Guinea - Boke Bauxite Projects

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 1589 PROJECT PERFORMANCE AUDIT REPORT GUINEA BOKE BAUXITE PROJECT (LOANS Si-GUI, 557-GUI, & 766-GUI) May 11, 1977 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.  FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT GUINEA BOKE BAUXITE PROJECT (LOANS Sl-GUI, 557-GUI & 766-GUI) TABLE OF CONTENTS Page No. PREFACE PROJECT PERFORMANCE AUDIT BASIC DATA SHEETS i - iii HIGHLIGHTS PROJECT PERFORMANCE AUDIT MEMORANDUM I. Background 1 II. Project Performance 3 III. Main Issues 5 IV. Conclusions 8 ATTACHMENT: PROJECT COMPLETION REPORT 1. Project Data A.1 2. Summary A.3 3. Background of Project A.4 4. Physical Execution of the Project A.5 5. Operating Forecasts and Results A.10 6. Organization and Performance of the Borrower A.12 7. Economic Impact A.13 8. Criticisms by Borrower A.15 9. Complaints of the Consulting Engineer A.20 10. Mission Conclusions A.21 APPENDICES 1. Rate of Return Calculations 2. Secondary Effects and Economic Indicators MAP Guinea Boke Bauxite Project This document has a restricted distribution and may be used by recipients only in the performance of their offcial duties Its contents may not otherwise be disclosed without World Bank authorization.  PROJECT PERFORMANCE AUDIT REPORT GUINEA BOKE BAUXITE PROJECT (LOANS Sl-GUI, 557-GUI & 766-GUI) PREFACE This report is an audit of performance under the Boke'Bauxite Project, for which Loans Sl-GUI of March 30, 1966 in the amount of US$1.7 million, 557-GUI of September 18, 1968 in the amount of US$64.5 million and 766-GUI of June 25, 1971 in the amount of US$9.0 million were closed, fully disbursed, on May 31, 1967, December 31, 1972 and March 31, 1974, respectively. UNDP Grant GUI/72/006 of US$49,000 for technical assistance for the management of railway, port and townsite,and for which the Bank is executing agency,is still being disbursed. This report is based on a review of the attached Project Completion Report (PCR)P/ of March 1976 issued by the Western Africa Regional Office, reviews of the project files and transcript of the Executive Directors' meeting which considered the project, and discussions with Bank staff members. A joint OED/Regional staff mission visited Guinea in October 1975 in connection with the preparation of the PCR and the performance audit memorandum. OED staff contributed to the preparation of the section of the PCR dealing with the economic impact of the project and commented on earlier PCR drafts. These comments are reflected in the attached PCR, with whose main findings OED is in agreement. The audit memorandum ex- pands on the question of the split of economic benefits of the project, the long-run returns for Guinea and the institutional development aspects. The valuable assistance provided by officials of the Guinea Government in the preparation of this report is gratefully acknowledged. 1/ Annexes on the Port, Railway Component, Mine and Factory, and OFAB Management and Organization are not attached and are available in OED on request.  PROJECT PERFORMANCE AUDIT BASIC DATA SHEET GUINEA BOKE INFRASTRUCTURE PROJECT (LOAN Sl-GUI) Amounts (in US$ mln) Original Disbursed Cancelled Repaid Outstanding Loan Sl-GUI 1.7 1.7 - Consolidated in Loan 557-GUI Project Data Oria Actual or P ana_ Revisions Est. Actual Conception in Bank 1965 during 1965 10/ 65 Board Approval 1/ 18 /66 3/ 7/ 66 3/ 29 /66 Loan/Credit Agreement -, 3/ 30/ 66 Effectiveness 6 / 30 / 66 6/ 27/ 66 Physical Completion 5/ 67 11/ 67 / of orig>inal project actually completed 80% 100. Loan/Credit Closing 5 /67 5/67 Total Costs (mln) US$ 1.7 us$ 1.7 Econ. Rate of Return not calculated not calculated Mission Data Month, No. of No. of Olate of Year Weeks Persons Manweeks Report Identification 4/65 1 5 5 5/13 & 5/25/65 Preparation Preappraisal Appraisal 8/65 1 3 3 8/11/65 Subtotal 2 8 Follow-on Proiect Loan 557-GUI of US$ 64.5 mln, signed in 9/68 for Boke Bauxite _ Project. Date of Conception in Bank is date Bank first recorded project was being considered for financing and began to follo'wup that decision in a serious continuous way (Project Negotiations or Country General Files). Actual Loan/Credit Closing Date is date of last disbursement out of the loan andalr credit, as given by Controller's Department data.  - ii - PROJECT PERFORMANCE AUDIT BASIC DATA SHEET GUINEA BOKE BAUXITE PROJECT (LOAN 557-GUI) Amounts (in US$ mln) As of 12/31/76 Original Disbursed Cancelled Repaid Outstanding Loan 557-GUI 64.5 64.5 6.5 58.0 Project Data Orignal Actual or Plan Revisions Est. Actual Conception in Bank 1966 6/66 Board Approval 9/ 17/ 68 Loan/Credit Agreement 9/ 18 / 68 9/ 18/ 68 Effectiveness 12 / 11 / 68 12/ 11/ 68 Physical Completion 12 / 72 3 / 73 '. of original project actually completed 50% 100j;, Loan/Credit Closing 4 / 1 / 72 12/ 72 Total Costs (mln) US$123 numerous (see 766-GUI) Econ. Rate of Return 18 % (see 766-GUI) Mission Data Month, No. of No. of I)ate of Year Weeks Persons Manweeks Report Identification 1966 1 2 2 Preparation 1966/67 7 4 28 Preappraisal 5/67 3 3 9 Appraisal 11/67 | 3 3 8/30/68 Subtotal 12 42 Supervision I 5/70 1.0 1 1 n.a. Supervision II 11/70 1.0 2 2 12/70 Supervision III 3/71 1.0 2 2 3/71 Supervision IV 10/71 1.0 2 2 11/71 Supervision V 2/72 0.5 2 _ 3/72 Subtotal 4.5 8 Follow-on Proiect Loan 766-GUI of US$9 min, signed in June 1971 for Boke Bauxite Project. Date of Conception in Bank is date Bank first recorded project was being considered for financing and began to follow up that decision in a serious continuous way (Project Negotiations or Country General Files). Actual Loan/Credit Closing Date is date of last disbursement out of the loan and/or credit, as given by Controller's Department data.  - iii - PROJECT PERFORMANCE AUDIT BASIC DATA SHEET GUINEA BOKE BAUXITE EXTENSION PROJECT (LOAN 766-GUI) Amounts (in US$ mln) As of 12/31/76 Original Disbursed Cancelled Repaid Outstanding Loan 766-GUI 9.0 9.0 0.8 8.2 Project Data Orifinal Actual or PIan Revisions I.st. Actual Conception in Bank 1966 numerous (see 557-GUI) 10/ 70 Board Approval 6/ 22 / 71 Loan/Credit Agreement - 6/25_/ 71 Effectiveness 9 /28 /71 9/29_/ 71 Physical Completion 4 / 73 7/ 73 I of original project actually completed 80% 100l . Loan/Credit Closing 3/ 31/ 74 3/31/74 Total Costs (mln) US$165- numerous US$275 /a Econ. Rate of Return for 26% 32% /a Guinea Mission Data Month, No. of No. of 1ate (4i Year Weeks Persons Manweeks _Reprort Identification 10/70 1 (headquarters) 2 10/23/70 Preparation ) Preappraisal ) 11/70 6 " 4 24 n.a. Appraisal 11/71 1 3 3 5/28/71 Subtotal 8 29 Supervision I 4/72 1 4 4 6/72 Supervision II 11/72 1 2 2 2/73 Supervision III 1/74 1 2 2 5/74 Supervision IV 9/74 1 1 1 12/74 Supervision V 10/75 2 4 8 3/76 Subtotal 6 17 Date of Conception in Bank is date Bank first recorded project was being considered for financing and began to follow up that decision in a serious continuous way (Project Negotia- tions or Country General Files). Actual Loan/Credit Closing Date is date of last disburse- ment out of the loan and/or credit, as given by Controller's Department data. /a Refers to total project financed by Loan 557-GUI and this loan.  PROJECT PERFORMANCE AUDIT REPORT GUINEA BOKE BAUXITE PROJECT (LOANS Sl-GUI, 557-GUI & 766-GUI) HIGHLIGHTS This project provided a total of US$75.2 million in Bank loans to the Republic of Guinea to help finance the infrastructure part of a bauxite mining project. This part included port and railway facilities, housing for the staff, and general services for the townships. The final cost of the project in constant prices is close to the appraisal fore- cast. First bauxite shipments took place about 13 months later than expected. The economic rate of return of the project is higher than ex- pected, mainly because of higher prices for bauxite exports. The project has given rise to some criticism by the Government, chiefly concerning the general Bank approach to the project and the townsite design. However, given the complexity of the project and all the major difficulties en- countered, the Bank's role can be considered a success. The following points may be of particular interest: Reinvestment of economic benefits (paras. 3.09-3.11 and PCR, Section 7 and Appendix 2); Issue of the split of economic benefits of the project (paras. 3.01-3.08, 4.02 and PCR, Appendix 1); Performance under the institutional agreements for the opera- tion of the project (para. 3.12 and PCR, paras. 4.03, 5.06-5.08 and 6.01-6.07); Expatriate manpower and inadequate training of local personnel (paras. 3.13-3.14 and PCR, Appendix 2); Question of townsite size and services provided (paras. 3.15- 3.16 and PCR, Appendix 2); Problems of procurement (PCR, paras. 4.11, 4.20, Section 9 and para. 12.09); and Government criticisms (PCR, Section 8).  PROJECT PERFORMANCE AUDIT MEMORANDUM GUINEA BOKE BAUXITE PROJECT (LOANS Sl-GUI, 557-GUI & 766-GUI) I. BACKGROUND 1.01 The Bank was first approached by the Government of Guinea in 1964 for assistance in developing the infrastructure for a bauxite mining project being then defined after a basic agreement signed in 1963 between Guinea and Harvey Aluminum Company, a private enterprise in the bauxite and aluminum business. The name of the company was changed to Halco (Mining) Inc. on April 1, 1964. The Government established in 1965 a semi-autonomous or- ganization, l'Office d'Amenagement de Boke'(OFAB), to carry out the engi- neering project, supervise construction of infrastructure, and subsequently operate and maintain the facilities (PCR, Appendix 1). An engineering loan of US$1.7 million (Si-GUI) was approved by the Bank on March 29, 1966 to finance the foreign exchange costs of detailed engineering of that in- frastructure project. Prior Loan (557-GUI) 1.02 The Compagnie des Bauxites de Guinee (CBG) was formed in 1966 with 51% participation of Halco and 49% of the Government, the latter in the form of mineral rights leased to CBG amounting to US$1 million. Halco was reorganized to include among its partners six big aluminum enterprises.- Bauxite sales contracts totalling 5.1 million tons per year between CBG and the stockholders of Halco were also signed in April 1967. The infra- structure/mining project was finally defined for an extraction and processing level of 6 million tons of bauxite per annum and appraised in October 1967. The mining project comprised an open cast mine, crushing, loading and cal- cining equipment and two townships (Sangaredi and Kamsar). The infrastruc- ture part included port and railway facilities, housing for the staff and general services for the townships. Together with the mineral rights, the infrastructure constituted the Government's contribution to the develop- ment. The total cost of the project was estimated at US$123 million; the cost of infrastructure was US$79.5 million. The prior loan for US$64.5 million was signed in 1968 to finance the foreign exchange cost of: (a) new port installation at Kamsar with a dredged access channel; (b) 136 km railway from Kamsar to the Sangaredi mining area; and (c) new town at Kamsar to house the personnel engaged in operat- ing the project designed for a maximum capacity of 6 million tons per annum. 1/ Aluminum Company of America, Alcan Aluminum Ltd., Martin Marietta Aluminum, Vereignite Aluminum Werke, Pechiney-Ugine and Mon tecatini Edison. The latter was replaced by Alumetal in 1967.  - 2 - Some ancillary buildings and works were also included in the loan together with housing for OFAB personnel in the townsite. USAID also made in 1968 a loan of US$21 million equivalent to finance the local costs of the project. 1.03 Guinea was not at that time considered creditworthy according to Bank standards, but the Bank decided to go ahead with the loan and to accept the guarantee offered by the six private aluminum companies' partners in Halco. A "White Book" contained the agreements resulting from complicated legal and technical arrangements between the partners: (a) loan and project agreements between the Bank and Govern- ment of Guinea and OFAB, respectively; (b) financial assurances and trust agreements between the Bank and Government, OFAB, CBG, Halco, each of the six participa- ting aluminum companies, and a private banking institution serving as trustee under the trust agreement; (c) railway and port agreements between the Government/OFAB and CBG for the use of infrastructure; (d) training agreements between Government and CBG/Halco; (e) technical assistance and training agreements between Government and CBG/Halco for mining installation; and (f) agreements governing participation in Halco of the six aluminum companies. Two other documents regulated CBG use of Guinean currency and the trans- fer to Guinea of bauxite deposits in Tamara Island. The financial assurance agreement included a force majeure clause relieving CBG and Halco's shareholders of their obligation under certain political events narrowly defined. 1.04 As there was no real world market price for bauxite, Guinea had agreed with Harvey Aluminum Company on a basic transfer price -- or price at which the country would sell the bauxite to foreign partners -- in 1963. A new transfer price for sales of bauxite from CBG to Halco was set in 1967 so that the debt of both CBG and the Government would be repaid over a certain period. The resulting transfer price (US$7/ton FOB as an average, with slight variations depending on ore quality) was similar to that of bauxite of similar grade (e.g., in Surinam). An escalation clause was introduced in the transfer price of bauxite, linked to US wholesale price index. Guinea would retain ownership of infrastructure, would charge to CBG all costs resulting out of its use, and would obtain revenues, amounting to 65% of net taxable profits of CBG. Guinea's debt service, and operation and maintenance expenses of infrastructure, would be debited against bauxite sales. Debt service  - 3 - would, in case of insufficiency, be covered by CBG advances recoverable from government's share of profits. Loan Extension (766-GUI.) 1.05 Halco and the Guinea Government agreed in November 1970 to in- crease the capacity of the project to 9.2 million tons per year. Additional contracts for 3.9 million tons were signed between Guinea and Halco's partners. The new cost estimate for the total project was US$165 million. The Bank was approached for further finance, which resulted in June 1971 in a loan of US$9.0 million for port extension, upgrading of railway track and construction of sidings to accommodate heavier and longer trains, road construction and other works to increase the capacity of the hospital and services at Kamsar. 1.06 The new project did not entail any major revision of the existing understandings and agreements, with the only exception of: (a) Government's agreement to improve road access between Kamsar and Conakry; (b) updating of financial guarantees and participation agree- ments; and (c) updating of sales' contracts to new production levels. Mine and processing plant were financed by Halco and private lenders. The Bank loan plus an AID loan of US$21 million, to finance local costs, covered 100% of infrastructure costs at the time of appraisal. 1.07 For the extension contract, covering the additional 3.9 million tons per annum, the transfer price would be US$9 per ton for the first 15 years and US$10 per ton for the remaining 5 years. The same adjustments for grade variations and price escalation clauses as applied to the initial contracts were extended to the additional contracts. II. PROJECT PERFORMANCE 2.01 Final construction cost of the project has been US$275 million equivalent compared with US$165 million equivalent (including contingencies) contemplated at appraisal of the project extension. The main factors ex- plaining the cost overrun were additional price escalation and varia- tions in exchange rates; the cost of the project in constant prices was -lose to the appraisal estimates. Bank's intervenL.on was instrumenta in reducing the costs of infrastructure works and of keeping them within tight budgetary limits (PCR, para. 4.26). The performance of the contractors has been generally satisfactory (PCR, para. 4.18).  - 4 - 2.02 Bauxite is being mined and shipped, and although original produc- tion targets are not yet being achieved (PCR, para. 5.09), the benefits anticipated at appraisal should be exceeded due to large price increases. First bauxite shipments took place in August 1973, about 13 months later than forecast; shipping targets are being reached two years later than forecast. The main reasons for delays were: (a) decision to extend the capacity of the project from 6 million tons to 9.2 million tons; (b) difficulties in recruiting labor and poor quality of middle management; (c) difficulties in communications; (d) the 1969 strikes in Italy; and (e) the November 1970 warfare situation in Guinea. Operating problems (mainly because of the destruction of the primary crusher) also contributed to the failure to meet shipping targets (PCR, para. 4.02). 2.03 OFAB has fulfilled its loar. covenants to a reasonable extent. But a chief accountant has not yet been appointed and a number of the recommenda- tions made by the consultants employed under the UNDP grant have not been implemented. The Government has complied with most of its loan covenants. The two main exceptions have been that the Government has not caused the road to Conakry to be properly maintained and that an additional tax on bauxite shipments was imposed resulting from a --ew special tax on exports of mining products. The effect of this tax was to change unilaterally the mechanism for the distribution of benefits of the project originally negotiated between Guinea and the foreign investors. 2.04 Excluding the export tax iimposed by the Government, financial and economic returns of the project for Guinea would have been 26% and 28%, respectively, compared with about 21% and 26% contemplated at appraisal of the extension project (PCR, Appendix 1). The additional revenues accruing to the Government in the form of export tax resulted in higher financial and economic returns for Guinea amounting to 30% and 32%, respectively. 2.05 The project, one of the most complex in Bank's history, covered a time span of ten years between identification of the engineering loan to completion. It has given rise to a number of criticisms by the Govern- ment (not always well founded) mainly because of dissatisfaction with the Bank's approach to the physical design of the project. In particular, considerable reductions in the standards of the original project made at the Bank suggestion have been criticized by the Government. These reduc- tions have not affected the viability of the project and are discussed and commented upon in Section 8 of the PCR. The audit agrees with those comments.  - 5 - 2.06 The Bank supervised closely both projects, in the field and from headquarters. Bank staff examined thoroughly bidding documents and bid awards. On this subject, the Guinean Government has shown impatience for the Bank's "lengthy procedures." Many problems resulting from difficult communications -- real or perceived -- between the contractors, consulting engineers and Halco, all having headquarters scattered in Europe and America, were ironed out by the Bank; cases in --oi- a3_ selection of con- tractors, consultants and the constitution of the Construction Coordinating Committee. 2.07 In view of the complexity of the project, the economic and politi- cal environment of Guinea and the inherent risks associated with it, the Bank's role can be considered a success. III. MAIN ISSUES Split of Benefits Between Guinea and the Foreign Investors 3.01 The project's total economic benefits can be measured by the value of the incremental aluminum production (net of operating expenses) made pos- sible by the exploitation of the Boke'bauxite fields. The split of those benefits between the foreign investor (Halco) and Guinea is an interesting issue raised by the project. 3.02 The most important variable for the benefit split is the transfer price for bauxite. Given a fixed price for aluminum, Halco's benefits (per ton) would be equal to alternative CIF aluminum plant prices for bauxite minus Kamsar transfer prices minus shipping costs between Kamsar and alumi- num plants plus Halco's participation in CBG's net revenue (35% of the difference between transfer price and operating expenses), while Guinea's benefits (per ton) would be its 65% share of the CBG net revenue (transfer price minus operating expenses). 3.03 Therefore, the lower the transfer price, the larger Halco's share of the project benefits. On the other hand, Guinea's share of bene- fits depends on the difference between transfer prices and CBG operating costs; and the higher the transfer price, the larger the benefits for the country. 3.04 There was thus a potential conflict of interest between Halco and Guinea which was also a reflection of their different positions vis-a-vis the market for bauxite. While Guinea was expected to supply about half the additional world supplies between 1969 and 1975 (a large part of the other half to come from other developing countries needing suibstantial foreign exchange for their development), Halco's six partners were among the largest bauxite buyers in the world and accounted for roughly 40% of world bauxite consumption. The CIF aluminum plant prices for bauxite depended mainly on FOB bauxite prices in other supplying countries; and these prices were con- trolled by a very small number of companies, most of them partners in Halco, which also controlled the marketing and processing of aluminum.  - 6 - 3.05 Before project appraisal, Guinea and Halco agreed on a transfer price of US$7/ton. The benefit split resulting from this transfer price was considered fair to Gutinea at appraisal, and an escalation clause for the trans- fer price was agreed upon to protect CBG against cost increase. This clause linked the transfer price to the variation in local wages, the cost of diesel and fuel oil, and the US wholesale commodity Price Index. It was expected that this escalation clause would help maintain the stability of the original agreement about the transfer price and that the resulting benefit split would therefore continue to be satisfactory for both parties. 3.06 However, this expectation did not materialize because of the lack of a direct link between the transfer price and aluminum prices and trans- fer prices from other sources, in the context of the strong world inflationary tendencies of the 1970s. In a period of price stability, this missing link with market forces would not have been a problem, but in the strong inflationary environment of the 1970s, it created a conflict of interest between the foreign partners and Guinea. Market prices increased more rapidly than production costs and as this was disadvantageous for Guinea's share in the project benefits, the Government reacted by imposing a new export tax on all mining exports in early 1975. The resulting transfer price increased by about 51%, and the revenues to Guinea increased to about 57% of the FOB value of bauxite exports. 3.07 A study of the fairness of the benefit split is outside the scope of this audit. However, this experience suggests that the mechanisms by which benefits are defined and shared merit further study, both at the methodological and the empirical levels. The findings of such research could be very useful in helping to develop terms of contracts between host countries and foreign partners that would, for any agreed benefit sharing objective, ensure a more stable basis for future operations and relationships. Long Run Returns for Guinea 3.08 The issue of maximizing the long run return for Guinea is related to an efficient reinvestment of Guinea's surplus funds from bauxite revenues. In developed economies an efficient reinvestment of surplus funds is assured by the ripple effect throughout the market forces. In less developed econo- mies such as Guinea, a special effort is required to internalize and multiply the benefits. 3.09 Guinea is transferring -- to transport, other infrastructure, and human capital -- resources proceeding from the mine. Transport infrastruc- ture, whose marginal operating costs and depreciation are covered and ensured by agreement between CBG and OFAB, has spare capacity which should be utilized. Townsites and services not covered by agreement between CBG and OFAB are unable, as of now, to generate sufficient funds for their simple reproduction and training is at a standstill (para. 3.13). There are, thus, slacks that reduce the maximum achievable efficiency of Guinean resources. 3.10 However, since existing physical infrastructure is a joint input necessarily localized, and human capital is almost exclusively tied to CBG and OFAB activities, the real question remains of how to allocate the cash revenues received by the Covernment in order to maximize the impact of the  -7- project on the overall economy of the country. This issue is explored in the attached PCR (Section 7 and Appendix 2) and we agree with its recommen- dations about possible efficient reinvestments in the Kamsar-Boke' area. Institution Building 3.11 OFAB is responsible for management of the infrastructure part of the project (para. 1). Even though OFAB is responsible to the Ministry of Economic Development, its areas of activity fall under the jurisdiction of various Ministries (for example, Health, Education and Transportation) and this could create coordination problems. Another problem is OFAB's dependence on CBG for overseas purchases because it has no foreign exchange account. As a result, CBG receives sylis (the local currency) every year from OFAB for these expenses, and the sylis are then used to pay local wages and local expenses. The processing of OFAB orders depends on CBG requirements for local currency. If CBG found itself with a surplus of sylis, there would be no incentive to process OFAB orders for overseas purchases, and the effect would be felt on OFAB operations. 3.12 The project could not have been carried out, and operations could not have proceeded, without the massive support of expatriate manpower. The Bank supported vital contributions of technical assistance such as the setting up of a Construction Coordinating Committee which had a major role in the successful completion of the project (PCR, para. 4.11). But, the success in other areas has not been as great. In some cases, Guinea has not absorbed operating and management techniques, brought about by expatri- ates, some of them either not having been replaced by nationals or not replaced at all, and the recommendations of others not having been followed up (like the UNDP management consultant, PCR, para. 6.06). 3.13 The Government's veto of some nationalities reduced the choice in expertise in both execution and transfer of technical assistance. Expatri- ate consulting firms and individuals employed by OFAB to run the railway and port respectively are performing well but are not training successors. Training is now at a standstill. CBG's training has suffered from a tight money policy and is not satisfying OFAB's expectations. OFAB itself lacks the budget and expertise to run any major training program. The Government is concerned about the situation and in recent Board meetings, CBG was asked to prepare an accelerated training and Africanization program for senior staff to make up for the delay in this area. Further training of Guinean nationals will: (a) introduce industrial skills necessary for the future development of the region; (b) add benefits to national income since more Guineans would have been employed; (c) reduce the payments in foreign exchange to foreign nationals; and  -8- (d) reduce the strategic dependence on expatriate manpower to exploit national resources. Townsite Sites 3.14 One of the issues in defining the project extension was to deter- mine the size of towns at Kamsar and Sangaredi and their financing. The reduction of townsite costs at Kamsar recommended by the Bank consisted of decreasing the number of houses and practically eliminating ancillary services and leisure activities. 3.15 The Government has criticized the extent of the townsite and its services as insufficient (PCR, paras. 8.03.7 - 8.03.12). However, OED agrees with the PCR that the Bank's decision was, in retrospect, sound. While a large and better equipped town for a 9.2 million tons/year production level would have added to the welfare standards and expectations of CBG/OFAB employees, it would have increased costs of the infrastructure project by about 17%/. More important, it would have been cumbersome and finally impos- sible to design the optimum sites and services package without explicit indications of the Government's or residents' preferences. Houses, although expensive compared with similar houses in more developed countries, were finally built to the minimum standards required to attract qualified staff to the Sangaredi and Kamsar sites. The trimming of townsite costs at Kamsar recommended by the Bank has not prevented the formation of a relatively large town, with areas of influence for health services well beyond Kamsar itself. It could have been expected, once the mining operations started, that some part of the mining revenues would eventually be reinvested in housing. IV. CONCLUSIONS 4.01 The project has been successfully completed. The final construc- tion cost in constant prices is close to the appraisal forecast; and even though the first bauxite shipment took place 13 months later than expected, the performance can still be considered satisfactory given the complexity of the project and the various extraneous circumstances affecting project implementation. The economic rate of return for Guinea is 32%, which is higher than the appraisal forecast of 26%. 4.02 The stability of the agreement about the split of benefits would have required that the transfer price be linked to aluminum prices. The escalation clause agreed upon was only cost-based; and in the inflationary environment of the 1970s market prices increased more than production costs, thus creating a disadvantage for Guinea. The broader issue of what is a fair split of benefits is outside the scope of this audit.  ATTACHMENT PROJECT COMPLETION REPORT GUINEA BOKE BAUXITE PROJECT (LOANS SI-GUI, 557-GUI & 766-GUI) 1. Project Data: 1.01 557 GUI 766 GUI Total Amount of Loan US$ 64.5 m US$ 9.Om US$ 73.5m Amount disbursed US$ 64.5 m US$ 9.Om US$ 73.5m Date of Loan Agreement Sept. 18, 1968 June 25, 1971 Effective date Dec. 15, 1968 - Closing date Dec. 31, 1972 March 31, 1974 Date of previous Supervision Mission Dec. 12, 1974 Exchange rate (at appraisal) $1- GF 250 GF 250 Exchange rate (Bank current) $1- Sylis 20.4628 (Sylis 1 = GF 10) 1.02 The project was carried out as part of the overall Bokh Bauxite Development Project which comprises an open cast mine, crushing, loading and calcining equipment and two townships - at Sangaredi and Kamsar. It was designed to provide the necessary infrastructure for the project, which to- gether with the mineral rights constituted the Government contribution to the development. The cost of this infrastructure was 42% of the total cost at the time of appraisal and about 40% at completion. The Bank loan plus AID loan covered 100% of the costs, including local costs on current account, at the time of appraisal. By completion this had fallen to 73% on capital account, with the difference being made up by overage loans from Halco. On current account there was a shortfall of about 15%, made up by advances from the Government.  - A.2 - 1.03 The infrastructure consisted of: (a) a new port installation, including a bulk-loading wharf, approach jetty, a new 17-km dredged approach channel, buoyage, service craft and some handling equipment; (b) a 136-km standard gauge railway from the mine at Sangaredi to the crushing plant at Kamsar, with locomotives and rolling stock re- quired for general, i.e., non-mineral, services (the other locomo- tives and rolling stock being supplied by CBG); (c) housing for the staff of the Office d'Amenagement de Boke (OFAB) including port, railway and municipal employees; and (d) general services for the townships, including water supply, hospital and schools. 1.04 The extent and quality of this infrastructure was considerably influenced by the Bank intervention, which resulted in such modifications as to reduce the cost from the $124m originally requested to the $65m ori- ginally accorded. Most of this is not, however, a permanent saving and will either have to be found out of profits (e.g. additional housing) or national funds (e.g. water supply extension to villages). 1.05 A UNDP grant of $492,000 provided for technical assistance for the management of the infrastructure. 1.06 Local costs were financed by a USAID Loan for Sylis 447m ($21.8m). Cost overruns and additional working capital requirements ($25m) were financed by overage loans from the Compagnie de Bauxite de Guinge (CBG), the exploiting company, on the same terms as the Bank loan and disbursed under the control of the Bank. 1.07 CBG is a mixed company, owned 51% by Halco, a consortium of aluminium manufacturers and 49% by the Government of Guinea. Halco has provided a nomi- nal equity, $46m loan capital and subordinated debt through direct subscription and has met the rest of its share of the project and operating costs ($211m) through loans guaranteed by Halco or its constituents. The 49% of the Govern- ment is a nominal valuation of the mineral rights. The investment arising out of the Bank loans is outside the company and is serviced by amortization charges to CBG, royalties and taxes. 1.08 OFAB is a semi-autonomous Government agency charged with the manage- ment of the infrastructure and with monitoring the Government interest in CBG. 1.09 Under the Loan and Project Agreement the following were the main special covenants: (a) the Government undertook to perform its obligations under various agreements with CBG, to which the Bank was not otherwise a party (557-5.08 (a) ). Of special significance in these agreements were: (i) the setting up of a Construction Coordinating Committee; and (ii) an agreement not to levy additional taxes on bauxite pro- duction or sales;  - A.3 - (b) OFAB undertook to employ qualified staff, approved by the Bank, as Directors of Railway, Port and Port City departments as well as a chief accountant and chief engineer (557 2.05 (c) ); and (c) the Government undertook the repair and maintenance of the main route to Conakry including both the road and the ferry crossing en route. 2. Summary 2.01 The project and its extension, with minor revisions, have been executed as envisaged at appraisal. 2.02 Bauxite is being mined and shipped and, although the original pro- duction targets are not yet being achieved, the benefits anticipated at apprai- sal should be exceeded due to large price increases. 2.03 OFAB has fulfilled its covenants to a reasonable extent but has not yet appointed a chief accountant nor implemented a number of the recommendations made by the consultants employed under the UNDP grant. 2.04 The Government has also fulfilled most of its covenants with the following notable exceptions: (a) it has raised an export tax on bauxite shipments; and (b) it has not maintained the route to Conakry. 2.05 The comparison between the rates of return at appraisal of the origiP nal project (i) of the extension (ii) and as of this time (iii) is as follows: (a) financial rate of return on the total investment: (i) 14.5% (ii) 19.5% (iii) 22.9% (b) financial rate of return on Guinea's investment: (i) 14.5% (ii) 20.8% (iii) 26% (c) economic rate of return for Guinea: (i) 18% (ii) 26% (iii) 27.7% 2.06 The project is a complex one, and this has given rise to a number of problems. The friction between the parties and the criticism of the Bank arising out of this fre largely the result of forgetfulness of the reasons for many of the decisions in the project design. These are, therefore, listed and discussed separately '(Section 8). None of these matters should significantly inhibit the success of the project.  - A.4 - 3. Background of Project 3.01 As early as 1964 the Government of Guinea approached the Bank for participation in the financing of a bauxite mining project which was to make an important contribution to the development of the country's economy. At that time the Government had already entered into an agreement with a US aluminum producer, Harvey Aluminum Company (later Halco Mining, Inc. was established by this firm for the Guinean project) to create the Compagnie des Bauxites de Guinge (CBG) which would mine some 1-1 1/2 million tons of bauxite in Guinea. The Government had already undertaken, in an agreement with Halco signed in 1963, to provide the infrastructure for the project, consisting of a railway, a port and a township; and Halco was to supply the mining, processing and bauxite train equipment. 3.02 At the time the Bank considered that much more detailed engineering would be needed before the project would be appraised, and in early 1966 made a US$1.7 million engineering loan (SI-GUI) for the infrastructure project. When in early 1967 several other large aluminum firms also became interested in purchasing Guinea's high grade bauxite, the project had to be enlarged so as to accommodata the mining and transportation of some 5-6 million tons annually. As a result thereof, Halco was reorganized by including the pros- pective buyers as shareholders in the company. The increased mining capacity called for revision of the engineering plans; this, and the reorganization of Halco, required the settling of many financial and legal issues before the Bank Loan 557 GUI was signed in September 1968. 3.03 In November 1970 Halco and the Guinean Government agreed to increase the capacity to 9.2m tons p.a. This resulted in the need for further Bank finance, provided under Loan 766 GUI in June 1971. Major redesign was not undertaken at this stage, there being sufficient spare capacity in many of the major items. For the factory the construction cost increase for the additional 50% output was estimated at 13%, for the infrastructure at 20%. The fact, however, that this extension was undertaken during the construction period created considerable difficulties for the consulting engineers, both in their design and management function and in the negotiation of the commer- cial content of the resulting variation orders. 3.04 Ever since the start of appraisal, the Bank has tended to treat the project as an enclave, while the Government has pursued wider regional objec- tives when dealing with loans but not when dealing with profits. The exception of this on the Bank side was the insistence that the Government rehabilitate and maintain the road to Conakry. The Government, however, remained consistent in its attitude and did not do it, no money therefore having been provided in the Loan. 3.05 While both approaches are theoretically questionable, that of the Bank was supported by commercial, of the Government by political, expediency. Both the commercial and political situation could be subjected to review  - A.5 - 4. Physical Execution of the Project General 4.01 Consultants Tractionel (Belgium) were retained for the design and supervision of the project. Contracts were awarded as the result of inter- national competitive bidding. The advantages of having the same consultant working for both OFAB and CBG and of using the same contractor wherever pos- sible for the work of the two owners, outweighed the difficulties. Two special measures were necessary, however, because of the lack of relevant expertise in OFAB: (a) a second firm of consultantq,NEDECO (Holland), was retained in the early stages to comment on the Tractionel designs. This was of limited value; and (b) during the construction period a Construction Coordinating Committee (CCC) was formed consisting of one representative each of the two owners and an independent chairman, to resolve possible areas of con- flict. This was both necessary and successful. 4.02 Execution of the project was delayed, first by strikes in Italy affecting the main contractor AMACEW (an international consortium led by Astaldi (Italy) with Constructions et Entreprises Industrielles (Belgium) and Westminster Dredging Co., Ltd. (U.K.) ) , which caused a loss of about 3 1/2 months in the first dry season; then by the political problems follow- ing the invasion of November 1970; by the bankruptcy of the subcontractor for the port and jetty; and by a cholera scare. Operating problems (particularly design problems in the primary crusher) also contributed to the failure to meet bauxite shipping targets. The first shipment was, in fact, made on August 2, 1973 (original target had been July 1, 1972); and total shipments to September 1973 approximated 100,000 tons, as opposed to 5 miLlion tons forecast. 4.03 Although the cost of the works at constant prices was very close to forecast and the time delays basically explicable in terms of special circum- stances such as those indicated above, the execution of the works has raised a number of issues, notably regarding: (a) the ways in which the contracts were split; (b) procurement; (c) bonuses; and (d) supervision; and there remain a number of questions outstanding regarding design. These are given a compendious treatment in a report from the consultants (Final Construction Report 1974, 2 vols.). The most relevant matters are discussed below (Sections 8 and 9).  - A.6 - Port 4.04 The principal civil engineering works for the port, including the dredging, were all completed by mid-May 1972, fifteen months before the first shipment. Early completion was necessary for the erection of the loader and for the handling of equipment required for other parts of the project. For technical reasons some of the specifications for the piles were changed after the main contract had been let. These reasons are fully documented and appear justified. 4.05 The pilot boat, 800 hp tug, buoys and other equipment were supplied according to specification. There have been complaints about the adequacy of the tug, lack of a suitable boat for the buoy master, failure of lights on the buoys, inadequacy of structure of buoys and corrosion of buoy chains. Of these, the problem of the tug and the buoy structures was foreseen and are discussed in &03.1 - 8.03.4, together with the question of the adequacy of the length of the wharf and the depth of the dredged channel which also require comment. Railway 4.06 The railway was substantially completed in November 1972 and was used from then on for the transport of equipment to the site at Sangaredi. The first ore train left Sangaredi on April 26, 1973. 4.07 There was a major derailment in July 1973. The damage was adequately covered by insurance and has been repaired. There have been a large number of welding failures which the contractor is currently correcting at his own ex- pense. 4.08 There has been trouble with some of the signalling equipment, which however was probably unnecessary in the first place and has since been removed without affecting operations. Townships 4.09 Progress in construction was too slow to allow the permanent buildings to make the contribution anticipated to the housing of contractors' staff, which at one time (January 1973) numbered 700. There were several changes in the design of the houses and their furnishings as the needs for economy were balanced against the harsh living conditions. The main current complaints concern sani- tation and water supply. The possibility of problems was foreseen and the rea- sons for the present situation are discussed in 8.03.7 - 8.03.12, 8.03.13, and 8.03.15. 4.10 The actual number of houses is also now considered inadequate. It must be remembered, however, that the original design was only intended to cover that fraction of the final requirements which were necessary for initial operations. There is also a lack of recreational facilities.  - A.7 - Performance of Consulting Engineers 4.11 The fact that the work was completed so close to forecast real prices and that shipments started only 13 months late in spite of various extraneous circumstances is the general measure of the consultants' per- formance. They considered themselves hampered by the Bank's requirements for international tendering, which meant that designs and specifications had to be left sufficiently indefinite to allow a reasonable number of firms to bid their proprietary equipment. This applied to the area of mining and processing as well as the infrastructure. They were critical of the employment of Nedeco as ad- visers to OFAB but appreciative of the work of the CCC in resolving areas of conflict between the owners. Performance of CCC 4.12 The CCC has worked very well, and much of the credit must go to its chairman, Mr. May. The decisions and cost allocations for which it has been responsible have been generally accepted as reasonable by all parties. Performance of Other Consultants 4.13 The experience with the chief accountant allocated to the project was not good. On his leaving, moreover, he was not replaced; Even now, after completion of the works, OFAB has not appointed its own chief accountant, in spite of repeated recommendations from Inbucon, the consultants appointed under the UNDP management project. However, in the opinion of the Government, this vacant position has not created a serious gap in OFAB's management because the functions of the chief accountant are carried out by OFAB's financial director in Conakry and by an expert in Kamsar who coordinate the financial aspects of the project. 4.14 Inbucon itself produced work of varied quality. Its output was no doubt affected by its expectation of an extension of its contract. The nature of the expected extension and the preparations undertaken for it themselves indi- cated a certain remoteness from the most immediate problems -- a concern for the more esoteric aspects of "management" before there was really something or someone to manage. .4.15 The employment of Nedeco to strengthen the Guinean Government in their dialogue with Tractionnel was virtually without effect, and the report they produced is never referred to. Generally, in cases of this sort it is better for the Government to recruit personnel, expatriates if necessary, who will have executive rather than advisory responsibilities over the whole period of the contract. The extent, however, to which such a person will be allowed to exercize such responsibilities depends on the attitude of the Govern- ment which may well be jealous of any diminution of its own authority. This can make the problem of owner/consultant dialogue difficult until the country has sufficiently qualified personnel inside the career civil service. 4.16 The various individual expatriates recruited for the port have per- formed well. There is, however, a problem in the recruitment of a Port Director, who the Government insists should be a civil engineer. In the meantime the Harbor Master is adequately taking care of the responsibilities. 4.17 Rather than recruit individuals, OFAB let the railway management to CANAC, a Canadian firm of railway consultants. While the functioning of the service is satisfactory, too little attention is being given either to training or to cost control. The shortcomings in this latter area are probably due to the relationship between the railway and CBG, which is discussed below (5 .06).  - A.8 - Performance of Contractors 4.18 The performance of the contractors has been generally satisfactory, again as evidenced by the degree to which schedules of cost and time were maintained. This to a certain extent contradicts the feeling expressed by Halco that the main contractor AMECEW 4.02) took advantage of its position both as regards rescheduling of works and as regards variation orders. The opportunities which existed in this respect should in any case be attributed not to superficiality on the part of the consulting engineers but to the re- quirements for partial design imposed by the tendering procedures In another sphere the ability of Astaldi to obtain high productivity from local labor, through piece work based on generous time content, has attracted favorable comment. 4.19 Inadequate inspection in the factory as far as bucket wheels and rail- way wagons were concerned and inadequate supervision of railway welds, have all caused problems which have been or are being rectified at the contractor's ex- pense. 4.20 The best standardization of equipment has not been achieved, partly as a result of one supplier going into liquidation, partly as a result of the way in which the contracts were split. There are undoubted operational benefits to be achieved by the grouping of bought-out items as well as of civil works, and methods of achieving this in a multiple supply and erect situation as in this project would be worth investigating. 4.21 Difficulties encountered in the erection of steel work supplied by Krupp when the Government refused entry visas to all West German personnel were successfully overcome through the cooperation of the various contractors con- cerned. Operation and Prospects 4.22 There are now no physical reasons why the target output of 9m tons p.a. should not be achieved. Both port and railway capacity are adequate, and the problems with the operation of the pan feeders in the plant appear to have been solved. The comments on inadequate preventive main- tenance in the report of December 12, 1974 (Completion Repprt of the Mine and Processing Plant - CW Lorenz) no longer apply, and there is sufficient capa- city to cope with the odd breakdown of unduplicated equipment such as the elevator/tipper. 4.23 The foreign currency requirements of OFAB for the maintenance of the infrastructure are being covered by CBG to the extent that they are not pro- vided for in trust accounts. Should CBG decide to withdraw this assistance, as they are entitled to, the main effects would be felt in the township rather than the production line.  - A.9 - Cost of Project 4.24 The forecast and the actual costs are compared in the table below. While the Bank loan is related to the infrastructure, its justification is based on the project as a whole. Construction (at 1968 prices including 13.5% inflation contingency) Mine Infrastructure Total $m $m $m May 1968 (for 6m tons p.a.) 65 58 123 Final (for 9m tons p.,a.) 77 55 132 General & Finance (1971) 19 14 33 Increases Additional Escalation 19 12 31 Delays 17 4 21 Exchange rates 18 9 27 IBRD interest 1 7 8 Claims 6 Engineering costs 16 Grand Total (plus rounding errors) 275 4-25 The explanations for the increases are as follows: (a) Additional escalation. Inflation was about 36% instead of 13.5% (appraisal figure); (b) Delays have been itemized above. Apart from the actual costs of the results of schedule changes, bonuses had to be negotiated to minimize the overall effect; (c) Exchange rates. This item shows the effect of the depreciation of the dollar; (d) IBRD interest. This came to 21.5% of the construction cost as opposed to 8.5% allowed for. The item also includes about $1m for the finan- cing of a number of mining project contracts; (e) Claims. These are the claims of contractors introduced during the works and accepted by the owners. It should be noted that these are under 4% of the cost of works plus escalation; and (f) Engineering costs. These include owners' expenses. They arise out of the extensions and improvements to the project, the additional escalation and the requirement for additional services.  - A.10 - 4.26 The reduction of the construction cost of the infrastructure arose out of a general scaling down of this element which the Bank insisted on at the time of appraisal. It has given rise to criticism of the Bank which is discussed in Sections 8 and 9. Operating Costs 4.27 It is difficult to make a fair comparison between forecast and actual operating costs because different outputs are being achieved at different points on the learning curve. The following table, therefore, gives only an approximate idea of the situation: Appraisal (at 5m Actual (at 6m tons tons p.a.), per ton p.a.), per ton Variable (labor, materials, etc.) $2.4 $5 Fixed 1.8 3.4 Interest 1.0 2.6 5.2 11.0 4.28 This increase in operating costs is highly comparable with the in- creased capital cost. There is, however, a feeling in Guinea that local wages are being held artificially low. Thus, although the increase in labor cost is in line with the general price increases, in this case it may be due not only to such wage increases as have been awarded but also to a lower-than-expected level of efficiency. On the other hand the exchange rate is artificially high so that the local cost component is overvalued in dollars. A real comparison could only be made if the appraised and current labor contents were expressed in man-hours, and if there were no shift in the labor mix. 5. Operating Forecasts and Results 5.01 There were a number of features of the appraisal report which make comparisons between original and current forecasts difficult and probably non- significant. For instance, the appraisal assumed that the entire construction costs were incurred the year before exports started, and it did not distinguish between the interests of the two groups of shareholders - Halco and the Govern- ment - but rather between the Government and CBG. These matters are dealt with more extensively in Appendix I . The figures given below are, accord- ingly, calculated on a different basis as well as using more recent figures. Appraisal (1971) 1975 Financial RR - Guinea 20.8% 25.95% (without tax) Financial RR - Halco (i) not 13.6 % t If It" (ii) calculated 21.25% Economic RR - Guinea 26% 27.65% (without tax) Project as a whole 19.5% 22.85%  - A.11 - 5.02 The second calculation of the financial RR for Halco assumes that there is an additional benefit to the Halco shareholders of $2 per ton of bauxite arising out of the way in which the transfer price is calculated. This, and the sensitivity analyses therefore, are also discussed in Appendix I- 5.03 The transfer price itself was originally based on the then current Jamaican price with provision for cost escalation only. Although this was specifically pronounced equitable in the appraisal report, market prices have in fact increased more than cost prices, and this would have disadvantaged the Guinean Government. The latter imposed an export tax, which probably over- compensated and resulted in the figures given above, at the same time breaking a covenant to the Bank. 5.04 The tax was, however, imposed not merely to compensate for the un- sensitivity of the price clause to market forces, which could have been achieved by renegotiation, but also to enable the Government to reap the benefits of the price increases over all the bauxite operations in the country. This necessarily calls into question the reasonableness of the expectation of any mineral exploiting company to make an agreement which will give them a differential price advantage over others, either extracting in the same country or even elsewhere in the world. It is not the object of this report to comment on the morality of this situation, which will presumably be increasingly taken into account by the private sector when making investments of this sort. 5.05 The above rates of return are all calculated on the basis of the mineral in the soil having a value of under $lm. The Government export tax, in fact, puts a value of about $50m on the local bauxite deposit. The ignoring of this value obviously has implications both for appraisal and for viable exploitation arrangements. These problems are discussed in Appendix I. 5.06 By separating the infrastructure from the mining operation and pro- viding for its servicing on a direct cost basis, all incentive for economy and control has been lost in this area. In fact, the railway produces no cost accounts at all. This could give rise to difficulties if ever the railway is used for other purposes - another mine or general traffic. The fact that the port charges have to be related to port costs in order to build up sinking funds does not relieve the system of this criticism.; The greater visibility of costs of the port do, however, enable a general judgment to be made. The port is expensive, but areas for saving are limited. The sinking funds, moreover, are being properly maintained. 5.07 Other difficulties are being experienced in those parts of the infra- structure where advance provision for foreign exchange requirements have not been made. OFAB itself has no independent foreign exchange budget and has to seek funds from the treasury. This means that, for all practical purposes, it depends on CBG for overseas purchases. These tend, therefore, to be made either on a crisis basis or on the basis of what CBG, rather than OFAB, considers to be necessary. 5.08 Partly as a result of this lack of independence, OFAB has been unable to engage in, or has been unsuccessful in pursuing, either extensions in its original responsibility (new houses, sanitary services, etc.) or additions thereto (food supply, fishing, etc.). This, however, is fully in accordance with Government policy.  - A.12 - Shipments 5.09 The difference between the appraisal forecast, actual shipments and current forecast is shown In the table below. The causes are partly the con- struction delays, non-availability of ships of correct tonnage and possibly lack of exertion because of the softening of the market. Millions of tons 73 74 75 76 77 78 79 80 Appraisal forecast 4.7 6.8 8 8.1 8.2 9 9 9 Actual .7 3.6 4.9 Current forecast 6.3 8 8.1 8.6 9 6. Organization and Performance of the Borrower 6.01 The Office d'Amenagement de Boke (OFAB) was established by Decree No. 425/PRF of December 31, 1965 and was the body responsible for the formu- lation of contracts and their execution, as well as the administration of the township, port and railway, arising out of the project. 6.02 The fact that OFAB is responsPble to the Minister of Economic Development has created problems with, for instance, the Ministry of Health which would like to have a direct line of authority regarding the hospital; the Ministry of Transport regatding the Port and the Railways; the Ministry of Education regarding the school and training programs; the Provincial Governor regarding planning matters in general. The political ability of the Director has enabled OFAB to discharge its obligations within the limits of the capacity of its personnel, but the existence of OFAB remains in constant jeopardy. 6.03 Its financial automony is rendered illusory by the fact that, although it can dispose of local funds, it has no foreign currency budget. It is thus dependent either on CBG or on specific allocations from the Ministry of Finance. This hinders both its ability to make proper provision for, say, maintenance as well as its ability to undertake improvements or expansion. The Government has no sympathy with the concept that either the project or the district has any preferential claim on the foreign exchange earnings which they generate. The theoretical correctness of this approach overlooks the practical fact that the project creates its own opportunities for invest- ment as well as the nascent human infrastructure for its management. There is therefore considerable frustration both in CBG and, though better hidden, within OFAB itself. 1/ However, according to the Government there have not been conflicts and OFAB has always received cooperation from all ministries whose assistance was necessary for the implementation and continuous progress of the project.  - A.13 - 6.04 OFAB's lack of adequate technical staff led to thetr representation on the CCC by an expatriate. Although the outcome of CCC activity was satis- factory, a Guinean would hav., provided a more visible preserce for the owners as well as continuity of involvement. The OFAB representaeive moreover either did not, was not allowed to, or could not carry Guinean suggestions regarding such matters as the design and furnishing of houses. The result was a depen- dence on the Consulting Engineers which while it may have given an adequate final output, has left the Guineans with the feeling that many things could and should be better than they are. 6.05 OFAB was also weak on the accounting side, and an expatriate was recruited to work as Chief Accountant. The recruitment was not successful, and the accountant was not replaced after he left. The gap remains. 6.06 Consultants (Iribucon) were engaged to assist OFAB ranagement. Many of their recommendations in tne areas of structure and recruitment, as well as of training, 'nave been ignored. The obvious wealmesses on the accounting side however must be attributed iln scme meacure to the lack of impact a management a,:counting systemr would have on the OFAB results. Since they charge CBG for their services at cost, i-i; is importLrit only to know the cost post facto. There is no visIbib