding, SODEPA actually lost money on the animals that were sold. This can be demonstrated by two methods of calculations using SODEPA's actual operating cost and technical coefficients. (Annex III, Tables 15 and 16): (i) Fattening Operation for 1,000 Purchased Steers: CFAF'000 Purchase Price 25,191 Maintenance Cost, (3 Years) 15,374 Sales Price 39,133 Loss 1,433 (ii) Breeding Operation for 1,000 cows CFAF'000 Annual Maintenance Cost for 2,931 Head 17,087 Annual Sales of 293 Heads at 15,008 CFAF 51,222/Head Loss 2,079. Future 3.44 During 1979, the Bank informed Government and SODEPA that based on available information, the ranches were not financially viable. The Bank recommended that, as part of the second livestock project, the ranches should be closed and the land subdivided to be made available for a credit scheme. Buildings and infrastructure could be utilized for extension purposes and SODEPA would assist in the implementation of this new credit component. Government did not accept this proposal and sent its own team to evaluate the ranch performance but the findings of their study did not differ significantly from the Bank's findings. However, in November 1979, MINEL concluded that it was politically unacceptable to Government to close the ranches. As a compromise Government agreed that, as a condition for continued Bank participation in the ranch scheme, certain achievable performance criteria would have to be met (Annex III, Table 17). As a first step and in order to cut operating costs SODEPA immediately reduced the number of staff employed on the ranches (Annex III, Table 11). A further joint Bank/SODEPA review of the latest ranch development is scheduled for the end of 1981. Slaughterhouses 3.45 Brief History of Events. During 1975 SODEPA hired a consultant to prepare bidding documents and represent SODEPA's interest in dealing - 29 - with contractors. Following an international tender, SODEPA proposed that the lowest bidder, an European contractor, should construct the two slaughterhouses in Yaounde and Douala. In August 1976, a combined Bank/SODEPA mission visited the contractor to verify the financial and technical capability of the company and found the company qualified to carry out the works. In February 1977, the company requested and received a downpayment of 30%, whereas only 20% were covered under a Bank guaranteed performance bond. In May 1977, the company declared bankruptcy and as a result SODEPA lost US$300,000. In August 1977 SODEPA signed a new agreement with the second lowest bidder also from Europe. Again the company was found capable of carrying out the works, but in 1978 it also ran into financial difficulties, requested an increase of approximately 40% of the total contract sum and when this was not granted, stopped construction work. In June 1980, SODEPA cancelled the contract. By that time the Yaounde plant was 95% complete and the Douala plant approximately 60% complete. SODEPA intends to complete the two slaughterhouses under its own supervision and has for this purpose hired a slaughterhouse engineer. Issues 3.46 First Contractor: When a Bank mission, together with SODEPA representatives, went to Europe to review the contractor's financial situation, the mission learned that the company: (i) had been bankrupt in 1970; (ii) had been in another business until 1975 when it joined forces with a slaughterhouse equipment company; (iii) held three contracts in the slaughterhouse sector, worth only about one-third of the the SODEPA contract; (iv) mainly had experience in conveyors and slaughterhouses; and (v) its financial situation had to be reevaluated by its bankers before they would assure the necessary financial support to the company. 3.47 Given the above, it appears in hindsight that it was doubtful that the contractor had either the technical expertise or the financial backing required to carry out the works. Furthermore it also proved ill- advised to allow a second downpayment of 10% over and above the contractually agreed downpayment of 20%. Consultant 3.48 During negotiations between the second contractor and SODEPA's consultant, it was agreed that payments would be made according to an agreed schedule and upon receipt of three documents: (i) the contractor's invoices; (ii) the consultant's certificate of works - 30 - accomplished; and (iii) a telex from SODEPA requesting payment of the pertinent application. The arrangements worked satisfactorily until 1978, when the contractor began to experience financial problems. As a first measure, the contractor issued payment requests for services not rendered and for equipment not delivered. Instead of drawing immediate attention to these facts, SODEPA's consultant continued to certify that works had been completed and goods had been shipped or delivered. The discrepancies became obvious at the end of 1978, by which time the contractor had received 97% of all payments due, but the Yaounde plant was only 90% complete while Douala had just reached the 50% mark. SODEPA realised that it could no longer rely on its consultant and cancelled his contract. Sites 3.49 The slaughterhouse in Yaounde was built on a site, which had been accepted during appraisal. The Douala site was changed following appraisal and the slaughterhouse was constructed approximately 20 km outside the town center. In discussions held by supervision missions in early 1980 butchers indicated that they would be reluctant to send their animals to be slaughtered so far out of town. In the Bank documents, the only reference to the Douala site appears to be the following from a supervision report of September 22, 1976: Douala: "The proposed plant site has been changed since appraisal, however the new location is still in the same area. SODEPA does not have a representative in Douala and I was unable to find anyone familiar with the site to show it to me. SODEPA and the consultant claim that the site is suitable and that the butchers have apparently no objections". Thus it seems that the Government and the early supervision missions had not paid sufficient attention to this issue, until it was too late to change the construction site. To overcome this deficiency in the future, trucks will have to bring animals to the slaughterhouse and transport the meat to the market. Costs 3.50 Construction costs were considerably underestimated and the total cost to complete both plants is still uncertain. During appraisal the construction costs for the two slaughterhouses were estimated at US$1.8 million (CFAF 448.2 million). The first contractor offered to build the slaughterhouses for US$2.8 million (CFAF 665.6 million) which was already 56% above the appraisal estimate. The second contract was for US$3.1 million (CFAF 742 million). Up to the moment when the contract was cancelled, the second contractor had received CFAF 683.2 million (US$2.9 million) or 97% of the contract sum, excluding the retention amount of CFAF 37.1 million. By June 1980, the slaughterhouse construction had cost in total US$3.3 million, but the two plants were still not operational. - 31 - 3.51 During the March 1980 supervision mission, the Bank employed a slaughterhouse consultant to evaluate the technical standards of the works performed by the second contractor and to estimate the additional funds required to complete both plants. The consultant's main findings were that: (i) certain items were not built according to contractual specifications; (ii) some works did not meet basic hygienic standards; (iii) equipment was missing; and (iv) costs to complete both plants were estimated at CFAF 253 million (US$1.2 million). 3.52 After this second unsuccessful attempt to construct the slaughterhouses with a foreign company, Government and SODEPA proposed to complete the slaughterhouses under their own supervision. The Bank accepted this proposal, under the condition that a qualified slaughterhouse engineer would be retained to supervise the work. Under the second project, CFAF 250 million (US$1.2 million) is therefore provided to complete the two slaughterhouses. SODEPA Headquarters 3.53 The objective of the headquarters office was to coordinate and manage activities of the ranches and slaughterhouses. However, two significant changes occurred during project implementation: (i) The appraisal mission had assumed that the infrastructure development on the ranches would be carried out by contractors. However, as qualified contractors were not available, SODEPA purchased its own heavy equipment to carry out the construction program. This expenditure was recorded under SODEPA headquarters. Actual investment costs (CFAF 76 million) were therefore 10 times higher than projected (Annex III, Table 18). (ii) SODEPA found that the ranches could not be managed with the degree of autonomy expected at appraisal. The ranches needed support in general administration, purchasing and selling, and had to be supervised regularly. The headquarters staff of 9 persons, as projected at appraisal, was not sufficient. By June 1980, headquarters employed a total of 33 persons. This increase in staff size and activities resulted in the increase in operating costs of 154% above appraisal estimates. - 32 - IV. PROJECT IMPACT 4.01 The overall impact of the project on the national meat production was less than expected since the livestock production system could not be improved as originally planned. Only a few technical innovations became available, but their impact was limited due to the lack of an effective extension service to disseminate the information. Tsetse Fly Extermination 4.02 Among the project components, the tsetse extermination campaign had the strongest impact on meat production. Out of an estimated 2 million ha of tsetse infested pasture land, the extermination campaign freed about 800,000 ha and by June 1980, local administration had reopened 570,000 ha of the freed area. About 60,000 head of cattle have re-entered the area in the meantime. The re-entry was controlled and followed agreed administration procedures, but it began too early to follow the recommendations of the Land-Use Plan (para 3.18). As a result, some herds have occupied lands which were planned to be used for other purposes. 4.03 During the time when the cattle herds had stayed outside the tsetse-infested Adamoua in the over-populated southern parts, their productivity had been reduced to a minimum. However with the return to the traditional pasture grounds in the north, the productivity improved. As a result of the project, the Adamoua will annually produce an incremental 9,000 head of cattle for sale which is equivalent to 1,500 t of carcass. 4.04 Another positive result of the project was the creation of the Tsetse Division within the Ministry of Livestock. This institution was established to implement tsetse eradication and control measures. During the first years, when it was still known as the "Special Mission for the Eradication of Tsetse Flies", it was less effective because of budgetary constraints. Following Bank recommendations, it was finally given the status of a full division, with its own specific budgetary allocations. Traditional Sector Development 4.05 The traditional sector credit scheme did not have as strong an impact on the national meat production as expected. At appraisal, it was expected that cattle would be purchased with credit funds and be distributed to new owners. The new owners would, with the help of technical innovations, equipment and feed, improve the productivity of the stock considerably beyond the previous production level. However, as SODEPA was unable to establish a livestock extension service and FONADER could only give very limited practical assistance (para 3.10) farmers and ranchers were very much left on their own. The few new methods which they used originated from FONADER's close contact with the farmers. The incremental beef production of the 7,500 animals financed under the project is estimated to be in the range of 100t carcass meat annually. - 33 - 4.06 The most important benefit of the credit scheme was that it demonstrated the positive impact that land allocation, together with credit and some technical advice, could have on traditional pastoralists. With the allocation of land, a number of pastoralists adopted basically new attitudes: (i) fencing their property to avoid trespassing of neighboring cattle; (ii) limiting transhumance; (iii) improving their pastures by planting grass and alfalfa; and (iv) offering supplementary feeding to their herds in the dry season. This change of attitude in a traditionally conservative population group willhave a significant long term impact on the sector. As a result of these encouraging developments, the credit scheme was therefore expanded under the second project. SODEPA 4.07 The creation of SODEPA and its operation of three ranches had no significant impact on the national meat production. The productivity of cattle which were moved from the private sector into SODEPA ranches, did not improve as expected. 4.08 One particular argument in favor of the creation of parastatal ranches had been that the ranches would serve as development centers for technical innovations. In fact, however, the ranches contributed little in terms of improved technology, mainly because during the early years of ranch development, the ranch managers were essentially occupied with the building of infrastructure, purchasing cattle, and establishing a management routine. By the time that general ranch management became well established, the expatriate ranch directors had to leave and were replaced by less experienced local counterparts. The technological packages which were introduced, included basic animal health measures (such as routine vaccinations and prophylactic treatments) as well as simple animal husbandry measures (e.g. small amounts of supplementary feeding during the dry season and the supply of mineral supplements or salt). V. RATES OF RETURN Assumptions 5.01 The financial rate of return was recalculated, using the following assumptions: (a) all costs and prices were expressed in 1974 terms. (Deflators in Annex 4, Table 1 and Exchange Rates in Table 2); - 34 - (b) in all projections, cattle purchase and sales prices taken are average actual prices expressed in 1974 terms; (c) the financial rate of return was recalculated for a project life of 20 years, with the exception of the credit scheme, which was recalculated for 15 years; (d) for the tsetse fly extermination campaign, it was assumed that herds entering the tsetse freed area would increase their productivity from 2.5% to 10%, due to improved pasture resources; and (e) as the financial losses on the SODEPA Ranches would grow with an increased cattle turnover, it was assumed that the ranches would maintain the number of 18,000 head of cattle. 5.02 The economic rate of return was recalculated basically using the same assumptions as in the SAR: (a) all costs and prices are expressed in the same terms as for the financial rate of return calculations; (b) for the traditional sector credit scheme, credit administration, investment and operating costs have been included for the first 6 years of the project; and (c) fifty per cent of SODEPA Headquarters costs have been included. 5.03 Based on these assumptions, the financial and economic rates of return were re-estimated (see also Annex 4, Table 3) as follows: Financial Rate of Return Economic Rate of Return Appraisal Reestimated Appraisal Reestimated Lraditional Sector Credit Scheme - 24 11 11 lodel I 23 40 - - lodel II 16 19 - - lodel III 17 18 - - rsetse Extermination Campaign 14 13 14 13 30DEPA - Negative - Negative Ranches 11 Negative - - Slaughterhouses 15 Negative - - Total Project - 6 13 4 1/ The appraisal report did not present separate economic rates of return for individual credit scheme models or for SODEPA ranches and slaughterhouses, nor did it estimate the financial rates of return for SODEPA or the traditional sector credit scheme. Also it did not distinguish between the financial and economic rates of return for the tsetse eradication program. - 35 - VI. INSTITUTIONAL PERFORMANCE MINEL 6.01 It was assumed at appraisal that the Ministry of Livestock would have overall responsibility for project implementation. Actually, due to the administrative approach followed by all Government agencies, FONADER and SODEPA soon began to manage the day to day affairs of their respective components on their own. However, while FONADER enjoyed full autonomy in its work, SODEPA remained under the Ministry's administrative supervision. This led to serious deficiencies (para 6.03). SODEPA 6.02 SODEPA was created with the intention to establish an organization that would be responsible for the commercial development of the livestock industry in Cameroon. Although the institution would be organized on a parastatal basis, it would benefit from administrative flexibility and operate as a commercial enterprise. In the initial phase, SODEPA would establish and manage ranches, slaughterhouses and a livestock extension service. Other tasks were to be added later on. SODEPA was unable to live up to these expectations. 6.03 The Government chose as SODEPA's Director General one of the most experienced, senior Veterinarians in the Civil Service. A team of expatriate specialists was hired to assist him in implementing the program. However, despite this strong management team, SODEPA suffered from a number of major problems: (i) Cattle ranching is an extensive production process. However, instead of adopting an adequately decentralized management approach, SODEPA established an intensive and costly administration; (ii) SODEPA was unable to generate profits in its commercial enterprises, because Government established the level of salaries and benefits to be paid to the employees; in addition, SODEPA was called to render public services such as research and extension services, which are not normally provided by a commercial enterprise; (iii) SODEPA was not authorized to recruit its own personnel, and this adversely affected the efficiency of its operation; and; (iv) whereas SODEPA was responsible for construction of the slaughterhouses, all relevant decisions were taken by Government. 6.04 This continued intervention by Government in the day-to-day management decisions were a major factor in SODEPA inability to play its intended role. In particular, the company should have been given more autonomy for the tasks of establishing the three ranches and constructing the two slaughterhouses. - 36 - FONADER 6.05 The flexibility adopted by FONADER management contributed significantly to the success of the credit scheme. When SODEPA proved unable to assist in loan processing, FONADER improvised and evaluated credit requests on its own and even provided some assistance in credit implementation (para 4.05). On the other hand, FONADER experienced serious problems in credit administration. Between 1973 and 1980, FONADER's overall disbursement volume increased by 546% and its personnel grew from 67 to 267 employees. This rapid growth of credit volume caused internal administrative problems FONADER, despite a 9-man expatriate technical assistance team and the increase of local employees, was unable to solve. The main deficiencies were in the book-keeping system. This led to a backlog at headquarters in sending out drafts and payment requests in time, which encouraged farmers to ignore their repayment dates and caused the low recovery rates (para 3.13). Despite some recent improvements, FONADER's administration still requires further strengthening. In preparation for this completion report, the Bank had requested from FONADER a list of all loans disbursed, the amounts repaid, the repayments delayed and a final balance. Up to date FONADER has not been able to supply this information. As proper account-keeping is the most essential requirement of any financial institution, this situation needs to be redressed further under the follow-up second project. Tsetse Division 6.06 As a result of the project, a Tsetse Division was established within the Ministry of Livestock. The creation of this new division is a significant achievement because, for the first time in Cameroon, all activities related to tsetse fly survey, eradication and control measures are coordinated by a single agency. It is also important that the Ministry of Livestock agreed to establish this division in the field at Ngaoundere, i.e. in the centre of all tsetse control activities. VII. CONSULTANTS PERFORMANCE 7.01 The consultants chosen by Government to prepare the livestock project were known as being fully qualified for this work. Since both Government and the Bank were satisfied with the preparation report, the Bank had no objections to Government's proposal to hire the same company for the implementation of the project. 7.02 The main responsibility of the consultants was to supply qualified personnel. The first Technical Director of SODEPA had excellent qualifications but left due to personal reasons. As his replacement, the consultants proposed one of the ranch managers. This was accepted by SODEPA and the Bank but he left soon thereafter. Thereafter the consultants proposed the last remaining ranch manager as the only available candidate. SODEPA and the Bank reluctantly accepted this third Technical Director. However, when he proved unable to meet Bank expectations, he cancelled his contract and left the country. These repeated changes made it difficult for SODEPA to overcome its technical problems with the ranches and slaughterhouses. - 37 - 7.03 Up to 1978, SODEPA's accounts were not well maintained. Although the accounts were audited annually, they did not permit any detailed cost analysis. It was only after the departure of the first expatriate financial controller and the arrival of his replacement that SODEPA's account fully met adequate standards. The new financial controller introduced cost accounting which now serves -s a useful management tool. At the same time, he established a tight control system covering all financial and administrative matters of the ranches. Thus, SODEPA had no difficulty in supplying all the financial and technical data requested for the preparation of this project completion report. 7.04 The same consultants who had prepared and implemented the first phase of the project was asked to evaluate the project and prepare a second phase. It now appears that the consultant was possibly biased in evaluating his own work. In particular, his report on the progress of the first project proved deficient as it failed to highlight the ranch problems (para 8.03). VIII. BANK PERFORMANCE 8.01 During preparation of the completion report, the Bank had asked the different organizations responsible for project execution to evaluate the performance of the Bank. Up to date, no comments have been received. 8.02 The Bank's limited in-house experience in tsetse eradication and at the time of appraisal also in parastatal ranching proved a handicap for project implementation. In designing these two components, the Bank had relied on the successful experience with similar programs in Nigeria and Kenya. Thus, no additional research preceded the beginning of field operations. The only modifications were the reduction in the numbers of ranches from ten to three and the introduction of a preparatory phase before the tsetse eradication campaign could be implemented. These measures alleviated to some extent the risks associated with the implementation of the two new project components. 8.03 In retaining for project implementation and evaluation the same consultant who had prepared the project, the Bank and Government possibly failed to realize the danger that the consultant could be biased in evaluating his own work. As a matter of fact the technical ranch problems were discovered relatively late: the consulants' evaluation report of the first project did not mention any serious problems in the ranching component, and it was only during appraisal of the second phase that the full extent of the difficulties became obvious. 8.04 In reviewing the supervision reports it appears that the Bank missions mainly focussed on technical issues of project implementation and less on implications that high mortality rates, extended fattening periods, purchase and sales price differentials and high labour costs have on the economic and financial viability of the ranches. It is also true, however, that a certain time lag is required to form definite views on the - 38 - technical coefficients. For example, it takes at least two years to realize if feeder steers need three years or only two years to finish. 8.05 A final issue relates to the Bank's direct involvement in the discussions concerning the slaughterhouse construction. It can be argued that the Bank should have kept a low profile with its involvement consisting only of technical advice to the Borrower as necessary. However, the Bank was led to play a more active role for two main reasons: the need to complete the slaughterhouses on schedule and the Borrowers' inexperience in dealing with such a complex matter. IX. CONCLUSION 9.01 In sum, the project did not quite achieve the anticipated objectives. The main shortfall was the failure to establish SODEPA as a viable commercial enterprise; the costs and efforts that went into the creation of SODEPA have not yet yielded tangible results. While the tsetse eradication has successfully been carried out so far, there are risks that following the withdrawal of the expatriate team and phasing out of external funding, the tsetse division will be unable to prevent a reinvasion of flies. The most successful project component was the traditional sector credit scheme, despite high administrative overheads and the lack of an extension service. 9.02 Credit facilities were provided by FONADER to mixed farmers and settled graziers to develop small-scale fattening and larger breeding fattening operations. The lessons to be learnt are: (i) land allocation in combination with credit will encourage owners to (a) fence their property to avoid trespassing, (b) stop transhumance, (c) improve their pastures, and (d) offer supplementary feeding in the dry season; (ii) loans for cattle in small farms have a good chance of success because animals are usually well attended and the mixed farming system can supply valuable supplementary feeding; (iii) credit funds should be used mainly for production-oriented components. The necessary infrastructure can often be provided by the owners themselves; and (iv) the credit recovery rate depends largely on the effectiveness of the credit agency. 9.03 The tsetse fly infestation was correctly identified as a major constraint to livestock development in the Adamoua. An eradication campaign was designed and successfully carried out, and a land use study prepared under the project will be implemented under the second project. The lessons to be learnt are: - 39 - (i) tsetse eradication projects should be based on a land use plan. This study should clearly spell out the potential benefits in terms of improved land utilization, and the risks and costs that the tsetse eradication campaign would involve; (ii) tsetse eradication should only be conducted when sufficient technical and financial resources would be made available in order to ensure that the freed area can be adequately protected against reinvasion of flies; and (iii) in regions where trypanosomiasis is the limiting factor for livestock development, the introduction and use of trypano- tolerant breeds should be considered as an alternative to tsetse eradication. 9.04 SODEPA was established as a parastatal organization to manage ranches and slaughterhouses and to operate a livestock extension service. The lessons to be learnt are: (i) the activities of parastatal organizations should be clearly differentiated between commercial activities designed to generate profits, and public services which should be financed by the Government; (ii) parastatal ranching faces serious difficulties because: - ranching is an extensive production method which cannot bear high overhead costs; and - the productivity of animal husbandry operations depends on a strong personal incentive for the man taking care of the animals. 9.05 Finally, the project has shown that the risk of failure is high when project implementation depends on several inter-related project activities. SODEPA was to supply land and implementation assistance to FONADER but proved unable to perform as planned. FONADER's credit program in the East depended on the implementation of the tsetse eradication campaign, but the campaign was unable to clear the flies from the Dibi- Tourningal area. The ranches were to supply meat to the slaughterhouses, but the production was less than anticipated. The private butchers were to receive meat from the slaughterhouses, but the completion of the slaughterhouses was delayed. FONADER was to finance private butchers, but private butchers were not established because the slaughterhouses had not been completed. This situation underlines. the need for simpler project design, with a limited number of inter-related project components.  - 41 - ANNEX 1 Table 1 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Total Project Costs (CFAF million) Actual as % Appraisal Actual of Appraisal Private Sector Credit Scheme 433.8 571.9 151 SODEPA 1,898.9 2,441.4 129 Tsetse Fly Extermination 1,195.7 1,729.6 145 Training 57.3 7.4 13 Project Preparation 57.3 68.0 119 Total Project Costs 3,643.0 4,818.3 132 - 42- ANNEX 1 Table 2 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Disbursement Schedule (US$ 000) Cumulative Actual as % Appraisal as Actual as Appraisal Actual of Appraisal % of Total % of Total 1974/75 Sept. 200 - 0 2 0 Dec. 500 135 27 4 1 March 1,100 147 13 10 1 June 1,900 239 13 16 2 1975/76 Sept. 2,500 306 12 22 3 Dec. 3,100 663 21 27 6 March 3,700 1,041 28 32 9 June 4,300 1,472 34 37 13 1976/77 Sept. 4,900 1,842 37 42 16 Dec. 5,500 2,466 45 47 21 March 6,100 4,004 66 53 35 June 6,700 4,171 62 58 36 1977/78 Sept. 7,200 4,751 66 62 41 Dec. 7,700 5,376 70 66 46 March 8,200 6,201 76 71 53 June 8,700 6,987 80 75 60 1978/79 Sept. 9,100 8,348 92 78 72 Dec. 9,500 8,980 94 82 77 March 9,900 9,747 98 85 84 June 10,300 10,199 99 89 88 1979/80 Sept. 10,600 10,199 96 91 88 Dec. 10,900 10,857 100 94 94 March 11,100 10,934 99 96 94 June 11,300 10,972 97 97 95 1980/81 Sept. 11,500 11,090 96 99 96 Dec. 11,600 11,239 97 100 97 March - 11,469 99 - 99 June - 11,534 99 99 - 43 - ANNEX II CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Amendment of Schedule 1 of the Loan Agreement 1/ (Expressed in Dollar Equivalent) Loan Amount Allocated Category Original Amendment I. Ranch and Slaughterhouse development under Parts A and B of the Project. (a) Consultants' services 600,000 975,000 (b) Slaughterhouse construction & equipment 1,600,000 3,500,000 (c) Livestock 2,300,000 2,200,000 (d) Ranch infrastructure 600,000 350,000 (e) Ranch equipment, materials & vehicles 100,000 175,000 Sub-total 5,200,000 7,200,000 II. Tsetse-Fly Extermination under Part C of the Project. (a) Land Clearing and Fencing 400,000 550,000 (b) Spraying 1,100,000 1,700,000 (c) Vehicles and equipment 800,000 450,000 (d) Consultants' services & training 100,000 200,000 Sub-total 2,400,000 2,900,000 III. Loans under the Agricultural Credit Program under Part D of the Project. (a) Loans 900,000 1,100,000 (b) Consultants' services, vehicles & equipment 300,000 - Sub-total 1,200,000 1,100,000 IV. Training under Part E of the Project 200,000 100,000 V. Technical Assistance under Part F of the Project 200,000 300,000 VI. Unallocated 2,400,000 - TOTAL 11,600,000 11,600,000 1/ In the letter dated March 3, 1978 the disbursement percentage remained unchanged. ANNEX III Table 1 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Traditional Sector Credit Scheme Processing of Loan Application Number % Loan Applications submitted 986 100 Returned as incomplete 131 13 Submitted after completion (from 2) 22 2 Evaluated in Regional Office 877 89 Rejected by Regional Office 378 38 Approved by Regional Office & submitted to HQ 499 51 Rejected by HQ 182 19 Approved by HQ 317 32 Disbursed 255 26 Undisbursed 59 6 Rejected by loan applicants 3 - -ANNEX III Table 2 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Traditional Sector Credit Scheme .Disbursements Schedule No.of loans per year Budget Year Project Year Appraisal Actual 1974/75 PYl - 1975/76 PY2 25 6 1976/77 PY3 36 43 1977/78 PY4 41 49 1978/79 PY5 48 51 1979/80 PY6 - 106 Total 150 255 ANNEX III CAMEROON Table 3 LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Traditional Sector Credit Scheme 1! Summary Loan Numbers Average Loan Amounts Total Loan Amounts Disbursed Actual as % Appraisal Actual Actual as % Appraisal Actual Actual as % Appraisal Actual of Appraisal CFAF 000 CFAF 000 of Appraisal US$ 000 US$ 000 of Appraisal Model I 35 172 491 200 245 123 28 182 650 Model II 90 57 63 1,400 1,559 111 504 384 76 Model III 25 26 104 3,400 2,016 59 340 227 67 Total 150 255 2/ 170 (1,100) (1,017) 8 1/ 12 Butcher loans, estimated at CFAF 5,520,000 or US$22,000 are not included. 2/ During the period July 1, 1975 to June 30, 1980 FONADER financed, in addition, 96 loans at a value of approximately CFAF 250 million. 3/ According to the amendment of Schedule 1 of the Loan Agreement, US$1.1 million (CFAF 275 million) were allocated). 4/ Out of the total of CFAF 247,080 (US$1,069,000) in current prices, only CFAF 235,065,500 (US$1,017,356.77) were reimbursed from loan 983-CM. The difference of CFAF 12,015,000 (US$52,000) was financed by FONADER. ANNEX III CAMEROON Table 4 LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Traditional Sector Credit Scheme Herd Structure and Cattle Purchases (Heads of cattle per ranch) Model I Models II and III Herd Structure Total animals purchased Animals purchased Before Development Animals purchased under Credit Scheme Appraisal Actual Appraisal Actual Appraisal Actual Appraisal Actual Bulls - - 5 3 1 1 Cows - - 52 68 - - Heifers 3 years - - - - 16 15 Heifers 2-3 years - - 8 16 - 3 Heifers 1-2 years - - 9 15 - 11 Males 4 years - - 5 - - - Males 3-4 years - - 7 13 - 2 Males 2-3 years - - 8 24 - 11 Males 1-2 years 6 12 9 19 21 22 Calves - - 19 34 - - Total No.per herd 6 12 122 192 38 65 Total No.of animals purchased under Credit Scheme 210 2,064 3,154 5,395 3,364 7,459 1/ Weighed average of Model II and III. ANNEX III CAMEROON Table 5 LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Traditional Sector Credit Scheme Loan Utilisation Comparison between Appraisal & Actual (in % of Total Loan) 1/ Model I Models II and III - Appraisal Appraisal Actual Appraisal Actual Model II Model III Models II/III Livestock 50.7 74.2 29.9 31.2 56.9 Fencing/Firebrakes 17.0 - 39.6 23.0 21.2 Dips/Sprayers - - 0.9 22.0 9.5 Watering facilities - - 15.7 11.0 2.1 Cattle crush, holding facilities 8.5 - 3.1 2.2 2.4 Land survey - - 9.5 9.9 - Equipment .8.5 - 1.3 0.7 0.6 Feeds/Minerals - - - 0.9 Pasture Improvement 15.3 - 0.4 Others - 25.8 - - 6.0 Total 100.0 100.0 100.0 100.0 100.0 1/ The loan Models II (250 ha) and III (750 ha) for ranch development have been grouped together because it became difficult to distinguish them. ANNEX III CAMEROON Table 6 LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Traditional Sector Credit Scheme FONADER Administration Costs l/ (CFAF 000) Total Actual as % Category Appraisal Actual of Appraisal Investment Costs Vehicles 3,200 19,067 596 Equipment - 24,865 - Sub-Total 3,200 43,932 1,373 Operating Costs Salaries - Livestock Development Officers / 67,600 125,009 185 Technicians 12,000 22,905 191 Other employees 10,000 15,600 156 Sub-total 89,600 163,514 182 Housing Allowances - Livestock Development Officers 9,000 4,401 3/ 49 Technicians 4,000 8,454 211 Sub-total 13,000 12,855 99 Other expenses - Vehicle running costs 6,250 22,283 357 Vehicle replacement 2,600 - 4/ - Office Expenses 1,000 12,364 1,237 Transport and per diem 2,750 5,007 182 Daily labourers - 8,246 - Sub-Total 12,600 47,900 380 Total Operating Costs 115,200 224,269 195 Total FONADER Administration 118,400 268,201 227 1/ Sources: FONADER and GTZ 2/ Expatriates (2) financed through FRG bilateral aid. 3/ Housing costs for last 3 years included in expatriate salaries. 4/ Not estimated as replacement vehicles are covered under investment costs. ANNEX III CAMEROON Table 7 LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Tsetse Fly Eradication Program Area Sprayed Railaimed and Costs Actual Total Actual as 2 Category Units PY3 PY4 PY5 PY6 Appraisal Actual of Appraisal Helicopter flying time h 407 912 828 1,000 1,800 3,147 174 Area sprayed per h flying time ha 53 56 53 53 60 54 90 Area actually sprayed ha 21,606 51,418 43.989 53,174 108,000 170,187 158 Area reclaimed ha 90,300 203,700 277,600 222,900 780,000 794,500 102 Reclaimed area actually sprayed % 24 25 16 24 10 21 - Area Resprayed ha 1,000 15,300 55,400 1/ 31,000 2/ 80,000 102,800 129 Resprayed area as % of reclaimed area % 1 8 17 12 10 13 - Insecticide applied per he reclaimed area 1 2.7 2.1 1.2 2.7 0.6 2.0 333 Total cost per ha reclaimed CFAF 1,092 2,177 199 Total cost per he excluding T.A., survey & consultants CFAF 1,040 1,666 160 1/ Includes 1967 he barrier spray 2/ Includes 1532 ha barrier spray. - 51 - ANNEX III Table 8 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Tsetse Eradication Campaign Total Costs (CFAF 000) Actual as % Appraisal Actual of Appraisal Category Land clearing and fencing 98,600 8,190 8 Spraying 308,600 449,366 146 Vehicles and Equipment 27,500 227,312 827 Insecticides 194,600 412,889 212 Aviation fuel 10,600 35,823 338 Personnel 104,100 174,026 167 Survey and Consultants 19,700 310,118 1,574 Training 10,300 15,249 148 Operating expenses 78,100 96,626 124 Total costs 852,100 1,729,599 203 Total cost excluding technical assistance, surveys & consultants 811,580 1,323,694 CAMEROONA III Table 9 LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-C) COMPLETION REPORT SODEPA Ranches Cattle Numbers - Summary (in head of cattle) AvR.held on 3 ranches Actual a Animals Purchased Actual as I Animals SoldActual as Appraisal Actual of Appraisal Appraisal Actual of Appraisal Appraisal Actual % of AppratsW PYl 5,120 - - 10,240 - - - - - PY2 11,960 3,700 31 8,150 7,832 96 3,864 59 2 PY3 14,840 8,899 60 5,040 4,509 89 5,636 639 11 PY4 18,260 8,251 45 6,560 5,201 79 6,880 2,317 34 L PY5 22,000 14,242 65 5,320 4,374 82 8,926 3,306 37 PY6 23,500 17,139 73 4,020 4,372 109 9,036 1,995 22 Total 39,330 26,288 67 34,342 8,316 24 1/ For three SODEPA Ranches 2/ On 6.30.1980 the actual number of animals held was 18,867 or 57% of the appraisal estimate. - 53 - ANNEX III Table 10 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT SODEPA Ranches Mortality & Calving Rate for different Stock Classes (in %) Actual Appraisal Dumbo Ndokayo Faro Mortality 1/ Bulls 3.0 4.8 4.5 - Bulls 2-4 years 3.0 - 4.4 19.6 Cows 3.0 6.9 5.6 9.0 Calves 5.0 6.3 23.5 44.0 Heifers 1-4 years 3.0 1.8 4.7 6.8 Steers 3-4 years 3.0 3.6 - 7.0 Purchased fattening stock 3.0 8.1 21.7 42.0 Mixed stock 3.0 10.4 8.8 12.6 Calving Rate Cows 73.0 47.0 38.3 -3/ 1/ For PY5 2/ For PY6 3/ Data unsuitable for analysis. - 54 - ANNEX III Table 11 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT SODEPA Ranches Growth in Personnel Numbers 1/ Average per Year Persons per Personnel Cattle 100 Cattle PY2 253 3,700 6.8 PY3 393 8,899 4.4 PY4 403 8,251 4.9 PY5 338 14,242 2.4 PY6 312 17,139 1.8 For Comparison Appraisal Estimate 168 32,928 0.5 ORD, Congo J 130 9,000 1.4 Van Lancker, Zaire 1/ 320 25,000 1.3 1/ Summary of 3 ranches. 2/ Bank Project. 3/ Private Enterprise. - 55 - ANNEX III Table 12 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT SODEPA Ranches Personnel Structure 1/ (Persons) Persons Actual as % Appraisal Actual of Appraisal Category Management Manager 3 3 Assistant Manager 3 - Veterinary Assistant 3 14 Clerk, Secretary, Storekeeper 3 8 Sub-Total 12 25 208 Technical Skilled Labor 36 66 Herdsmen, Laborers, Guards 120 221 Sub-Total 156 287 184 Total Personnel 168 312 186 1/ For 3 ranches as of June 30, 1980. - 56 - ANNEX III Table 13 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT SODEPA Ranches Investment - Income & Operating Cost Summary (CFAF 000) Actual as % Category Appraisal Actual of Appraisal I. Investments Buildings 70,700 109,470 155 Fencing 38,700 1,810 5 Stock handling 38,700 11,426 30 Roads, Firebrakes 20,300 19,393 96 Watering Facilities 15,400 10,279 67 Vehicles & equipment 31,500 73,044 232 Purchase of cattle 2/ Total Investment Costs 215,300 225,422 104 II. Income Cattle sales 1,166,942 339,991 29 III. Operating Costs Salaries and Wages 346,666 315,627 91 Maintenance 37,575 23,091 61 Running cost of vehicles and equipment 59,250 78,045 131 Replacement costs 45,900 71,923 156 General administration 15,250 58,013 380 Animal Health 144,200 21,743 15 Sub-total operating costs 648,841 568,442 88 IV. Purchase of cattle 726,204 624,154 86 Total operating costs 1,375,045 1,192,596 87 Total Investment and Operating Costs 1,590,345 1,418,018 89 1/ Summary for Dumbo, Ndokayo and Faro 2/ Cattle purchased, covered under Operating Cost due to high proportion of feeder steers. ANNEX III Table 14 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOM 983-CM) COMPLETION REPORT SODEPA Ranches Cattle Prices Al Average Price Actual as Z PY2 PY3 PY4 PY5 PY6 Appraisal Actual of Appraisal Purchase price per head 22,954 22,486 24,061 24,625 25,192 21,160 23,864 113 Sales price per head 34,473 37,952 35,669 44,753 51,222 44,850 40,814 91 Price difference (Margin) 11,519 15,466 11,608 20,128 26,030 23,690 16,950 72 1/ 1974 prices. 2/ Purchased stock contains 90% steers and 10% heifers. Animals sold are 90% steers and 10% culled cows. - 58 - ANNEX III CAMEROON Table 15 LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Model Calculation Steer Fattening (CFAF 000) Remaining Cattle Cattle Expenses Receipts Activity Year 1 Purchase of 1000 steers 1,000 1,000 25,192 Tracking losses of 5% 2/ 50 950 Maintenance cost CFAF 5830/head 5,539 Annual mortality 8% 4 76 874 Year 2 Maintenance cost of CFAF 5830/head 5,095 Annual mortality 7% 61 813 Year 3 Maintenance cost of CFAF 5830/head 4,740 Annual mortality 6% 49 764 Sales price CFAF 51,222 39,133 Balance 40,566 39,133 Loss 5/ 1,433 1/ Prices used from PY6 in Table 24. 2/ Tracking losses of 5% assumed, see para 3/ Operating cost from PY6 in Table 19, CFAF 104,900,000 for 18,000 head of cattle. 4/ See Table 20 5/ Not included are investment cost, interest, depreciation. CAMEROON ANNEX III LIVESTOCK DEVELOPMENT PERJECT (LOAN 983-CM) Table 16 COMPLETION REPORT Model Calculation 1000 Cow Breeding Herd Opening Transfer Transfer End of Year Stock - (Mortality + Sale + Out ) + (Purchase + in ) + Birth = Stock BREEDING BULLS .... 6 + . ) + (....... + ..*17 60 1,000 40 131 171 1,000 cowS ........ - ....... . ....***** *+*.*.* ) A 182 HIES >3182 11 1 171 + ...+ .82. 182 1 98 16- ...+ 18 9 HEIFERS 2-3 .... ... . 6 8 220 - 22 + 198 + 22.~q 22. HEIFERS 1-2 .... ...... + 220220 0 171.....+ 17 + + ...+ 11 171 MALES > 4 ... 7 1 1. . 182..... + 171 + (......... 1A82. 18 MALES 3-4 ....182 . 1.22 198 16 12 + 198 98 MALES 2-3 ........ ...............+.......) 9 220 22 . . ..... . 198 + ... . 220. 220 MALES 1-2 ..220.. ....22 20 . 22 CALVES 0-1 500 - 6 440 ) + 500 . 2,931 207 293 1,559 1,559 500 2,931 TOTAL ANIMALS ......... . ...( ...... + ...... **+* . ****.***...****.**** 2,422 145 325 1,067 1341 200 2,422 A.U. ........ -.HE..... .... . ****** K = ****** ************* ..0.% HERDGROWTH + ..].Ae. % OFFTAKE = .3,.4 % HERDPRODUCTIVITY ANNEX III CAMEROON Table 17 LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT SODEPA Ranches Proposed Herd Coefficients Current Situation (11/78-10/79) Targets to be reached by all ranches Dumbo Ndokayo Faro 1980 1981 1982 1983 1984 Mortality in % Total Herd 6.3 5.5 8.9 5.5 5.0 4.5 4.2 4.0 Cows 8.2 3.5 3.1 4.0 3.5 3.0 3.0 3.0 Bulls 6.3 5.5 - 3/ 5.0 4.5 4.0 3.5 3.5 Heifers 1.5 1/ 6.5 2.2 1/ 5.0 4.5 4.0 3.5 3.0 Fattening Stock 5.5 6.5 7.0 6.0 5.5 5.0 5.0 5.0 Calves 10.7 6.3 2/ 8.6 8.0 7.0 6.5 6.0 6.0 c Calving Rate in % 45 42 44 45 50 55 57 60 Steer Fattening period in years 3 3 3 3 2.5 2.0 2.0 2.0 First calving age in years 3-5 3-5 - 3/ 3-5 3-5 3-4 3-4 3-4 1/ Unrealistic, probably wrong 2/ Probably low because calves are not recorded at birth 3/ No sufficient data available. - 61 - ANNEX III Table 18 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT SODEPA Headquarters Investment - Income & Operating Cost (CFAF 000) Actual as % Category Appraisal Actual of Appraisal I. Investments Vehicles 4,600 59,724 1,299 Office furniture & equipment 1,800 8,099 449 Radio and telephone 600 696 116 Housing furniture & equipment 7,544 - Others 397 - Total Investments 7,000 76,460 1,092 II. Income Subsidies 441,554 Others 14,337 Total Income 455,891 - III. Operating Costs Personnel related costs & general administration 198,200 452,799 228 Vehicle running costs 7,950 29,054 367 Vehicle replacement 4,920 44,157 901 Animal health costs - 9,372 - Total operating costs 211,070 533,382 254 -62- ANNEX IV Table 1 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT Implicit Price Deflators Recorded Inflation Deflator 1974/75 (PYl) 1.000 9.1 1975/76 (PY2) 1.091 1.4 1976/77 (PY3) 1.215 1.5 1977/78 (PY4) 1.355 9.2 1978/79 (PY5) 1.480 8.0 * 1979/80 (PY6) 1.598 9.0 * 1980/81 (PY7) 1.742 9.0 * 1981/82 (PY8) 1.899 9.0 * * Estimate -63 - ANNEX IV Table 2 CAMEROON LIVESTOCK DEVELOPMENT PROJECT (LOAN 983-CM) COMPLETION REPORT 1/ Exchange Rates (per US Dollar) FF DM CFAF 1972 5.044 3.180 252.21 1973 4.454 2.673 222.70 1974 4.481 2.588 240.50 1975 4.286 2.460 214.32 1976 4.780 2.518 238.98 1977 4.913 2.322 245.67 1978 4.513 2.009 225.64 1979 4.255 1.833 212.72 1980 4.226 1.818 211.30 Average Exchange Rates 2 1 US Dollar was equivalent to 4.517 2.224 227.21 1 DM was equivalent to 102.16 1/ Source: IMF Supplement on Exchange Rates No.1, 1981 2/ July 1974 to June 1980. 1.MES10CK DEVEIPNET PRO.1BC (~PR MM 983-CK) Rate. f Return 1974/75 1975/76 1976/77 1977/78 1978/79 1979/80 1980/81 1981/82 1982/83 19853/84 1984/85 1985/86 1986/87 1987/88 1988/89 1989/90 1990/91 1991/92 1992/93 1931g4 Traditional Sector Crdit Schmme (Financial Rate of Return 241 and Econoeic Rate of Return 111 Cr-dit Ad1.ni.tration Invstnt & Ope.ating Costa 21.54 41.80 51.20 52.80 51.10 49.80 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.80 0.00 ModelI 5~11 Mi-d ParS (Fi.nncal Rat. of Return 40%) -nve9tant & incratal operating -ot. 1/ 0.00 0.00 4.77 4.33 16.03 39.11 15.23 29.58 29.58 29.58 29.58 29.58 29.58 29.58 29.58 29.58 29.38 27.00 24.96 17.f9 ncrntal benefite 1/ 0.00 0.00 0.00 3.42 6.85 16.75 42.34 47.13 47.13 47.13 47.13 47.13 47.13 47.13 47.13 47.13 47.13 43.02 39.73 26.50 Model II dLm-tae ranches (Financial Rat, of Return 19%) Invstnt. & incr~atal operating Costa 1/ 0.00 12.20 27.62 57.97 13.55 44.37 23.54 25.20 26.76 26.84 23.84 22.71 18.45 18.27 16.76 16.76 14.99 11.17 3.53 3.53 Incrntal benfit. 1/ 0.00 0.00 2.77 9.70 24.51 29.67 37.96 39.63 46.84 52.98 56.08 57.60 54.20 54.74 52.61 68.21 80.87 102.67 11.08 42.28 Model III Large rabnohe (Finencial Rate of Return 28) Inveant and fncremantal opeattng coet. 1/ 0.00 0.00 40.02 34.87 16.17 21.25 21.68 22.06 21.18 18.83 16.30 13.96 12.97 12.97 12.2.9 7 .97 12.97 5.49 0.00 0.80 Inrntal benefita1/ 0.00 0.00 1.80 16.32 28.85 31.84 33.52 37.96 45.97 51.25 53.28 46.68 41.29 41.29 41.29 41.29 30.29 46.07 0.00 0.00 Testse ExtereinAtion Caespetn (7inancial ate of Return 13%) InStement 4 peAting costa 22.00 80.00 430.00 377.00 412.00 408.00 130.00 130.00 130.00 130.00 130.00 130.00 130.00 130.00 130.00 130.00 130.00 130.00 130.00 130.00 Incremental cattle Dal.. 0.00 0.00 0.00 0.00 39.50 102.00 239.50 412.00 412.00 412.00 412.00 412.00 412.00 412.00 412.00 412.00 412.00 412.00 412.00 412.00 Inc tal hard valu. - - - - - - - - - - - - - - - - - - - 2237.00 S0DEPA (FinanIal 6 Econoi. Rate of Return .gativ) e.dquart.r. Invetne t Coate 45.00 0.00 6.00 23.00 1.00 0.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 Operattng co.t. 81.00 86.00 85.00 98.00 94.00 92.00 78.00 78.00 78.00 78.00 78.00 78.00 78.00 78.00 78.00 78.00 78.00 78.00 7A.00 78.00 Ranch.. In-tn.t. 30.00 31.00 15.00 110.00 39.00 0.00 15.00 - 15.00 15.00 15.00 15.00 13.00 15.00 13.00 15.00 15.00 15.00 15.00 15.00 • 15.00 Cattle purchaees 0.00 180.00 101.00 125.00 108.00 110.00 126.00 126.00 126.00 126.00 126.00 126.00 126.00 126.00 126.00 126.00 126.00 126.00 126.00 126.00 Opeatitng cot, 33.00 78.00 112.00 120.00 120.00 105.00 - 105.00 105.00 105.00 105.00 105.00 103.00 105.00 105.00 105.00 105.00 105.00 105.00 105.00 105.00 Cattei Dales 0.00 2.00 0.00 83.00 153.00 102.00 247.00 247.00 247.00 247.00 247.00 247.00 247.00 247.00 247.00 247.00 247.00 547.00 247.00 247.00 Incrøental herd value - - - - - - - - - - - - - - - - - - - 800.0 Slauhterh (Financial Rate of Return negative) De nt 0.00 7.00 244.00 190.00 955.00 0.00 0.00 146.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 Operting cost. 0.00 0.00 0.00 0.00 10.00 10.00 11.00 45.00 130.00 121.00 121.00 121.00 121.00 121.00 121.00 121.00 121.00 121.00 121.00 121.00 Inc- 0.00 0.00 0.00 0.00 0.00 0.00 0.00 50.00 185.00 185.00 185.00 185.00 185.00 185.00 185.00 185.00 185.00 185.00 185.02 185.00 1/ Phad. M A L J ' 5 N I G E R C H A D SUDAN - UPPE R_ - - , s VOLTA r\( " 1 Z) CENTRAL AFRICAN T REPUBLIC N I G E i iic - H A D - ~ ~ -o . - ¯ ¯ ¯ ¯ ¯0 1 UNITED REPUBLIC OF CAMEROON LIVESTOCK PROJECT LOCATION OF PROJECT ACTIVITIES COmmERCI AL R ANCH SITES TRADITIONAL LIVETCK ETOR DEVELOPMENT AREAS ¯ e TSETSE FLY E TERMINATION GRAVEL RAD - -i t- EARTH ROADS Li.RL- ~ADI,iL ... INTERNATIONAL.I POYT INTERNTI-N L B3CUNDCARIES. . .. . . . . . . . N Gr J,: 6° . .AL ZONE.5 n - -QHE - - .LDANIAN HIGHLANDS - - UDANIAN-GUJINEAN - -UDAN-GUINEAN TRANS] 1N - 6 V. ý1 INEAN -5- SeASED ON IBRU 10586 S..1 ER 1973. LLNL Y SOME PLACE NIo E1o , ESEEN C1HANGE0. {J- ÅC. - Gt- s . iN .E1 A. - L,l- 7L E,; ij L'1 ii vsojL 11 11 HI-1