Ghana - Livestock Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 3972 PROJECT PERFORMANCE AUDIT REPORT GHANA LIVESTOCK DEVELOPMENT PROJECT (CREDIT 500-GH) June 16, 1982 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. ABBREVIATIONS APMU - Agricultural Project Management Unit ERR - Economic Rate of Return GLC - Ghana Livestock Company Ltd. IEMVT - Institut d' Elevage et de Medicine Veterinaire (French Livestock and Veterinarian Institute) PMWA - Permanent Yission to West Africa SEDES - Societe d' Etudes pour le Developpement Economique et Social (French Consultancy Firm) SFC - State Farms Corporation COUNTRY EXCHANGE RATES Name of Currency (Abbreviation) Cedi (0) Year: Exchange Rate: Appraisal Year Average US$1 = 01.1538 Revision Year Average US$1 = 1.1538 Completion Year Average US$1 = 02.75 FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT GHANA LIVESTOCK DEVELOPMENT PROJECT (CREDIT 500-GH) TABLE OF CONTENTS Page No. Preface .......................................................... i Basic Data Sheet ..................................................... ii Highlights ...................... ................................ iii PROJECT PERFORMANCE AUDIT MEMORANDUM I. PROJECT SUMMARY ........................................ 1 II. MAIN ISSUES - ......................... . 4 A. Bank Performance .......5............ ....... 5 B. The Country Situation ...................... 7 C. A Successful Livestock Development Project ......... 8 D. Strategy for the Livestock Sector .................. 9 E Cofinancing with Local Banks ................... 11 F. Disbursements, Suspension and Closing.............. 12 Disbursements and Suspension........ 12 Credit Closing and Cancellation .1.......14 Decision-Making Around Project Closing ........... 15 G. Conclusions -................................... . 19 Annex 1 - Bank Operations in Agricultural Sector in Ghana ........ 21 Annex 2 - Herd Coefficients ......................... 23 Annex 3 - Estimated and Actual Disbursements and Disbursement Profiles ......................... 24 ATTACHMENT: Comments of the Executing Agency ................... 25 PROJECT COMPLETION REPORT I. Background .......................................... 44 II. Project Formulation and Appraisal ..................... 44 A. Identification ............. ............... 44 B. Project Preparation ............................... 46 C. Appraisal ................................... .... 46 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. TABLE OF CONTENTS (Continued) Page No. III. Project Implementation ................................ 48 A. Effectiveness and S:art-Up, .................... 48 B. Revision ........................................ 48 C. Physical Implementa:ion ............................ 50 IV. Project Costs and Financing ............................58 V. Project Impact ...................................... 61 VI. Financial and Economic lesults .................... 61 VII. Institutional Performance and Development .............. 63 A. Board of Directors .............................. 63 B. Accounts and Audits ............................... 64 C. Technical Assistance ............................... 64 VIII. Bank Performance .................................... 65 IX. Conclusions ............................................ 66 Tables: 1 - Data Changes over Project Life ..................... 70 2 - Stock Numbers on Ra.ches ........................ 71 3 - Accra Meat Prices ............................... 72 4 - Investment Expenditares ............................ 73 5 - Financial Cash Flow ................................ 74 6 - Financial Rate of Return ...................... 75 Maps: IBRD 10484 (PPAR) Location of Project Ranches IBRD 12805 IBRD Agricultural Projects PROJECT PERFORMANCE AUDIT REPORT GHANA LIVESTOCK DEVELOPMENT PROJECT (CREDIT 500-GH) PREFACE This is a performance audit of the Ghana Livestock Development Project, for which Credit 500-GH in the amount of US$2.0 million was approved in June 1974. The credit was closed in June 1981 as originally scheduled. The final disbursement was made in November 1981, and the remaining balance (US$0.65 million) cancelled on January 21, 1982. The audit report consists of an audit memorandum prepared by the Operations Evaluation Department (OED) and a project completion report (PCR) dated November 12, 1981, slightly updated February 5, 1982. The PCR was prepared by the Western Africa Regional Office following a visit to Ghana in June 1981. The audit memorandum is based on a review of the Appraisal Report (No. 277-GH) dated May 28, 1974, the President's Report (No. P-1390-GH) of June 6, 1974, the Credit and Project Agreements dated July 26, 1974 and the PCR. Correspondence with the Borrower and internal Bank memoranda on project issues as contained in relevant Bank files have also been consulted, and Bank staff associated with the project have been interviewed. An OED mission visited Ghana in October/November 1981. Discussions were held with officials of the Ministries of Agriculture and Finance and Economic Planning, the Ghana Livestock Company (GLC) and the commercial banks which participated. in the project. A field visit to the Branam ranch was also undertaken. The information obtained and the observations made during that mission were used to evaluate the conclusions of the PCR and are also reflected in the audit memorandum. The draft.report was sent to the Borrower and the executing agency as well as the commercial banks who participated in the project for comments on March 15, 1982. Detailed and lengthy comments and clarifications have been received from the executing agency and are attached to this audit memorandum. In most cases, these duplicate, but also expand upon, the PCR as GLC presents its complete views on the topics selected for comment. Variances with PCR conclusions exist, but are small in number and mostly a matter of interpreta- tion more than fact. GLC's comments can stand mostly on their own. Because of this, and extensive references in GLC's letter, only the most important dissenting opinions and cross references have, therefore, been footnoted in the audit memorandum. The PCR provides an excellent overview of project formulation and implementation and accurately analyzes the reasons for project failure. The audit expands on some of the issues which arise from the implementation experience. OED gratefully acknowledges the valuable assistance provided by the Government and project staff, as well as other banking officials involved in the project. Their assistance greatly contributed to this report. s 1 - ii - PROJECT PERFORMANCE AUDIT REPORT GHANA LIVESTOCK DEVELOPMENT PROJECT (CREDIT 500-GH) BASIC DATA SHEET KEY PROJECT DATA Appraisal Revised Actual or Actual as Expectations Expectations Current Estimate % of Estimate Total Project Costs (US$ million) 4.5 3.8 4.0 89 (105)/a Credit Amount (US$ million) 2.0 2.0 1.4/b 70 (70) Date Physical Components Completed 12/31/80 12/31/80 12/31/81/c 117 (117) - Economic Rate of Return (%) 12 - -4.1 Negative Financial Rate of Return (%) 14 8.5L 16 (maximum) 114 (188) Number of Direct Beneficiaries 50 20 0 0 (0) CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS (US$ thousands) FY76 FY77 FY78 FY79 FY80 FY81 FY82 Estimated (Appraisal) 550 950 1,400 1,800 2,000 - Estimated (Revised) 60 600 1,200 1,900 2,000 - - Actual 61 503 876 1,071 1,339 1,349 1,355b Actual/Estimated Revision (%) 102 84 73 56 67 68 68 Date of Final Disbursement: 11/09/81 Date of Cancellation : 01/21/82 Amount Cancelled : US$0.65 million MISSION DATA Date No. of Manweeks Specializations Performance Types of Mission (Month/Year) Persons in Field Represented/e Rati Trend/& Problems/h Identification (Bank) 12/70-01/71 1 14 L - - - Identification (FAD/CP) 05-06/71 (4) (70) L FE - - - Appraisal (Bank) 02-03/73 3 108 L F E - - - Total project formulation 122 Supervision I 06/75 1 16 L 3 2 M 0 Supervision II 09-10/75 2 26 L F 3 2 M F Supervision III (Revision) 03/76 1 13 L 3 1 I F Supervision IV (Follow-up) 05/76 1 4 L - - - Supervision V 01-02/77 1 7 L 2 1 M Supervision VI 05-06/77 2 20 L F 2 2 M T Supervision VII 01/78 1 10 L 3 3 M P T Supervision VIII 05-06/78 2 18 L F 3 1 M P T Supervision IX 02/79 1 9 L 3 1 M F T P Supervision X 01/80 2 16 L F 3 2 M F T P Supervision XI 09/80 1 7 L 3 2 M F T Supervision XII 02/81 1 5 L - - - Supervision XIII (Completion) 06/81 2 24 L E 3 3 M F T Total 175 OTHER PROJECT DATA Original Actual or Item Plan Revisions Est. Actual First Mention in Files - 09/12/69 Government Application - - 11/72 Appraisal 10/72 - 02-03/73 Negotiations - - 05/74 Board Approval 03/74 - 06/18/74 Credit Agreement Date - - 07/26/74 Effective Date 10/25/74 01/31/75 05/31/75 05/14/75 Closing Date 06/30/81 06/30/81 06/30/81 Last Disbursement Date 12/31/79 12/31/80 11/81 Borrower Government of Ghana Executing Agency Ghana Livestock Company (GLC) Fiscal Year of Borrower July 1 - June 30 Follow-on Project None /a An overrun of 5% if measured against revised estimates. 76 US$0.65 million cancelled January 21, 1982. /c Development not completed at Closing Date and continued at a low level beyond. /d Figure would increase to 27% if feeder steer fattening program is included. /e Specialization represented: L = Livestock Specialist, F = Financial Analyst, E = Economist /f Performance rating: 1 = Problem-free, 2 = Moderate Problems, 3 = Major Problems /g 1 = Improving, 2 = Stationery, 3 = Deteriorating /h F = Financial, M = Managerial, T = Technical, P = Political, 0 = Other 個 - iii - PROJECT PERFORMANCE AUDIT REPORT GHANA LIVESTOCK DEVELOPMENT PROJECT (CREDIT 500-GH) HIGHLIGHTS The project was designed to create the beginnings of a modern com- mercial cattle industry in Ghana through the development of three ranches managed by a limited liability company - Ghana Livestock Company, Ltd. (GLC) and to assist smaller privately owned herds (50 eatimated at appraisal) in areas surrounding the ranches. The project was to improve breeding stock and slaughter cattle and included preparation of a second stage project. Project costs were estimated at US$4.5 million; the economic rate of return was estimated at 12%. Project implementation was held up initially due to delays in ful- filling conditions of effectiveness. Subsequent land acquisition problems, delayed management appointments and cost increases resulted in a major project revision in 1976: the implementation period was reduced from six to five years, the number of ranches from three to two and the number of private farmers from 50 to 20. Physical development was much delayed at the ranches, and adequate water supplies were only available after much pressing by the Bank. Herd build-up will be much lower than expected. The private herd component was not pursued, and a follow-up project was not prepared. The project, originally expected to be completed by December 1980, continued beyond that date. The estimated time overrun is, therefore, at least 20%, but costs were ll%' below appraisal estimates and 5% higher than the 1976 project revision. The reestimated rate of return is negative com- pared with the appraisal estimate of 12%. However, due to major economic distortions which make importation of cattle and their sale on the local market highly profitable, GLC's financial rate of return can be reestimated at 16% maximum (appraisal estimate 14%) provided certain assumptions will be realized (PPAM, paras. 9 and 53). Management and national economic problems have beset the project since the beginning, and it has been rated a problem project from 1975 until closing (June 30, 1981). The project's impact *has been limited, and the project can be classified as a qualified failure: a failure because of its negative reestimated economic rate of return; qualified because a livestock company has been established which has been estimated to be, in theory, financially viable. However, this is dependent on GLC obtaining additional funds and foreign exchange, selling price increases and improved technical coefficients. Although realizable, this might be optimistic in view of GLC's history, the current economic situation in Ghana and GLC's precarious finan- cial situation at project completion. Urgent steps have to be taken by the Government and the local banks who cofinanced the project to ensure that GLC remains a viable undertaking. - iv - Lessons which emerge from the project's implementation experience include: (i) land acquisition should be completed before project implementa- tion starts (PPAM, para. 6 and PCI, para. 41); (ii) a commerical company might not be the proper vehicle for improving the smallholder livestock industry (PPAM, para. 31 and PCR, para. 38 and 76); (iii) good management and its reasonable autonomy are important for commercial ranching (PPAM, paras. 37 and 25-26); (iv) the Bank should monitor more closely and follow up the mechanics of loan/credit closing (PPAM, p&ras. 43-44; and (v) the Bank should ensure that Borrowers' perceptions about their own and the Bank's responsibilities in project implementation are realistic and that false expectations about these roles are dispelled (PPAM, para. 16). Other points of interest are: - an innovative feature of the project was the cofinancing with a consortium of local banks. Although their financial participa- tion was lower than envisaged, these banks remained interested in the management of the company and will remain a prominent source of company funding in the immediate future (PPAM, paras. 33-37; PCR, paras. 16, 58, 66 and 77); - commercial livestock development in Ghana is possible, as has been shown by the successful livestock investment made by the Bank of Ghana in the Shai Hills Ranch (PPAM, paras. 24-28); - if a training program is to be mounted, it is necessary to identify at appraisal who is to conduct the training, what are the sources and qualifications of trainers and what the training program should consist of (PCR, para. 76); and - although they werE represented on GLC's Board, the Ministry of Finance and Economic Planning felt that they had not been adequately informed about project progress and problems by the Bank (PCR, para. 73, but see also GLC comment, para. 30). - 1 - PROJECT PERFORMANCE AUDIT MEMORANDUM GHANA LIVESTOCK DEVELOPMENT PROJECT (CREDIT 500-GH) I. PROJECT SUMMARY 1. The Livestock Development Project was the Bank's fourth operation in Ghana's agricultural sector and the first in its livestock subsector. Financed by Credit 500-GH of US$2.0 million, the project was designed to lay the foundation for a modern cattle industry in Ghana. 2. The possibility of a livestock project was first mentioned in April 1969. Subsequently, at the end of 1970 a Bank reconnaissance mission identi- fied a livestock project and proposed the establishment of four state ranches in various parts of the country and a pilot extension component for some private cattle owners. Ghana's Ministry of Agriculture prepared a commercial ranching project proposal which was submitted to the Bank in April 1971. The project was prepared by consultants in mid-1972; their proposals were almost identical to those of the earlier identification. The appraisal mission of February 1973 agreed with the preparation mission's concept but reduced the proposed number of ranches and reintroduced the pilot component for small private ranches; a new feature was the possible participation of local commer- cial banks. Negotiations were held in Washington in May 1974, and the Board approved the Credit in June 1974 under special procedure arrangements. 3. The project, as approved, was to provide the basis for a commercial cattle industry in Ghana. A company - Ghana Livestock Company Ltd. (GLC) - was to be formed to manage three commercial ranches and to assist in develop- ing smaller private herds in areas surrounding the ranches. The project was to produce improved breeding stock and slaughter cattle. Total project costs were estimated at US$4.5 million; the economic rate of return was calculated at 12%. The project would comprise:1/ (i) the development of three breeding/fattening ranches (Tadzewu, Pong Tamale and Branam - see map 10484 (PPAR) attached to this report) by rehabilitating and expanding two existing ranches and a maize farm which would be taken over from Government. The main development would include: land clearing, fencing, provision of watering facilities, improved pastures and the purchase of breeding cattle; 1/ See Appraisal Report - Appraisal of Livestock Development Project, Ghana, Report No. 277-GH, dated May 28, 1974, para. 3.01. - 2 - (ii) the improvment of some 50 privately-owned herds, initially of about 100 head each, in areas surrounding the ranches. The owners would be provided with technical advice and finance for improved breeding stock and, where necessary, spray races, fencing materials, land clearing and pasture improvement; (iii) the training of Ghanaiar. managers; and (iv) the identification of further ranch sites and the preparation of a second stage project. 4. Project start-up was held up from October 1974 to May 1975 by delays in fulfilling the conditions of effectiveness, mainly the difficulty in find- ing a suitable person for the position of GLC managing director.11 Subsequent cost increases for materials and cattle, serious reductions in the expected profitability of growing maize, lower than expected cattle availability, settlement of smallholder rice growers on prospective ranch land (Pong Tamale) and delayed management appointments necessitated a project revision._/ This revision took place in May 1976 and included reductions in implementation period from six to five years, :he number of ranches to be developed from three to two, and the number of private farmers to be supported from 50 to 20. The Board approved the necessary amendments to the Credit and Project Agreements.3/ 5. Initial discontinuity in management affected project progress. In addition, national economic problems dislocated procurement procedures and also affected the development in general. This, together with the need to 1/ In its comments (paras. 4 ar.d 6), GLC states that one of the conditions was not fulfilled at the time of effectiveness as adequate arrangements for the transfer of assets had not been made at that time. However, de jure what was required was the entering into of an agreement for the transfer of property and assets (Credit Agreement, Section 5.01 e). De facto GLC makes a valid point namely that the actual transfer took a long time and was an important reason for implementation delay. This has been recognized and the lesson that transfer of property, especially land acquisition, should be completed before project implementation starts, has been highlighted in this report (Highlights and PCR, para. 41). 2/ Adequate provision had not bEen made in the Appraisal Report for salaries and other benefits of the Managing Director and the three Ranch Managers; however, this was provided for under the project revision (GLC comments, para. 3). 3/ President's memorandum: Ghana - Credit 500-GH (Livestock Development Project). Proposed amendments to Credit Agreement and Project Agree- ment, dated August 6, 1976 (IDA/R76-66). In the absence of objections by August 18, 1976, these revisions were implemented. For more details see PCR, paras. 20-24. - 3 - revise the project, was the reason to rate the project in June 1975 as having major problems. The project continued to be rated as such until the Closing Date (June 30, 1981), except in 1977 when some progress was recorded. 6. Physical development at the Branam ranch (one of the two remaining ranches under the project) was seriously delayed on account of not having full rights of occupancy. Title to the land was only obtained in June 1979, although ranch development had started earlier. The dip and handling yards took two years to complete; road construction took about one year and although well constructed proved expensive; water development was much delayed, and adequate water supplies were only available after much pressing by supervision missions in middle-1981; no buildings were constructed. Legalities of the Tadzewu ranch acquisition were only completed in 1980;1/ existing spray race and some fences were repaired and some additional paddock fencing con- structed; no buildings were constructed; roads were not constructed either, but some firebreak establishment was carried out; one dam was completed in 1981. 7. The private farmer component was not pursued. This component was seen as a public service function and given less priority by GLC relative to ranch development.2/ Overseas training of Ghanaian staff never materialized either; priority was given to on-the-job training, and suitable staff could not be found for overseas training. Preparation of a second stage was never undertaken as it became apparent that a follow-on project would be unrealistic in the light of the project's unsatisfactory performance. 8. At appraisal, the project was expected to build up the company herds to a total of 14,000 head and increase private herds from 5,000 to 13,000 head. GLC now projects the company herds at full development to total only 7,600 head (at present,. the total GLC herd numbers 2,100 head), and the increase of the private herds will be nil. The demonstration effect from GLC practices has been insignificant because GLC operations were themselves generally unsatisfactory or not replicable on small farms. On the positive side, it must be mentioned that the project, in addition to fattening cattle, imported some 1,000 breeding stock with the potential of improving the local herd. The project also sold about 450 head of breeding stock to local farmers. 1/ For Branam ranch land, acquisition was only completed in 1979 and for Tadzewu, only in 1980. Much of GLC management's time was taken up with the problems of land acquisition. During negotiations, the Bank had accepted verbal assurances that land would be available. With the benefit of hindsight, land acquisition should have been made a condition of negotiations or Board presentation (see PCR, para. 41 and also GLC comments, paras. 6-10). 2/ In the attached comments (para. 26), GLC provides its detailed reasoning for not undertaking the private farmer component (see also PCR, para. 39). 9. Despite the delays in project implementation and the financial difficulties facing the company, GLC could nevertheless achieve a financial rate of return of about 16 percent maximum (appraisal estimate was 14 percent) if they did everything right. The major reason for this is that at the official exchange rate for the Cedi the importation of cattle and their eventual sale in the local market is highly profitable.-/ The economic out- come is quite different when the real exchange rate for the Cedi is applied: the reestimated economic rate of return for the project is negative (appraisal estimate was 12 percent). II. AAIN ISSUES 10. The problems afflicting the project were both project-specific and country-specific. Bank supervision missions classified the project as having major problems throughout most of its implementation; the types of problems indicated ranged from technical, managerial and financial to political and other. The PCR (see para. 75) identifies the major project problems as: (i) weak conceptual featu:es such as: scattered location of the ranches; establishment of the lead ranch (Branam) in a tse-tse infested, and thus low cattle density, area and use of a private company as a medium for a service and credit function; (ii) difficult political circumstances; (iii) weak management at different levels in the project; and (iv) financial problems due to high inflation and adverse exchange rates. 11. Borrower and participat:Ing commercial bank officials agreed gener- ally with this assessment, but some felt that the Bank was mainly responsible for the poor project concept and the initially unsatisfactory management situation as well as for funding problems. Other officials were, however, more inclined to see the country's difficult political and economic situation as the main cause of the problems. These issues are discussed below. 1/ Realization of this rate of return is, however, dependent on various factors: GLC obtaining additional funds and foreign exchange; selling prices increasing; and tech-ical coefficients improving to reasonable levels. While all these assumptions are realistic, some must be regarded as optimistic in view of GLC's operations history and the current eco- nomic situation in Ghana (see PPAM, para. 53). Furthermore, at project completion, GLC's financial situation was precarious; see PCR, para. 58. -5- Bank Performance 12. Bank performance must be seen in the light of the difficult country situation. As a consequence the Bank was often faced with the dilemma of whether or not to suspend disbursements. Some aspects of Bank performance merit special attention.!/ 13. The project had a number of weak conceptual features (PCR, para. 75): poor location of ranches (PCR, paras. 42 and 69) with the three ranches widely scattered throughout the country and one of these in a tse-tse infested area, and entrustment of a public function to a profit making company (PCR, para. 39). The reason for the scattered location of ranches was to benefit from existing investments. The actual project requirements in terms of assets, the valuation of the existing investments, the suitability of the areas for cattle and maize growing and the requirements for land acquisition were, however, not carefully assessed, and problems resulted. The Bank bears some responsibility for the conceptual mistakes because it was heavily involved in project identification and preparation from the earliest stage. 14. Later, the search for and appointment of company management was fraught with difficulties. The first managing director, an expatriate, performed unsatisfactorily. The second managing director, also an expatriate, spent too short a time in the job to get much done. The third and current managing director lacks experience and had to cope with the general ineffi- ciencies of the civil service. Ranch managers have not always been found competent either. 15. Bank staff had identified the lack of livestock expertise in Ghana and hence rightly recommended use of an expatriate consultancy firm. GLC Board members made it clear to the audit mission that they had originally relied on the Bank's judgement of capabilities of prospective managing directors, even though they themselves had at the time expressed explicit agreement with the appointments. The poor results and the high cost of expatriate management, however, made the GLC Board suspicious of the Bank's judgement and even its intentions in this respect, and the GLC Board started to look for a Ghanaian managing director. The GLC Board did therefore not accept the later Bank recommendation to employ an expatriate management consultancy firm. One of the last Bank supervision missions blamed the Board for this decision and argued that this and the Board's apparent inability to turn management in the right direction was the reason for continued weak project performance. 16. These events caused mutual ill-feelings. The audit concludes that the respective project implementation responsibilities of the Borrower and of the Bank have been perceived differently by each party and that this 1/ GLC's view on these aspects of Bank performance can be found in the attached GLC comments (paras. 29 and 31-34). - 6 - caused problems. The Bank's responsibility for providing or arranging for qualified management and, in general, to carry the project through was per- ceived as high by the Borrower. It expected the Bank to be a partner in development rather than merely a provider of finance.l/ A lesson to be learned from this project is that the Bank should ensure that Borrowers' per- ceptions about their own and the Bank's role in project implementation are realistic and that false expectations are dispelled. Most of the Bank's supervision efforts have never:heless been appreciated by the Borrower. 17. The GLC Board perceived the Bank's disbursement procedures as inflexible. While in most cases, the Bank rightly stood by its procedures, the audit mission has doubt in one case. After repeated requests to do something about water development at the Branam ranch, the Bank in February 1980 indicated to the GLC Board that, because of increasing stock numbers in relation to water availability, reallocation and further disbursement of funds for cattle purchases would deper.d on sufficient water supply at the Branam ranch. The February 1981 supervision mission noted slow progress on water supplies and continued to judge project and management performance as very poor. It also noted that breeding stock numbers had declined; foreign ex- change to purchase fattening steers (to improve the ranch cash flow) was lacking and country conditions were worsening. The Bank, therefore, decided to close the project as scheduled on June 30, 1981. In June 1981, a Bank supervision mission found that construction of additional watering facilities had just been completed and that this would justify the introduction of more cattle at Branam (PCR para. 27); the mission informed the GLC Board that the Bank might consider extending the Closing Date. In fact, the mission did not base its recommendation on the additional water finally becoming avail- able, but rather in consideration of the likely proper utilization of the still undisbursed funds. However, the Bank saw no evidence that the project had improved2/ and decided to stick to the decision on credit closure. The audit has doubts on the appropriateness of closing the project at that moment. First, after seven yeari/ of implementation efforts and problem- solving, the project had finally reached a stage where disbursements for cattle purchase were warranted..7 Second, there have been moments in the 1/ Regional staff insist that they properly informed the Ghanaians of their responsibilities. 2/ Regional staff argue that the provision of water after such a long delay was not indicative of :Improvement and that there was no evidence to suggest that the new stock would be better looked after. Mortality figures had risen significantly. 3/ The Closing Date, June 30, 1981, was almost seven years after the sign- ing of the Agreements. 4/ Especially critical for GLC's continued viability is the availability of foreign exchange to import breeding stock. - 7 - project life when credit closure would have seemed more justified than in June 1981. Issues regarding disbursements, suspension and closing are further reviewed in this audit memorandum (PPAM paras. 38-53). B. The Country Situation 18. When the project was still under preparation, the country situation was already reported as being difficult. Not much real agricultural develop- ment had occurred since independence. Large-scale investments in state farms, mechanized crop production schemes, etc. had for the most part proven expen- sive failures. However, there was no modern cattle industry in Ghana. Beef was in short supply with per capita consumption in 1973 low and declining. In addition to the favorable market conditions, large areas in the country were suitable for cattle and there was interest in commercial cattle production. Thus, conditions were favorable for a successful project if it could be isolated from the more general country specific difficulties. 19. Possible problem areas were already identified at preparation and appraisal: availability of land and of suitable breeding stock, valuation of ranch assets to be transferred, price controls and the overvalued Cedi. Central Projects Staff felt at that time that removal of price controls by the Government and preferably an adjustment of the Cedi exchange rate too should have been made conditions for approval. This was, however, not done since: (i) immediate removal of all price controls relative to this particular project was considered inappropriate; (ii) the sharp retrenchment of the import program had put large pressures on food prices;.and . (iii) the Government seemed quite determined at the time to pursue sound price policies. At negotiations, and at Government request, it was agreed that specific references to Government controls on cattle or meat prices be deleted from the legal documents. 20. The Government also assured (verbally) that the land required for GLC would be made available with satisfactory title and security and further- more that import licenses would be issued to the company to procure necessary supplies and breeding stock. Another issue at appraisal was the percentage of company shares to be controlled by the Government. Initially 25% was pro- posed, but at Government request it was agreed during negotiations to increase this percentage to 70%. 21. The project was thus subject to Government control and thereby greatly influenced by country conditions. And yet these country specific conditions were a major reason for project failure. Country problems were - 8 - conducive to low staff morale, irregularities, inefficient supporting ser- vices, funding problems, foreign exchange shortages, rampant inflation, distorted cost-price relationships, shortage of supplies and generally poor macro-economic management. These country problems also plagued other projects and have been discussed in earlier OED audits.1/ 22. These country specific problems may be summarized in the words of two Borrower officials. One expressed to the audit mission his opinion that management of Government sponsored projects has been mostly problematic. Such projects are not considered very important because staff feel no direct relation with the results. The tendency is then to be somewhat negligent, and such negligence is normally not discovered or accepted as normal. Another official stated that civil servants have to spent as much as half their official working hours attending :o personal business to supplement their low insufficient salaries. At only slightly higher than Government salaries, Ghanaian staff in Bank-supported projects cannot, therefore, be expected to work diligently for the projects.-! 23. What the country needs, said a Ghanaian in the private sector, is tax reform to provide better procuction incentives and reduce smuggling into neighboring countries. This should be achieved by harmonizing Ghana's tax system with those of its neighbors, by lowering taxes on exported products like cocoa and on personal incone and by increasing taxes on common goods and services. The audit mission agrees that tax reform should feature promi- nently in a redress of the country problems. Tax reform measures to alleviate Ghana's revenue shortage have earlier been suggested by OED../ Such measures would need to be complemented by other structural adjustments. It is beyond the scope of this audit to present detailed suggestions along these lines. It appears, however, that Bank-supported projects in the agricultural sector so far audited by OED have had little development impact due to the country's general problems. C. A Successful Livestock Development Project 24. In contrast to the project's experience, an example of a successful experience with livestock development exists: the Shai Hills Ranch Company, established and financed by the Bank of Ghana (see also PCR, paras. 9 and 1/ PPAR, Ghana-Eastern Region Cocoa Project (Credit 205-GH), OED Report No. 3526, dated June 24, 1981 (paras. 14 and 15) and PPAR, Ghana-Sugar Rehabilitation Project (Credit 354-GH), OED Report No. 3525, dated June 24, 1981 (paras. 10 and 18-22). 2/ Tentative calculations in the late seventies by Bank supervision missions for the Ghana Upper Region Agricultural Development Project (Third Window Loan 1291-GH) revealed that staff salaries would only cover 50% of the cost of living requirements of the official's family. 3/ PPAR, Ghana - Eastern Region Cocoa Project (Credit 205-GH), OED Report No. 3526, dated June 29, 1981, para. 16. -9- 76(c)). This ranch was properly located, implemented in a logical sequence of : provided with sufficient funds, foreign exchange and strong management and was with the full support of the Bank of Ghana to a certain extent immune from civil service inefficiencies. 25. A somewhat more detailed review of the Shai Hills experience reveals that the implementation approach applied there carried success. The ranch started activities in 1976, when the company was incorporated. Before that, however, land for the ranch had been acquired over the 1970-76 period, and management had been identified in 1974. Funding the company has not been a problem. The ranch is located at a two-hour driving distance from Accra, facilitating supply of inputs and selling of cattle. The area is suitable for grazing even though the existing pasture is only of fair quality. Management is of high standard, resident on the ranch and expatriate. Nevertheless, the company has also experienced some difficulties in procuring inputs and cattle; its herd has also suffered from cattle diseases (skin disease and pleuropneu- monia), but these problems have not been major impediments. 26. The Bank of Ghana considers good management the key to the success of the ranch. Good management is particularly important for selection of breeding stock, proper herd management, provision of water at all times, round the clock grazing and supplementary feeding. Expatriate management holds an advantage as it is less subject to internal domestic pressures than local management. 27. Actual financial performance indicators could not be made available to the audit mission, but the Bank of Ghana recovered the funds it invested in the company, and the company itself is making profits. The herd coefficients on the ranch are good (calf mortality rate for example is only 3%). An interesting feature is also that no cash crops are grown for initial cash flow improvements, but instead cattle are fattened and sold. 28. In summary, the Shai Hills Ranch Company has established a viable livestock enterprise simply by applying some basic common sense criteria. The purpose of this livestock venture according to the Bank of Ghana has been to demonstrate lending opportunities to other commercial banks, and as such it was successful. D. Strategy for the Livestock Sector 29. Pilot extension operations involving a limited number of nearby private cattle producers were planned at appraisal. This component was never initiated, however.1/ The project as a whole gives, nonetheless, some indications about what should be included in a livestock sector strategy. 1/ See also GLC comments, para. 26. - 10 - 30. Any development of the cattle industry in the country must center on the promotion of private farners, as is concluded in the PCR (see PCR, para. 78). Several Borrower officials hold the same opinion for the obvious reason that virtually all cattle in Ghana is in the hands of smallholders. The audit mission also concurs with this view. 31. The second lesson from the project in this respect is that a company like GLC is not the proper vehicle for improving the smallholder livestock industry. Any efforts on GLC's part to improve smallholder cattle may con- flict with its own profit-making objective._/ Substantial economic benefits would result from smallholder Livestock development. These would mainly accrue to the smallholders themselves in the form of increased income and to the country in the form of a growing economic sector, a broader tax base and foreign exchange savings. The only way a profit-oriented company could benefit from this development effect is through a fee to be imposed on the smallholders for services rendered, apart from profits in the form of cattle sales to smallholders. But since these services are more of a public nature they are more appropriately carried out by public services, funded through the public sector budget. The Anima.l Husbandry and Veterinary Divisions in the Ministry of Agriculture have rendered such services to smallholders and are a more logical channel for these services. The dilemma is, of course, the present low efficiency of these divisions. 32. As an alternative to developing smallholder livestock through public efforts, current development activities in the sector could be continued. These ongoing activities include: (i) commercial ranching, including the Shai Hills ranch, which appears to be very profitable (in financial terms), and GLC's ranches, which may turn out to be only marginally viable (also in financial terms); (ii) keeping small herds of about 50 head of cattle each as additional sources of income for fixed salary and wage earners (this is report- edly an increasingly popular engagement for civil servants); and (iii) improvements in smallholder herds through sale of improved breeding stock and demonstration effects of (i) and (ii). Any strategy for Ghana's livestock sector should in any case take into account the country problems and the faat that under the prevailing overvalued Cedi livestock development is heavily subsidized at the expense of the export sector, particularly cocoa. 1/ It should be noted, however, that the Shai Hills Ranch Company, with the support of the Bank of Ghana, recently established a training and support program for smallholders; it undertook such program only after it had established its own ranch. - 11 - E. Cofinancing with Local Banks 33. An innovative feature of the project at the time of appraisal was the arrangement with four local banks to participate in funding the project. The consortium of local banks would finance the equivalent of US$938,000 or 21% of estimated project costs: US$530,000 would be for infrastructure and equipment of GLC, US$184,000 would be for loans to private farmers, and US$224,000 was reserved as a contingency allowance. At negotiations, it was envisaged that the (local) banks would provide finance by way of equity contributions of 0600,000 (US$520,000 equivalent at the time), the purchase of 0200,000 (US$173,000) of 8% debentures and overdraft facilities at 9% rate of interest./ The four participating local banks were: Ghana Commer- cial Bank; Barclays Bank of Ghana; Standard Bank of Ghana; and Agricultural Development Bank (PCR, para. 16). 34. The opportunity to have local banks participate arose because of Government policy which required commercial banks to invest part of their liquidity in agriculture. In this case, the Bank made good use of this opportunity by opening investment prospects for the commercial banks. That such investments can be profitable has been demonstrated by the Bank of Ghana's investment in the Shai Hills Ranch (PPAM, paras. 24-28). 35. The local banks' participation turned out to be lower than and different from appraisal estimates for two reasons: (i) the private farmer component of the project was not implemented, and it was in sub-loans to farmers that the commercial banks were expected to play a major role; and (ii) because of their typical risk aversion attitude, the commercial banks did not provide their contribution in the form of equity but only as debentures and overdraft facilities. 36. Although their financial contribution had therefore been reduced, the commercial banks remained interested in project implementation. They were all represented on GLC's Board, which played an important role in the com- pany's management (for further details see PCR, para. 66) although this was insufficient for proper implementation. Furthermore, the commercial banks have an important role to play in allowing GLC to weather the next four years, when cash flow problems will become acute (for further details, see PCR, paras. 75 and 58). 37. Experience with this project has demonstrated the importance of good management and its reasonable autonomy, availability of foreign exchange and inputs and location of the investments. Provided that such basic requirements are available, more local bank co-lending could be organized for commercially- oriented Bank assisted agricultural projects or project components. 1! .Appraisal of Livestock Development Project, Ghana, Report No. 277-GH, dated May 28, 1974, para. 4.05. - 12 - F. Disbursements, Suspension and Closing Disbursements and Suspension 38. At appraisal, it was envisaged (see PCR, para. 15) that the US$2.0 million credit would finance 53% of all project costs during the first four years of the six-year development periodP (this would be equivalent to 44% of total costs estimated for the six-year development period). The project completion date was expected to be December 31, 1980;2, thus, the project development period was expected to cover the six-year period 1975-80, while the credit was expected to cover costs in the four-year period 1975-78. Because of normal administrative delays, the disbursement schedule in the appraisal report (Annex 7) covered a four and a half-year period, and on the basis of this original plan (see also the Basic Data Sheet) disbursements should have been completed by mid-FY80, or December 31, 1979. However, the Closing Date in the legal documents was set at June 30, 1981-3/ rather than December 31, 1979.I Not a critical issue, but apparently either a lack of coordination or a misunderstanding occurred when the legal documents were drafted5/, or the legal guidelines which provide that the closing date be set six months after the completion date were simply applied automatically. However, this "mistake" enabled project financing to continue even with the implementation delays without going through the formalities of extending the Closing Date. 39. The Bank has recently instructed operational staff to use disburse- ment profiles as a basis for projecting disbursements during appraisal. Actual disbursements under the project are compared with these profiles in Annex 2. 1/ See also Appraisal Report, Appraisal of Livestock Development Project, Ghana, Report No. 277--GH, dated May 28, 1974, para. 4.04. 2/ See Schedule 2 of the Development Credit Agreement dated July 26, 1974. 3/ Credit Agreement dated July 26, 1974, Section 2.04. 4/ At first glance, the June 30, 1981 seems logical given the December 31, 1980 completion date. However, this is not so. As sometimes happens in other treecrop projects also, the Bank decided only to finance costs over the first four years of the project and therefore expected its disbursements to be completed well before all physical investments were completed. The closing date should, in such cases, be linked to the disbursement schedule, not tD the physical completion date. 5/ A more detailed discussion of the lack of coordination in preparing loan documents is contained in PPAR, Ethiopia, Lower Adiabo Development Project (Credit 516-ET); OED Report under preparation. - 13 - 40. Disbursements since mid-1980 have been slow as can be seen from the table below (see also Basic Data Sheet for disbursement details): Disbursements as of: June 30, 1980 US$1,339 thousand June 30, 1981 US$1,349 thousand December 31, 1981 US$1,355 thousand Thus, only about US$10,000 were disbursed in FY80 and only US$6,000 in FY82..- The reason for this was simple: as of June 1980, the Bank had not allowed any disbursements under the credit to be made except for improved water supplies.!V 41. During final meetings with the comyletion mission, GLC complained about the Bank's handling of disbursements.. GLC stated that it was only in January 1981 that they had been made aware that funds allocated for private farmer loans could be reallocated for GLC use. GLC also felt that by restrict- ing disbursements to water development, the Bank had also affected disburse- ments which had nothing to do with the water supply issue such as the pro- curement of a weighbridge.A/ 42. What was the actual situation? Inadequate water supplies remained the principal bottleneck to effective development of the Branam ranch. After the seriousness of this situation had been emphasized by various earlier missions, the January 1980 supervision mission recommended that disbursements on cattle purchases, ranch development, vehicle and machinery purchases, other than those required for water supply, be suspended.! A draft letter 1/ In CY1980 and 1981, the following disbursements were made: November 9, 1981, Australian Dollars 6,000; 10/28/80, Australian Dollars 8,430 (both payments to water supply consultants); March 3, 1980, CFAF 3.3 million to a vehicle suppliers in Ivory Coast; and March 24, 1980, CFAF 5.2 million to a transport enterprise in Senegal to cover freight costs. 2/ For further details, see also PCR, para. 27. 3/ Report on meeting between the Directors of Ghana Livestock Company Ltd. and the World Bank Supervision/Project Completion Mission, held on June 8, 1981 at the office of Ghana Livestock Company Ltd., Accra. 4/ In a letter to the supplier, dated August 1980, the Bank informed the supplier as follows: "I regret that we are unable to give you the answers you require regarding payment for the weighbridge as our dis- bursement against the purchase is dependent upon certain actions to be taken by GLC of which they are aware." 5/ In its Aide Memoire, the mission informed GLC that it would make such recommendation to the Bank. - 14 - was prepared in February 1980 ia which two failures (continued failure to provide water and the procurement of the wrong sort of cattle) would have led to the decision to suspend disbursements on all claims other than those associated with the provision of water supplies, dipping facilities and expatriate staff.1/ However, these letters were never sent when, in March 1980, another Bank mission reported that GLC had made arrangements with a consultant for study and implementation of water supply at Branam. Although there had been some positive action, the situation regarding water and stock on Branam remained unsatisfactory. GLC, being aware of the issue, had in the meantime not submitted any reimbursement requests. However, a letter was received from the Ministry of Finance in late May requesting reallocation of proceeds to allow GLC to purchase further cattle. The matter was then re- viewed in the Bank, the Legal Department consulted, and it was determined that, under the circumstances, a suspension was not needed. This would not, in fact, have been necessary as Category 1 (investment in infrastructure, equipment and breeding and incremental fattening stock for ranches) was already overdrawn, and IDA, therefore, had the right not to reimburse any further under this category. Furthermore, reallocation is at the discretion of IDA and, therefore, this would only be effected after sufficient water supplies had been assured. Government was so informed.2/ Government was also requested not to send any farther disbursement applications under Cate- gory I other than for items required for ranch water supply development..3/ Thus, and in line with the position taken, when the question of disbursements for the weighbridge was raised, IDA informed GLC that it regretfully could not disburse against this purchase as Category 1 was overdrawn, but that this position could be reviewed when adequate water supplies had been provided at Branam.4/ The above case has only been elaborated upon by the audit to illustrate that there are various ways by which IDA can use disbursements as a "remedy." Credit Closing and Cancellation 43. As early as March 19E1 the Bank had decided not to extend the Closing Date beyond June 30, 1931. That decision implied, already at that time, cancellation of the unused balance. A final disbursement was made on November 9, 1981, four months after the Closing Date. In fact, such disburse- ments after the Closing Date are not unusual. First, the Closing Date is the 1/ Please note that detailed guidelines regarding suspension of disburse- ments (OMS 3.31) were only issued in March 1980. 2/ Letter to Minister of Finance, dated July 24, 1980, and telex dated June 13, 1980. 3/ Telex to Ministry of Finance, dated June 13, 1980. 4/ Telex to GLC, dated July 22, 1980. - 15 - date "as of which the Association may, by notice to the Borrower, terminate the right of the Borrower to withdraw from the credit account."! Present informal internal guidelines stipulate that whenever an undisbursed amount remains in the loan/credit amount at the Closing Date and it is expected that only one or two withdrawals are still to be made in the near future (e.g., three months) such withdrawals may be made without a formal extension of the Closing Date.. 44. By mid-January 1982, however, the remaining credit balance of about US$650,000 had still not been cancelled nor had the Borrower been formally notified of the intention to cancel. Not a critical issue, but simply weak administration by the Bank. It was not that the Bank did not wish to cancel the Credit balance,3/ it simply failed to do so. A lesson which emerges from this experience is that the Bank should more closely monitor and follow up on loan/credit closing. Decision-Making Around Project Closing 45. Decision-making regarding project closing, especially for problem projects, is difficult, and there are no fast and easy ground rules which can be applied uniformly. The decision-making regarding the closing of this project provides some insights. 46. Project closing is a complex process, and this is demonstrated in the various approaches chosen in other Ghanaian projects. (i) In the case of the Ghana Fisheries Project, protracted discussions were allowed to take place to reach agreement on the final alloca- tion of the credit funds not committed. By 1975, about US$600,000 of the US$1.3 millioncredit had remained uncommitted, but realloca- 1/ General conditions applicable to Development Credit Agreement, Article 2.01(12). 2/ For a more detailed discussion of issues related to disbursements after the closing date, see PPAR, Ethiopia Addis Ababa Dairy Development Project (Credit 269-ET), OED audit report under preparation. 3/ On May 11, 1981, the Loan Department informed the Programs Department about the status of funds and the upcoming closing date; a reminder was sent on June 26, 1981. On July 20, 1981, the Projects Department re- quested the Programs Department to close the credit. On November 19 and 25, 1981, the Loan Department informed the Programs Department again of the status of funds. On December 8, 1981, the loan officer informed the lawyer, disbursement officer and projects officer that the remaining balance could be cancelled. -- 16 - tion was only agreed upon by end of 1977, five years after the original Closing Date, and the credit was closed, fully disbursed, by end 1978.1/ (ii) In the case of the Ghana Eastern Region Cocoa Project/ the project was extended from five to nine years in order to achieve planting targets. In the end, the credit was fully disbursed in mid- 1980, four and one-half years after the original Closing Date of December 31, 1975. (iii) The project period was not extended in the case of the Ghana Sugar Rehabilitation Project.z/ There were major implementation prob- lems, and the credit was closed as scheduled (March 31, 1979) with the project termed a failure. Due to administrative bottlenecks, the final disbursement was only made in mid-1980, and the balance (US$1.1 million) cancelled. 47. The major issue is really at what time a project becomes such a problem that it should no longer be supported by the Bank. 48. As early as 1979, the Bank was considering closing the project. At that point in time, however, there was some revived hope that GLC might succeed in turning the project around, and there was also a change of Govern- ment that year. With the benefi' of hindsight it might well have been ap- propriate to close the project at that time. 49. The February 1981 supervision mission concluded that the project was seriously mismanaged, that Drospects for dam construction before the departure of the water consultant were bleak and that the inspection of the Branam herd gave an appalling impression: "The evidence from this mission indicates that there is no point in continuing further disbursements except to finish off existing commitments for water development. We should, therefore, no longer consider the possibility of extending the project from June 30, 1981 to December 31, 1981 to complete the water supplies and purchase the remainder of the breeding stock. An extension would have been necessary to give GLC time to purchase the cattle (in 3 batches) and 1/ See Project Performance Audit Report, Ghana Fisheries Project (Credit 163-GH), OED Report No. 2794, dated December 28, 1979. 2/ See PPAR, Ghana Eastern Region Cocoa Project (Credit 205-GH), OED Report No. 3526, dated June 24, 1981. 3/ See PPAR, Ghana Sugar Rehabilitation Project (Credit 354-GH), OED Report No. 3525, dated. June 24, 1981. - 17 - move them to the ranch after the dams had been completed and filled by June 1981. We asked Government to consider this possibility subject to adequate water supplies and management in our letter of December 19, 1980, which would have involved reallocation of funds between categories. This possibility emerged after it seemed that GLC was about to take action since they had gotten the heavy machin- ery on the ranch and had promised to provide another manager in Branam. The good faith in which we have continued to support the necessary water development and stocking of the ranch in the hope that the faint gleam of management improvement we thought we saw in 1978/79 and water supply efforts in 1980 would continue, is no longer tenable in the present circumstances of GLC or Ghana's present economic malaise. To utilize the necessary water supplies properly and to prevent the loss of further new breeding stock requires a level of management which clearly cannot be provided in the short timespan of one year. Apart from the fact that the project is now over eight years old, the main reason for not extending the Closing Date can be summarized as follows: GLC's unwillingness to accept outside management advice, which had led to incompetent management and the present poor state of Branam ranch; inadequate water supply, insufficient fodder resources, and a low level of herd management, leading to high calf mortality, cattle in poor condition,... declining herd numbers and poor handling of the feeder steers operation. Also, most of the recommendations made by previous supervision missions have generally not been followed."17 The Bank accepted the mission's recommendations and informed Government2/ that it would not be justified to extend the Closing Date of the credit beyond June 1981. 50. As mentioned above, following the November 1980 supervision a few months earlier, the Bank had recommended reallocation of Credit funds provided certain conditions were met. "There is now the question of the best use of the remaining IDA funds. These funds will be far too short to fully stock the ranches but sufficient to achieve the breeding cattle targets set out in the 1976 revised development program. The remnants of the credit US$650,967 could be used for water and pasture development US$40,000, purchase of breeding cattle US$445,500, and small equip- ment and building materials all under Category 1 of the Credit Agreement; for management consultancy US$105,000 and training 1/ February 1981 supervision mission, Back-to-Office Report, dated February 24, 1981. 2/ By telex, dated March 10, 1981. - 18 - US$6,000. The subcomponent for provision of credit to 20 farmers could be dropped becaus& at this stage, it is too late to reinstate this component. The funds saved (US$260,000) could be used to complete the development. We would propose to make the necessary reallocation to the credit accordingly, but we would not like for you to commit funds for purchase of breeding cattle until we are both satisfied that there is sufficient water at Branam to take the extra stock."I/ 51. Following the completion mission, the Bank reported its findings in a letter to GLC.1 In it, the Bank inter alia remarks: "The mission has reported that two dams are now full at Branam and that a further two can be confidently expected to fill with the coming rains. You are now, there- fore, in a position where additional cattle can be brought into the ranch without fear of malnutrition leading to uneconomic performance. In projecting future company cash flow, we have assumed that you will be able to obtain a line of foreign exchange for the purchase of 500 head of breeding stock and a continuous injection of 1,000 feedersteers over the next nine years." In the audit's view, the Bank had two options at that time. 52. The first option was to extend the Closing Date for the following reasons: (i) there was an undisbursed balance available in the credit account; (ii) GLC was in need of foreign exchange for the purchase of breeding stock and feedersteers; and (iii) foreign exchange is a scarce resource in Ghana and GLC had had earlier problems in .obtaining foreign exchange allocations (see also PCR, para. 44). 53. The alternative was to close the project for the reasons spelled out above (PPAM para. 49). The Bank chose to close the project. In doing so, the Bank basically judged that there was no hope that the project and GLC could or would be turned around. Therefore the Bank, by implication, dis- qualifies its own analysis of the financial viability of GLC (PPAM, para. 9) and does not expect GLC to achieve a financial rate of return of 16% because it did not think that GLC would achieve the technical coefficients or manage- ment necessary. 1/ Letter to GLC, dated December 19, 1980, received by GLC in January 1981. 2/ Letter to GLC dated July 9, 1981. - 19 - G. Conclusionsl! 54. The failure of this project is not incidental: most projects in Ghana have suffered, or still experience, major problems. The Fisheries Project (Cr. 163-GH), the first Ghana agriculture project audited by OED/, was listed a problem project for most of the implementation period (although the problems were mainly caused by wrong project design). The Eastern Region Cocoa Project suffered from poor pricing policy and poor institutional perfor- mance and achieved acceptable economic returns only because of higher than expected economic cocoa prices.1 The Sugar Rehabilitation ProjectL ended in failure, implementation having suffered from a multitude of problems including political and economic problems, erratic rainfall and overcast weather, borer infestation, frequent breakdowns of equipment, labor problems, cost escalation, inadequate electricity and water supplies and particularly 1/ OPS staff notes that "PPARs have now been prepared on a number of projects of this type (ranching by parastatal corporations) e.g., in Madagascar (Beef Cattle Development Project - Loan 585-MAG, PPAR, OED Report No. 1559 dated April 11, 1977), Tanzania (Beef Ranching Develop- ment Project - Credit 132-TA, PPAR, OED Report No. 994 dated January 28, 1976), Zambia (Livestock Development Project - Loan 627-ZA, PPAR, OED Report No. 1362 dated November 24, 1976) and Indonesia (Beef Cattle Development Project - Credit 355-IND, PPAR under preparation), and they have all recorded very poor implementation performance. This is true also for a number of other similar projects, for which PPARs have not yet been prepared. Problems common to a number of these projects are delays in acquisition of land, in carrying out on-ranch investments such as water facilities, fencing, etc., and in acquisition of livestock; inabil- ity to recruit and/or retain good managerial and technical staff; political interference in marketing; serious over-staffing; serious shortfalls in productivity coefficients vis-a-vis appraisal estimates; and in some cases, stock losses by theft." 2/ PPAR, Ghana Fisheries Project (Credit 163-GH), OED Report No. 2794 dated December 28, 1979. 3/ PPAR, Ghana Eastern Region Cocoa Project (Credit 205-GH), OED Report No. 3526, dated June 24, 1981. 4/ PPAR, Ghana Sugar Rehabilitation Project (Credit 354-GH), OED Report No. 3525, dated June 24, 1981. - 20 - lack of foreign exchange. Of four other completed projects in the public utilities sector, three achieved marginal (financial) returns and only one was successful.i/ 55. Experience from this and other projects also highlights the limits the Bank faces when confronted with counter-productive Government policies. This raises a wider question on how effective projects - and Credit Agreement Covenants - can be when used for leverage in solving sector and macro-economic problems. There seems to be vicious circle: on the one hand, it is extremely difficult to resolve economic and sector issues through projects; on the other hand, it seems impossible to implement projects when the economic and/or sector environment is unsound. Thus, a further observation needs to be made. If the various projects in an operational program are not to fail, then matters external to a project, bat essential to project outcome, need to be addressed, not just by the project, but also by the country program dialogue. 1/ The presently estimated incremental financial rate of return of the Second Volta River Authority Power Project (Loan 618-GH) was about 20% as compared to 26% in the appraisal (PPAR, OED Report No. 1363, dated November 23, 1976). The actual rate of return on the combined invest- ments for the First and Second Electricity Company of Ghana power proj- ects (Credits 118-GH and 256-GH) was 9% compared to the 10% and 13% respectively, estimated at appraisal. The 9% incremental financial rate of return on the water part of the Accra/Tema Water Supply and Sewerage Project (Credit 160-GH) was substantially lower than the 25% envisaged at appraisal (PPAR, OED Repo:t No. 1683, dated July 29, 1977). - 21 - Annex 1 Page 2 of 2 BANK OPERATIONS IN AGRICULTURAL SECTOR IN GHANA Fisheries Project Cr. 163-GH, US$1.3 million A Sep. 23, 1969 Appraisal Report PA-7a, 09/08/69 fully disbursed S Sep. 25, 1969 President's Report P-74u, G9/11/69 E Jan. 21, 1970 PPAR, OED Report No. 2794, 12/26/79 CL June 23, 1978 The credit provided funds for building 40 fishing vessels and spare parts, providing management staff for the Boatyard Division and technical assistance to the Agricultural Development Bank, and for studies of one or more new fishing ports. Project design was sub- stantially changed during implementation. As a result of price increases and technical design changes, only 10 vessels were built, but their engines proved underpowered for trawling. Funds for management staff and technical assistance were not utilized, and only the first stage of the fishing port study was financed under the project. After protracted discussions, the remaining credit proceeds were utilized in purchasing additional spare parts and equipment for the Boatyard Division. The nine project vessels (one sank) are landing only 1,400 tons of fish per year (only 2% of the increase in fish landings in Ghana since the project started). The economic rate of return was reestimated at 10% (well below the appraisal estimate of 50%). Eastern Region Cocoa Project Cr. 205-GH, US$8.5 million A = June 23, 1970 Appraisal Report PA-43a, 06/03/70 fully disbursed S June 26, 1970 President's Report P-845, 66/11/70 E = Mar. 12, 1971 PPAR, OED Report No. 3526, 06/24/81 CL = Dec. 31, 1979 The project was to help Ghana rehabilitate its cocoa industry which had declined in the late sixties as a result of unsound Government policies concerning cocoa taxation and pricing. The project was to support cocoa rehabilitation (up to 51,000 acres) and replanting (up to 36,000 acres) through farm credit, farmer training, and project administration. Achievements under the project were disappointing. Most of the area targets were met, but cocoa yields are expected to remain substantially below expectations. The economic rate of return has been reestimated at 12% compared to 26% as estimated at appraisal. Institutional results were worse than expected and the project experience shows how difficult it is to implement projects when major sector issues remain unresolved. Ghana's cocoa production meanwhile declined from a record 566,000 metric tons in 1965 to 416,000 metric tons in 1970 and 249,000 metric tons in 1979. Sugar Rehabilitation Project Cr. 354-Gh, US$15.6 million A = Dec. 21, 1972 Appraisal Report PA-1OUa, 11/15/72 fully disbursed US$14.5 S = Jan. 29, 1973 President's Report P-1159-Gh, 12/0b/72 cancelled (05/20/80) LS$1.1 million E = May 31, 1973 PPAR, OED Report 3525, 06/24/81 CL = Mar. 31, 1979 The project aimed at increasing Ghana's sugar production and making the industry efficient, thus reducing imports and eliminating the heavy losses incurred. Specifically the project provided for rehabilitating two sugar estates: replanting (16,050 acres) and expanding (7,425 acres) estate and farmers' cane fields, rehabilitating factories, and improving support services. The project ended in failure. Implementation suffered from a multitude of problems, including political and economic problems, erratic rainfall and overcast weather, borer infestation, frequent breakdowns of equipment, labor problems, cost escalation, inadequate electricity and water supplies and, particularly, lack of foreign exchange allocations. Sugar production in 1978, the last project year, reached a low of 18% of the appraisal target for that year. The project's economic rate of return is negative. Livestock Development Project Cr. 500-Gh, US$2.0 million A = Jun. 18, 1974 Appraisal Report 277-GH, 05/28/74 fully disbursed US$1.4 million S = Jul. 26, 1974 President's Report P-1390-Gh, 06/06/74 cancelled (01/21/82) US$0.6 million E = May. 14, 1975 PPAR, OED Report under preparation CL = Jun. 30, 1981 The project was designed to create the beginnings of a modern commercial cattle industry in Ghana through the development of three, later revised to two ranches, and by assisting smaller privately owned herds (50 later revised to 20) surrounding the ranches. Physical development was much delayed at the ranches. The private herd component was not pursued and a follow-up project was not prepared. The project is a qualified failure; a failure because of its negative rate of return, qualified because a livestock company has been established which could become a viable enterprise. Oil Palm Development Project Cr. 531-GE, USS13.6 million A = Feb. 18, 1975 Appraisal Report 173a-GH, 02/03/75 disbursed (09/30/81) US$8.8 million S = Mar. 05, 1975 President's Report P-1431-Gh, 02/06/75 E = June 24, 1976 CL = Dec. 31, 1983 The first phase of integrated oil palm development in the Eastern Region of Ghana comprising: establishment of a 4,000 ha nucleus plantation; 1,200 ha of outgrower/smallholder plantations; construction and operation of a 10 ton ffb/hr palm oil mill; training; field trials; some road construction and technical assistance. The project is making satisfactorily progress and planting targets are likely to be met, albeit with some delay. Training, so far, has been unsatisfactory. However, serious issues arose about mill procurement. After considerable delay. Training of higher level Ghanaian staff to take over project management has so far been unsatis- factory. After considerable delay over mill contract award and financing, construction is now proceeding satisfactorily. Issues have arisen about appointment of the Board Chairman and land acquisition to complete the planting program. Land has now been made available by Government, but a Board Chairman has still not been appointed. Lessons learned from the project have been incorporated in Phase Il (appraised in April 1981). - 22 - Annex 1 Page 2 of 2 BANK OPERATIONS IN AGRICULTURAL SECTOR IN GHANA Ashanti Region Cocoa Project Ln. 1181-GH, US$14.0 million A = Dec. 23, 1975 Appraisal Report 827a-Gh, 11/18/75 disbursed (09/30/81) US$10.7 million S = Dec. 23, 1975 President's Report P-1687-GH, 12/08/75 E = Fcb. 23, 1976 CL = Dec. 31, 1982 The project as appraised included replanting and maintenance during the project period of 42,500 acres of cocoa with high yielding varieties; providing credit in cash and kind to farmers; farmer and staff training; equipment for feederroad improvement; project evaluation through a field survey; establishment of a project administration; and taking various measures and implementing certain studies to put the industry as a whole on sounder footing. At appraisal it was envisaged that planting targets would be met in three planting seasons. By 1973 only 51% of the planting target was achieved and the loan closing date was extended by 3 years. .o date, the project has planted 38,500 acres or 91% of appraisal target. The loan Closing Date has been extended to December 1982 mainly ior sector strategy reasons, but planting targets can be expected to be met by that date. The major problem has been the lack of interest on the part of the farmers in doing any of the work themselves. Other significant problems have been migration of large numbers of project labor and junior staff; shortages of almost everyi:hing have hampered staff mobility and thus project implementation; feederroad -aintenance continues to be unsatisfactory. pper Region Agricultural Development Project Third Window Loan 1291-Gh, US$21.00 million A = June 22, 1976 Appraisal Report 106la-Gh, 06/03/76 disbursed (09/30/81) US$13.80 million S = June 28, 1976 President's Report P-1868-Gh, 06/10/76 E = Apr. 11, 1977 CL = Dec. 31, 1982 A five year project to increase agricultural production and establish permanent farm support services; to expand and improve training and extension, animal health and husbandry; establish 90 service centers; rehabilitate or construct 220 small dams; provide farm inputs; improve grain storage; establish 10 ranches and 7 research facilities; expand broadcasting; establish functional literacy schemes; and improve nutrition. Project implementation has been problematic for a very long time, but there have been some improve- ments recently although the project still faces substantial problems. Volta Region Agricultural Development Project Cr. 1009-GH, US$29.5 million A = A>r. 22, 1980 Appraisal Report 2527-GH, 03/20/80 disbursed (09/30/81) US$0.2 million S = Aug. 27, 1980 President's Report P-2762-Gh, 04/02/80 E = Mar. 31, 1981 CL = Dsc. 31, 1985 The project aims at: increasing production of major food:rops through provision of farm inputs, credit facilities and extension; con- structing and improving houses, buildings, farm access ani feeder roads, and boreholes for water supply, establishing pilot irrigation schemes and collecting hydrological data; providing institutional support including strengthening of the regional MOA office, training, nutrition extension and applied research faci;ities, and 3rovision of consultancy services for extension, monitoring, evaluation, etc. The project has only recently become effective and progress has since been made. - 23 - ANNEX 2 PROJECT PERFORMANCE AUDIT REPORT GHANA LIVESTOCK DEVELOPMENT PROJECT (CREDIT 500-GH) HERD COEFFICIENTS The Ghana Livestock Company Ltd. (GLC) provided the audit mission (but see also attached GLC comments, paras 20 and 25) with herd coefficients substantially different than those presented in the PCR (paras. 31 and 37). Further review revealed, however, that the differences between these coeffi- cients are based on differing definitions of their content and time period. A. Herd coefficients prepared by GLC Branam Ranch June 1979 June 1980 June 1981 Adult Mortality (%) Ia 6.1 3.7 4.8 Calf Mortality (%) 19.6 12.1 33.8 /c Tadzewu Ranch Adult Mortality (%) /a 3.1 3.8 4.4 Calf Mortality (%) /b 5.1 3.6 12.6 /a Mortality, not including losses or thefts, as percentage of average monthly adults for that year. /b Mortality as percentage of calves held for that year (not as percentage of only the calves born for that year). /c High calf mortality that year, primarily attributable to poor bull management, with many calves born in unfavorable seasons. B. Herd coefficients in PCR Branam Ranch (see PCR, para. 31) 1979L.a 1980 1981 Revised Estimates Adult Mortality (%) /c 19.6 12.1 - /b 4% falling to 37 . Calf Mortality (%) Id 45.9 43.3 - /b 10% falling to 6.Le- Tadzewu Ranch (see PCR, para. 37) 1978/79. 1979/80 1980/81 Revised Estimates Adult Mortality (%) /1 3.1 3.8 4.4 3% Calf Mortality (%) Lh 5.1 3.6 12.6 10% falling to 6%.e /a Calendar years. /b Technical coefficients for 1981 not available. /c Mortality, including losses or thefts, as percent average monthly adults for that year. /d Calf mortality as percent of calves born that year. /e Falling to that percentage by maturity. /f Years are May-June, except 1980/81 which is May-April. /j Calculated as adult mortality and losses as a percent of total average monthly herd less calves. /h Calculated on calf mortality as percent of calves born. - 24 - ANNEX 3 PROJECT PERFORMANCE AUDIT REPORT GHANA LIVESTOCK DEVELOPMENT PROJECT (CREDIT 500-GH) Estimated and Actual Disbursements and Disbursements Profiles (Cumulative percentages) Profiles Year Appraisal Revised Actual Subsector.Lb SubsectorkL Sector-d Sectore PY 1 27.5 3.0 3.0 1.2 2.9 3.7 3.2 PY 2 47.5 30.0 25.2 8.3 13.3 16.8 13.6 PY 3 70.0 60.0 43.8 20.9 29.4 35.9 29.5 PY 4 90.0 95.0 53.6 36.1 48.8 57.1 48.5 PY 5 100.0 100.0 66.9 52.4 67.7 75.9 67.1 PY 6 - - 67.5 68.4 83.1 89.4 82.5 PY 7 - - 67.&-a 82.3 93.7 97.5 93.4 PY 8 - - - 93.0 100.0 100.0 100.0 PY 9 - - - 100.0 - - - Py 10 - - - - - - - /a Remainder cancelled. /b East Africa and West Africa livestock subsector disbursement profile. /c Livestock Bankwide disbursement profile. /d West Africa Region, agricultural sector disbursement profile. /e Bankwide agricultural sector disbursement profile. - 25 - ATTACHMENT GHANA LIVESTOCK COMPANY LIMITED RANCHES AT: BRANAM (B.A.) TADZEWU (V.R.) Telephone 29704 P. 0. Box 4608 Accra, Your Ref. Ghana. Our Ref. GLC/VOL/VII Cables: Savannabeef. 20th May, 1982. Mr. Shiv S. Kapur Director Operations Evaluation Department The World Bank 1818 H Street N. W Washington D. C. 20433 U. S. A. Dear Mr. Kapur, REs PROJECT PERFORMANCE AUDIT REPORT ON GHANA LIVESTOCK DEVELOPMENT PROJECT (CREDIT 500-GH) I refer to the draft Audit and the Project Completion Reports despatched to us for our study and comments. We have studied both reports quite comprehensivelly and pre- sent our comments and facts on areas of the reports especially PCR which either conflict with our observations and decisions which were mutually arrived at or conclusions which are at variance with our own observation. In as much as we would have liked to have made comparison and to bring out some salient features of the Shai Hills Ranch opera- tions vis-a via the Ghana Livestock Company, we feel that, we do not have sufficient and in depth details to make a more meaning- ful contribution here. Yours sincerely, FOR: GHANA LI7STOCK COMPANY LIMITED P. A. Ocran CHAIRMAN Attachment Directors-J. K Agyekumhene, D. 0. Andah, A. K. Aryee (Managing), P. A. Ocran (Chairman) T. 0. Sunkwa-Mills, Aa uff h. 4 - F< - 26 - GHANA LIVESTOCK DEVELOPMENT PfCJECT (Cr. 500-GH) GLC - PROJECT IMPLEENTATION (COMMENTS) 1. Effectiveness and Etart ups Although the Company came into formal existence in June, 1974, the credit did not became effective until the appointment of the first Managing Director in May, 1975. By the time the project became effective only 3 out of the 4 conditions stated within the Credit Agreement had been fulfilled leaving one of the most crucial aspects of the start-up arrangements (OED para 4 and PCR para 19). 2. Staffing: The GLC has been saddled with staffing pro- blems since its very inception. Initial efforts at attracting suitable personnel for the positions of Managing Director and Ranch Managers proved difficult as potential candidates pre- sented to the Company through the World Bank's Agricultural % prn1/ Projects Management Unit #GW either failed to meet the GLC's requirements or declined such appointments altogether after selection. 3. By January, 1975 both the Board and the Bank's super- vision Missions had recognised that although funds had been provided for under the project revision, adequate provision had not been .nade.in the World Bank Appraisal Report to cover salaries and other benefi-s of the Managing Director and the 3 Ranch Managers. (of OED para 4) 4. The first expatriate 1lanaging Director was among a group of potential candidates (7) whose curriculum vitae had been despatched to the GLC from the AM1. Although he had a personal disability, requisite information so far as his competence, referees and recommendatiDns from an earlier interview held in London for the World Bank indicated that such disability as he had, had not diminished his working capacity. In October, 1975, a Supervision Mission's observation that the Managing Director was unsuitable for the position and should be replaced as early as possible confirmed the Board's previous misgivings about his overall competence. Consequently, his services were terminated in January, 1976. 1/ Explanation added by audit. Throughout this letter GLC refers to AMPU as the main actor for the'orld Bank. Although the reference to the Agricultural Projects Management Unit is generally correct for staffing and recruitment, the reference is incorrect in other cases, when the reference should rather have been to the Bank or sometimes to supervision missions. ..../2 - 27 - Page 2 A tempor replacement (consultant) was appointed through the in November, 1975. He served until his departure in July, 1976. In April, 1976 an expatriate Ranch Manager also proposed and recommended through the AMP joined the GLC and served until March, 1977 when he resigned over disagreements on Ranch dev- lopmeht, cattle breeding, wage rates, mechanisation generally marketing and finally transfer from Pong Tamale to Branas. (PCR para 68) In February, 1977 the expatriate Ranch Manager stationed at Pong Tamale was moved down to the Branam Ranch to initiate development of that Ranch as Pong Tamale had been deleted from the Company's operations. After about a week's stay, he retu- rned to Pong Tamale and tendered his resignation. The sudden- eas of the action taken confirmed observations that he may haveresigned more because of the conditions and the enormity of the task ahead rather than the reasons advanced. (PCR para 68). 5. As a result of the sudden resignation of the expatriate manager in April, 1977, fresh efforts had to be made to reoruit replacements. Ploement of advertisements and interviews did not yield and fruitful results until the second interview around September, 1977 when another Ghanaian Ranch Manager was selected. The latter performed very badly and was inter- dicted in December 1979 for being unable to account for some 52 heads of cattle which had been reported missing. He was subsequently dismissed from the services of the Company since the police were unable to prefer any theft charges against him (PCR para 32). After fruitless effort and search, a Senior Ranch Manager was finally selected for the Branam Ranch. He finally assumed his post at the begining of March, 1981 and is still at post. 6. Land acquisition and Assets taken over: Although adequate arrangements for the transfer of pro- perty and assets to the GLC had been one of the pro-requisites demanded for effectiveness, this condition had not been met by May 1975. - 28 - 7. IgUJL Although the land areas earmarked for the Reaches had been identifitd earlier, no positive efforts had been made to proee*d with the administrative processes involved in the actual governmental acquisition between project identification (1969) and start-up (1975). It had wrongly been assuied that such arrangements for land titles and other legal matters would be completed within 3 - 6 months of start-up. (PCR par& 41). 8. Tadseu Ranchs An area (indeterminable) but often stated at 1,955 acres were available for utilisation al- though the State Farms (SPC) had no legality to that land. After frequent alterations to the size of the area to be aequl- ired (1975 and 1976), a final draft plan was submitted in Sept- ember, 1976 for the actual acquisition. The Rights of entry to the area finally acceeded to by the Site Advisory Committee- at 6,920 acres in March, 1978 was signed by the Regional Comm- issioner in December, 1978 and the lands formally acquired in the name of the government in February, 1980. The GLC did not make any pvivate arrangements with any landlords for grasing rights outside the original available lands. The grazing cattle, however, had access to a private dam on the western border of the demarcated area for which it sought permisqion to water cattle grazing around that area. (PCR paras 35 and 41) 9. Branam Ranch: The State Farms Corporation which was operating on some parts of the land did not have any legality to the land which was being used. However, it had a private arrangement with the Braaam stool for utilisation of those areas under consideration. In November 1975, the GLO opted to establish a new Ranch altor gether in view of the high value placed on infrastructural facilities mostly buildings on that property. Whilst the app- raisal report had put the value of buildings at 086,000.00 in 1973, the actual valuation was put at 01,205,400.00 in 1975. As a result of this decision, new demarcated area of 40,000 acres was presented for processing in May, 1976. Rights of entry to the area acceeded to by the Site Selection Committee was signed in June, 1977 and final acquisition by government was made in June, 1979. The land area demarcated for the operations of the Branam Raneh do not have any permanent settlements within its body (confirmed by valuation report of February, 1980). ..../4 ~-29 ~ Tenant farmers operating within the area live in the outlying village@, (3ranam, Ayorya, Tingak:-oz *to) which are outside the ranch boundaries. These do have temporary farm huts utilised only during the farming season. After initial attempts to nego- tiate directly with some of the chiefo within the area, the Boad directed that the line of approach be discontinued so as to allow the government to take up the issue. Thus, by May, 1976 the GLC had stopped any direct dealings with the landowners. Further dealings with regional authorities involved appeals to them to hasten the tempo of land acquisition processes, payment of com- pensation and counsel against tresspass, destruction of tence lines, burning of grassland for hunting purposes especially during these preparatory phases. (PCR para 28) 10. Assets taken over: The GLC decided that assets to be taken over from the SFC and the Animal Husbandry Department (AHD) should await the arrival of the Managing Director who would determine the assets to retain and those to be returned to the original owners. At start-up in May, 1975, the conditions had changed dramatically. Liason work between the government set-up machinery on the one hand and the SFC and the AHD on the other hand seems to have kept at a very low level with the result that in the case of the SPC there was misunderstanding and apprehen- sion at the mode of valuation and the assets to be taken over. 11. Tadzewu: The appraisal report indicated a take-over of a total of 1,790 heads. j3y June, 1975, the SFC had moved out and or sold very large numbers of cattle with the result that only 331 heads were finally transferred tG the GLC. This deple- tion forms by far the biggest shortfall (1,459) in the actual numbers of cattle finally taken over by the GLC. 12. Branam: Buildings formed by far the biggest asset on that property. Although the appraisal report put the total at a value of 086,000.00 (1973), later valuation in 1975 by a firm of valuers put the actual value at 01,205,400.00. In view of the value put on the buildings and its effects on the overall capital structure, the GLC opted to develop a new ranch alto- gether. Later estimations show, however, that the buildings had been grossly underestimated for within the appraisal report. ...45 - 30 - 13. Pong Tamale: Cattle.: At appraival it ws estimated that a total of 1,086 heads would be transferred to the GLC. The AHD indicated in November, 1975 that it vould hand over all 763 heads counted and listed to GLC and that the shortfall of 323 was due to out- break of disease on that property. The GLC insisted on the need to have all cattle tested e.nd to reject all reactants. Finally, some 660 heads were actually transferreC to the GLC bringing the actual shortfall to 426 heads. (POR para 200) 14. Land: The Pong Tamale lands were in fact part of the Northern State Lands which had been vested in the State. The private rice farmers who were operating in the area had no legal access to these lands and in actual fact could be regarded as tresspassers. From the point of view of acquisition, the site was easier to acquire than any of the other two which were either Stool lands (Branam) or a uixture of Stool and private lands (Tadzewu). Due to the various problems encountered the project had to be revised culminating in the reappraisal report of May, 1976. Physical Development 15. Branam Ranch: Formal approval of the actual area designw ated was given by the Site Advisory Committee in December, 1976. Prior to this period, therefore, placement of structures which had been formulated earlier (if any) had no bearings on the actual. Design plans and drawings were supposed to have been discussed in detail with the Senior Ranch Manager at Pong Tamale who was due to have moved down to Branam to implement the actual construction. 16. Tracts (water and roads), receiving yards and dip without adequate design specifications and placements which were under construction by inexperienced foremen and designed by the consult- ant managing director had to be discontinued whilst fresh ones were formulated for implementation. These included dips, handling yards and collecting pens, roads, firebreaks and fencing. Actual construction after the discontinuation excercise started after the arrival of the first Ghanaian manager in December, 1977 and not in 1976 as previously stated. (PCR para 29) [1oad construction started in April, 1978 and was completed in March, 1979. The cost u? this elenent cu-d not have been paid for by the Company had it not been for a government grant of 01.0 million pro- vided under the 1978/79 financial budget estimates. .... /6 - 31 - Page 6 One dam holding approximately 800,000 gallons was constructed in August, 1978 at #25,900.00 whilst a second dam with a capa- city of 2.5 million gallons was constructed in July, 1980 at 044,800.00. Two of the six additional dams designed and marked in February, 1981 with an average holding capacity of 2.5 to 3 million gallons were completed at the end of June, 1981 at 050,000.00 each. (PCR para 28) 17. In November, 1975, the GLC was duly informed by the con- sultant Managing Director that the World Bank was sponsoring a water consultant to investigate all aspects of water supply and irrigation prospects to GLC's ranches. Without waiting for the arrival of the water consultant, a piped and pumped water system at a base cost of 0120,000.00 was substituted instead of the original surface dam system at appraisal by the consultant Man- aging Director. Tenders for the supply of piping and pumps based on inadequate data and outdated survey maps were placed with supplie-sin March, 1976 for delivery. The pipes at a cost of DM69,693 which arrived in 1977 are of little value in so far as stock watering is concerned. (PCR para 28) 18. Variations in the actual cost of development as against estimated development costs have arisen mainly because of the level of underestimation of the various cost components and to a lesser extent on rise in inflation generally. The extent of underestimation in the cost of provision of infrastructural facilities is illustrated in the case of roads. Whilst at app- raisal (1973) costs were estimated at 0500.00 per mile, at re- vision in 1976 the cost was stated at 0360.00 per mile, (PCR paras 28 and 29) 19. Since 1978, the Ranch has been cultivating between 100 - 150 acres of maize. The cultivation of maize on the Ranch was initially undertaken as an excercise in conjunction with pasture development. Yields have been low not because of poor cultural practices (except in 1978/79) but primarily due to unavailability of fertilizers and poor weather conditions especially in 1979 and 1980 reflecting in the overall price of maize in the country. Production costs have increased significantly since 1975 in line with general increased labour charges, costs of fuel and other charges. It in ;herefore, wrc, ascribe the profitability on maize production simply in terms of high maize price. ..../7 - 32 - Page 7 If the singular crit rion of hlj-h rice is used as the basis of determining the value of an operation, then the inclusion of feeder steer operation based on an over-valued cedi for import- ation and resale on local market at black market prices becomes even more questionable. (PCR paras 20a and 30) Attempts to establish pasture have failed mainly because of in- ability to procure seeds through importation. In both 1977 and 1981 ploughed fields had to be left unplanted because of unavai- lable pasture seeds whilst the 1978 plantings from locally pro- cured seeds failed because of poor germination. 20. Movement of cattle to the Ranch from Pong Tamale started in October, 1978 and was ccmpleted in February, 1979. Altogether a total number of 1,542 heads including 333 steers/bulls and a some cull cows with calves at foot were moved out. Since these animals were introduced on the ranch no further heifer intro- ductions have been added. In July, 1980 a consignment of 500 feeder cattle were purchased for the unit. The productivity coefficients of the herd since its transfer from the AHD at Pont Tamale and their subsequent transfer to Branam is @s recorded below: (PCR para 31)1/ Pong Tamale (PT) PT/Branam Branam June, 1978 1979 1980 1981 Calving %1 74.7 59.7 54.3 68.7 Calf Mortality Y62 13.1 16.5 12 33.8 Adult Mortality %3 5.1 6.3 3,7 4.8 1. Calving percentage as a percentage of cows at the end of calving year 2. Calf Mortality based on percentage of calves held for that year 3. Adult Mortality as percentage of average monthly for the year excluding theft 21. Management has beer one of the most problematic areas in the development of the Ranch. The first manager an expa- triate recruited through the AMPU resigned after a week's stay on the Ranch in April, 1977 (para 4). The abruptness of his resignation notice caused the GLC to protest strongly to the AMPU on personnel seconded to the GLC. 1/ Added by audit. For herd coefficients, see PPAM, Annex 2. - 33 - Page 8 In so far as the quality of his work is concerned, confidential information from the Managing consultant to the AMPU and the AMPU's observations and remarks on his performance in January, 1977 give an impression quite at varidnee with the PCR (PCR pa.ra 68). As a result of previous frustrations in the recruit. ment of managers, the GLC had to make more strenuous efforts to ensure filling of the position as early as possible. After an interview involving 4 such potential candidates the first Ghana- ian manager was appointed. His curriculum vitae was despatched to the Ministry of Finance and Economic Planning as the Bank had requested. Since there was no adverse reaction to the can- didate presented, the GLC employed him in December, 1977. The first serious limitation to his ability was observed in January, 1979 on the maize cropping operations. In December, 1979 he was interdicted for not being able to account for 52 heads of cattle. His appointment with the GLC was terminated in March, 1981 for mismanagement. After the interdiction of the Ghanaian manager in Desember, 1979 it became absolutely necessary to look further afield for more competent and honest candidates. Curriculum vitae of potential candidates submitted through ATU for the URADEP project in the Upper Region were considered and none were found suitable for the position. In addition, the GLC despatched invitations to GRM International which manages the Shai Hills Ranch of the Bank of Ghana. In June 1980, GRM representatives who met with the GLC indicated preference for management consultancy based on a "total deal package" preferably over a 6 year period and sub- ject to review every 2 years. In view of the GLC's inability to guarantee finance for a total package deal, the GLC opted for consultancy based on supervisory role which the GRM International subsequently turned down in view of the imminent closure of the Credit in June, 1981. Following further search for candidates, a new manager (a pro- fessional agriculturist with specialisation in animal science and manager of the Animal Husbandry Cattle Breeding Unit at Winneba) was appointed. He met with the AMPU Representative in October, 1980 and finally assumed office in February, 1981. In as far as the ranch is concerned, since its establishment it has had the services of a resident manager for a total of 27 months up to June, 1981. - 34 - Page 9 22. Tadzewu Ranch: The revised programme for the ranch reduced its size from '15,000 acres to 1,955 acres thugh neither the government nor the SPC had legality to the latter land. The programme also projected a reduction in herd numbers from 2,000 to 600. At this level of production, the ranch was not a viable economic entil;y and could not have survived on its own. Consequently, at the projected level envisaged at reappraisal in 1976 there was abso.utely no justification for its inclusion unless it was merely to increase the number of GLC ranches. After protracted negot:.ations between the Volta Regional Admini- stration and the local chiefs, a total land area of 6,920 acres were demarcated in 1978 and these were formally acquired by the government in 1980. 23. Provision for the development of infra-structural faci- lities on the ranch wa! placed at 062,950.00 for houses and buildings, fences, repairs to spray race and stock handling facilities, firebreaks and roads, watering facilities and pastures. Repairs to ;he spray rcce and fences, construct- ion of additional paddock facilities at a total cost of 066,267 have been made,, An additional dam has been congtruc- ted at 027,290.00. Apart from these facilities, the rest are yet to be developed primarily because of the poor funding position of the GLC and the need to have valuation effected on the sites selected before any buildings etc can be constructed. (CPR para 36) 24. Cultivation of maize was introduced into the ranch's p- gramme to improve the cash flow and to increase overall viabia lity. 'This operation 'has turned out to be a disappointing exe cercise. Apart from tie 1978/79 and 1979/80 crops, the remain- ing years have turned out to be either non-profitable or ttal losses altogether. Poor cultural practices (1978/79) and late planting (1976/77) and general bad weather conditions (1977/78 1979/80 and 198)/81) are to main causes of the disappointing results. Observations over the span of 5 years show conelu" sively that maize was the wrong choice of crop for the area and initial evidence on the issue is shown by the relative absence of cultivation of maize within the area generally. ... 140 - 35 - Page 10 Table 2 shows the comparison of the performance of maize pro- duction at Tadzewu viz-a-viz Branam. TABLE 2 1976/77 1977/78 1978/79 1979/80 1980/81 1981/82 TAD BR TADI BR TAD BR TAD BR TAD BR TAD BR 1. Acreage 55 - 62 - 62 120 40 140 40 100 - 78 2. Yield (Bags- 220 lbs) 83 - 63 - 260 175 139 658 17 523 - 603 3. Average (Bags/ acre) 1.5 - 1 - 4.2 1.5 3.5 4.7 0.4 5.2 - 7.8 Apart from the low yields experienced at Tadzewu, production costs per bag of maize produced has generally been very high. The inference from these observations is that if there were doubts on the economics of maize cultivation at Branam then that at Tadzewu needed the least consideration. Attempts to establish pasture at Tadzewu equally failed due to lack of import licence to support the importation of seeds. (PCR paras 20a and 36). 25. The GLC took over a total head of 331 animals including some streptothricosig infected ones from SFC in November, 1975. Between 1977 and 1978 some 750 breeders purchased from Senegal were added to the original stock. The ranch suffered its first major catastrosphe in 1980 from an outbreak of lumpy skin dis- ease which had erupted all over the Volta Region and emanating from trade cattle. Productivity of cattle on the ranch impro- ved considerably (up to 1979) after the initial disastrous figures but have started to deteriorate after the initial impr- ovements. (PCR para 37) TABLE 3.1/ 1978 1979 1980 1981 Calving Percentage 65 70.2 48.5 55.7 Calf Mortality 15.9 5.1 3.6 12.6 Adult Mortality 3.1 1.3 3.8 4.8 ..../11 1/ Added by audit. For herd coefficients, see PPAMN, Annex 2. -36 - Page 11 26. Private Farmer Lending: The GLC has considered that the pilot private lending component is a responsibility fraught with risks and which is essentially a service scheme. Although the Company was initially willing to appoint an officer to undertake the assignment, it later be3ame evident that the scheme as ori- ginally conceived was far too risky and costly for a commercially oriented project. The basig reasons underlying these conclusions were: (a) In attempting to operate the scheme, the GLC had to bear an initial cost equivalent to 71% in- terest (being the cost of onlending from government) on the IDA loan committed to this component. With the risky nature of the scheme and the administra- tive costs involved this was too heavy a burden for a company with a poor funding position to shoulder. (b) Interest charges on loans from the local commer- cial banks channellel through the GLC to the private farmers would make these loans more expensive than those committed through the banks directly and unless these was concrete evidence of possible performance above the levels of those who obtained their loans directly from the commercial banks, then in the end these loans would prove more expensive to the opera- tors themselves. (c) The scheme did not take cognisance of the gen- eral ownership pattern of the national herd in which a herdsman normally looked after cattle belonging to several people or families thereby making decision making on vital economic issues involving the outlay of capital more difficult. (d) In most cases cattle owners did not own lands (to be uFed as collateral) on which the cattle grazed and the-efore could only offer the cattle as security. This was unacceptable because of the risks involved and (e) Unlike the private farmer lending approach in- volving cash crops like sugar cane, oil palm, rubber, cocoa etc ,,here the sponsoring body becomes the main rarketing agency, the -tle market in the country is dominated mainly by the private cattle traders. - .... /12 - 37 - Page 12 With such a system of marketing there is very little assurance that loans could be repaid after the sale of cattle. In view of the foregoing, the GLC indicated to the AMPU that it would only be willing to super- vise the administration of the IDA Credit compo- nent through the purchase and sale of much needed inputs rather than the extension credit scheme. The AMPU however, pointed out that it would rather channel this resource through the commercial banks which were already involved in lending to private farmers, a decision which the GLC found nothing wrong with. As a result of this decision, the GLC did not incorporate this component into any of its development schemes. (PCR para 39) 27. Training From the GLC point of view, it was essential that poten- tial assistant managers receive field training on the ranches before any supplement3tion overseas or in other Bank sponsored projects. It is very incorrect to state that the GLC did not make efforts to get the calibre of personnel for this excercise. The first assistant manager appointed in early 1976 was dismissed in December, 1976. An AMPU representative was on the interview panel in July, 1977 when the first assistant manager for the Tadzewu Ranch was appointed. Although it was agreed that the general standard of candidates was low, it was conceded that the candidate be given a chance because of his apparent enthusiasm. A year later he was dismissed for unreliability in his approach to work. During the years 197b through to 1980, the company conducted further interviews for these positions. Response to these have been poor and the general performance has not been satisfactory either. Following these unsuccessful attempts at recruitment, it became evident that the most reasonable method was to apply for numbers of new graduates through the National Service scheme who would undergo basic practical training for the whole of Ihe national service year (9 calendar months). On satisfactory completion of their assignments interested personnel could then be appointed. .... /13 - 38 - Page 13 The first of such applications wa- made in 1979 and in 1981 the Company was allocated two newly qualified agricultural gra- duates one of which joined the Branam Ranch in October, 1981. It has taken some time to arrange for this mode of recruitment rather than through the noimal advertiserments and interviews etc which had been employed earlier. 28. Pricing: The GLC has not had any interference from any govern- mental agency in the determination of price for the company's products. Prices which arE invariably far above the control price for meat are fixed annually before the major season sales around October and are also used as a basis for the valuation of the Company's assets which normally takes place around the end of June each year. Because of the distance between the main consuming centres of iiccra, Kumasi etc from the actual ranch sites, it becomes qu:ite necessary to get the wholesale price of meat at the slaughter houses of the main centres of consumption and to work these prices backwards for the farm gate price. The company has discarded bargaining as a mode of pricing because of differentials which exist on the various ranches and to avoid problems in accounting with a project which has more than one producing centre. By and large, the pricing structure forulat?d by the Company has been reasonable and considering the distances which traders have to travel be- fore reaching the market, this has at times proved to be pro- hibitive. Comparison with the marketing arrangements of the Shai Hills Ranch Unit is untenable as the conditions are entirely different. Initially, the Shai Hills Ranch is only one unit and does not form part of any company. Consequently, all procedures are centralised and are easily verifiable. Secondlly, the ranch markets its stocks mainly at Ashiaman, one of the main cattle maikets for Accra and some 5 miles (8 km) away from the Ranch. These conditions are not exactly compar- able to those of Branam ar.d Tadzewu which are 280 miles (448 km) and 110 miles (176 km) respectively from Accra. 29. Project Concept: The GLC considers the wenakne:s of the project in several areas viz; ... ./14 - 39 - Page 14 (a) Between the period cl' identification (1969) and effectiveness (1975) very little (if any) initial ground work was made as a follow-up to identify the actual sites and to take the appro- priate administrative steps towards the acquisi- tion of the land areas involved. (b) As a result of the weak link in (a), under- estimation of assets and liason with the govern- mental organisationo whose assets were being used as a basis for the production, problems did arise in the early stages culminating in the project being revised. (c) Data and information regarding areas of activities such as cropping were inadequate and have been proven wrong at both appraisal and revision. (d) The ranches had been spread along the length of the country making physical supervision diffi- cult and arduous especially with the poor road conditions. (e) Pong Tamale and Tadzewu ranches were obser- ved to be maxginally profitable at appraisal (1973). At revision (1976) Tadzewu had become an uneconomic proposition in view of the reduc- tion of its size from 2,000 head to 600 heads and justification for its inclusion remained questionable. (f) As a pilot project, the private farmer lending scheme should not have been attached to the whole project. Apart from the rcasons advan- ced earlier (para 36), the deletion of Pong Tamale from the project at revision made the scheme even more irrelevant considering the fact that about 809o of the national herd is located in the Northern and Upper Regions. Finally, the choice of such a scheme as a module for lovestock improvement in the country is very suspect taking into consideration, failures which have been experienced with livestock development projects generally (vide ILCA Bulletion 10; December 1980; v'y,j F i - 40 - Page 15 30. Ghana Government; Relations between the GL1 and the various government agencies have been cordial but Leave a lot to be desired as some of the problems and failures of government emanate from the very nature of the initial relationship instituted between the Company and the government institutions. Firstly, the choice of the Ministry of Finance and Economic Planning as a supervisory ministry and which has very little to do with agricultural projects has been less beneficiary generally.1/The Ministry does not allocate import licence and neither does it get any licences allocated to it. Except in its role as a permanent member on the import licence allocation committee, its direct influence on such allocations for the Company is limited. In part, the placement of the project under the Ministry's supervision accounts for some of the problems on licences encountered by the Company. It is a mistaken belief that because of the World Bank involve- ment in certain projects, these do have uninterrupted and unli- mited access to foreign exchange. In the case of the Company, there have been instances where impressions have been given by some visiting Bank missions that there have been unutilized funds from the loan which the Company can fall upon. In view of this sad impression, the Company's applications for foreign exchange allocations have in some instances not been treated with the seriousness that the applications deserve. 31. The Bank: Although relationship between the GLC and AMPU started initially on a sound footing, -here is no doubt that this later became strained as time went on. The main issues leading to this state of affairs are: 32. Staffing: The GLC rig'atly assessed in the earlier stages that it will be unable to provLde and search for the staff re- quired for implementing the prDject and initially came to rely heavily on the AMPU with its wide net work for the supply of such staff. In addition, it was very widely belived that such staff would be vetted and shortlisted before the curriculum vitae were despatched for final selection. The experience of the Company in the staff appointed through the AMPU has, how- ever, not been very favourable. 1/ Explanation added by audit. Js usual, certainly in the case of Ghana, the Ministry of Finance is always the representative of the Borrower under Loan/Credit Agreements or Guarantee Agreements. Furthermore, both the Ministry of Finance and the Ministry of Agriculture were represented on GLC's Board (see also PCR, para. 73). - 41 - Page 16 33. Project Concept and Funding: As far back as June, 1975 when doubts were expressed on the economics of maiza production on the Branam Ranch, the Company started having misgivings about the basic facts used in designing the project and advised on the need to get more local involvement in the initial stages of conceptual formulations in order to avoid such problems. By the time the project was re, vised in 1976, it had become very apparent that confidence in the whole project had waned considerably. The main issue centred on how much credence had to be placed on the revised programme in view of increased project costs, areas of underestimation of costs (eg. salaries, infrastructural costs) and the poor cash flow position. Consequently, it was argued that in view of the increased costs and foreign exchange requirements, the level &f IDA funding should be increased. Since the IDA loan was kept at the original level, there was particular reluctance on the part of shareholders to pump more funds into the project. 34. Disbursements: Although the GLC followed the disbursement procedures as indicated within the various categories, it became very apparent in the early stages of implementation that there was the need to restructure these. As far back as January, 1977 the Company requested the AMPU to allow disbursement for the purchase of a bulldozer to undertake some of the assignments and thus reduce the overall cost of the project instead of using constractual and other hiring arrangements which tended to make the overall cost higher and generally retard the rate of progress. The issue became especially more relevant when it came to the development of water facilities. In as much as the Company did not pessess a bulldozer, it became very difficult to hire these machines at very high cost just to try and see that possibilities of such construction did exist. The response of the AMPU in this area has been to say the least quite discouraging with its res- olution being liied to various side issues (provision of foreign exchange for spares, construction of dams towards private farmo:., lending). If this had been resolved even allowing for the fact that the cost had not been originally structured, it could have saved the Company consideracle costs in road construction, firebreaks dam construction and time generally. ...1P -42 - Page 17 Inflexibility in approach to variation of ideTas, structuring of disbursements and sticking to what has been stated within an appraisal report whose concept is generally classified by all to be weak has been one of the greatest drawbacks to the pro- gress of the project. Whilst management may not possess the expertise in solving all problems, at least it has shown flexibility and openeness in its receptability of reasonable ideas. This cannot be discribed as the attitude of the AMPU and the exam-le illustrated by the Credit closure testifies to this assertion. - 43 - GHANA LIVESTOCK DEVELOPMENT PROJECT (CREDIT 500-GH) PROJECT COMPLETION REPORT February 5, 1982 Western Africa Region - 44 - GHANA LIVESTOCK DEVELOPMENT PROJECT (Cr. 500-GH) PROJECT COMPLETION REPORT I. BACKGROUND 1. The Livestock Development Project was the Bank Group-s fourth operation in the agricultural sector 1/ and the first project for livestock development. Financed by Credit 500-GH (US$2.0 million), the project was designed to create the beginnings of a modern cattle industry through the development of three ranches, managed by a limited liability company. Also, 50 privately owned herds in areas surrounding the ranches were planned to be assisted under the project. 2. The project was appraised in early 1973 and Credit 500-GH was negotiated in May 1974 and approved by the Board on June 18, 1974. It was signed in July 1974, declared effective in May 1975, revised in 1976 and, although not physically comDleted, the last disbursement was made in November 1981. The credit was closed on June 30, 1981. 3. This report is based oa the findings of Bank supervision missions, a major project revision in 1976, and a review of files. II. PROJECT FORMULATION AND APPRAISAL A. Identification 4. The project originated in April 1969. A Bank mission visited Ghana and proposed commercial ranch development on the Accra plains and 1/ Before 1975, there had been three IDA Credits for agriculture in Ghana: Fisheries Project (163-GH) in 1969, Eastern Region Cocoa (205-GH) in 1970, and Sugar Rehabilitation Project (354-GH) in 1973. The Fisheries Project took 8.5 years to complete and following disagreements between Government, the Bank and the consultants a second stage was not financed. The proceeds of the credit (US$1.3 million) were fully disbursed. A Project Performance Audit Report (No. 2794) was published on December 28, 1979. The Cocoa Project was fully disbursed in May 1980 and a Project Performance Audit Report (No. 3526) was published on June 24, 1981. Achievements under the project were disappointing. The project experienced poor management, time and cost overruns. The Sugar Project achieved none of its major economic objectives. The main problems were low production, yields and inefficiency. Also the project had inadequate management, shortage of foreign exchange and high production costs. The Credit was closed on March 31, 1979 as scheduled and the-undisbursed amount ($1.1 million) was cancelled in May 1980. A Project Performance Audit Report (No. 3525) was issued on June 24, 1981. - 45 - possibly in central and northern Ghana. At the end of 1970, a Permanent Mission to West Africa (PMWA) reconnaissance mission identified a livestock project which proposed four state ranches in various parts of the country and pilot extension operations for a limited number of private cattle producers. 5. The Planning and Coordination Unit of the Ministry of Agriculture, prepared "A Proposal for a Cattle Development Project in Ghana", which was submitted to the Bank in April 1971. Two alternative proposals to make Ghana self-sufficient in beef production were examined: (a) the development of commercial, large-scale ranches; and (b) the provision of an extension service to small cattle owners. Commercial ranching was chosen, as extension work among small cattle owners was not considered possible as it woula be too lengthy and costly and little was known about cattle owners response. A three phase program was proposed of which only the first phase would be undertaken and presented for financing. 6. The first phase, a four-year project, would be the conversion of four existing Government ranches to serve as nuclei for the establishment of cattle development through extension districts in the surrounding areas. These ranches would: (a) adapt improved technologies to local conditions on a commercial basis; (b) produce improved breeding stock; and (c) conduct pilot extension operations with a limited number of nearby private cattle producers. 7. The first phase, to be managed and implemented by an autonomous administrative unit, the Ghana Cattle Development Commission, within the Ministry of Agriculture, would produce both improved breeding stock to private farmers and cattle for slaughter. The initial herd of 3,700 head on the four ranches was expected to increase to over 21,000 head on 24,000 ha, and annual sales to reach 5,000 head. The cost of the project was estimated at US$1.8 million of which US$1.3 million was proposed to be obtained from IDA. Main project costs would be livestock purchases (US$0.9 million). Government would contribute land, buildings and some machinery and livestock, all of which were existing assets on the four state farms to be taken over. 8. The second phase would provide support services to private livestock owners in the four districts around the ranches, including pasture improvement, improved nutrition during the dry season, mineral supplements, veterinary care, improved bulls and herd selection, construction of water points, fire- breaks and fencing. The third phase, covering a period of 3 - 5 years, would be expansion to a nationwide extension service. 9. At about the same time, the Bank of Ghana also commenced commercial cattle development, at the Shai Hills ranch on the Accra Plains; This enterprise has successfully demonstrated the viability of commercial beef production in Ghana (para 76(c)). - 46 - 10. In May 1971, Government agreed with a PMWA mission that the prepara- tion should be carried out by consultants and in early 1972 a team of five experts from SEDES 1/ and IEMVT 2/ financed by France commenced project preparation, monitored by PMWA. 11. During preparation, in April 1972 a Bank mission to Ghana expressed concern over the availability of land, suitable breeding stock, and the extent of settlement on the ranches. At appraisal, the project design was thought realistic in relation to cattle availability and verbal assurances were given by Government in regard to settlement and land title concerns. Nevertheless, during project implementation, these became issues. B. Project Preparation 12. The preparation, which was completed in mid-1972, included studies in agrostology, ranching, animal nutrition, animal health, hydrology and extension services. A six-year project was proposed having objectives almost identical with the proposals of ?MWA and the Ghanaian identification team. However, herd numbers on four ranches at full development were projected to rise from 4,090 head to 16,730 head on a total ranch area of 53,640 ha (132,500 acres). To attain this increased stocking, it was proposed to establish 3,300 ha (8,250 acres) of improved pastures (Stylosanthes). Branam ranch (28,600 ha), in a tsetse a:-ea, was proposed as a breeding and selection station for trypano-tolerant N-Dama cattle. The other three ranches - Pong Tamale (12,230 ha), Demon (7,300 ha) and Tadzewu (5,510 ha) were in tsetse free areas, but Tadzewu was considered unsuitable as a breeding center because of its small size. N'Dama breeding stock were to be imported, supplemented by the purchase of local heifers, mostly of the West-African Shorthorn breed. Management would be provided through a central coordination unit, independent of the State Farms Corporation, and directly responsible to the Minister of Agriculture. Extension services to local cattle owners were not included. Total project cost was increased to US$2.7 million, of which US$1.4 million was for ranch development, US$0.7 million for operating costs and US$0.6 million for technical assistance, The internal rate of return was estimated at an average of 10.6% for the four ranches. C. Appraisal 13. At appraisal in February/March 1973 the mission agreed with the conceptual plan made by the preparation mission but reduced the number of ranches to be developed to three (Branam, Tadzewu.and Pong Tamale). However, it recognised that one of the main problems would be the supply of good quality heifers of trypano-tolerant taurin breeding stock which were to be imported from West African suppliers. A new feature of the project was the possible participation of local, commercial banks, and the reintroduction of the original suggestion (para 6(c)) to include a component for small private ranches. 1/ Societe d'Etudes pour le Developpement Economique et Social. 2/ Institut d-Elevage et de Medicine Veterinaire. - 47 - 14. The six-year project, as proposed by the appraisal mission would produce improved breeding stock and slaughter cattle and would include the following components: (a) rehabilitation and expansion of the two existing state farms, Tadzewu in Volta Region Southeast Ghana and Pong Tamale (in the North) into commercial ranches, and Branam ranch (in Brong Ahafo Region, in the West) on which maize would be retained as a component; (b) the improvement of some 50 privately owned herds of about 100 head each; (c) training Ghanaian ranch managers; and (d) the identification of further ranch sites and the preparation of a second stage project. It was estimated that at full development the annual output of the ranches and private herds would be 950 breeding heifers, 600 male yearlings, 1,000 tons of carcass meat and 2,000 tons of maize. The target for an average private farmer cattle herd was to increase from about 100 head to about 250 with about 25 steers being finished annually. The project would be carried out by a new company, Ghana Livestock Company Ltd (GLC). The shareholders would be Government, the Agricultural Development Bank, and commercial banks. A Board of Directors would be responsible for laying down and implementing the company's policies. The posts of Managing Director, Chief Accountant and three ranch managers would be recruited internationally. Despite the lack of experience of commercial cattle production in the country, the project was not expected at the time of appraisal to encounter serious technical difficulties. 15. Total project costs were estimated at US$4.5 million, including US$0.5 million for assets taken over, with a foreign exchange component of US$2.0 million which would finance 53% of all project costs during the first four years of the six-year development period - the equivalent of the total foreign exchange costs and 44% of total project costs, including the assets taken over. The IDA credit would be for US$2 million. Government would on-lend US$1.52 million to GLC at 8% for a term of 16 years, including six years of grace. The breakdown of project costs by categories is shown in para 47. 16. Appraisal Follow-up. In December 1973 the Bank discussed project proposals with Government and the commercial banks in Ghana. At this stage the principal issues were: (a) financial arrangements for the project; (b) the status of the organization which would carry it out; (c) price controls; and (d) interest rates for farmer sub-loans. The interest rate proposed by the Bank was 10%, but Government policy was that all agricultural lending should be at 6%. Four domestic banks, Ghana Commercial Bank, Barclays Bank of Ghana, Standard Bank Ghana, and the Agricultural Development Bank, confirmed their willingness to participate in the project, and a steering committee was established to prepare for implementation. As cattle and meat prices had risen since appraisal from 0 0.35/lb to 0 0.40 - 0.50/lb. 1w, pricing was no longer an issue and was, therefore, not made a condition of Board presentation. - 48 - 17. Negotiations were held in Washington from May 6 - 8, 1974. The only matter on which agreement was not reached was in regard to interest rates on farmer loans. The Ghana delegation argued that the on-lending interest rate to private cattle farmers should be either 6%, which was the present official rate for agricultural credit, or a rate equal to the Bank of Ghana discount rate, which also was 6%. The Bank representatives pressed for a rate that would be 2% above the Bank of Ghana's ruling discount rate. Subsequently, in early June, Government approved all credit documents, including an interest rate of 8% on loans to private farmers. Changes to appraised staffing recom- mendations (para 14) were made as follows: the post of Development Manager was omitted; the post of Deputy Managing Director was merged with that of Chief Accountant; and the three ranch managers would handle relationships with surrounding farmers promoted under the project. 18. Board Approval. The Board approved the credit on June 18, 1974. Explanation was provided on the numbers and types of cattle to be imported and the potential sources for these purchases. III. PROJECT IMPLEMENTATION A. Effectiveness and Start-up 19. The start of the project was held up because of delays in fulfilling the conditions of effectiveness, namely: (a) the appointment of a managing director for GLC; (b) arrangements, satisfactory to the Association, between GLC and its shareholders for the provision of finance to GLC; (c) agreements between GLC and Government/State Farms Corporation for the transfer of property and assets to GLC; and (d) legal opinions. The deadline for effectiveness was extended twice, from October 25, 1974 to January 31, 1975 and again to May 31, 1975; the main cause of delay being the difficulty in finding a suitable person for the position of Managing Director of GLC (para 68). Conditions were finally fulfilled in early 1975, and the credit was declared effective on May 14, 1975. B. Revision 20. The Bank reviewed the project in the course of supervision missions in October 1975, March 1976 and May 1976 in view of changes which had taken place in 1975 following the start of the project (para 19). The major changes which had occured were: - 4q - (a) Due to rising import costs of fuel, spares and repairs the profita- bility of maize production at Branam was now doubtful; an intensive fattening operation with larger numbers of feeder steers would now have to replace maize cultivation in order to maintain a satisfactory cash flow; (b) at Pong Tamale about 10% of the area originally planned to be taken over by the Pong Tamale ranch had been cleared for large scale rice production by local farmers; (c) The number of cattle available for transfer to GLC ranches at Pong Tamale and Tadzewu had declined from about 2,800 head at the time of appraisal to about 1,000 head, due to the desire of the Animal Husbandry division to retain larger numbers of cattle under their control. The acquisition of breeding stock posed major problems since the Sahelian countries had stopped export of breeding stock following the drought. The best trypanotolerant cattle were the N'Dama breed, but due to shortage of these cattle in West Africa, prices had inflated. An alternative was to purchase a nucleus herd of N'Dama cattle for bull breeding and upgrading. 21. In August 1975, the GLC Board had reported to the Bank their dissatis- faction with the performance of the expatriate Managing Director, who was appointed in April 1975. The managing director-s physical disability made it difficult for him to discharge his duties fully. Also GLC considered that he did not have the necessary executive ability. The October 1975 supervision mission agreed with this assessment and had the following main concerns: (a) dismissal of the managing director and his replacement by a more suitable candidate; (b) preparation of a short-term operational program for the ranches and reassessment of the assets to be transferred to GLC; (c) preparation of changes in the long-term development plans for the ranches; and (d) possible need for additional financing. 22. The March 1976 supervision mission reported significant improvement in project management since a new interim managing director had been appointed in November 1975. However, it appeared then that substantial cost overruns would occur if the physical program planned during appraisal was carried out. Total estimated project costs had increased to US$7.3 million, compared to US$5.4 million estimated in October 1975 and US$4.5 million during appraisal. This was partly attributable to inflation (para 28) and partly to underesti- mation of costs at appraisal. The mission proposed scaling down operations, with costs being met by available finance. The revised proposals would: (a) reduce the investment period from six to five years; (b) develop a breeding herd of about 4,000 trypanotolerant cattle at Branam which would be increased from 30,000 to 40,000 acres and fatten purchased feeder steers during the early years of herd build-up; - 59 - (c) develop the 1,955 acre farm at Tadzewu for 600 beef cattle but defer the development of an additional about 15,000 acres to a follow-up project; (d) hand Pong-Tamale back to Government, because it would not be commercially viable; (e) transfer the Pong Tamale cattle to Branam and Tadzewu, in order to save on foreign exchange that is otherwise required for the purchase of cattle in Mali and Upper Volta; (f) establish the GLC headquarters at Branam; and (g) use the proceeds of the credit entirely for foreign exchange expenditures for goods, cattle and services. Before getting involved in additional investment programs the mission con- sidered that GLC should demonstrate commercial viability with existing resources of 03.9 million comprising the IDA credit of US$2.0 million (02.3 million equivalent), share capital of 01.0 million, bank debentures of 0475,000, and debentures issued in return for assets transferred to GLC by Government and SFC estimated at about 0140,000. Changes in the Credit Agreement 23. The May 1976 mission reached agreement with Government and GLC-s Board of Directors on the new ranch development proposals, initially prepared in March 1976. 24. On August 6, 1976, the Board approved the amendments to the Credit and Project Agreements to reflect the agreed changes. At that time, almost all of the credit (US$ 1,939,000) remained undisbursed. Section 1.02 a, b, h. (definition of the ranches) and Section 3.07 (transfer of cattle) and Schedule 1 of the Credit Agreement required to be changed. Government agreed to the proposals in their letter of April 2, 1976. The table in para 49 sets out the Categories of items to be financed out of the proceeds of the Credit as revised, the allocation of the amounts of the Credit to each Category and the percentage of expenditures for items so to be financed in each Category compared to appraisal estimates. C. Physical Implementation 25. General. Ghana Livestock Company was formed in June 1974 and a steering committee operated until the General Manager arrived in May 1975. As noted in para 68 the General Manager was dismissed five months later and was replaced by an expatriate from Kenya on a temporary basis. He left in July 1976, but by this time development had been initiated with planning and some physical work. A Ghanaian General Manager qualified in animal husbandry and range management was appointed with Bank approval in August 1976 and continued for the remainder of the project. He did not, however, possess a proven record in commercial ranch development. This early discontinuity in - 51 - management affected project progress thereafter and influenced GLC to appoint a local ranch manager for Branam even though there was financial provision for this post to be filled by an expatriate (para 32). 26. The Branam ranch manager's appointment was made in December 1977 prior to Bank approval being sought. The Bank approved the appointment in April 1978. The Bank raised no objection to the appointment of a Ghanaian ranch manager to Tadzewu Ranch even though he possessed inadequate commercial ranch experience. As planned at revision the small herd of 600 head did not justify expatriate management. However, in practice the cattle bought in Senegal and destined for Branam were subsequently retained at Tadzewu thus changing the management requirement. In addition to these management problems, political factors have seriously affected project progress. National economic problems which occurred before and after the revolution of June 1979 dis- located procurement procedures, and disrupted generally development progress. This was further exacerbated by the need to revise the project (para 20) and to close down the operations at Pong Tamale and reduce those at Tadzewu. 27. The project was first rated a problem project in June 1975 and continued thus throughout project life except for 1977 when some progress was recorded. Principal problems were: failure to construct adequate water supplies at Branam without which the ranch could not be properly stocked, and management failings. In addition, sale prices for cattle, fixed by GLC's Board, frequently lagged behind best market prices (para 55(c) and 62). This can be attributed partly to GLC taking a Government line of trying to limit increases of consumer prices and partly to the lack of autonomy allowed to management. On the Bank of Ghana's Shai Hills ranch, for instance, the ranch manager, operating with adequate autonomy, negotiated sale prices of cattle personally at best current market levels (para 76(c)). The most serious aspect of slow development was-in regards to water supplies on Branam. At revision, a pumped and piped supply was envisaged. However, in late 1977, a local engineering firm carried out a detailed study of a pumped and piped supply, estimated to cost about 0768,000. The firm was inexperienced in provision of ranch-type water supplies and the terms of reference provided by GLC, without consultation with the Bank were inadequate. The proposal which was over-elaborate, was rejected by GLC Board; the consultants bill being met from internal funds by GLC. Meanwhile much valuable time had been lost. Successive supervision missions recommended the employment of an experienced ranch water engineer to be engaged to site and construct surface water points. Although some contacts were established with overseas consultancy firms, GLC was indecisive and it was not until February 1981 that a consultant engineer surveyed six sites and, following his departure, two dams were constructed. Meanwhile, with inadequate water supplies, cattle held on the ranch suffered, resulting in low technical coefficients (para 31). In January 1980, a super- vision mission informed GLC of its intention to recommend to the Bank that disbursements on cattle purchases and ranch development other than for water supplies be suspended until a further and satisfactory water supply had been provided, additional to the one dug-out constructed in 1978. In July 1980, the Bank wrote to the Ministry of Finance and noted, in the context of pos- sible reallocation of credit proceeds, that the Bank would delay action on doing so until assurance had been obtained by the Bank that Branam had a - 52 - sufficient water supply. It was not until the final supervision mission in June 1981 that the Bank was able to record that adequate water now existed to justify the introduction of further cattle. However, by this time management problems had become more acute and technical coefficients were very poor. It was therefore decided to close the project as scheduled on June 30, 1981. Ranch Development - Branam Ranch 28. At the time of project identification it is likely that little settlement existed within the main body of Branam ranch. During subsequent years, however, settlement increased, resulting in protracted negotiations by GLC with local farmers and regional authorities which were not finally resolved until title was obtained in June 1979 (para 41); compensation costs being borne by Government. Phys*cal ranch development at Branam has been seriously delayed on account of not having full rights of occupancy and proved much more expensive than envisaged when the project was revised in March 1976. In general, the higher actual costs reflect annual inflation levels between 1975-1980, ranging from 42.5 to 94% and averaging 71%. Latterly, local costs were also affected by the over-valuation of the currency. Delays affecting development are attributable to :ndecisive management and political factors which resulted in difficulties in obtaining materials. The dip and handling yards took 2 years to complete; the roads were well constructed over about 1 year by a local contractor but proved expensive and were financed by a 01 million grant from Government: no buildings were constructed. Water development is discussed -in para 27 above. One dug-out (dam) holding approxi- mately 800,000 gallons was constructed in August 1978. An additional dam holding 2.5 million gallons was constructed in July 1980. The two dams cost about 050,000 each. Base costs of 0120,000 was provided for a piped and pumped water supply at revision. 29. A comparison of actual against estimated development results is as follows: Physical Cost 0 Revised Revised Appraisal Actual Appraisal Actual Date of Completion Dips, yards 1/ 2 1 15,500 40,200 1978 Roads 2/ 48 km :36 km 1,800 505,900 1979 Firebreaks 111 km 51 km 24,150 73,400 1979 Fencing 76 km 56 km 50,200 45,650 1979 30. About 100 - 150 acres naize were grown each year despite this com- ponent being excluded at revision. On each occasion poor yields were obtained (never exceeding 5 bags/acre). However, prices rose dramatically over the project life reaching 0350/bag in 1981; thus, even with the low yields 1/ Includes 2 loading ramps and night yards. 2/ Includes 7 culverts. - 53 - obtained, maize production became profitable. The poor yields can be attri- buted to poor cultural practices, including a lack of timeliness of operations and poor quality seed, exacerbated by climatic factors. Attempts to establish pastures failed due to poor seed quality. This also reflects the indecisive- ness of GLC management in procuring adequate quality seed, Government's failure to comply with covenants (para 44) in regard to granting a license for seed purchase; and ranch management s inability to carry out the timely cultivations required. 31. The first serious introduction of cattle onto Branam occured between October 1978 and February 1979, when 1,542 head of cattle were transferred from the company-s Pong Tamale ranch prior to its closure. To this were added purchased heifers and fattening steers. Since then, herd numbers have declined continuously. Herd build-up or decline is recorded in Table 2 and summarized below. 1979 1/ 1980 1981 2/ Revised Estimate Adult mortality 3/ 19.6% 12.1% 4% falling to 3% by maturity Calf mortality 4/ 45.9% 43.3% 10% falling to 6% by maturity Calves born 5/ 36.3% 79.3% 60% rising to 70% by maturity Total number head 1,499 1,330 1,066 6/ 2,920 by Year 6 Total breeding cows 7/ 563 425 322 Herd technical coefficients were seriously below expectations. Adult mortality can be attributed primarily to poor grazing conditions with grazing areas restricted due to inadequate water supplies, but also to the effects of transferring cattle from Pong Tamale to Branam and theft. Calf mortality is primarily attributable to poor nutrition, with cows having been unable to satisfactorily rear their calves under the restricted grazing conditions prevailing due to poor water supplies. Total herd numbers have steadily declined due to the poor technical coefficients. Culling has been ques- tionably high (75 head in 1979 and 103 in 1980). 32. The first resident ranch manager, a Ghanaian, who had previously been employed in the Veterinary Services Department took up post in December 1977. Para 25 records the circumstances of the manager's appointment; the Bank accepting a fait accompli on the grounds that the manager did not appear sufficiently unsuitable to justify demanding a reversal of GLC-s decision to appoint him. In March 1981, his employment was terminated by GLC on grounds 1/ Calendar years. 2/ Technical Coefficients for 1981 not available. 3/ Mortality including losses/thefts as percent average monthly adults for year. 4/ Calf mortality as percent of calves born for that year. - 5/ Calves born as percent of average monthly cows for that year. 6/ As at April 1981 there were 1,066 head on Branam ranch. 7/ Breeding cows in December each year. - 54 - of cattle theft (he had been suspended from duty in December 1979). His management performance was throughout not satisfactory. At this point, GLC acknowledged to the Bank the importance of having a suitably experienced ranch manager at Branam and in view of the inadequacy of candidates available in Ghana, considered the appointment of an expatriate. This became inter-related to the need for employment of management consultants who would provide ranch management services. However, serious delays occurred in taking action on this possibility and eventually it became apparent that GLC could not offer an expatriate an adequate contract period financed by IDA relative to the closing date of the Credit (June 30, 1981). Therefore in March 1981 a replacement Ghanaian ranch manager, also previously employed by the Animal Husbandry Department, was engaged. Neither of these ranch managers had experience in developing and managing a commercial ranching operation as such experience is not readily available in Ghana. Management supervision for the 15 months preceeding the replacement ranch manager's engagement was provided on a periodic basis by the Managing Director based in Accra. Since the journey takes about 8 hours over very poor roads, management effectiveness was poor and costs high. 33. The ranch has a low carrying capacity (estimated at revision at 13.5 acres/a.u) on account of there being much thin to rocky grazing. Natural pastures are mostly tall species (Andropogon spp), and are much affected by fire hazard in the tall state. Improved pastures are required on the better grazing areas to provide an all year round supply of nutritious grazing. No improved pasture was established and for most of each dry season, grazing was very scarce (due to burning). T'e position was exacerbated by inadequate water availability resulting in grazing being restricted to the proximity of the two surface water points - only one dam was available in early years. In sum, the poor nutrition of stock has been the major contributory factor to poor animal health, high'mortali:y and low weight gain. 34. Future herd build-up is dependent upon improving technical coeffi- cients to reasonable levels and upon GLC obtaining access to foreign exchange for the purchase of breeders and fattening cattle (para 55). The herd would then stabilize in 1995 at 5,800 head with annual sales of 940 head. 35. Tadzewu Ranch. This ranch was reduced at revision in 1976 from 15,000 to 1,955 acres on account of difficulty in obtaining the additional land. Total projected herd number was reduced from 2,600 to 600 head. The executive instrument for the acquisition of Tadzewu Ranch was not signed until early 1980 when rights to the full 6,900 acres were obtained. Mean- while, by private arrangement with local land owners the ranch was allowed grazing rights and the use of a nearby dam on land outside, equivalent to about 13,000 acres. This permitted the ranch to carry a higher stocking rate than the carrying capacity of the base unit (1,955 acres) would allow. At revision this was estimated at 4,7 acres per a.u.. Since the ranch has, in fact, been carrying about 900 a.u (1,100 head) it would have been stocked at 2.2 acres/a.u. if the outside grazing rights had not been obtained. 36. The project provided for the following development to be carried out: houses and buildings 032,000; fercing and stock handling 07,475; firebreaks - 55 - and roads 04,085; watering facilities 016,200; and pasture improvement 93,190. There is no housing on the ranch and no housing was constructed under the project. This has resulted in detached management and an inadequate control of ranch activities. An existing spray race and fences were repaired under the project and some additional paddock fencing constructed at a total cost of 066,627. No roads have been constructed though some annual firebreak construction was carried out. One dam was constructed in 1980/81, additional to the dam taken over from State Farms Corporation. No improved pasture has been established despite attempts to do so. The lack of development progress on Tadzewu reflects in part on indecisive management and in part on an apparent reluctance by GLC to invest in such a small entity which was surviving with little additional development input. The failure to establish pasture reflects poor quality seed and poor management. Maize crops were grown in each year until 1981, but were unsatisfactory (averaging about 3 bags/acre) and marginally profitable only on account of high producer prices (para 30). 37. In 1978/79, GLC purchased 967 head of generally good quality breeding cattle--217 from Upper Volta and 750 from Senegal. These were originally destined for Branam where funds were provided under the project for the purchase of 1,561 head of breeding cattle. No funds were provided under the project for cattle purchases for Tadzewu. However, the cattle were not transferred to Branam on account of the inadequacy of water facilities and poor grazing conditions there. Many of the cattle which were purchased in Senegal were, in the view of supervision missions, not suitable for transfer to Branam on account of their not being of trypano-tolerant breeds. However, the Senegalese supplier asserted that all cattle exported from Senegal to Ghana were of trypano-tolerant strain. In any case, with the increased area that subsequently became available on Tadzewu (6,900 acres), GLC preferred to retain these cattle at Tadzewu where it was considered they would perform better and where there was a large local body of farmers who could readily benefit from improvement to their herds by the introduction of improved breeding stock and by technical advice received from the parent ranch. However, this created the problem of Branam remaining understocked and was contrary to development proposals. Herd build-up at Tadzewu is recorded in Table 2 and summarized below: 1978/79 1/ 1979/80 1980/81 Revised Estimates Adult Mortality 2/ 5.7 5.0 5.9 3% Calf Mortality 3/ 14.9 11.7 25.7 10% falling to 6% by maturity Calving Percentage 4/ 55.6 49.9 53.0 60% rising to 75% by maturity Total Number Head 846 1,187 1,033 Total Breeding Cows 5/ 326 381 351 1/ Years are May - June except 1980/81 which is May - April. 2/ Calculated as adult mortality and losses as a percent of total average monthly herd less calves. 3/ Calculated on calf mortality as percent of calves born. 4/ Calculated on calves born as percent of average monthly cows in herd. 5/ Breeding cows in June each year. - 56 - Herd performance has been unsatisfactory relative to revised estimates. No major disease outbreak occurred though losses were recorded to tickborne diseases attributable to faulty tick control procedures (spray race operating) and there were incidents of streptothricosis. The main factors of poor herd performance were poor nutrition and poor management. There have been three managers in charge of the ranch over the project life. None has been ade- quately experienced and two performed especially poorly. 38. Assuming that herd tec'nical coefficients are improved to reasonable levels (para 55), the Tadzewu heed will stabilize in 1990 at 1,400 head with annual sales of 290 head. Private Farmer Progress 39. The Board of GLC was reluctant to take action on this component which was seen as a service function and unprofitable relative to the develop- ment of the ranches which were given first priority. At one stage, with Bank support, GLC considered the appointment of an officer to be in charge of the private farmer program. 'However, it was not pursued on account of GLC's general attitude to the component, noted above. It was in the field of sub-loans to private farmers that the commercial banks were to have played a significant role. The decision not to proceed with the component largely negated the value of commercial bank participation in the Project. Had management consultants been employed early in project life (para 68) they could have been made responsible for undertaking the promotion of private farmers without diluting ranch management efforts. However, it is doubtful whether a private ranching company is the best medium through which to promote such activity. Nevertheless, GLC sold some 450 head of breeding cattle to local farmers between 1978-1981 at market prices plus a premium of 25-30 percent for quality. The exact number is difficult to determine as a male can be sold for slaughter by GLC but used for breeding by the purchaser. The number involved suggests that GLC sold off some breeding animals which it should have retained in view of its slow herd build-up. Training 40. The training of Ghanaian ranch managers was regarded in the appraisal document as fundamental not only to the success of the project, but also to the greater development of the cattle industry. In practice no training of staff occurred. GLC Board proposed that no overseas training should take place until staff had been appro3riately trained in the practical aspects of ranch management at field level. This was a sound approach but came to nothing as, it was claimed, suitable staff could not be obtained for training. The real problem seems, however, to have been that the quality of ranch managers was inadequate to undertake the training function since the ranch managers themselves had little or no training or experience in commercial ranch development and management. It is a reasonable assumption that had management consultants been employed to provide management services to the project (para 68) they could have been made responsible for ensuring the sound implementation of this component. Without such experience, which is not available in Ghana, GLC was unable to carry out the training program as envisaged at appraisal. - 57 - Land Acquiral 41. Much of management's time in the first year (1976) was taken up with the problem of obtaining land tenure for the original three ranches. Eventually, as a result of failing to obtain the required land area at Pong Tamale this ranch had to be dropped for development by GLC. On Branam, disputes arose with local cultivators of the land, leading to ill-will, trespass and arson of grazing (para 28). On Tadzewu, the uncertainty of what land would be acquired resulted in a sub-economic unit (1,955 acres). It was through the goodwill of local residents around Tadzewu that ad hoc grazing rights were permitted to GLC outside the official boundaries. It was not until 1980 and 1981 that compensation and rights of occupancy issues were resolved for Tadzewu and Branam respectively. The project was seriously affected as a result of land acquiral issues being unresolved at the time of project effectiveness. The first two Managing Directors spent much of their time over land acquisition and consequently less on physical aevelopment issues. This led to Board frustration with the speed of ranch development and contributed to Board dissatisfaction with management (para 68). During appraisal and at negotiations the Ghanaian delegation had stated that they envisaged no problems over land acquired. The Bank had accepted this. It can be concluded that rights of access and development should have been obtained before project commencement. Rights of access should have been made a condition of negotiations and payment of compensation should have been made a condition of effectiveness. Ranch Concept 42. The selection of the four locations for ranch development was originally made by the Ministry of Agriculture at project conception. At the time, State Farms Corporation (SFC) was in serious difficulties and likely to be wound up. This influenced the Ministry-s selection of sites - all of which were either ongoing SEC or Animal Husbandry Division operations. No alterna- tive sites were examined at feasibility study stage although the site selec- tion had clearly been heavily influenced by these considerations. The widely scattered nature of the three ranches selected - Pong Tamale in Northern Region, Branam in Brong Ahafo and Tadzewu in Volta Region (see map) resulted in a heavy strain on management-s supervisory time and high transport costs. The location of Branam was further complicated by the presence of tsetse fly and the rather difficult nature and low carrying capacity of the ranch (para 33). Had a ranching complex been identified, with ranches in closer proximity to each other and having good access to private farmers, development progress would likely have been cheaper and limited management resources would have been better used (para 76(b)). (See also para 76(c)) which discusses the Bank of Ghana's Shai Hills ranch, which was successful, and where the above conditions were met). Reporting 43. Progress reports have been submitted in conformity to reporting formats provided to GLC at project revision. Reporting was carried out with adequate regularity until September 30, 1978 after which only two other - 58 - reports were submitted. Part of the reason for reduced report submissions in later years was on account of the Managing Director functioning as a part-time ranch manager following the suspension and eventual termination of the Ranch Manager of Branam and the failure of GLC to find a suitable replacement until early 1981 (para 32). Procurement 44. There have been serious delays or failures on the part of Government to fulfill the conditions of the Development Credit Agreement which states in Section 3.04 "The Borrower shall issue promptly as needed import licenses for supplies required for purposes of the project and shall take all action required to facilitate the procurement and import of breeding stock for the project." Up to the end of 1979, GLC submitted its applications for foreign exchange facilities to the Ministry of Trade and Tourism which in turn sought application from the Central Bank. GLC submitted one application in May 1978 for vehicles, machinery and pasture seed, but the license was received by GLC from Government so late that it could not be processed. In November 1979, an application for the purchase of breeding and feeder cattle, veterinary drugs and a vehicle was eventually rejected in favor of allocations to the poultry industry. Subsequently, GLC were obliged by Government to direct their foreign exchange applications through the Ministry of Agriculture. A request for purchase of breeding cattle and veterinary drugs in September 1980 was rejected. Another request for breeding and feeder cattle and drugs in January 1981 remained unresolved in June 1981. Government has not complied with its covenanted condition (3.04) to issue promptly as needed import license for supplies required by the project, nor has it complied with Schedule 2B of the amended credit agreement under which the Bank of Ghana was to make available US$270,000 over the first four years of project life as foreign exchange for the purchase of feeder steers and foreign exchange for the purchase of 1,000 feeder steers per annum thereafter until Year 12. Government-s failings in these matters must be seen in the light of the general scarcity of foreign exchange in Ghana. GLC made little follow-up to its applications, and other industries either with a better actual or potential performance record or those which made a stronger follow-up to their applications were favored ahead of GLC in award of foreign exchange facilities. 45. GLC has experienced difficulties in obtaining contractor operated machinery for the construction of ranch roads and dams. A public works roads contractor was engaged by GLC and released some machinery when not employed on nearby public works construction to work on ranch development. 46. With the exceptions noted above, GLC has conformed to the Develop- ment Credit Agreement in respect of procurement requirements. IV. PROJECT COSTS AND FINANCING Cost and Financing 47. A breakdown of actual and estimated (at revision) project costs by major expenditure category is shown below: - 59 - Actual and Estimated Project Costs 1/ (- 000) % of Category Cost Estimates 2/ Actual Costs Estimated Costs Ghana Livestock Co. Ltd. Branam Houses 72.5 94.6 124.5 Other Buildings 3.5 Fencing and Stock Handling 72.7 202.9 279.0 Firebreaks and Roads 26.0 583.1 2,242.7 Watering Facilities 120.8 401.6 332.4 Pasture Improvement 51.4 116.1 225.9 Vehicles and Equipment 161.1 196.9 122.2 Cattle Purchased 1,281.6 550.3 42.9 3/ Total Branam 1,789.6 2,145.5 119.9 Tadzewu Houses and Farm Buildings 32.0 31.2 97.5 Fencing and Stock Handling 7.5 85.8 744.0 Firebreaks and Roads 4.1 - - Watering Facilities 16.2 32.8 202.5 Pasture Improvement 3.2 1.3 40.6 Vehicles and Equipment 55.0 69.6 126.5 Catle Purchased - 1,079.8 - Total Tadzewu 118.0 1,300.5 1,102.1 4/ GLC Headquarters Fixed Investments 62.5 58.8 94.1 Expatriate Salaries 331.1 119.6 36.1 Total GLC Costs 393.6 178.4 44.0 Value Assets Taken Over 5/ 220.4 378.8 171.9 Private Herd Development 611.8 - - Training Component 21.0 Project Preparation 84.0 - - Total Base Costs 3,238.4 4,003.2 123.6 Contingencies 611.1 - - Total Project Costs 3,849.5 4,003.2 104.0 1/ Figures for dollar equivalent are not given as exchange rate moved from US$1.00 = 01.5 at end August 1978 to US$1.00 = 02.75 thereafter. 2/ 1976 Review costs. 3/ If Tadzewu cattle purchases charged to Branam figures become 127.2% and 118.6% respectively. 4/ Applying 2 above Tadzewu becomes 187%. 5/ Buildings and cattle taken over by GLC from state farms. - 60 - 48. The standard of accounts and financial control has been generally good. Most ranch development expenditure items are heavily overspent (paras 29 and 36). 49. The share of Bank and Government financing was reviewed in 1976 and the credit reallocated based on the various changes. A comparison of appraisal estimates, the 1976 reallocation and final disbursements is shown below: Appraisal 1976 Revision Amount of Percentage Amount of Percentage Final the Credit of Expen- the Credit of Expen- Disburse- Allocated ditures Allocated ditures ment Category (US$ equiv) Financed (US$ equiv) Financed (11/30/1981) Investment in Infra- 970,000 60% 1,160,000 100% of 1,240,570 structure, equipment foreign and breeding and in- expendi- cremental fattening tures stock for ranches Credit to farmers for 260,000 60% 260,000 100% of - livestock development foreign expendi- tures Expatriate Staff 210,000 100% of 330,000 100% of 98,612 Costs foreign ex- foreign penditures expendi- and 70% of tures local expen- ditures Preparation of 90,000 100% of 90,000 100% 15,568 Second Stage foreign ex- Livestock Pro- penditures ject and Overseas and 70% of Training local expen- ditures Reallocated 470,000 - 160,000 -- 2,000,000 2,000,000 1,354,750 50. The main components of the project were either substantially delayed or not initiated at all by closing date - June 30, 1981. On January 21, 1982 the remaining balance in the credit, US$645,250, was cancelled and the project closed. - 61 - V. PROJECT IMPACT 51. The goal of the project was to provide a basis for the development of a commercial cattle industry in Ghana through the establishment of a company which would manage three commercial ranches and assist the development of small private herds in the surrounding areas. There has only been a very small benefit in this secondary objective. Some 450 head of breeding stock (mostly bulls but also some females) have been sold by GLC to local farmers at meat price plus a premium on breeding quality of 25-30 percent. The effect of learning from practices being adopted on GLC ranches has probably been negligible - partly because GLC operations were themselves generally un- satisfactory or inapplicable to small farmers and secondly because there was no means of extending to farmers the technical concepts of GLC ranches - either through the Ministry of Agriculture or through GLC's own extension agents, as such a service did not exist. 52. The project has imported 967 breeding cattle - mainly heifers with some bulls, from Senegal and Upper Volta. These cattle have the potential of being improvements on local stock. In particular, those of trypano-tolerant strain will be of importance in the stocking of areas subject to tsetse infestation. VI. FINANCIAL AND ECONOMIC RESULTS General 53. Both financial and economic rates of return were reestimated in constant 1980/81 prices. Cost and benefit streams were divided into their foreign and domestic components and appropriate inflation indices applied. Further details are in Table 6. Financial Rate of Return 54. At appraisal it was estimated that GLC's operations would generate an after tax financial rate of return of about 14 percent. Despite the delays in project execution and the financial difficulties facing the company at project completion, GLC could nevertheless achieve a financial rate of return of about 16 percent. The major reason for this is that, at the official rate of exchange, the importation of feeder steers and their subsequent fattening and sale on the local market, is a highly profitable operation, with a net return per head of about 01,500. 55. Realization of this rate of return is, however, dependent upon: a) GLC obtaining additional funds(estimated at 06.7 million over the next four years) to finalize its ranch development; b) GLC obtaining access to foreign exchange to import 1,500 breeders in 1981/82 and 1,000 steers each year for the next nine years; c) selling prices for the 1981/82 season being increased from 06 to 07 per lb. liveweight; and d) technical coefficients being improved to reasonable levels. 56. While all these assumptions are realizeable, some must be regarded as optimistic in view of the history of the company's operations and the current circumstances existing in Ghana. For example, given the current severe foreign exchange shortage and the low economic viability of the project (para 61), the agreement of Government to release foreign exchange for cattle imports cannot be assured. Also, the improved technical coeffi- cients will require better management. If this does not materialize and technical performance remains at existing levels, the financial rate of return falls to about nine percent. 57. As originally appraised it was envisaged that 50 privately owned herds would be developed under the project. The pre tax financial rate of return for these projects was estimated at 17 percent. For reasons outlined in para 39, this component of the project was never implemented. 58. At project completion, GLC's current financial position was pre- carious. As a result of projected low sales revenue for some years, coupled with continuing funding requirements to finalize ranch development, the company will incur substantial cash deficits on its operations, estimated to reach a cumulative peak deficit of 06.7 million in 1983/84. Economic Rate of Return 59. At appraisal the economic rate of return of the project was esti- mated to be 12 percent. Apart from the usual allowances for duties, taxes, subsidies, etc., the key assumptions in the economic analysis were: a) a shadowexchange for the Cedi 50 percent above the official rate; i.e., $1.725 to the dollar at appraisal; and b) actual domestic meat prices were consistent with international prices and thus could be used unmodified in the economic analysis. 60. While the assumption with regard to meat prices still seems valid, in retrospect, the shadow exchange rates used since appraisal have been considerably above the official exchange rate. At project completion, the shadow rate was estimated at 012 to the dollar, or 340 percent above the official rate of 02.75. The unofficial rate was about 032 to the $. 61. The economic rate of return was recalculated using shadow rates of exchange used in IBRD project documents over the period since project imple- mentation. Other assumptions used at appraisal were unchanged. On this basis, the reestimated economic rate of return for the project is negative. - 63 - 62. Marketing and Meat Prices. While meat prices are officially controlled by the Government, prices in the larger unofficial market reflect actual supply and demand conditions. In response to limited domestic supplies and restrictions on imports, prices have risen sharply in recent years. Depending upon the availability of foreign exchange, restricted amounts of frozen meat are imported by the Ghana National Procurement Agency, and marketed by the National Meat Marketing Board. Only a few head of live slaughterstock are now imported. 63. Table 3 gives a breakdown of the detailed price structure from liveweight through to the retail level, for April 1981. The June 1981 retail prices were said to be moving toward $30/lb, suggesting that relative liveweight prices would then be in the region of 07/lb. 64. The imported price of meat should be roughly comparable with the domestic wholesale price, with some premium being paid for fresh meat (say 20%). The 1981 import price of 03.60/lb is, however, at the official exchange rate of 02.75 = US$1.00. At a shadow exchange rate of 012.00, the imported meat price jumps to nearly 016/lb. With a 20 percent premium for fresh meat, this would give a comparable domestic wholesale price of over 019/lb. This is roughly equivalent to the actual domestic price of about 021.00. 65. Previous Bank appraisal reports 1/ have also concluded that domestic and imported meat prices were consistent with each other. This analysis assumes likewise. VII. INSTITUTIONAL PERFORMANCE AND DEVELOPMENT A. Board of Directors 66. The Board of Directors has had the same composition throughout the project with representatives from Standard Bank of Ghana, Barclays Bank of Ghana, Ghana National and Commercial Bank, Ministry of Agriculture, Ministry of Finance and Economic Planning, and the National Development Bank. The Managing Director was the seventh Board member. The Board has met 71 times since July 1974, an average of nearly once a month, and has maintained a close relationship with project management. Nevertheless, as commented upon by supervision missions the frequency of Board meetings has occupied too much of the Managing Director-s time which would have been better utilized in field development activities. Visits to ranches have been made on about 6 occasions by Board members and there have been 16 attendances by Bank staff at Board meetings. Attendance at Board Meetings throughout project life has been 79% and reflects the interest shown in company affairs by the Board members. The Board has made sound decisions as when it rejected a revised development plan produced by the Managing Director and requested that a more realistic one be 1/ Livestock Development Project; 277-GH, May 1974. Upper Region Agricultural Development Project; 106a - GH, June 1976. - 64 - produced instead and when it rejEcted an over costly water development plan (para 27). On the other hand, muIch delay has occured in some extremely important decision making, notably: appointment of key staff; support to the Managing Director in obtaining foreign exchange facilities for neected farm inputs (e.g., pharmaceuticals anc pasture seed); drawing up a plan for the company's long term prospects in view of its weak financial performance and unsatisfactory financial structure. As noted in para 39 the role of the commercial banks on GLC's Board was much reduced by not undertaking the private farmer component of the project. The Ministry of Agriculture has played a low key role in the enterprise though it was represented on the Boara by the Director of Animal Husbandry. However, a close relationship has been maintained with the Department of Veterinary Services which has in general provided adequate support to company activities. B. Accounts and Audit 67. The present Chief Accountant took up post in 1976 and has per- formed creditably throughout. Accounts have been generally satisfactorily maintained with ranch accounts centralised in the Accra headquarters. Auait of accounts has been finalised by the auditors (a Ghanaian company) and received by the Bank between 4 - 11 months later than the 4 months stipulated in the Project Agreement. No adverse comment has been made by the auditors in their accompanying reports. C. Technical Assistance 68. The credit agreement required that senior management posts of the company, namely the Managing Director, Chief Accountant and Ranch Mianagers of each ranch operated by GLC would be persons with qualifications and experience, and be appointed on terms and conditions of service acceptable to IDA. Funds for the employment of expatriates were provided under the project revision for the positions of Managing Director and Development Manager for 5 years. At negotiations in May 1964, the nanes of three potential managing directors and five development managers/ranch nanagers were given to the Ghanaian delegation by the Bank-s Agricultural Projects Management Unit (APMU). A number of good potential candidates were either contacted or travelled to Accra for interview in September 1974, but none accepted and the names of five new Managing Director candidates were provided in November 1974. In addition, the services of South American and British consultant groups were offered to provide a complete management team, but wer:e rejected by GLC. In January 1975, GLC management decided to appoint a well qualified expatriate veterinarian, experienced in livestock developnent in East Africa. This man was physically handicapped and operated from a wheel chair. GLC management made the appoint- ment despite the clear notificat:on by the Bank to GLC of the officer's disabilities. The appointment was made on a Bank secondment contract through APMU. Appointment of other officers was delayed pending the arrival of the Managing Director. By July 1975 there were indications given to a supervision mission that GLC was not satisfied with the work output and Board/Management relations of the Managing Director. In October 1975 he was dismissed. An expatriate ranch manager for Pong Tamale was appointed in August 1975 and served satisfactorily until March 1977 when he resigned on account of - 65 - disagreement with GLC regarding the over-centralisation of management. In October 1975 an expatriate from Kenya who was an experienced livestock manager took over as managing director, originally on a 6 months contract, to put GLC back on the rails. The contract was subsequently extended since no suitable expatriate replacement had been found. During this period a supervision mission noted the remarkable improvement in project management which had been effected (para 22). In May 1976 GLC decided not to advertise the post and to appoint a Ghanaian Managing Director whilst retaining the expatriate as a managing consultant for one year, during which he would be expected to carry out about four missions, each of not more than four weeks duration. In July 1976 the expatriate had a serious injury and returned to Kenya. In August 1976 a Ghanaian Managing Director (the only candidate) was appointed with IDA approval and continued as such for the remainder of the project. The performance of the Chief Accountant is noted in para 67. Various supervision missions recommended the employment of other expatriate assistance: a water development specialist (July 1975); and a farm management specialist (June 1975), but no appointments were made by GLC. VIII. BANK PERFORMANCE 69. At Appraisal. As originally designed, the project would have been innovative in using the establishment of commercial ranches as the nuclei for the promotion of private ranching enterprises. Government had originally proposed that four nucleus ranches be established and that these be located where existing farming operations were being conducted either by the State Farms Corporation or by the Ministry of Agriculture-s Department of Animal Husbandry. The three ranch sites finally selected are very widely dispersed (one in the north, one in the west central and one in the far south - see map). This wide dispersal of enterprises has contributed severely to logistical, cost and management control problems. In addition, the ranch in the west central area (Branam) lies in an area where few cattle are kept (less than 10% of the national herd) and it is unlikely therefore that much benefit would accrue to local farmers. This ranch is also heavily covered with bush and has a low carrying capacity (para 33). The cost of development would therefore be higher and ultimate profitability would be lower than in other more traditional cattle keeping areas. At appraisal the problem of land acquiral and compensation to existing settlers was not considered serious, though concern on the matter had been voiced by the Bank's preparation mission in April 1972. In fact it was, and as a result, one ranch (Pong Tamale) had subsequently to be abandoned since inedequate unoccupied land remained to justify an economic proposition; and Tadzewu ranch was greatly reduced in extent for the same reason and title was only obtained after iuch effort and expense in 1980/81. 70. Prior to Effectiveness the Bank provided through its Agricultural Projects Management Unit prompt and appropriate assistance to GLC in its search for an expatriate Managing Director and ranch managers.- The Bank emphatically forewarned GLC of the implications of its decision to employ a physically handicapped though technically competent General Manager (para 68). - 66- GLC were able to obtain, with Bank assistance, a replacement expatriate General Manager (para 68) but only on a short term basis. Thereafter the Bank approved the appointments of local staff for the posts of General Manager and Ranch Manager Branam. 71. During Implementation it became apparent at a fairly early stage that the project required revision and the Bank's response was timely. A supervision mission in September 1975 revised financial and economic projec- tions and the project was further and fully revised in March 1976. This was six months after the Managing Director had been appointed by when serious problems of management, finance, land acquiral and ranch development had become apparent. 72. Bank supervision throughout project life has been regular and adequately diversified. There have been 14 supervision missions involving seven Bank staff and two consultants. This represents one supervision visit for 5 months operation. All supervisions have contained a livestock specialist. The remaining supervisory role has been provided by financial analysts or economists. 73. During supervision the major problems of the project were identified in good time and GLC Board informed of the implications. Alternative lines of action were put forward but in few cases did GLC take appropriate and prompt action. The Ministry of Finance and Economic Planning feel that they have been inadequately kept informed of project progress and problems. The Ministry were represented on the Board of GLC and the problem is therefore probably a matter of internal Ministry communications. The Bank kept the Ministry informed of its views through the Ministry's Board member and by copying correspondence. Some of the most important issues raised by Bank supervision missions were: - The need for management support consultants and water consultants. First raised in January 1977 when draft terms of reference were provided by a supervision mission, the issue was raised by each succeeding mission. Revised terms of reference were drafted by a supervision mission in February 1979 and amended and agreed to by GLC's Board at the same time. - Cost overruns and problems of cash flow. First raised in June 1978 and revised :In February 1980 during a major financial review of company prospects. - The essential need for water development before stocking could be introduced. First raised in 1976 and raised as a major issue by each succeeding mission. - The need to establish improved pasture on the ranches - appro- priate action to be taken to obtain suitable seed was first raised in June 1978. - The need for a revolving fund to be established for feeder steer purchase and a line of foreign exchange established. - 67 - - Possible suspension of loan unless appropriate development action taken on water development and management issues was first mentioned in January 1980. - The need for livestock purchase for Branam to be of trypano- tolerant stock was first emphasised as a result of unsatis- factory purchases in June 1978. 74. The project was first rated a problem project in June 1975 and continued thus throughout project life with the exception of two supervision missions in 1977 when some project progress was recorded. IX. CONCLUSIONS 75. The project has had only very limited success and shows a negative economic rate of return (para 61). However, it has been the means of importing into Ghana some good quality breeding stock (para 51), some of which has been of trypano tolerant strain suitable for use in areas of tsetse challenge. Some breeding stock, mostly bulls, has been sold to private farmers and some improvement can be expected in the quality of their herds. Physical ranch development has been very slow and some aspects remain uncompleted at project close (paras 28, 29 and 36). The most important conceptual component of the project - the promotion of private farmers - was never attempted (para 39). Similarly, the training of ranch managers was never attempted in a formal manner (para 40). Because of the high producer prices for cattle (Table 3), commercial cattle ranching in Ghana is nevertheless attractive financially and GLC could show a good financial rate of return provided management is improved, leading to satisfactory herd technical coefficients and provided funds can be provided from Government or the commercial banks to permit the company to weather the next four years when company cash flow problems will be acute (para 58). The only extraordinary circumstances which affected the project during its life were political and financial. The difficult political circumstances through which Ghana passed during project life affected, to some extent, management s ability to obtain rapidly: goods, services, and decisions when those were affected by Government policy or decision making (para 44). Furthermore, the high inflation rate (para 28) and adverse exchange rate resulted in estimated project costs being exceeded for most expenditures. The project had a number of weak conceptual features as originally appraised: a) A very scattered set of ranch development foci (paras 42 and 69). b) The establishment of the lead ranch (Branam), in an area of low cattle density, tsetse challenge, few private cattle owners, having a low carrying capacity and being rather difficult to develop as a start-up ranch (paras 33 and 69). c) The use of a livestock development company (GLC) as a medium for a service and credit function - the promotion of private farmers. - 68 - 76. Several lessons from th.e project should be learned: a) The necessity of ensuring rights of access and development to land before project commencement. As a condition of negotiations for the IDA Credit, rights of access should have been obtained. As a condition of effectiveness, compensation to former occupants of the land should have been fully paid (para 41). b) In a country such as Ghana, with no history of commercial ranch development, the task cf developing large scale ranches would have been made easier by gecgraphically concentrating the foci for development and selecting, for the first stage, the easiest areas to develop and to ranch (para 69). c) The key factors to the success of the Bank of Ghana's Shai Hills Ranch were: the employment of a firm of management consultants (from Australia) to prcvide the services of a Project Manager and back-up support services; adequate foreign exchange to purchase ranch development and recurrent inputs; reasonable autonomy of management answerable to a Bank of Ghana Board; rapid herd build-up; sales of stock at best market prices; and the location of the ranch having easy accessibility to Accra. d) Demanding of a private livestock development company that it also provide an extension service and enter a high risk field by providing credit to private farmers conflicts with its commercial basis and is therefore likely to result in failure of the private farmer component (para 38). e) If a training program is to be mounted it is necessary at appraisal to identify who is to conduct the training, what are the sources and qualifications of trainers and what the training program should consist of. f) Management is the key component to successfully developing an enterprise such as Ghana Livestock Company. Section 3.02 of the Project Agreement provided for the Managing Director, Chief Accountant and the Ranch Manager of each ranch to be a person of qualifications and experience acceptable to the Association. Stricter criteria should have been followed in assessing potential candidates. In particular, experience of commercial ranching practice should have been made of especially high significance in rating candidates (para 68). g) If parastatal ranches are to be used as catalysts for the promo- tion of private farmer development they should be selected primarily for their suitability as ranches and secondly for their location relative to anticipated potential private farmer development. These do not seem to have been the principle selection criteria for this project (para 42). 77. Although the company established to develop commercial ranching in Ghana (GLC) had participation of commercial banks through debenture holdings, it has not been successful. The commercial ranches have not been effectively established and private farmers not promoted. Probably the concept which would have had most chance of success was a combination of that put forward by the Ministry of Agriculture in 1971 (para 5) and preliminary thinking of the appraisal mission in 1973. A small development group could have operated from an independent commercial base - possibly the Ghana Commercial Bank, to provide technical advice to private cattle farmers and to provide a bridge for these farmers to obtain financial support from the commercial banks and livestock from Government-run farms. The project could have provided technical advice and some financial support to improve upon the perfomance of existing Ministry of Agriculture ranching operations. 78. The project provided funds for the preparation of a Phase II project but it became apparent at an early stage that consideration of a second phase was unrealistic in the light of the unsatisfactory performance of Credit 500 GH. No clear indicator emerges from the project as to where any future involvement of the Bank in Ghana-s livestock industry may lie. However, it can be concluded that any development of the cattle industry must center on the promotion of private farmers, though for the most part these may not be the poorer sections of the community. Furthermore, parastatals are not a necessity when fostering private farmer development. In fact, as the project has shown, they are likely to be expensive and have little impact as demonstra- tions, training areas or sources of improved breeding stock. -Tabl-- LTVEUTOCK DE10g~T PIWXT (500-0H nJET CG#ET10 REM1T IA Changs oec Projet Lif. BRÅAN TADEEU 0 ~ 2-TA1 DEOM TOTAL 1' IAentiflcation Mist1en (1970 and marly 1971) F'est pha.. erect State Far st. p.e Anial Ru-bndry State Fee- 0- nersh,ip Corporation U'P-tior Diviio Corpor.tlon -teetil eurface 25 41 aval..le '. 7.756 11.500 16.000 25.000 60.756 ea) (1.102) '4 600) /6.400) '10.000) '74,10) -Actual carrIe S umberg '3eal) 5' 0 13"5 1.019 1.971 3.685 - ro ert.' ratt le nurber *t full evel. 'has) - - - 11-000 Pro-ert,? an~~l output 'ea) - - - ' 5.000 6' prepario n m-~g1on (early 1972) -Etimatel eur'ace available va) 0.600 10.000 7.700 5.500 5-.100 number et f,1l level 'head) 10.500 5.400 5.400 5.400 '6.700 7/ - vreettn rowe 3.500 1.500 1.500 1.500 8.000 Pro,ete' annua) output . 6! .t fullevel. 'hesl 7.470 1.260 1.760 1.260 6.750 P-~oecel surface of ~rroved haroe'h 850 950 980 530 3.310 Appraptsa ieson (Feb./%arch 1973) -v n (inches) 54 40 43 Rn season ..r-ov Mr-Nov May-Sept Dry season Dec-Mr D •-F.1b Oct-Apr Teneratre - v.' C 71 24 ?3 Cavemax C1 78 33 - Stockung rate on ex sting r.t,,ral paeturce (.c/Anime Ur' tf 6-9 '8 10 Po4lation ene tv-people',q ' 35-30 1)0-'0 50-100 CatIe leve ty lo Mgh high Ranch are. (av) st tull develoet1 1 en '0,000 15,000 31,400 - of chr sesonally noodard 5, r000 16,500 - of wrtch cle.rel 3,000 5,000 700 - of wh'rh fenred 3,000 5,000 7,000 9/ - attlenu.r. (heal at Appraial 0 1.790 1.086 - 2.876 Pro erre' anuoal1 ootput at tull levelop.enr -- Breeding heifars (head) Breeding bolin (head) - 600 Careasn meat (tons) -1 000 miehi (rona) FIrst Supervtsion Mission (June 1975) Cattle Nt-bera (head) 0 400 850 0 1 '50 Site of Ranches taken over by GLC (neren) 30 r 5 rr 500 Review mon (Feb. 76) Estimted surafce availabl (v) LO,000 1,9m Stoki.g rt' on eristi natural - pas,_r. (a /Animal Unit 11.' 40 - - Actual numbors tarttle 5 7 sapervisio m~ssor - Copletion (Ju-e 1981) Estimated surfae anailable (nc) 40,MO 3000 - stoking vete n existing ntutral - pastures (a/Animal lni) 1 9 - AO -ual nunors attlo 1/ A PMWA mission identifted the project in early 1971, later a te-t of far experts pr.pared the tdentificatton report. The tean proposed livestock developent program lo thr-e phaes only data for the firat phaa. Projeot in shon. 2/ of whirh 3,078 arres cleared land, mt of the land idle, w cithot rattle lxt eome orop production (maize?) 3/ The najority of West African Shorthorn type 4/ o fOur lora=tins (Taderw, Oha-u, Afite, Akatsi) of 11,500 norr. about 9,100 anree cleared land. 5/ a of Janury-1971. 6/ A rea of ive Fronrh enrerta from IE4VT nd SEDES worked on the preparation fro early 1972: the preparation ra.oeploted by eid-1972. 7/ 1000 N-Daa row1 td 2,500 West Afrioan Shorthorn too' 8/ all estimate. at full development are for year 25, agrieole tong of meat/beef. 9/ Pog--Taale and Ted~eeu renöhes had about 2,600 head of rattlet about 200 Zebu - types (White Fulani and Sokto Gudalt), 1,000 Sanga. (rosbreedn beten Zebus and West African Shorthorns), 600 pure 'Danw, and 800 West African Shorthorna. 0ANA - thana tjestockt Company Stok cumters on Panchs Pre-Dev. 1975 1976 1977 1978 1-79 580 1981 Appraisal- Herd Projections (Year 1) (Year 2) (Year 3) (Year 4) ,er5) (Year 6) (ear 7-20) (Year 8-20) (Year 9-20) Pong Tamale Ranoh Total Aninala 1,086 1,773 2,242 2,875 3,580 4,093 4,460 4,647 4,73/ 4,748 Br'eding Cows 45, 721 957 1,204 1,440 1,548 i,548 1,s48 1,s48 1,548 Prchases Breeding Age Heifers 30 100 100 100 100 100 - - Tadzewu Ranch Total Animals 1,790 2,103 2,290 2,556 2,627 2,o51 2,618 P,618 - Breeding Cows 700 676 879 977 977 977 977 977 - Purchases Immature Heifers - 50 - - - - - - Bulls 1 10 5 2 3 3 6 - Branam Ranch Total Animals - 620 1,450 2,502 3,413 4,197 4,994 5,810 6,956 - Breeding Cows - 505 978 1,323 1,4.7 1,657 1,980 2,388 2,,78 Purchases Inature Heifers - 100 100 100 50 50 50 -- Breeding Age Heifers - 500 400 300 100 100 100 100 - Bulls - 15 16 14 10 14 20 25 - All Ranches Total Aninals 2,826 4,498 5,982 - 7,933 9/20 10,041 12,072 13,075 - Total Cows 1,150 1,902 2,844 3,509 3,864 6,707 6,o55 4,013 -- PROJECT REVISION 2/ - Herd Projection Tadzeso Ranch 1976 (06/30) Total Anilale - - 350 . 41 4o 455 492 521 555 596 Breeding Cows - - 130 178 16o 119 177 198 208 217 Purchases - - - - - Brana Ranch (Year 14-20) Total Animals - - - 970 1,629 2,275 2,594 2,920 3,111 3,251 1,441 3,891 Breeding nom - 430 800 1,131 1,029 1,049 1,165 1,224 1,284 1,387 Purchases ý/ Breeding Bull - - - 15 22 24 --- Breediog Females - 500 500 500 - - Peeder Steers 500 500 250 - - - PREVIOUS AND ACTUAL SITUATIOV Appraisal 1973 Pong Tamale Rach Total Animals i,086 · 850 380 1,343 ?,114 - Breeding Cows - - - 446 - - - - Sa .. - 67 809 - - Transfer Tadzewu Ranch Total Animls 1,790 - 400 340 621 727 1,207 1,127 - - Breeding Coos - - - - 140 - 304 - - - Puýrchases Heifers - - 66 225 364 4/ 240 (total 895 heifero) - - Bulls - - - 25 36 10 (total 71 bulls) - Steers - - - - -- Branam, Ranch Total Anissals 0 - 0 - 4 7 1,216 1,258 - Breeding Cot - - - - 3 - 480 - - Purchases 7 5 Steer 100/ - - - - 978 3/ - 500 / - Heiferc - - - - 62 - - - - Transfer of Cattle - - - - - All Rancheo Total Animls 2,876 - 1,250 - 1,968 2,1i 2,728 ?,385 - Boredilg Cos - - - - 589 - 784 - - Total ales - - - - 9 - - - A/ A coparison o 0 oexnors at oppraial (1974), at proaet revion (197() aad actual, including purenases of breding rattle and Yeeder stees. / lowing th project revision in 1976, it was decided to drap Paong Taale Ranch from the project, li.it raorh develapmnat Tadzew0 to aborl , aes 0 0arry a breedin! herd of about 601 head f attle (of which 217 breeding cows), and develop r- am to about 40,000 0re0 and carry a trypanotolerarl reeding herd of of 4,00 head (of wh.ch 1,357 breoding cows), continuing the fattening of feeder steers doring the herd buI-up 2h.s. Il as also d ,rioded ta r e t rat-le fro P,og Tamale to Branam. 2/ Total purchses o breedingstoco: 1 bull and 1,300 heiforo. / D whbi 225 trom Senegal: somee: Back-to-Office and Full 0eport daltd April 71, 1580. Ogt rulling rate prior to trao.fer V^ rattla ta Brana,s Ra End of Jure 1979; stock tranferred froa Pong, Tamale Ran1 (transferfro December 1978-January 2379); Information fron iC's annual report For 197 -1uly 1979. i mprt from Uper Volta in June/July 1987; only 137 remaining on rhe mranc as of September 19e . From pper Volta. Jf which 151 head sold .. to utbraof CPP. 1 Furtler data about steer purclaiss w11 bo collected during the next msoll. 72- Table 3 ACCRA YlAT PRICES April, 1981 O/1b. 1. Liveweight 6.00 1/ 2. Transport to Accra (from Branam) 0700 per animal at 550 lb. liveweight 2.65 Accra Liveweight Price: 8.65 3. Dressing-out at 48%, gives equivalent bone-in price 18.00 4. Slaughter fees, etc. 0.50 18.50 5. Wholesalers' margin (15%) 2.75 6. Wholesale price 21.25 7. Retailers' margin (20%) 4.25 8. Retail Price, Accra 25.50 1/ At appraisal (March 1973) bota the economic and financial price of cattle was 00.35 per lb liveweight. In view of subsequent price increases (between appraisal and Board presentation) a price of 00.42 was used for both the financial projections and the economic evaluation of the project. Livestock Developent Project (¢ 000) ACTUJALý 8ý,MAN ES WTTEIS TIMAr APPRATSAL ESTIATES Foreign Exchange Percentage 1975/76 1976/77 1977/78 1976/79 1979/6 1982/81 1981/82 1982/83 1963/84 1984/85 g000 U '003 US$'000 h-ana Livestok Company A. Branam Rnch Hens an- Buildings 20 5.4 8-.1 0.7 3,4 433.7 26.5 286.5 1,101.3 ' Fencing and stockholding 30 4.2 22.0 68.2 34.1 21.8 54.6 128.0 30.0 360.9 Firnbreaks and Roads 35 3.4 137.0 352.7 18.7 71.3 60.2 15.0 2.0 660.3 Watering Facilities 45 0.5 4.6 6o.4 38.9 2"1.7 250.0 555.1 Watering Consultancy 90 96.3 96.5 Pastorn Emprovement and Land Clearing 6, 30.5 2.6 3.6 8.1 71.3 30.0 10.0 10.0 166.1 Vehiles and Equip~ent 8o 7.8 15.6 42.3 49,2 16.5 65.5 151.2 348.1 Tsetsa Control 50 80.0 80.0 80.0 80.0 320.0 Sub Total: . 45.9 38.7 261.1 589.6 167.9 492.0 1,133.1 421.5 378.5 80.0 3,608.3 Cattle 90 0.5 0.5 551.3 2,001.8 1,151.0 1,151.0 1,151.0 5,006.1 Total Brnam Rnoh: 46.4 39.2 261.1 589.c 167.9 1,042.3 3,134.9 1,572.5 1,529.5 1,231.0 9,614.4 3,837.5 ß. Tadzewu Ranch Honcin and cildknga 20 30.0 1.1 0.1 150.0 169.6 350.6 Firebreaks and Roads 30 45.4 1.1 17.8 9.5 3.5 8.5 30.0 115.8 Watering Facilities 5 9.0 89.0 Pasture Improvement 5 2.5 5.3 25.0 3?.8 Vehicles and Equipnent 65 o.4 o.6 0.3 10.0 10.0 10.0 31.3 Sub Total 80 3.3 0.4 11.2 0.1 25.6 25.5 95.1 Cattle tl o0.6 2.6 29.6 9.8 9.0 59.1 304.5 179.6 10.0 714.8 Total Tadzw Ranch: 90 90.3 178.8 22.2 588.5 - - - - 79.8 200.9 181.4 251.6 598.3 9.0 59.1 304.5 179.6 10.0 1,794.6 1,14i.B C. Peng Tamale Ranch House and Building. 20 Fencing and Stockholding 30 0.4 Firebreaks and Roads 304 Watering Facilities .1 Pastue Isproeent 45 1.3 1.1 2.4 Vehicles and Equipnent 8017.6 4. 2.1 -14 Sub Total: 8178 43 21 -24.e Cattle 19.5 4.3 2.1 1.1 27.0 Total Peng Tale: 90 128.4 331.4 387.8 283.9 1,131.5 147.9 335.7 389.9 285.0 1,158.5 943.4 D. 1GC Headquartera Teclding and Vila , eIn. 30 21.2 1o.6 19.5 7.0 0.5 164.0 244.6 150.0 90.0 707.4 651.0 Teshuical Asaiatans. 90 79.0 39.5 0.7 0.4 120.0 243.5 240.3 _ 100.3 83.6_____ ) Total GLC Cota: 495.4 606.4 923.0 1,481., 177.4 1,101.4 3,723.4 2,236.7 1,929.5 1,1441.0 14,114.5 6,573.7 3,154 2,7314 Value Asneta Taken Ove 50 378.8 378.8 329.4 600 520 Private Herd Development - - 659 571 Training Ctaponent- -21i - -21 18 Project Preparation __________________________________________ - 614 73 Total Hann eo.ta: 874.2 606.4 923.0 1,49o.3 17/.4 1,101.4 3,723.4 2,236.7 1,929.5 1,44i.0 14,493.3 (,903.1 3,918 3,396 C.ntingenciesi/450 e,. 1,4. 95oi,4 -----. ..-485.0 960.0. -,2-.0 9-45.0 3,635.0 1,321.8 1,304 1,130 Total P-Jent Cost: 874.2 606.4 9230 48.3 _ 1724 101.4 4,208.1 3,196.7 3,174.5 2,386.0 18,129.3 8,224.9 5,222.3 o 4,52>.3 LeOal Currency Caponent 40 319.4 83.3 243.4 466. 84.4 3.3 1,627.1 1,550.0 1,512.6 60.1 7,047.0 3,290.0 2,913.3 2,521.5 Foreign Exchange Component 6o 554.8 523.1 679.6 1,013.7 93.0 601.1 2,581.3 1,6 .7 1,1.9 1,525.9 11,081.1 4,934.9 2,309.0 2,000.0 Until end Aug. 1978 exchange rate was US$1.00 0 01.5 From that time. rate hao been US$1.00 = 02.7 :/ Ass-e-d inflatione trats: 1981/82 1982/63 1983/84 1994/8 International 7.0 7.1 7.0 7.0 DontLic 65 5' 35 215 OTE: Tonese uosts include feeder steers. while the text of the Appraisal Report states that only breederu ner: ineldei in project cu-ta, this in inconsistnt wth the actual tables in whih feeder steers appear to have been included. Exclusio of the steers would reduce costs by about US$2.0 m. -"A m Livetock Develont Project kana ILnvestOk Cæxpaly ACT '<il FINATEP 17 /7 197'/77 1977/79 1978/79 1979/80 190131 1981j2 1982. 1983/8. 1984/8'- 198L/8. 1996/7 1987/83 1939/89 1989/90 1990/91 1991/9P 1992/93 1993/94 1994/95 F7MU0E Equitv: ;hana Commercial Bank 22 .0 Agricultural Development Bank 200.0 Standard Bank 155.0 Barclava Bank 150.0 onvernmennt 27'¯.3 hovernent Debentures 1/ 27.0 overnent Cash irant 990.1 9.9 IDA Credit 34., 493., 429.0 003.4 29'.1 1,102.3 Sales Reve,e 3. 3)».> 3f6.0 1,044.7 '30.1 1,984.4 868.7 3,974.' 3,843.3 5,02< .1 ,,455.8 5,992.' o,143. A .295.0 6,531.8 5,132.0 4,915.3 5,136.4 4,679.2 4,943.5 TTAL0REVE4M': 823.1 1.o4.4 794.0 1,053.1 1,.3 -3 3,09(. 8,'.7 3,374.' 3.134. 5.02(.1 5,405.3 5932. ,143.8 6,28 50 6531.8 5,132.0 4.915.3 5.135.4 4,679.2 4,943.5 Capital 49.4 r ,4 r,-3. i,43.3 177.4 1,101.4 3,723.4 5,53 . 1,929. 1,441.0 1,330.0 1,300.0 1,500.0 1,300.0 721.0 150.0 150.0 150.0 150.0 150.0 ,pnraTon. 2-.0 299.' 37.4 '32.2 "52.? 1,303.' 1,492.1 1,51<,2 1,523.0 1,531.3 1,533.0 1,533.0 1,933.0 1,533.0 1.533.0 1,033.0 1.533.0 1,533.0 1,533.0 1,533.0 IDA Lon Repaveos: ,rin_pal 17(.4 199.9 223.5 247.0 270.5 294.0 317.6 341.1 364.6 380.0 Interst 2/ 4 59.0 235.2 211.7 188.2 164.7 141.2 117.6 94.1 70.6 3).0 Interest Arrer 3/4. L'ben turen: Intereah t 32.u 32.3 32.3 32.3 32.3 32.3 32.3 32.3 32.3 32.3 17.1 2.0 Interest Arrears / 3.2 3.3 3.0 32.0 32.0 -epavme,' 190.0 190.2 25.0 Estimated Company iax 910.5 1,23-.0 1,030.0 1,130.0 ,iM'.. 1,575.0 1,710.0 1,965.0 1,505.0 660.0 66.o 660.0 660.o T')TAL E'PE'fD:T7<: 730,4 <.3.9 1.230.4 2,012.5 329.. 2,901.1 5.,82.2 202.1 5,l7.0 4502.3 4,43. 4,477. 4,7. 5.010.5 4.590.5 363.3.3 2,550.1 2.533.0 2,368.0 2.343.0 17 -E/N<: 72.7 18. (49: .4) (159.4 1,0/.7 195.. (4,313.: (1.327.4' (1,343.2' 23. 93.3 1,490.1 1,2 9 1,274. 1,941.3 1.501.7 2,3.5.2 2.,oi.4 2311.2 2.600.5 GN'ATI'E 1-:' : 70.7 .21.1 (22,.2) (384.1 7 '72.1 >7.. (3,945.9) (0.273.3 2(.(5. ,09.7) (5,19.4< (3,> 34.3)(2,36u.0) (1,091.5) 849.8 2,3)1.5 4.716.7 7.318.1 9.8e9.3 12,229.8 1/ Senur> tapilal fron Animal tnbandrv Departoent a-1 StatnI anno 'rpration, represe'neisset-.ake' vor. Did nul ome 1_ as a . L pu.n.suve.nuin. ^'rre--. arrearn 81- 20. Assumeo repavoont over enar', ronnei.- 12/2 . GHANA Litesoegent_ eoao. Projec Project Completion Report Reestimated Fjannial Rea of Return - _LC ( 000; cxnstant 1980/81 terms) 1975/76 1976/77 1977/78 1978/79 1979/ 198/81 191/82 194/85 1985/86 1986/87 1987/88 1988/89 1989/90 1990/91 1991/92 1992/93 1993/98 1998/95 INFLOW Debentures: Department Animal Husbandry 134.7 State Fa=s Corporation 244.0 Govereant 25.0 Govereent Grant 990.1 9.9 IDA Loan 94.6 483.6 429.0 808.4 296.1 1,102.3 Salen Revanc 3.5 300.8 365.0 1,044.7 6oo.1 1,984.4 868.7 3,874.7 3,843.8 5,026.1 5,405.8 5,982.6 6,143.8 6,285.0 6,531.8 5,132.0 4,915.3 5,136.4 4,679.2 4,943.5 Resiual Valu7 79 23,000.0 Total Inflw: 476.8 809.4 794.o 1,853-1 1,886.3 3,o96.6 868.7 3,874-7 3,843.8 5,026.1 5,405.9 5,982.6 6,143.8 6,285.o 6,531.8 5,132.0 4,915.3 5,136.4 4,679.2 27,943.5 Cedi Caponent 382.2 325.8 365.0 1,044.7 1,590.2 Inflation Adjostment Factor 5.9 10.8 21.1 34.6 52.6 Cedi Component, 1981 Ters 6,478.0 3,016.7 1,729.8 3,019.4 3,023.2 Foreign Eochange Coponent 2j 94.6 483.6 429.0 808.4 296.1 Inflation Adjustment Factor 57.3 60.2 68.4 79.6 90.3 Foreign Compound; 1981 Ters 165.1 803.3 627.2 1,015.6 327.9 Total Inflow; 1981 Temo: 6,643.1 3,820.0 2,357.0 4,035.o 3,351.1 3,096.6 868.7 3,874.7 3,843.8_ 5026.1 5,kO5.8 5,982.6 6,143.8 6,285.0 6,531.8 5,132.0 4,915.3 5,136.4 4,679.2 27,943.5 O0FLOW Investnent Expenditore 495.4 606.4 923.0 1,480.3 177.4 1,101.4 3,723.4 2,236.7 1,929.5 1,441.) 1,300.0 1,300.9 1,300.0 1,300.0 725.0 150.0 150.0 150.0 150.0 150.0 Operational Kpenditure 255.0 289.5 357.4 532.? 652.2 1,353.5 1,492.1 1,516.2 1,523.0 1,531.8 1,533.0 1,533.0 1,533.0 1,533.0 1,533.0 1,533.0 1,533.0 1,533.0 1,533.0 1,533.0 Principal and Interest Repayments 446.2 466.7 499.2 499.5 499.5 499.5 499.5 467.5 467.5 467.5 442.3 17.1 2.0 , 25.0 Company Tax 959.0 1,235.0 1,030.0 1,100.0 1,165.o 1,575.0 1,710.0 i,865.o 1,505.0 660.0 660.0 - 660.0 660.0 Totai Outflw: 750.4 895.9 1,280.4 2,012.5 829.6 2,901.1 5,682.2 5,24.1 5,187.0 4,502.3 4,432.5 4,497.5 4,875.5 5,010.5 4,590.5 3,630.0 2,360.1 2,345.0 2,368.0 2,343.0 Cedi Component 400.1 327.2 539.5 905.4 625.8 Inflation Adjustment Factor 5.9 10.8 21.1 34.6 52.6 Cedi Coponent; 1981 Terms 6,781.3 3,029.6 2,556.9 2,616.8 1,189.7 Foreign Exchange Coaponent 350.3 568.7 740.9 1,107.1 203.8 Inflation Adjustment Factor 57.3 60.2 68.8 79.6 90.3 Foreign Copound; 1981 Termos 611.3 944.7 1,083.2 1,390.8 225.7 Total Outflow; 1981 Terms: 7,392.6 3,974.3 3,640.1 4,007.6___ 41.4 2,901.1 5,682.2 5,292.1 5,187.0 4,502.3 4,432.5 4,497.5 4,875.5 5,010.5 4,590.5 3,630.. 2,360.1 2.345.0 2.368.0 2.343.0 NErt: (749.5) (154.3) (1.283.1) 27.4 1.935.7 195.5 (4,813.5) (1,327.4) (134 523.8 973.3 i485.1 1.268.3 1,274.5 1,941.3 1501.7 2.365.2 2,601.4 2311.2 25.600.5 oj Herd: l5.2m; Other assets: 97.8 m. oj Equal to IDA Credit Financial Rete of Return 16.2% 괩 E R V 0 L T A ýA TI j T, 3 j G H A N A LIVESTOCK DEVELOPMENT PROJECT LOCATION OF PýOJEC- ýANCHES R Y ýlý;."ý* ------- 0 A S T A NAM,' 0 G 0 \,,KeeK h W-h, Aýebýb. 2 J-k- SUNYANI Aý-kr. To !c ýc KUMASI HO T'j Ak-.. Ob-i ý/TADZEW, SOFOR1rUA- - T er A -CRA App,- ,,,,e < ie,,,d,, 2, 'Iep,.d. ýý1 0- D., R.p---, 2W T.,k- W-b. (D R Rýg-.; B.und-e, H.lý A-n, QIAPE COAST C.P,t.l KONDI S.-~h A.i, T.k..d, F.-t Z- 닛 픽 . · SE'EGAy I ALI ý,P E R R GHANA VWA REGION AGRICULTURAL ýDEVELOPMENT PROJECT I BRCi AGRICULTURAL PROJECTS v.it. Reg,bý A9-.11-1, De-lopneni, P,,olect JW Proiect Headquartere 0 R T H E TAMALE Pn.teet' 0,1 Pý(. p'.I.,,1 E.,ýen, Rog- C.,,o. Pr.j.,t A,ho,,1, Region C- P,.1-t Alh(GI:,,,fý FiýKng for Sa~ 4,- F"ý - p' 0,' N Se- F,,h,ng f., 3-dnell. A D-elopment Pn.fec[ N Upper Region De,eýopment P qr- Prolect Headq-rlers Reg-.1 Bo.nd-e,, Reg-.1 H..dqu.,,i-e,, M- R.. d BRANAM Inle-t-al 0 R E G > /v Iný Lo-, R E G 0 N 42 AýHA. I I lNDUS-ýýAL ýU ESTý. E'<ý K,bi . .22 E R N EAS 0N p,- R E Y T. AP-