Cameroon - Oil Palm and Rubber Consolidation Project

)ocument of The World Bank FOR OFFI(IAI I tSE ONLY Report No. 9430 PROJECT COMPLETION REPORT CAMEROON OIL PALM AND RUBBER CONSOLIDATION PROJECT (LOAN 2160-CM) MARCH 18, 1991 Agriculture Operations Division Occidental and Central Africa Department Africa R ional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency Unit = CFA Franc (CFAF) Fixed Parity = 50 CFAF = 1 French Franc WEIGHTS AND MEASURES Metric System FISCAL YEAR July 1 to June 30 ABBREVIATIONS AND ACRONYMS CAMDEV Cameroon Development Corporation CCCE Caisse Centrale de Cooperation Economique (France) CDC Commonwealth Development Corporation (U.K.) EIB European Investment Bank FED Fonds Europeen de Developpement FFB Fresh Fruit Bunch FONADER Fonds National de Developpement Rural HEVECAM Societe Hevea-Cameroun SOCAPALM Socifte Camerounais de Palmerais SOFINCO Consulting Firm  FOR OFmICIAL USE ONLY THE WORLD BANK W.shngion D C 204J3 u S A OfViCe Of O,inKG.(WraC O9maIm Iv&1 Mt,"e- March 18, 1991 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT Project Completion Report on Cameroon Oil Palm and Rubber Consolidation Proiect (Loan 2160-CM) Attached, for information, is a copy of a report entitled "Project Completion Report on Cameroon - Oil Palm and Rubber Consolidation Project (Loan 2160-CM)" prepared by the Africa Regional Office. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  FOR OFFICIAL USE ONLY CAMEROON OIL PALM AND RUBBER CONSOLIDATION PROJECT (Loan 2160-CM) PROJECT COMPLETION REPORT Table of Contents Table No. Preface . ................................................... Evaluation Summary .........................................ii PART I. PROJECT REVIEW FROM BANK PERSPECTIVE .............. 1 A. Project Identity. 1 B. Background. 1 C. Project Objectives and Description 2 D. Project Design and Organization. 2 E. Project Implementation. 3 F. Project Results. 7 G. Project Sustainability. 9 H. The Bank's Performance. 9 I. The Borrower's Performance .10 J. Project Relationships .10 K. Consultant Services .11 PART II. PROJECT REVIEW FROM BORROWER'S PERSPECTIVE ... .... 12 PART III. STATISTICAL INFORMATION ......................... 13 Related Bank Loans ......................................... 13 Project Timetable .......................................... 14 Loan Disbursements ...................................... 14 Project Implementation ........ ............... .... 15 Project Costs and Financing ................................ 17 Project Results ............................................ 20 Economic Impact ............................................ 21 Status of Covenants ........................................ 27 Use of Bank Resources ...................................... 29 This document has a restricted distribution and mas be used by recipients only in the performance of their official duties Its contents mm riot othersise be disclosed without W'orld Bank authorization  - i - CAMEROON OIL PALM AND RUBBER CONSOLIDATION PROJECT (Loan 2160-CM) PROJECT COMPLETION REPORT PREFACE This is the Project Completion Report (PCR) for the Oil Palm and Rubber Consolidation project in Cameroon, for which Loan 2160-CM in the amount equivalent to US$ 50.8 million was approved on May 27, 1982. The Loan was closed on January 9, 1989, about six months behind schedule. At closing US$16 million was canceled. The Preface, Evaluation Summary and Parts I and III of the PCR were prepared by the Agriculture Operations Division of the Occidental and Central Africa Department of the Africa Regional Office. The two beneficiary companies had prepared PCRs, in conformity to the old model, which did not adequately reflect all the points raised in the Guidelines to the new format. Part II, therefore, contains a summary of their main conclusions with respect to the project drawn from the material received. Preparation of the PCR was done at Bank headquarters over a year after the project was completed. It is based, inter alia, on the Staff Appraisal Report and associated working papers; the loan and project Agreements; the "Contrat Plan" for CAMDEV and associated working papers; supervision reports; interviews with the last Task Manager of the project; correspondence between the Bank and Borrower; and internal Bank memoranda. The draft PCR was sent to the co-financiers (the EIB, CDC, and CCCE) in June 1990. However, no comments have been received.  - iii - PROJECT COMPLETION REPORT CAMEROON OTL PALM AND RUBBER CONSOLIDATION PROJECT (Loan 2160-CM) EVALUATION SUMMARY Proiect Objectives 1. Financial and managerial problems under preceding Bank-financed projects in support of SOCAPALM and CAMDEV gave rise to the Oil Palm and Rubber Consolidation Project which aimed to allow the two corporations to complete development already initiated and consolidate their financial positions. Specifically, they were to be given the means to bring to maturity rubber and oil palm plantations, most of which had been established under the previous Bank-supported projects. complete the necessary infrastructure and construct processing facilities needed to handle the additional production. Strengthening of management and continuation of smallhol.der programs were other key components. During implementation, the Government was required to review with the Bank its strategy for future development of the oil palm and rubber subsector. In this way, while the project's short term goal was to safeguard the return on past investments, it was also intended to serve as a vehicle for rational long-term planning. (PCR para. 4) Implementation Experience (PCR paras. 9-23) 2. In spite of difficulties, plantations established under the preceding projects were maintained and brought into production in accordance with objectives. Revised rubber tapping procedures permitted signific economies in CAMDEV's replanting program, 3. Implementation proceeded satisfactorily with SOCAPALM. The CAMDEV component had a late start due to management problemss introduction of a "Contrat Plan' became a condition of disbursement for the CAMDEV component, which was only met in December 1984. Disagreement between Cameroonians and external financing agencies on the necessity of a management contract with an expatriate agro-industrial firm was only resolved when, following significant improvements under the Contrat Plan, the Cameroonian General Manager accepted full responsibility for its implementation. Although ,etter managed, SOCAPALM suffered from endemic labor shortages and the impact of the drought in 1983. The crash in world prices of palm oil and rubber in 1986 together with the strengthening of the CFA franc against the dollar wrecked the precarious financial situation of both corporations. Despite efforts to increase productivity, financial health was never recovered. Actual project costs of CFAF 30 billion were 732 of the CFAF 41 billion estimated at appraisal. The main reason was the reduction in the development cost of CAMDEV, with a significant cut in its investment and operating expenditures which amounted onily to 622 of appraisal figures. 4. Smallholder programs were not coordinated among the companies involved and suffered from a focus on planting targets rather than careful selection of participants with the result that many plots were abandoned. The development credit of FONADER (National Rural Development Fund) was effectively distributed to farmers through the implementing public companies  - iv - SOCAPALM and CAMDEV. Credit recovery was made directly by the parastatals through 3tention of dues on farmers' sales proceeds. S. A consultant's report containing proposals for a subsector strategy which involved expansion of production was rendered obsolete by the evolution in world prices for rubber and palm oil. Government never formally adopted a strategy but appears to be supporting greater exploitation of the domestic market for palm oil and maintenance of existing plantations for both commodities. This is a reasonable strategy under the present circumstances, and the project has helped to execute it. 6. Loan disbursements were particularly low in the early years and only amounted to 70 percent of the planned amount over the life of the project. Main causes were the strengthening of the dollar with respect to the CFA franc in the first th'oe years of the project, and savings on the CAMDEV component with respect to field establishment costs. Project expenditure was also constrained by the participating organizations' limited capacity to generate funds, which led to a shortfall in financing expected from them of over 70 percent. Actual project costs of CFAF 29,968 million (equivalent US$ ) were 732 of costs at appraisal, estimated at CFAF 41,000 million (US$152 million), inclusive of taxes. Results (PCR paras. 29-29) 7. Production targets (modified by the Contrat Plan in the case of CAMDEV) came close to being met. In addition, management of both organizations improved although there was still some reliance on expatriates in key positions. However, financial results were far below expectations. Almost 502 of project financing was intended to be provided by the two companies which were also projected to become financially independent by the early 1990s. Over the life of the project, SOCAPALM's self-generated funds amounted to two-thirds of appraisal estimates and those of CAMDEV to only 30%, the situation being particularly critical in che last two years of the project. As costs were contained, this can be attributed to the drastic downward movement in world prices of oil palm and rubber in association with the falling value of the dollar vis-A-vis the CFA franc. Separate financial analyses for oil palm (SOCAPALM; CAMDEV) and rubber (CAMDEV) show that even though oil palm may be financially feasible with the prevailing protectionist policy, it is not economically viable when border prices are applied, while the financial rate of return for rubber is negative, given the overvalued CFA franc. A simulation on rubber production, using a 40? devaluation of the CFA franc to the dollar between 1986/87-1988/89 and a 5OX devaluation thereafter to calculate the projected cash flows, reveals a positive rate of return of around 8Z. Findings and Lessons Learned 8. Although major physical and organizational strengthening objectives were largely met, the project was less successful in financial terms. Yet it had been well supervised and, given external circumstances, it is not clear how results could have been radically different. With their high debt/equity  v ratios, the companies were in no position to withstand the decline in international commodity prices and the CFA franc exchange rate. A sounder financial structure should have been put in place at the outset. With respect to future development, Government might now be more open to the proposal, originally put forward by financing encies at the time of appraisal, to divide CAMDEV into separate units according to the crops produced. This would make it more amenable to eventual privatization, if and when prices recover and/or macro-economic decisions on the exchange rate change. The scope for management contracts (such as the one for CAMDEV's bananas) should be explored with a view to encouraging equity participation from the private sector. 9. Main lessons learned from the project are: (a) given the difficulty of accurately projecting commodity prices, more attention should be given to consideration of alternative financing plans, including debt versus equity financing; (b) comprehensive structural and managerial reforms of complex companies are best developed and implemented on the basis of consensus between all parties concerned; adequate time should be allowed for their introduction; (c) projects aiming to develop financially viable enterprises should cover all operations of the organization concerned; particular attention should be paid to their financial structure; (d) ac it has proved difficult for parastatals, subject to Government intervention, to operate successfully in a commercial environment, autonomous management, preferably private management, should be promoted; (e) smallholder programs for the same crop run by different agro- industrial enterprises should be coordinated; careful selection of participants and identification of suitable land is more important than meeting physical targets.  CAMEROON OIL PALM AND RUBBER CONSOLIDATION PROJECT (Loan 2160-CM) PROJECT COMPLETION REPORT PART I. PROJECT REVIEW FROM BANK'S PERSPECTIVE A. Project Identity - Project Name: Oil Palm and Rubber Consolidation Project - Loan No.: 2160-CM - RVP Unit: Africa - Country: Cameroon - Sector: Agriculture - Subsector: Tree crops B. Background 1. Since the late 1960s, the World Bank has been assisting the Government of Cameroon to diversify exports, which were heavily dependent on coffee and cocoa, by promoting state-owned agro-industrial corporations producing palm oil and rubber. The two supported under the present project (SOCAPALM-Societe Camerounaise de Palmeraies and CAMDEV-Cameroon Development Corporation) had each been the executing agencies of two previous Bank- financed projects. 2. The first phase projects for the two organizations, initiated in the late 1960s, both focussed on oil palm and, despite cost overruns, were considered successful not least because economic returns benefitted from considerable increases in world prices. In 1975, second-phase projects for both SOCAPALM (US$ 38.5 million) and CAMDEV (USS 39.3 million) were prepared. At this point, national production of palm oil was 78,000 tons with 62X produced in plantations and the rest from wild palm groves, while about 18,000 tons of rubber were produced with Government plantations accounting for about 75Z. For CAMDEV, the emphasis shifted to rubber planting but both projects aimed to extend planting and processing facilities with provision of related infrastructure. An innovation in the SOCAPALM II project was the mounting of a smallholder oil palm program with the objective of spreading the benefits of oil palm cultivation among a wider group. A separate division for smallholder management was also envisaged under CAMDEV II. These components reflected a concern on the part of Government to pay closer attention to the linkages between agro-industrial development and traditional farming (PPAR No. 6019 of December 31, 1985).  - 2 - 3. As the expected yields failed to materialize l,, costs increased, alternative financing plans were not available and Government failed to increase its participation in CAMDEV's and SOCAPALM's working capital, the corporations faced severe financial shortages which further affected project maintenance, yields and social com.ponents. Poor management, especially in the case of CAMDEV, aggravated the problems. Faced with this situation, the Bank and cofinanciers CCCE and CDC were prompted to provide supplementary financing in the form of the Oil Palm and Rubber Consolidation Project, which is evaluated in this PCR, even before the two second-phase projects were -ompleted. C. Proiect Obiectives and Description 4. The project aimed to balance the current and expected deficits of SOCAPALM and CAMDEV and improve a number of their administrative and technical shortcomings. Specifically the enterprises were to be enabled to bring to maturity rubber and oil palm plantations most of which had been established under the phase II projects, complete the necessary infrastructure and construct processing facilities needed to handle the additional production. Strengthening of management and continuation of the smallholder programs initiated under the preceding projects were additional key components. In view of the serious doubts raised about Cameroon's comparative advantage in oil palm production and the uncertain prospects facing rubber exports, it was agreed that SOCAPALM, CAMDEV and HEVECAM 2/ would not undertake any expansion of oil palm or rubber plantings beyond those included in the project. During implementation, the Government was required to review with the Bank its strategy for future development of the oil palm and rubber subsector. 3/ D. Project Design and Organization 5. Although SOCAPALM and CAMDEV are geographically and managerially distinct, the conceptual foundation of the project was clear and understood by all parties. The project as designed was oriented to the tree crop sub- sector and allowed discussions with the Government on sectoral issue,, such as production and marketing of palm oil, realization of a tree crop sector study etc. The fact that the two major companies in the tree crop sector were included in a single project prompted the Government to resolve sectoral problems affecting both companies. 6. Given the deterioration of the financial situation under SOCAPALM and CAMDEV II, project design should have paid more attention to preventing 1/ Due to poor clones and inefficient exploitation system for rubber and poor quality field work and water deficit for oil palm. 2/ Societe Hevea-Cameroun, another parastatal with rubber plantations. 3/ Details of the SOCAPALM and CAMDEV compone..ts including appraisal objectives and achievements under the project are given 'r Pa III, Table 4 Project Implementation.  -3- the recurrence of a similar situation. For example, the debt/equity ratio could have been reduced through the injection of more capital. While improvements occurred at the start of the project due to some recapitaliza- tion of the companies and provision of external finance, too much reliance - s placed on potential self-generated funds (the two corporations were expected to provide almost half of total project costs and make substantial progress towards financial independence). 7. Management of both companies had long been acknowledged to be unsatisfactory, although there had been some improvement under SOCAPALM's new General Manager. In the case of CAMDEV, the phase II project had demonstrated the need for major reorganization. As a result, the new project required the appointment of 5 expatriate managers in line positions. A reorganization study, commissioned under the project and done in 1983, found that only 50. of estate managers and 302 of assistants were competent. This underlined the fact that it was not simply a case of bringing in a few expatriate managers in line positions to improve the situation. The study recommended a total overhaul of the company by contracting its management to a private foreign firm. However, this recommendation was contested, as it was politically unacceptable to implement such an unpopular reform. 8. Because of the financial crisis affecting the preceding projects, preparation appears to have been rushed. Perhaps understandably under the circumstances, appraisal functioned as both preparation and appraisal, a single mission giving rise to both an Issues Paper and the Appraisal Report, although it was necessary to carry out a post nppraisal mission to supplement data. While the short-circuiting of the project cycle may be justified in so far as staff were familiar with the corporations and the project was in the nature of a rescue operation, the result was a rather unimaginative project with no innovative fiatures. E. Proiect Implementation 9. Progress in project implementation with respect to the objectives set at appraisal are given in Part III, Table 4. Essentially, plantations established under the previous projects were maintained and brought into production as planned and management at CAMDEV was significantly improved. However, both companies suffered from serious financial problems. More detailed information on project implementation is provided below. 10. Disbursements of Loan 2160-CM are shown in Part III, Table 3. The particularly low levels in the early years is due to the delays in starting, especially with respect to the CAMDEV component. The strengthening of the dollar with respect to appraisal estimates permitted also some savings. Total disbursement was only about 70Z of planned and the balance was canceled. While the SOCAPALM component was almost totally exhausted (partly due to an agreement by the Bank to finance the full foreign exchange cost of the factory extension at Kienke), only about 60Z of the CAMDEV component was disbursed. Major savings were made with respect to field establishment costs due to the revised schedule for rubber replanting. In addition, CAMDEV's positive cash-flow in the beginning of the project allowed the company to finance its operating expenditures without having to draw on the Loan's  -4 proceeds. 11. As might be expected from the pattern of disbursements, proiect expenditure (Part III. Table 5A) was lower than planned. Actual project costs of CFAF 30 billion were 73% of the CFAF 41 billion estimated costs at appraisal. This was mainly due to the CAMDEV component where expenditure was only about 62 percent of appraisal estimates and 75 percent of revised (Contrat Plan) estimates. Investment and operating cost estimates for CAMDEV were cut down in its "peLformance plan". In addition, CAMDEV's management decided to finance part of its development program from own sources instead of drawing on the Bank loan, considered too expensive. While costs for plantation and smallholder development were significantly lower than expected, those for civil works, equipment and vehicles were higher. In CFA francs, expenditure on SOCAPALM was greater than appraisal estimates with costs of civil works (including construction of Dibombari Headquarters) substantially in excess of planned amounts. 12. Data on project financing (Part III, Table 5B) highlight the serious shortfall in domestic funds of over 70 percent. The participating organizations' capacity for generating funds was significantly less than expected (Part III, Table 5C). 13. Although the Loan Agreement was signed in July 1982, the Loan did not become effective until one year later principally owing to problems connected with CAMDEV. In the interim, SOCAPALM was obliged to rely on short-term bank overdrafts to cover day-to-day expenditures. The financial situation improved somewhat when project financing commenced, although the severe drought in 1983 was the main cause behind a significrant drop in production and consequent deficit in self-generated funds in that year. 14. The fundamental problem affecting operations had long been recognized to be personnel and its low productivit'. Because of the remoteness of most of the estates, workers were for the most part recruited from other provinces and turnover and absenteeism were high. This was aggravated by labor legislation in Cameroon which sanctions pro-rata payments, with the result that a few hours work each day entitles a worker c. an estate to participate in all the social facilities (housing, medical etc.). Because there has been no substantial change in prevailing legislation, attempts to introduce incentive schemes for workers have had little impact. On the other hand, the level of management improved under the excellent cooperation between the General Manager and the Chairman of the Board of SOCAPALM. As a result the company was able to gradually recover from the effects of the drought and improve its efficiency of both production of FFB and extraction of oil (see Part III, Table 6 Project Results). 15. The financial situation of SOCAPALM eased somewhat in 1984 after the Government agreed to the company's proposal, supported by Bank supervision missions, to raise the domestic price of palm oil 4/. However, 4/ Although the Government exchanged views with the Bank on methods for setting producer prices, no mutually satisfactory method was agreed.  - 5 - the positive trend was abruptly reversed in 1986 when prices of palm oil on international markets crashed (almost 60% in real terms) at the same time SOCAPALM's increased production and obligations under an agreement of producers forced the company to export an increased quantity of crude oil. Between 1984/85 and 1985/86 the quantity exported increased by 352 but receipts from exports actually declined by 20Z. The situation was aggravated by a price war among producers for the local market and the unrestricted import of refined vegetable oils. The Government, at that stage, was unresponsive to requests to protect the domestic market via a system of levies on imports and SOCAPALM had no interest in taking on another loan to finance refining capacity. Although an effort was made to increase productivity by retiring less productive areas and minimizing maintenance expenditure, the financial situation deteriorated further. Price of palm oil In CFA/Kg Projections 1983 1984 1986 1988 1987 local 167 177 188 200 212 export 139 183 189 210 234 kernel 64 68 72 76 80 Actual local na n 274 240 230 export no no 239 77 76 kernel na na 167 66 40 16. In view of CAMDEV's serious management problems, conditions of loan effectiveness included the appointment to line positions of five internationally recruited specialists. By May 1983, four of the five specialists were in place and the Government and CAMDEV Board of Directors had approved the report submitted by the consultant. It was agreed that recruitment of the fifth expert, the smallholder manager, could becc-e a disbursement condition against the CAMDEV component so that the loan could become effective. However, the situation at CAMDEV continued to deteriorate and although the final disbursement condition was met in early 1984, the Bank and other cofinanciers were reluctant to start fi:.anc,ng an organization which was so poorly managed and inefficient. The Bank and the co-financiers requested that a study be done on the reorganization of CAMDEV's management and procedures. As a result of high-level discussions between the Bank, cofinanciers and Government, agreement to submit and implement a comprehensive "Contrat Plan" incorporating the recommendations of the  - 6 - consultant's report became the new condition for disbursement. 17. From the point of view of the Bank and other cofinanciers, a key element of the proposed reorganization was to be the replacement of individually recruited experts by a team from an experienced agro-industrial firm, under a management contract. Despite Cameroonian opposition, the tender was issued but there was only one bid proposing a combined SOFINCO/CDC team. It might have been adviss'c to reissue the tender at that stage (particularly as CDC was one of the cofinanciers) in order to solicit more bids. This was not done, the cofinanciers were enthusiastic about the proposal but it was rejected by the Government, ostensibly on the grounds of excessive cost. Although the Contrat Plan was signed in December 1984, permitting initiation of disbursements, the impasse over the management contract was only broken in March 1986 when the consultants were persuaded to withdraw their proposal and the General Manager of CAMDEV undertook full responsibility for implementation of the Contrat Plan. In fact, under the Contrat Plan, significant improvements were made to management and most of the objectives achieved. 18. The financial situation of CAMDEV was adversely affected first by the late start of the project and subsequently by similar factors to those affecting the performance of SOCAPALM. CAMDEV was even more dependent on export revenue than the other company and although world rubber prices did not decline as steeply as those for oil palm, the trend was similar. Losses were also made on tea and bananas, crops which were not included in the project although they formed an integral part of CAMDEV operations. Price of Rubber in CfA/Kg ..~ - = . =.. .. - - _ . 1 1983 1984 1986 1986 1987 Projections 396 477 637 818 700 Actual 364 482 377 310 289 19. In 1983 the tapping policy promoted by the expatriate Rubber Manager was criticized as being exhaustive of bark reserves and quite inappropriate in the light of the financial situation which required their conservation. A new package of tapping recommendations was introduced and the whole replanting program revised, in some areas replanting being postponed by as much as ten years. 20. In a move not foreseen at appraisal, on the job training was carried out by a training officer recruited specifically for that purpose. General training in management was later undertaken by another CAMDEV staff member, who was designated Management Development and Training Manager. Management personnel in the rubber and oil palm units benefitted from overseas studies. Overall, the training program, particularly in rubber, was  -7- successful. 21. Smallholder Programs 5/ differed among the three companies involved: SOCAPALM, CAMDEV and HEVECAM (under a separate project). There were different policies with regard to respective obligations of the company and the smallholder, valuation of labor (used to calculate the bonus paid by FONADER-Fonds National de Developpement Rural), and even producer prices. The differences were the result of adaptation to local circumstances and technical reasons, such as land preparation techniques. SOCAPALM achieved 100Z of its oil palm smallholder program while CAMDEV achieved 402 of the rubber program and 52Z of the oil palm program. Problems with CAMDEV were that often its smallholders were dispersed on physically unsuitable land (farmers giving priority to food crop ir. this area) with difficult access. As a result some planted areas (10-15U. in poor physical condition and were subsequently neglected. 22. FONADER, responsible for J _r the necessary credit to smallholders under the programs, cont arctid operations to SOCAPALM and CAMDEV. While it might have been appropriate for the technical organizations to handle payments to producers and supply of inputs on receipt of funding from FONADER, the latter should have been fully in charge of credit recovery. However, given FONADER's inefficiency elsewhere in Cameroon, the system adopted was probably more effective. Over four years, credit recovery by SOCAPALM averaged 73.52 although it was lower at CAMDEV, with less than 50Z. This is a relatively good recovery rate by African standards in agricultural credit projects. There is also some evidence of willful delinquency encouraged by lack of sanctions for non-payment. 23. A consultant's report on strategy for the oil palm and rubber subsector, presented in late 1985, which recommended expansion of production of both crops, was soon rendered obsolete by the evolution in world prices. A formal strategy for the subsector has not since been adopted by Government, although there appears to be a tacit understanding to maintain existing agro- industrial plantations and exploit more fully the domestic market for oil palm by developing a domestic reftning capacity. In the circumstances, this appears to be a reasonable strategy from an economic, financial, and social point of view. It could be improved upon by sales of assets or even contracting out of management to the private sector. F. Proiect Results Direct Benefits 24. Production performance with respect to targets is given in the table on Direct Benefits in Part III. Actual production of rubber and oil by CAMDEV came very close to estimates made under the Contrat Plan (although somewhat lower than original appraisal estimates) underlining the commitment 5/ The conclusions of this section draw heavily on the analysis in the "Note sur le ddveloppement des plantations villageoises au Cameroun", May 1986, to be found in Project Files.  - 8 - of personnel to achieving those objectives. Yields of FFB varied enormously among CAMDEV's different estates from 5 to 15 t/ha supporting the policy of discontinuing production in the less productive areas. Rubber yields averaging about 1,200 kg/ha at project completion, were significantly below appraisal estimates but this may be due to the age distribution of trees in production. 25. After overcoming the effects of the 1983 drought, which affected all plantations, actual production of FFB and palm oil at SOCAPALM increased to exceed appraisal estimates in 1986/87. Yields in the last two years of the project were equivalent to estimates of peak production made at appraisal. 26. Planting and production targets of the SOCAPALM snmallholder program had been met by project completion but only about 45Z of the area planned for smallholder production under CAMDEV had been planted. The project benefitted 1920 farmers in SOCAPALM and 450 in CAMDEV. Economic/Financial Impact 27. At appraisal the overall economic rate of return for the project was estimated to be 12.4Z. It was not possible to recalculate the rate of return on the same basis as at appraisal because there were neither data distinguishing between enterprises such as first and second generation rubber nor information on sunk investment costs. However, separate financial analyses on the incremental net benefit have been carried out for oil palm (SOCAPALM; Table 6C(l)) and oil palm and rubber (CAMDEV; Tables 6C(2),(3)). Except for the first couple of years in the case of SOCAPALM, the (financial) cash flow for oil palm is positive throughout the period of study (up to 2007) for both companies (Tables 6C(l), (2)). This is principally due to the protection of the domestic market and the resulting high local prices. The same analysis, using export parity prices, would yield very different results, with negative figures for most of the period. This shows that oil palm production in Cameroon is not economically viable at prevailing world prices. Rubber production at CAMDEV shows a financial rate of return of -14% (Table 6C(3)). The weakening of the dollar to the CFA franc eroded the sector's competitiveness, with prices in CFA franc falling from 400-450/Kg in 1983-84 to under 300/Kg in 1989/90. The result was substantial negative net flows for the CAMDEV rubber group when heavy investments were being undertaken. In financial terms the project was a failure, although, as is evident from the discussion under project implementation, major causes were beyond the control of project management. A simulation based on a devalued CFA franc from 1986/87 throughout the period of projection (presented in Table 6C(4)), reveals a significantly different result with a rate of return of 82. Differences between financial and economic returns suggest how much of this is due to exchange rate over-valuation and other price distortions. 28. Over the life of the project, self-generated funds averaged two- thirds of appraisal estimates in the case of SOCAPALM and only 30Z in the case of CAMDEV, the situation being particularly critical in the two final  - 9 - years of the project 6/. Whereas they were expected to be close to financial independence by the early 1990s, thanks to support provided during the project, the two institutions instead represent a heavy financial burden which the Government ot Cameroon is in no position to support. Other 29. In the course of the project, personnel was cut at both companies from about 20,000 which was considered to be close to the number required for efficient operations. On the other hand, there are about 3,000 smallholders which rely on the two companies to purchase their produce. These people have benefitted from the employment opportunities provided by the project. G. Project Sustainability 30. Without an improvement in world prices of the commodities produced and an exchange rate devaluation, no net financial benefits will be realized. Substantial external financial support is needed in order to sustain direct benefits in terms of output. This has been negotiated in the framework of new performance contracts for both organizations. H. Bank's Performance 31. Bank performance was weakest in project preparation and design. Little consideration was given to flexibility (for example in financing plans) or to alternative approaches. By contrast, the project was closely supervised, which permitted appropriate modification of project design and achievement of some physical objectives. Key to supervision was continuity of Bank staff, with an overlap of the two task managers involved. In retrospect, the following lessons may be learned from experience with the project: (a) the fact that a project is a follow-up to a preceding one does not cbviate the need for careful preparation and design; project design should have taken alternative approaches into account, particularly with respect to the smallholder program, and given the difficulty of accurately projecting commodity prices, more attention should be given to the consideration of alternative scenarios (sensitivity analysis) even at the expense ef the level of detail; this, in turn, should lead to a more flexibly designed project which would permit prompter reaction to unexpected events; (b) adequate time should be allowed for the introduction of comprehensive structural and managerial reforms into complex companies, particularly when these are in the public sector and the problems are long outstanding; project expenditure and activities should be scheduled accordingly; (c) external donors should not try to force management contracts with 6/ See table on Self-Generated Funds in Part III.  - 10 - foreign firms on unwilling recipients but explore together with local management and Government the scope for alternative means of achieving the desired objectives (in the case of CAMDEV, close involvement of all relevant parties in elaboration of the Contrat Plan resulted in a mutually agreed program to which local management was highly committed); (d) a smallholder (outgrower) program for oil palm and rubber should not focus on achievement of physical targets; instead particular attention should be paid to selection of participants, identification of appropriate land (with respect to location, physical characteristics and tenure), establishment of consistent policy with respect to financing and prices and close supervision; (e) as a smallholder program is a peripheral activity to a major corporation principally concerned with running industrial plantations, commitment of top management should be sought prior to establishment of the program, if this commitment appears weak, the smallholder unit should be made as independent as possible. (g) these commodities, especially rubber, appear to be economically viable if well managed, but cannot be financially viable at the present over-valued exchange rate. I. Borrower's Performance 32. The Borrower complied promptly with the initial loan effectiveness conditions by making substantial financial contributions to SOCAPALJ4 and CAMDEV, which provided immediate relief to their critical financial situations. With respect to senior staff, pension arrangements were improved. However, Government never submitted proposals to enable SOCAPALM and CAMDEV to improve incentives, although it did support the managerial reforms introduced in both institutions. The situation with respect to estate and factory workers was less satisfactory as labor legislation remained an obstacle to their effective use. The Government also exacerbated the problems of the two companies and other domestic producers of palm oil when it failed to restrict imports of vegetable oils and to promote development of a domestic refining capacity when export markets deteriorated drastically. 33. Management problems were a major issue for CAMDEV, although SOCAPALM was not immune. The lessons to be learned are those indicated under (c) and (d) above, namely that any substantial reform of a large organization takes time to implement and consensus is required from all parties involved. J. Project Relationships 34. Bank-Borrower relationships were generally satisfactory apart from the issue over the management contract which was eventually resolved. The continuity and expertise of personnel on the supervision missions contributed significantly to the good relationships with SOCAPALM and CAMDEV. The  - 11 - missions were carried out jointly with representatives of other main cofinanciers (CCCE and CDC) and close contact was maintained between the different external financing agencies throughout the project. This collaboration undoubtedly helped to resolve issues arising during the course of project implementation. K. Consulting Services 35. Foreign experts in line positions were a key element of managerial reform at CAMDEV and the existing management structure at SOCAPALM. With a few exceptions performance was satisfactory. Problems arose with a rubber manager at CAMDEV and the manager of the smallholder program, who was recruited with some delay and failed to satisfactorily reorient the program. 36. All rubber factory civil works and associated infrastructuv , except factory machinery, were constructed by CAMDEV's own engineeL.ng services and two local firms at lower cost than foreseen in project estimates. SOCAPALM also took advantage of local firms. Factory equipment was not always satisfactory. For example: CAMDEV's Idenau oil mill constructed under the project was out of service for six months in 1987/88 for technical reasons while serious problems were experienced with the boiler supplied to the extension of SOCAPALM's Kienke mill. Proiect Documentation and Data 37. The Staff Appraisal Report itself proved of limited use to the Bank and Borrower during implementation, but the detailed working papers provided valuable yardstick information against which to assess progress. In the case of CAMDEV, the detailed papers accompanying the Contrat Plan replaced the appraisal working papers. 38. Supervision reports were the most valuable source of information for the PCR, nonetheless use of indicators was inconsistent, for example in December 1984 there was a switch from data on areas under the project to production and yield information. There is no full set of annual reports or audits of the two companies available at Bank headquarters. While information on loan disbursements was readily available, this was not the case for data on total project costs, specifically, expenditure by the two companies on the project did not figure separately in their accounts and had to be derived from analysis of requests for reimbursement to financing agencies. Part of the problem probably lies in the fact that the project did not c'ncern the totality of operations of the two companies and it was difficult to maintain separate accounts. When a project depends to such a large extent on contributions by the recipient organizations, it may be preferable to consider all operations as part of the project even if the Bank is only financing some of them.  - 12 - PART II. PROJECT REVIEW FROM BORROWER'S PERSPECTIVE 7/ 39. SOCAPALM considers that the project enabled it to complete the developments initiated under SOCAPALM II. In spite of the unfavorable world economic environment and other unforeseeable events like the drought, all essential components of the project were carried out. Although accounts at end-June 1988 showed a deficit of CFAF 2,733 billion, the company felt that this should not be given undue weight since the accomplishments under the project ha,' served to strengthen SOCAPALM's development potential. The company was optimistic that application of the new performance contract would help improve the financial situation provided that the Government maintained its obligations. In addition, the new sales strategy which emphasized the domestic market rather than exports was expected to contribute to the successful future of the company in the longer term. 40. CAMDEV points out that the Loan was contracted to enable completion of the CAMDEV II development program which had run out of funds. It considers that the project progressed satisfactorily to completion. Nevertheless, the consistent and dramatic fall in commodity prices during and after the project period, together with the financing agencies' policy of reimbursing expenditure only upon presentation of claims 8,, drove the corporation to finance investment operations initially with high interest overdraft funds. The combined effect of the above, together with the impact of debt service and loan repayments during the project period reduced the working capital of CAMDEV from CFAF -26 million in June 1982 to CFAF -12,037 million in June 1988. Because of the project, CAMDEV believes that if commodity prices on world markets were to improve, it would be in a position to earn substantial revenue. Local farmers and businessmen have benefitted from the increased worker population in Kompina, Penda Mboko and Malende areas, while local government has gained from taxes paid on employee earnings as well as on sales of produce. 7/ The executing agencies prepared their PCRs according to the old format, their conclusions with respect to the Project are general in nature and do not cover all the points given in the Guidelines for preparation of Part II of the new format. 8/ CAMDEV proposes that the financing agencies consider providing advances towards the acquisition of specific and costly capital items.  - 13- PART III. STATISTICAL INFORMATION 1. Related Bank Loans and/or Credits Loan/Credit Year of Title Purpose A2proval Status Ln. 1391-T-CM Oil palm plantation 1977 Completed Ln. 1392-CM maintenance and development, SOCAPALM II processing facilities, smallholder plantations. Ln. 1508-CM Continued rehabilitation 1977 Completed CAMDEV II of CAMDEV. Emphasis on rubber planting and maintenance, processing facilities, oil palm planting, smallholder plantations. Comment Ln. 2160-CM was intended to consolidate investments under SOCAPALM II and CAMDEV II. Although there has been no follow-up project, World Bank staff are advising on the Performance Contracts for the two companies in the context of parastatal reform of the Structural Adjustment Program.  - 14 _ 2. Protect Time.sble Date Date Item Planned Actual Identification 01/81 01/81 Preparation 04/81 04/81 Appraisal Mission 04/81 04181 Post Appraisal 09/81 09/81 Loan Negotiations 12/81 03/82 Board Approval 02/82 05/82 Loan Signature 07/82 07/82 Loan Effectiveness 06183 06/83 Loan Completion 06/88 06/88 Loan Closing 12/88 01/89 Co_mments There was no standard project preparation. A single mission provided the data for an Issues Paper and related Decision Memorandum as well as for the Appraisal Report, which was finalized by May 1982. Disbursements against the CAMDEV component were not initiated until signature of the "Contrat Plan" in December 1984. 3. Loan/Credit Disbursements Cumulative Estimated and Actual Disbursements (USS Million) Bank FY 1983 1984 1985 1986 1987 1988 1989 Appraisal Estimate 3.0 13.0 23.0 35.0 45.0 50.0 Actual 0.8 4.0 9.1 16.9 20.9 31.0 34.8 Actual as 2 of Estimate 21 31 40 48 46 62 70 Date of Final Disbursement 01/09/89 Amount Canceled USS 16 Million Comment: Additional US$ 0.8 Million loan fee included in actual amount.  - 15 - 4. Project Implementation Appraisal Actual (or Indicators Estimate PCR Estimate) SOCAPALM Component Maintain and bring into 5691 ha 6132 ha production oil palms. Complete housing and Villages for workers at social infrastructure Kienke and houses for at Kienke, Mbongo and officials at all 3 sites Dibombari estates. constructed as well as three schools. Two additional 20t/hr Dibombari Operational 01/85. processing lines. 20t/hr Kienke Contract signed 05/87 9/. New company head- Essentially completed quarters at Dibombari. by end-1985. Additional storage 12000 t 6000t constructed at capacity. Douala port. Smallholder Program: Maintain and bring into production oil palm plantations. 1660 ha 1600 ha Carry out new plantings. 1000 ha 1028 ha CAMDEV Component Maintain and bring into 8286 ha Target met in full. production rubber plantations. Maintain and bring into 1159 ha 1198 ha production oil plantations. 9/ In view of SOCAPALM's grave financial situation, the Loan Agreement was amended so as to permit the World Bank to finance 100Z of foreign exchange (852 of total costs for this item).  - 16 - Appraisal Actual (or Indicators Estimate PCR Estimate) Plant and maintain 1719 ha 2332 ha lo/ rubber to complete CAMDEV II. Replant with young rubber 2931 ha old rubber 2470 ha of old rubber and maintain. 500 ha old oil palm replanted, old oil palm area not replanted. New rubber processing New factories built facilities at Kompina, except for Kompina and Pendamboko, Malende, Tombel which were Tombel and MBonge. considered unnecessary. Completion of houses Accomplished. and social infrastructure at CAMDEV II estates, road rehabilitation. Smallholder Program: Maintain and bring into production: 570 ha oil palm, At project completion 844 ha rubber, there were: 557 ha Plant and maintain 500 ha oil palm, oil of which 382 ha 1000 ha rubber. mature; 730 ha rubber, of which 429 ha mature11/. 10/ Targets for planting and replanting modified following recommendations of rubber consultant which permitted longer exploitation of old trees. 11/ Program suffered from lack of interest by top management, delayed recruitment of smallholder expert, poor selection of participants (dispersed locations, insufficient family labor) and their inadequate motivation.  - 17 - 5. Project Costs and Financing A. Project Costs Appraisal Revised Actual A algl Actual EstTmat- Estimate utimate CFAF Million USS Million CAMDEV Ag.Development Rubber 8292.0 7366.2 5479.1 30.7 14.3 Oil Palm 146.0 89.9 77.4 0.6 0.2 Other(l) 2972.0 1741.7 0.0 11.0 0.0 Sub-totcl 11409.0 9198.8 6666.6 42.2 14.6 Vah.1 Equlpment Fctory equipment 2400.0 3743.9 6366.8 8.9 14.0 Other equIpment 362.0 81.2 237.6 1.3 0.6 Vehicles 928.0 1123.1 662.2 3.6 1.7 Sub-total 3890.0 4948.2 6266.6 18.7 16.8 Civil Works 6uildings 1600.0 2048.8 8131.9 6.6 8.2 Water supply 607.0 842.1 682.4 1.9 1.8 Roads 1240.0 1664.7 919.4 4.6 2.4 Sub-total 3247.0 4446.6 4733.7 12.0 12.4 Mar;gement team, studies A training 796.0 814.8 1322.6 8.0 3.4 Smallholders 1298.0 962.3 869.8 4.8 0.9 Unallocated 9161.0 4117.6 0.0 33.9 0.0 Total 29600.0 24476.2 18238.0 109.6 47.5 SOCAPALM Plantations 1734.0 1734.0 2136.0 6.4 6.6 Civil works,equtp. ;Os.0 2008.0 4008.1 7.4 10.4 vehicles Oil mills 2268.0 2268.0 2316.1 8.4 6.0 Dibombari HQ 568.0 668.0 2020.8 2.1 6.3 Oil terminals 417.0 417.0 89.6 1.6 0.2 Studies, training 191.0 191.0 146.5 0.8 0.4 Smallholders 1077.0 1077.0 1012.9 4.0 2.6 Unallocated 3139.0 3139.0 0.0 11.6 0.0 Total 11400.0 11400.0 11730.0 42.2 30.6 TOTAL 41000.0 36876.2 29968.0 161.8 78.0 (1) Headquarters general charges. Comments: Accounts of the two companies did not reflect expenditure in accordance with the categories indicated in the appraisal document. In order to provide a basis for comparison of projected ard actual expenditure, the latter has been derived from analysis of requests for reimbursement to the financing agencies. In the case of SOCAPALM, discrepancies exist between the figures so derived and those in the completion report prepared by the organization, pending clarification, figures resulting from the analysis (which are more complete) have been used. The revised estimates of expenditure for CAMDEV date from establishment of the Contrat Plan in 1984.  - 18 - B. Proiect Financing Source Planned Final Final/Planned US$,000 2 IBRD SOCAPALM 16300 15096 93 CAMDEV 25600 15197 59 Smallholders 8100 3799 47 Fee 800 751 94 Total 50800 34843 69 CO-FINANCIERS For SOCAPALM EIB 4100 2416 59 For CAMDEV CCCE 7400 5898 80 EIB 2500 2500 100 CDC 12300 10787 88 Total 26300 21601 82 DOMESTIC Government 4500 204 5 SOCAPALM 16300 11179* 69 CAMDEV 54900 10255* 19 Total 75700 21638 29 TOTAL 152800 78082 51 * Resources actually provided from self-generated funds. Other expenditures were financed through bank overdrafts and supplier credit. This helps to explain the discrepancy between the total figures for project financing and project costs. Comments: IBRD figures relate to disbursements as recorded by World Bank Loan Department. Other financing as reported by SOCAPALM and CAMDEV. Analysis of self-generated funds (below) helps to explain the shortfall in funding by the two companies.  - 19 - C. Self-Generated Funds Analysis Year SOCAPALM CAMDEV Appraisal Actual Appraisal Actual CFAF Million Z CFAF Million Z 1982/83 1545 496 32 3441 297 9 1983/84 1953 2274 116 3949 2949 75 1984/85 2311 3146 136 4845 6904 142 1985/86 2208 2189 99 4921 2352 48 1986/87 2345 176 8 4937 -3034 -61 1987/88 2818 557 20 5404 -1307 -24 Total 13180 8838 67 27497 8161 30  - 20 - 6. Project Results A. Direct Benefits Ind4cator- Unt 1982/3 196S/84 ,1984/85 19,85186 1986687 1981168 SOCAPAIJM Production Ap po . i44 FF8 ton. 161599 186328 207399 225487 238663 N A. Actual FF8 ton- 144621 143100 162891 201048 256S20 248620 Act.al/Aporseial S 89 77 74 80 107 NA. R*al;i d Yield t/h. 8.S 7.6 7.3 9 6 12.1 11.8 Palm Oil Appraisal tons 33232 38279 42843 46681 49840 NA. Actual tons 27850 28021 32492 46162 62096 62792 Actual/Appraissl S 84 73 76 97 106 N.A. Extraction Rate S 19.3 19.6 20.2 22.2 21.0 20.6 Smllhold9r, number 1919 Production Appraisal FF9 tons 15720 Actual FFB tons 15130 Actual/Appraisl S 96 CAMDEV Rubber Production Appraisal tons 13S17 13683 14309 15665 1m7 N.A. Contrat Plan tons 11882 12383 138S6 14698 N.A. Actusl tons 11575 11409 13263 1S861 16694 17856 Actual/Appraisal S 87 83 93 88 04 NA. Actual/C. Plan 3 96 106 100 114 NA. Reali-ed Yield kg/ha 1061 986 1094 1029 1148 1178 Oil Production Appraisal FF8 tons 124224 129885 131866 131527 129957 N.A. Contrat Plan FFB tons 81000 109917 123077 122397 NA. Actual FFB tons 103987 89633 106314 118278 122516 101408 Actual/Appraisal S 84 69 80 90 94 N A. Actual/C. Plan S 111 96 96 100 N.A. Realised Yield t/he 6.8 6.8 6.8 7.8 8.3 8.0 Palm Oil Appraisal tons 25392 26737 27663 27.45 27624 N.A. Contrat Plan tons 21185 16200 21983 23015 24479 N.A. Actual tons 14275 17914 22734 28211 26701 23479 Actual/Appraisal S 66 67 82 102 97 N.A. Actual/C. Plan S 67 111 103 113 109 N.A. Extraction Rats S 19.9 20.0 21.6 22.4 20.5 21.1 Sal lholders Rubber Produced tons 384 FFId Purchased tons 10906 Comments: Results are for targets indicated in Appraisal Working Papers which cover all rubber and oil palm production of SOCAPALM and CAMDEV. The latter's targets were substantially revised under the Contrat Plan established in late 1984 with the support of the Bank.  - 21 - B. Economic Impact Comments: The economic rate of return could not be calc-lated on the saiuie basis as at appraisal, primarily because data were not kept for the different estates of various crop enterprises (first generation oil palm, second generation rubber etc.). There was also no information on the previously incurred investment costs which were taken into account in the appraisal analysis for components supporting the bringing into production of already planted areas (66Z of project costs).  - 22 - C. Financial Impact Financial Rate of Return SOCAPALM: Oilpalm: 29X CAMDEV: Rubber: -14? Oilpalm: 232 In order to assess the financial impact of investments in the two commodities, separate analyses were done on the incremental net benefit (i.e. gross benefits' stream minus capital investment and operating costs net of without-project net benefit) of oil palm production at SOCAPALM and oil palm and rubber production at CAMDEV, (see Table 6C(l), (2), (3)). Investment costs are net of plantation expenditure, which is reflected in production costs, and the smallholder component for which supplementary information on output was lacking. As the project was the third in support of CAMDEV and SOCAPALM and essentially concerned with consolidation of previous investments, attribution of production is particularly difficult. In the absence of information to the contrary, production in the without project situation has been assumed to be 25,000 tons and 22,500 tons of palm oil in SOCAPALM and CAMDEV respectively and 10,000 tons of rubber at CAMDEV. Oilpalm has a positive financial IRR of around 20X in both CAMDEV and SOCAPALM (Tables 6C(2), (1)). This is largely due to the high prices of oil in the protected domestic market. However, using border prices of CFAF 90/Kg instead of the CFAF 180/Kg included in the financial analysis, would yield a negative return for CAMDEV and a much lower rate for SOCAPALM. This is mainly due to low world prices which are below costs of production during most of period under study. This clearly indicates that oil palm production in Cameroon is not economically viable. Rubber production at CAMDEV shows a negative financial rate of return of 14Z (Table 6C(3)). Overvaluation of the CFA vis-k-vis the dollar has eroded the sector's competitiveness, with prices in CFA falling from 400- 450/kg in 1983-1984 to under 300/Kg in 1989/90. The result has been substantial negative net flows for the CAMDEV rubber group during a period when heavy investments were being undertaken. However, a simulation using a 40? devaluation of the CFA franc to the dollar between 1986/87-1988/89 and a 50? devaluation thereafter, improves tremendously the cash flow and increases the rate of return to 8? (Table 6C(4)).  23 _ .3 a, .33: 330 3 :3~ 33 ew 2, -',8.o .o o3 33 C3 0 3) 03 o0 { o. ~~ 8 .S o 0 0 0o 033g > , o 8 sg