Turkey - Kayraktepe Hydropower Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 15244 PROJECT COMPLETION NOTE TURKEY KAYRAKTEPE HYDROPOWER PROJECT (LOAN 2655-TU) DECEMBER 29, 1995 Infrastructure Division Country Department I Europe and Central Asia 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.  1986 CURRENCY EQUlIVALENTS Currency Unit = Turkish Lira (TL) TL I 100 Kurus (krs) US$1 = TL 674.5 (annual average) TL I = US$.0015 WEIGHTS AND MEASURES kVA = kilovolt ampere kW = kilowatt kWh = kilowatt hour GWh (Gigawatt hour) 1,000.000 kWh MW (Megawatt) = 1,000 kW GLOSSARY AND ABBREVIATIONS DSI Devlet Su Isleri (State Hydraulic Works) GOT = Government of Turkey KGM = General Directorate for State Highway MENR = Ministry of Energy and Natural Resources PCN = Project Completion Note SAR = Staff Appraisal Report TEK - Turkiye Electrik Kurumu (Turkish Electricity Authority) Turkey - Fiscal Year - January I to December 31  FOR OFFICIAL USE ONLY The World Bank Washington, D.C 20433 U S.A. Office of the Director-General Operations Evaluation December 29, 1995 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Note on Turkey Kayraktepe Hydropower Project (Loans 2655-TU, B014 and B015) Attached is the Project Completion Note (PCN) on Turkey: Kayraktepe Hydropower project (Loans 2655-TU, B014 and B015, approved in FY86) prepared by the Europe and Central Asia Regional Office. No comments were obtained from the Borrower or co-financiers. The project was to be financed by loan 2655-TU (US$200 million) and two "B" loans: B014 (US$33 million) and B015 (US$19.2 minillioni), related to two commercially syndicated loans in Yens (about US$133 million) and Eurodollars (US$233 million). The three loans were closed on December 31, 1994. While the "B" loans were fully disbursed, only US$11.7 million was disbursed from loan 2655-TU, and the balance was canceled at closing. The project's objective was to develop indigenous energy resources by building a 420 MW hydroelectric plant which would also provide flood protection to the city of Selifke and 5,000 hectares of irrigated agricultural land. The project was not implemented and its objectives were not met. This is because projection of demanld for expanded power proved overoptimistic, leading the Government to postpone the project's execution. The "B" loans, however, helped meet Turkey's balance of payments needs. The Operations Evaluation Department (OED) rates the project outcome as highly unsatisfactory and its institutional development as negligibie. Since the project was not implemented, its sustainabilitv is not rated. OED also rates Bank performance as highly unsatisfactory because Bank staff overestimated the demand for power in Turkey and failed to build into project design appropriate environmental anid resettlement safeguards. The PCN presents a concise account of events that led to project cancellation. It highlights one important lesson: the need for a realistic economic analysis on which to ground the power demand forecast defininig the least-cost expansion of the generation system. No audit is planned. 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 TURKEY KAYRAKTEPE HYDROPOWER PROJECT LOAN 2655-TU Table of Contents Preface ........................................... i Evaluation Summary .. Part I: FINDINGS AND LESSONS . I A. Background ..................................... I B. Project Objectives and Definition ....... ................ I C. Achievement of Project Objectives ....... ............... 2 D. Major Factors Affecting Project ........ ................ 3 E. Bank Performance . ................................ 6 F. Government Performance .......... .................. 7 G. Lessons to be Learned . .............................. 8 Part 11: STATISTICAL INFORMATION ....... ................. 8 Table 1: Summary of Assessments ........ ................ 9 Table 2: Bank Power Loans to Turkey Since 1985 .... ........ 10 Table 3: Project Timetable ............ ................. 13 Table 4: Disbursements by Category ....... ............... 14 Table 5: Cumulative Disbursements ....... ............... 15 Table 6: Statue of B Loans (August 1994) ...... ............ 16 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 wiLhout World Bank authorization.   PROJECT COMPLETION NOTE (PCN) TURKEY KAYRAPTEPE HYDROPOWER PROJECT LOAN 2655-TU Preface This is the Project Completion Note (PCN) for the proposed Kayraktepe Hydropower Project (420MW), for which Loan 2655-TU in the amount of US$200 million was approved on February 25, 1986 and made effective on March 26, 1986. The borrower was the Government of Turkey (GOT). The Bank also helped finance this project by participating in two commercially syndicated co-financing operations (dated April 2, and April 7, 1986, respectively)-- a Eurodollar loan (US$233 million) and a Japanese Yen loan (Yen 23.4 billion or about US$133 million based on the then prevailing rate of exchange: US$1.00:Y175.94). The Bank's participation in the Eurodollar loan amounted to US$33 million; in the Yen loan, about US$19.2 million. As the project was never carried out, except for some consultant studies, the loan was closed on November 23, 1993--about thirteen months ahead of the original closing date of December 31, 1994--and the undisbursed balance of US$188.3 million was canceled. Preparation of the PCN was begun during the Bank's final supervision mission (May 9- 17, 1994). It is based on material in the project file researched by Antanasije Kocic, Task Manager, ECIIN. The PCN was edited by Andrew Waldrop, LAIIE. The GOT provided no comments to an earlier draft of the PCN.   ii PROJECT COMPLETION NOTE (PCN) TURKEY KAYRAPTEPE HYDROPOWER PROJECT LOAN 2655-TU Evaluation Summary Note: Because the Government of Turkey (GOT) did not execute the proposed Kayraktepe Hydropower Project, this report is a Project Completion Note (PCN), the focus of which is on the respective performances of the Bank and the GOT. A. Background Power Loans to Turkey. Since 1985, the Bank has made eight power loans to Turkey amounting to US$1.57 billion. In connection with the Kayraktepe Hydropower Project, the Government of Turkey (GOT) was the Borrower of Loan 2655-TU (US$200 million, 1986, an "A" Loan) and of two commercially syndicated co-financing loans (US$366 million, 1986, "B" loans.) (para. I of Part 1). B. Project Objectives and Definition. Guidelines, Objectives and Definition. In 1985, the GOT established guidelines for sector expansion, including that, inter alia, priority should be given to domestic sources of energy, especially hydro and lignite, provided that they were economically justified. The objectives of the project were in line with this guideline: they intended to maximize use of indigenous energy resources by developing generating facilities at Kayraktepe on the Goksu River and to provide flood protection for the city of Silifke and adjacent farmland. The project was to consist of a dam and a powerplant (Part A) and technical assistance (Part B). On the basis of the information available at appraisal, the project objectives and definition were compatible with sector guidelines. However, if the project had been carried out on the timetable initially conceived, it would have turned out to be premature, unresponsive to substantially altered circumstances, and therefore inappropriate (paras. 2-5). Project Design. The primary implementing agency was to be the General Directorate of State Hydraulic Works (DSI). As coordinated by the DSI, other governmental units were in charge of resettlement, environmental protection and road relocation. Upon project completion, the DSI was to transfer ownership of the power plant to the Turkish Electricity Authority (TEK).  iii C. Achievement of Project Objectives Parts A and B. Except for some engineering studies under Part A, the project was never executed. Hence, this report has no basis for commenting on its economic achievement and sustainability of its objectives, or its overall outcome (paras. 7 & 8). Timetable. The Loan Agreement became effective on March 26, 1986. During the next three years, the Bank and the Borrower took three years (1986-1989) to reach agreement on the civil works bidding documents; and during the next three years (1990-1992), they unsuccessfully wrestled with the problems of restructuring the project. In 1994, having accepted the Bank's views on the need to reappraise the project in connection with the approval of a new Bank loan, the GOT requested that the Bank close Loan 2655-TU as of November 23, 1993. Total disbursements had amounted to US$11.2 million for finance charges and US$0.5 million for engineering studies. The balance of the US$200 or $188.3 million was cancelled (paras. 10-12). D. Major Factors Affecting the Project Demand Shortfall, " B" Loans, & Organizational Coordination. The most important factor accounting for abandonment of the project was the shortfall in the growth of demand. This led the GOT to the view that completion of approved generation projects, including the proposed Kayraktepe project, might be deferred because of continuing macroeconomic difficulties and a much lower demand growth rate than was expected when the project was prepared. This is also the reason why it took three years to reach agreement with the Bank on civil works bidding documents (para. 9). The most important factor accounting for the long delay on the part of the Borrower and the Bank in cancelling Loan 2655-TU was the Government's interest in not stimulating the co- lenders to seek accelerated repayment of their loans. In connection with "B" loans, it was standard practice for the proceeds of such loans to be transferred very rapidly to the Borrower's central bank once the loan agreements were signed. This happened in the case of the "B" loans associated with Loan 2655-TU and acted as an incentive to restrain the Bank from cancellation of the "A" loan, as the Bank was sensitive to the country's balance-of-payments problems. The Bank then spent three years (1990-1992) trying to restructure the project. Finally after some delay in 1994 the Turkish Treasury no longer viewed the "B" loans as a problem and the Bank proceeded to cancel the "A" loan (paras. 13-18). On the part of the GOT, there was insufficient institutional interest in completing the project--that is, in addition to the shortfall of demand, the project suffered from poorly structure arrangements for implementation. As project coordinator, the DSI did not perform strongly and its leading role was not much supported by other government agencies or their ministries because of their reduced or qualified interest in completing the project. The sum and substance is that the DSI had insufficient stake in completing the project (para. 19).  iv E. Bank Performance Appraisal. The Bank did not prepare: a thorough economic justification of the project, including an adequate demand forecast or realistic arrangements for project implementation. Also there were not, by current standards, adequate provisions for resettlement and environmental protection but standards were not as rigorous at that time (1980's). Further, the "B" loan agreements constrained the Bank from taking timely action with respect to closing its loan (paras. 20-25). Implementation. The Bank tolerated poor performance on the part of the GOT. At the heart of the Bank's indecision was its sensitivity to Turkey's constrained foreign exchange situation which would be worsened by a possible acceleration of repayment of the "B" loans (para. 26). F. Government Performance Lack of Commitment After Appraisal. The attitude of the GOT toward the project varied, reflecting, apparently, its changing expectations with respect to trends in electricity demand and its changing needs for foreign exchange, including needs arising from non-project transactions. In retrospect, it seems clear that, from an early date--perhaps at appraisal, perhaps a little after--the GOT planned to avail itself of the option of using the proceeds of the "B" loans for non-project purposes if circumstances permitted such use (para. 27). G. Lessons to be Learned This PCN confirms the importance at appraisal of the Bank's ensuring that: (i) the power project is grounded on a proper economic justification, including a sound demand forecast and least-cost exercise; (ii) a detailed implementation plan is included in the SAR, including observance, where applicable, of existing norms for resettlement and environmental protection; and (iii) the agency most concerned with implementation has a larger stake in the completed project (paras. 6, 19). The PCN also confirms the shortcoming of the former practice of transferring the proceeds of the "B" loans outside the control of the lenders prior to disbursement for project needs.   PROJECT COMPLETION NOTE (PCN) TURKEY KAYRAKTEPE HYDROPOWER PROJECT (LOAN 2655-TU) PART I: FINDINGS AND LESSONS A. Background 1. Bank Power Lending to Turkey. The Bank has made twenty power loans to Turkey amounting to US$2.1 billion, including eight loans since 1985 amounting to US$1.57 billion. In connection with the Kayraktepe Hydropower Project, the Government of Turkey (GOT) was the Borrower of Loan 2655-TU (US$200 million, 1986, an "A" loan') and of two related commercially syndicated co-financing loans (about US$366 million, 1986; "B" loans-- para. 13). B. Project Objectives, Definition, and Design 2. Energy Sector Guidelines. In 1985--the year of project appraisal--the GOT, to overcome a deficit in the supply of electricity, established the following sector expansion guide-lines: (a) priority was to be given to domestic sources of energy, especially hydro and lignite, provided that they were economically justified; (b) imported energy was to be considered to meet the energy gap; (c) renewable and nonconventional resources were to be supported; and (d) private capital was to be sought for participation in energy development. The proposed Kayraktepe Hydropower Project was identified as one of several options compatible with sector guidelines because of its comparative advantage relative to other generation projects and because of the reduction of dependence on imported energy which it would result. But, its selection did not involve the measure of certainty which would have prevailed if it had been identified as the "next step" of a system-wide, least-cost expansion plan. The Bank referred to Loan 2655-TU as an "A" loan and the two commercially syndicated co-financing operations as "B" loans. The Bank established the "B" loan program in the early 1980s to provide credit enhancement on loans by commercial banks to developing countries. The main feature of the program was that, in addition to making its own loan for an investment project or program (the "A" loan), the Bank would participate along with commercial banks in a second loan (the "B" loan) for the same project or program.  2 3. Project Objectives. The objectives were to assist the GOT to maximize use of indigenous energy resources by developing hydroelectric generating facilities at Kayraktepe on the Goksu River in south-central Turkey and to provide flood protection to the city of Silifke and 5,000 hectares of adjacent irrigated agricultural land. An unstated but evident objective was to improve the selection process for power projects by the adoption and implementation of an appropriate technical/economic methodology. 4. Project Description. The project had two parts. Part A was to consist of a rockfill dam and power plant (420MW). The expected construction time, completion date, and cost were. respectively, 90 months, December 1993, and US$537.3 million. Part B was to consist of technical assistance, primarily studies. Benefiting the Ministry of Energy and Natural Resources (MENR), the first study (or group of studies) would complete the national energy profile of river basins and rank hydroelectric sites in order of priority on the basis of a uniformly applied technical and economic methodology. The second study would provide the agency mainly responsible for project implementation (see para. 6) with appropriate cost accounting procedures. plus electronic data processing equipment and software. The expected completion date and cost were December 1989 and US$5 million. 5. Evaluation. On the basis of information available at appraisal, the project objectives and definition were compatible with the sector guidelines, and, as set forth in the loan documents, they were clearly stated. However, if the project had been carried out on the timetable initially conceived, it would have been premature, unresponsive to substantially altered borrower circumstances, and therefore inappropriate. By not carrying out the project, the GOT avoided this error, but its performance was characterized by indecision and delay. 6. Project Design: Government Agencies Concerned with Implementation. The Loan Agreement assigned the responsibility for implementing different project components to different state agencies. For example, the General Directorate of State Hydraulic Works (DSI), a department under the Ministry of Public Works, was the main implementing agency for Part A. Besides being responsible for construction, DSI was to co-ordinate other government agencies involved with implementation, including: (i) the Ministry of Archaeology, Forestry and Rural Affairs (MAFRA)--responsible for resettling people displaced by the project; (ii) the Ministry of Environment--responsible for environmental matters; and (iii) the General Directorate of Highways (KGM)--responsible for relocating about 100 km of road. If the project had been completed, the power plant would have been transferred to and operated by the Turkish Electricity Authority (TEK), which is under the MENR. C. Achievement of Project Objectives 7. Part A. The GOT became hesitant and indecisive about Part A, which, after much waivering, it did not construct. The initial factor causing the GOT to hesitate was a short-fall in the growth of electricity demand (para. 9). Later, additional factors--financial problems (paras.  3 10 & 13-18) and poor organizational coordination due to faulty project design (para. 19) -- contributed to the GOT's indecision. As Part A was never implemented, except for some engineering studies, this report has no basis for commenting on its economic justification, achievement and sustainability of its objectives, or its overall outcome. 8. Part B. The GOT did not implement Part B; hence, this report has no basis for assessing the achievement and sustainability of the corresponding objectives. D. Major Factors Affecting the Project 9. Demand Shortfall. At appraisal, having experienced chronic supply shortages in the early 1980s (and before), the GOT was strongly committed to implement the proposed project in the shortest possible time. However, shortly after Loan 2655-TU became effective (1986), the GOT came to the view that completion of approved generation projects, including the proposed Kayraktepe project, might be deferred because of continuing macroeconomic difficulties and a much lower demand growth rate than that which was expected when the project was prepared (see Schedule 1). Besides moving very slowly on procurement2, the clearest indicator of the GOT's attitude was its not allocating funding for Kayraktepe in the annual budget. Schedule I Actual and Forecast TEK Sales Levels, 1986-1994 (in thousands of GWh) Growth Rates 1986 1987 1988 1989 1990 1991 1992 1993 1994 1986-89 1989-94 Actual 30.0 33.9 36.8 41.1 44.0 47.9 52.9 58.6 64.8 10.9 9.5 Forecast 30.7 37.9 45.6 49.5 51.6 55.3 59.1 64.8 71.1 17.2 7.5 % Diff. from -2 -11 -19 -17 -15 -13 -10 -10 -9 Forecast 10. In February 1989, the GOT requested a reappraisal of the project. Because such action might lead to complications with co-lenders, the initial Bank reaction was not favorable (see para. 13), but later (February 1990), the Bank reviewed the project and confirmed its economic validity. In 1990, as demand growth began to recover, the GOT showed more interest in the project and tried to identify a private investor in connection with a Build-Operate-Transfer (BOT) scheme. After about two years (1991-1992), this effort proved unavailing with the result that the GOT decided to continue with the project as originally agreed; but, as before, it did not 2 In 1986-1989, the DSI took 21 months (to December 1987) to complete the prequalitication process for the main civil works contractor and 34 months (to December 1988) to reach agreement with the Bank on the draft civil works bidding documents. The Bank and DSI also reached agreement in 1 987/88 on the terms of reference for the hydro ranking study and had successful discussions on procurement of computers.  4 allocate funding in the annual budget and DSI achieved very little in terms of actual project progress.3 This wavering attitude prevailed until 1994 when the GOT requested the closing of Loan 2655-TU, retroactive to November 23, 1993. At loan closing, US$11.2 million had been disbursed for interest and finance charges and US$0.5 million for engineering studies for Part A; the balance of the US$200 million or $188.3 million was canceled. 11. Underlying the request were the GOT's: (i) concurrence (July 1993) with the Bank's views on reappraisal and the issue of related Bank financing; and (ii) decreasing interest in the "B" loans, since both were being repaid on schedule (para. 18). Concerning re-appraisal, it was the opinion of the Bank that, given the seven year delay with respect to construction start-up, a re-appraisal was necessary to take account of cost increases, new and stringent environmental standards; and the increased scope of the resettlement program. Concerning Bank financing, the issue was whether to extend the closing date on Loan 2566-TU or to make a new loan. The GOT understood that a total reevaluation of the project was required no matter what the decision was on Bank financing. 12. It was the opinion of the Bank that a new loan would be preferable to extending the closing date of Loan 2566-TU because the loan terms no longer suited project needs. For example, that portion of the loan allocated for financing interest and other charges (US$55 million) would not be available after April 1994; and, under the existing amortization schedule, repayments would be occurring. 13. Co-financing Arrangements. In addition to the "A" loan--Loan 2655-TU--the Bank also helped to finance the proposed project by participating in two commercially syndicated co- financing operations--a Yen loan amounting to Yen 23.4 billion or about US$133 million4 and a Eurodollar loan amounting to US$233 million. The Bank's participation in the Yen loan amounted to about US$19.2 million; in the Eurodollar loan, US$33 million. The co-financing loans were signed on April 2 and April 7, 1986, respectively. Within days of the signing, the total proceeds of the "B" loans were transferred to project accounts at the Central Bank of Turkey. For "B" loans, such rapid transfers were normal. 14. Under the "B" loan agreements, the Bank was obligated to make available to the other "B" lenders, upon their request, information received from the Borrower concerning the service of the "A" loan and the implementation of the project. The Bank was also obligated to afford In 1991, having updated the project cost estimate, DSI began to prepare new prequalification documents to prequalify bidders again. The prequalification process was completed in April 1993. Also, a substantial effort was made to prepare a revised resetdement plan; and an Environmental Assessment Impact Report was initiated. Otherwise, by November 23, 1993, the loan closing date, except for completion of DSl's construction camp, no further project progress was achieved. 4 The prevailing rate of exchange was US$1.00 = Yen 175.94.  5 them a reasonable opportunity to exchange views prior to the Bank's taking any action which might lead to the suspension, cancellation or accelerated repayment of the "A" loan. Also, the Bank was to review the annual audit of the "B" loan accounts at the Central Bank. If the Bank determined that any sum had been disbursed for an ineligible expenditure, the Borrower, following notice from the Bank, was to restore such sum. However, the "B" Loan Agreements specifically exempt the Bank from "any responsibility as to the Borrower's Application of the Loans. " 15. The Bank's initially unfavorable response to the request for reappraisal (February 1989) had its origins in concern about the possible response of the co-lenders. For example, if the Bank decided to cancel the loan, such action might cause the co-lenders to require accelerated repayment of their loans. As Turkey was already experiencing balance-of-payments problems, the Bank did not want to complicate an already complex situation by creating the potential for additional pressure on Turkey's limited store of foreign reserves. Reinforced by frequent requests from the GOT, this approach and its rationale--deferring action in order to accommodate the foreign exchange needs of the GOT--came to prevail for the next several years. 16. Notwithstanding the above, the Bank sent a letter (dated June 26, 1989) to the Treasury, requiring receipt of the audited project account within 30 days. In January 1990, the Bank received such report prepared by a reputable international accounting firm It confirmed that the loan proceeds had been used to pay for financial expenses and that the balance of the loan proceeds had been placed in the project account. This report was also sent to the co-lenders. 17. Bank files contain five audit reports. The first one is dated April 28, 1989 and the last one is dated October 15, 1991. The audit report dated April 28, 1989 stated that there had not been any payment made from the accounts by the DSI. The second audit report dated November 6, 1989 stated that the accounts were audited for the years 1986, 1987, and 1988 in conformity with generally accepted accounting principles, and that Yen5.826 billion and US$68.3 million had been disbursed as of January 18,1990, and that these disbursements were found to be in conformity with the terms and conditions specified in the loan agreements. The third audit dated April 18, 1990 stated that an authorized DSI official declared that no amount was spent by DSI in 1989 on the project. The audit report dated May 3, 1991 concluded that DSI effected no spending from the accounts in 1990. However, the auditor could not reach an opinion as to whether the disbursements had been utilized in compliance with relevant loan agreements. The last Bank supervision mission (June 1993) found an apparent shortfall in the account of US$122.8 million and requested the Treasury to ensure that loan accounts were reimbursed to the original disbursed amount and confirm to the Bank that it had taken this action as soon as possible. The Bank project files reveal no evidence of any Treasury action. From the preceding, it could be concluded that part of the "B" loans were utilized for purposes other than the Kayraktepe project. 18. As already stated, by 1994 the Treasury no longer viewed the "B" loans as a major problem because: (a) more than half of the loans had already been repaid; (b) continuing to use  6 their proceeds would only be useful in conjunction with a new complete project financing plan, with or without Bank financing; and (c) the key co-lenders confirmed that they would give favorable consideration to a request not to exercise their option to accelerate the repayment of the "B" loans. 19. Organizational Coordination. As project coordinator, the DSI did not perform strongly; and its leading role was not much supported by other government agencies or their ministries because of their reduced or qualified interest in completing the project. For example, the interest of the agencies charged with resettlement, environmental protection, and road relocation was limited, as they had no stake in the completed project. Even the DSI had no more than a bureaucratic interest in completing the project, as it would not share in the project benefits after the transfer of the power plant to the TEK. Further, the TEK, which would own and operate the power plant, had less incentive to support hydro projects, the revenues of which had to be shared with a fund promoting infrastructure development--the Public Participation Fund. Since it did not have to share the revenues of thermal plants, the TEK gave priority to their completion over the completion of hydro plants. E. Bank Performance 20. Preparation and Appraisal. Looking at the Bank's performance with the benefit of hindsight, there are a number of elements where the appraisal effort was inadequate--demand forecasting and least-cost planning, "B" loan arrangements, project design, and planning with respect to the issues of environmental protection and resettlement. For example, one of the main causes of the project failure was the shortfall of demand as compared with what had been expected (para. 9). Demand forecasts are always a critical issue in the planning of power projects. However, the Staff Appraisal Report (SAR) for the Kayraktepe project stated that " ..a major constraint in identifying the costs and benefits of a loan management program has been the inadequate data base upon which to prepare demand projections." It is surprising that after several Bank loans, this was still an issue and during appraisal more effort was not spent on such a critical subject. Instead, a consultant was hired under the technical assistance component of another Bank-financed project to ensure that pricing policy, demand management and system planning were fully integrated. The work was started in March 1985 and the completion was expected by the end of 1987. There are no references in the documentation to the results of the study. 21. The SAR reports on the lack of knowledge and experience in power system planning as well as a lack of system data. However the Bank did not prepare a least-cost investment program (para. 2) in order to see the role and rank of the project within the context of system needs. At present, a large generation project that lacked such analysis would probably not be considered acceptable. 22. When the Bank inaugurated the "B" loan program in the early 1980s, it did not foresee that the transfer of the proceeds of the "B" loans to the Borrower's Central Bank (or equivalent) might, in certain circumstances, constrain the Bank from taking timely action with respect to its loan, as in fact happened (paras. 13-15).  7 23. The lack of GOT commitment to the project was at least partially attributable to the organizational arrangements which placed the primary responsibility for project construction on an agency (the DSI) which was not a stakeholder in the completed project. At appraisal, the Bank should have found a way to involve the TEK more closely in the construction of the project (paras. 6 & 19). 24. At appraisal, the Bank's environmental and resettlement guidelines were less demanding than they are today, and there were no Turkish environmental requirements. In the early 1990s, as the Bank and the GOT were trying to start the project, the Bank required the preparation of an environmental impact assessment report, which was also required by the Ministry of the Environment (established in 1992) (para. 10, footnote 3). The SAR did not provide adequate safeguards for the execution of proper enviromnental safeguards or the preparation of a comprehensive resettlement program. 25. It is important to note that the Bank made an extra effort to increase lending to Turkey in general and to the energy sector in particular during the period 1985-1988 and that in doing so, some projects were appraised in a relatively short period of time, This may help explain but does not justify why some of the policy and project aspects were not analyzed as thoroughly as is required by standard practice. 26. Implementation. The Bank's performance during implementation was characterized by indecision and delay in the face of an admittedly complex situation. Clearly, the Bank should not have accepted that three years were taken by DSI to reach agreement on the civil works bidding documents. The same is true with respect to the three years during which the Bank and the GOT discussed the problem of restructuring the project. Further, the Bank should have made a more timely and determined effort to audit the "B" loan project accounts at the Central Bank. At the heart of the Bank's indecision was its sensitivity to the problems for Turkey which might result from a possible acceleration of repayment of the "B" loans. The Bank's indecisiveness on the cancellation of the loan was influenced by repeated Government requests to postpone the decision in order to avoid a possible acceleration of repayments of the "B" loans. F. Government Performance 27. Lack of Commitment After Appraisal. The attitude of the GOT toward the project varied, reflecting, apparently, its changing expectations with respect to trends in electricity demand and its changing needs for foreign exchange, including needs arising from non-project transactions. In retrospect, it seems clear that, early on, the GOT considered the option of using the proceeds of the "B" loans for non-project purposes if circumstances allowed such use.  8 G. Lessons to Be Learned 28. This PCN confirms the importance at appraisal of the Bank's ensuring that: (i) the power project is grounded on a proper economic justification, including a sound demand forecast and a least-cost exercise (paras. 20-21); (ii) a detailed implementation plan is included in the SAR, including observance, where applicable, of existing norms for resettlement and environmental protection (paras. 24); and (iii) the agency most concerned with implementation has a larger stake in the completed project. Clearly, item (iii) also requires that the Borrower's commitment to the project and financing arrangements not be qualified or conditioned by other objectives (para. 23). The PCN also confirms the shortcomings of the former practice of transferring the proceeds of "B" loans to the country prior to the need for such loans for project purposes (paras. 22 & 26-27).  9 PART 1I. STATISTICAL INFORMATION Table 1: Summary of Assessments A. Achievement of Objectives Substantial Partial Negligible Not applicable - Part A X - Part B X B. Project Sustainability Likely Unlikely Uncertain Not Applicable - Part A X - Part B X C. Assessment of Outcome X Highly Highly Unsatisfactory Satisfactory Unsatisfactorv UnSatisfactory D. Bank Performance X E. Borrower Performance X  10 Table 2. Bank Loans for Power in Turkey Since 1985 Amount T Loan Number (in US$ Year of & millions) Purpose Approval Status Project I I I 2586-TU; TEK 142.0 To expand the high 1985 Loan closed on 12.31.91, a Transmission voltage interconnected year late. Project was IV system of the Turkish completed in 1992; almost two Electricity Authority years late. Delay was due to (TEK) and thereby poor compliance with Bank ensure supply of procurement guidelines. electricity in response to Project cost amounted to growing demand and to US$268 million or 29% more increase TEK's network than expected. Due to low planning and design electricity rates, economic capacity (lines & benefits of project are substations). underestimated. The outcome of the project is rated marginally satisfactory. The institutional development impact of the project is rated as negligible. 2602-*TU; 140.0 To assist TEK in (i) 1985 Loan closed 12.31.93, three Power System achieving efficient years late. The project was Operations utilization of its power completed in 1994, four years Assistance generation, transmission, late. Delay was due to and distribution facilities; frequent changes in TEK and (ii) ensuring management resulting in - lack reliability of electricity of commitment to the project supply by upgrading and its objectives. Also operational and contributing to delay were maintenance procedures. financial constraint. The PCR could not establish the final project cost. While the physical objectives of the project were partially met, the project outcome is rated as marginally unsatisfactory and its institutional development is .________________ .___________ __________________________ _______n eg lig ib le .  11 Table 2: Bank Loans for Power in Turkey Since 1985 (cont'd) Loan Number Amount & (in US$ Purpose Year of Status Project millions) Approval 2650-TU; 10.0 To complete construction 1986 Loan closed 12.31.88, six Elbistan and completion of months late. The project Operation & Elbistan units 3 & 4. consisted of the third and Maintenance fourth units of the Elbistani Assistance power plant and improvements in TEK's operaticn and maintenance capability at the plant. The overall outcoilie of the Elbistan Project, which was initially supported by Loan 1023-TU (US$148 million, 1974) is rated unsatisfactory due to uncertaini operational pertformance and negligible institutional | developmient. 2655-TU, B014 200.0 To help the Government 1986 Loan closed 11.23.93, about & B015; 33.0 to maximize the 13 months ahead of schedule. Karaktepe 19.2 utilization of indigenous T'he project was never Hydropower energy resources to implemented because the Project provide flood protection Government, in circumstances to the city of Silike and of declining electricity growth adjacent 5,000 hectares due to poor macroeconomic of irrigated farmland. conditions, becarne increasingly less interested in the project. 2750-TU; Sir 132.0 To assist CEAS in 1986 Loan closed 12.31.92; as Hydropower jmaxiniizing the planned. Project completed Project utilization of its I December 1991 or about 9 indigenous energy imonths late. Project cost resources through the US$306 million, an 18% development of overrun, mainly due to price hydroelectric generating rises double the estimated on the Ceyhan River in 15 % price contingency. The response to growing overall project outcome is demiiand for additional rated as satisfactory and its power; and to encourage sustainability is likely. the development of Institutional development is i private sector rated as substantial. I participation in the sector. .~_ _ i _ _ _ _ , ,_ _ ___ _ _ _  12 Table 2: Bank Loans for Power in Turkey Since 1985 (cont'd) Amount I Loan Number (in US$ Year of & millions) Purpose Approval Status Project I I I I 2856-TU; 325.0 To support the 1987 Loan close 09.30.93, three Energy Sector continuation of the years late. Bank loan Adjustment Government reform consisted of two tranches, Loan program in the energy fast disbursing component sector, especially policy (US$175 million) supporting measures which improve the purchase of general the efficiency of energy imports; and a project- production and usage and financing component (US$150 which assist the million) financing physical Government in its effort investment, studies and to facilitate private technical assistance to eleven investment in the sector. energy sector state enterprises. Associated with Bank loan were two "B" loans amounting to US$500 million. Reforms in the energy sector were not achieved; the project outcome is rated as unsatisfactory. Institutional development was negligible. 3345-TlJ; TEK 300.0 To support TEK Loan to be restructured to Restructuring corporate restructuring as reflect division of TEK into part of large state two enterprises. enterprise reform program and to implement least cost expansion plan. 3476-TU; 270.0 To assist the Government 1992 Project not completed; loan Berke to accelerate development closed and undisbursed Hydropower of hydro potential and to balance canceled. Project expand the role of private capital. TOTAL 1,571.2 J  13 Table 3. Project Timetable Event Date Planned Date Revised Actual Date Identification/ 01/855 Preparation Appraisal 04/85 04/85 Negotiations 09/85 12/85 12/17/85 to 01/28/86 Board Approval 12/85 03/86 03/03/86 Signature 03/25/86 Effectiveness 03/26/86 Implementation 02/86 09/86 NA (Construction start-up) Loan Closing 12/31/94 11/23/93 11/23/93 Last Disbursement 10/19/94 Project Completion 12/31/93 12/31/95 NA Prior to the mission, the Electric Power Resources Administration had contracted with a foreign engineering consulting firm to complete a feasibility study and final designs for the project  14 Table 4: Actual and Forecast Disbursements by Category (in US$ millions) Category Actual Forecast 1. Civil Works. including materials and equipment: 0.0 124.00 2. Consultant Services 0.50 10.00 3. Data Processing Equipment 0.00 1.50 4. Interest & Other Charges I I.2() 55.00 5. Unallocated 0.00 9.50 Total 11.70* 200.00 * Does not include IUS$ 10.0 millioni disbursed to special account, which amount was repaid when loan was closed.  15 Table 5: Actual & Forecast Cumulative Disbursements Year End Actual * Forecast Act./Fore. (%) (---in US$ millions---) 1986 0.68 1.2 57 1987 2.29 17.4 1 3 1988 4.78 48.6 10 1989 7.29 92.6 8 1990 9.17 133.2 7 1991 10.26 165.1 6 1992 10.73 187.6 6 1993 11.19 198.6 6 1994 11.70 200.0 6 * Does not include disbursement of US$ 10.0 million to special account.  16 Table 6: Statue of B Loans (August 1994) LOAN AGREEMENT LOAN AMOUNT PREPAID [ PRINCIPAL DATE | 233 MILLION $ PRINCIPAL ($) |_OUTSTANDING ($) April 7, 1986 200 Million $ (The 195,000,000 5,000,000 part leaded by ABC including NIB contribution) 33 Million $ (IBRD 33,000,000 Part, Loan B-15) r Loan Agreement Date Loan Amount 23.4 Repaid Principal Principal Billion Yen (Yen) Outstanding (Yen) April 2, 1986 Tranche A+B 10,800 Billion 9.2 Billion Totalling to 20 Billion Yen Tranche C, IBRD 3.4 Billion Part Totalling to 3.4 Billion Yen (Loan B-l . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 1 4 ) m \ak\t.r\pcn 621   IMAGING Report No: 15244 Type: PCN