Turkey - Istanbul Water Supply and Sewerage Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 18419 IMPLEMENTATION COMPLETION REPORT REPUBLIC OF TURKEY ISTANBUL WATER SUPPLY AND SEWERAGE PROJECT (Loan No. 2888-TU) September 22, 1998 Infrastructure Sector Unit Europe and Central Asia Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS (Yearly Averages) Currency Unit Turkish Lira 1987 - 855 = US$ 1.00 1988 - 1,421 = US$ 1.00 1989 - 2,120 = US$ 1.00 1990 - 2,606 = US$ 1.00 1991 - 4,168 = US$1.00 1992 - 6,864 = US$ 1.00 1993 - 10,965 = US$ 1.00 1994 - 29,668 = US$ 1.00 1995 - 45,731 = US$ 1.00 1996 - 80,785 = US$ 1.00 1997 - 150,000 = US$ 1.00 1/31/98 Closing Date 218,600 = US$ 1.00 WVEIGHTS AND MEASURES Metric System US System 1 hectare (10,000 square meters) 2.47 acres I meter 39.37 inches 1 square meter 10.76 square feet 1 cubic meter 264 US gallons FISCAL YEAR OF BORROWER January 1st to December 31st ABBREVIATIONS AND ACRONYMS ISKI - Istanbul Water and Sewerage Administration DSI - State Hydraulic Works IB - Iller Bankasi SAR - Staff Appraisal Report SPO - State Planning Office KHGM - General Directorate of Rural Services PMU - Project Management Unit PPF - Project Preparation Facility Mm3 Million m3 Vice President: Johannes F. Linn, ECA Country Director: Ajay Chhibber, ECCO6 Sector Director: Ricardo A. Halperin, ECSIN Sector Leader: Walter A. Stottmann, ECSIN Team Leader: Bemardo Gomez, ECSIN  FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT REPUBLIC OF TURKEY ISTANBUL WATER SUPPLY AND SEWERAGE PROJECT (Loan No. 2888-TU) CONTENTS PREFACE .......................................i EVALUATION SUMMARY .......................................... PART I. PROJECT REVIEW FROM BANK'S PERSPECTIVE ........................................1 A. Introduction ........................................1 B. Project Objectives ........................................2 C. Project Financing Plan .........................................3 D. Implementation Experience ........................................4 E. Project Accomplishments .7 F. Factors Affecting Outcome ......................9 G. Assessment of Outcome ....................... H. Borrower Performance ..................... 12 I. Bank Performance ..................... 12 J. Sustainability ...........13 K. Future Operations ...................... 13 L. Lessons Learned ...................... 14 PART II. STATISTICAL TABLES ..................... 15 Table 1: Summary of Assessments .16 Table 2: Related Bank Loans/Credits .18 Table 3: Project Timetable .19 Table 4: Loan Disbursements: Cumulative Estimated and ActuaL .19 Table 5A: Key Indicators for Project Implementation .20 Table 5B: Status of Completion and Operation .21 Table 6: Key Indicators for Project Operation .22 Table 7: Studies Included in Project .22 Table 8A: Project Costs. 23 Table 8B: Project Financing .23 Table 9: Project Costs and Benefits .24 Table 10: Status of Legal Covenants .25 Table 11: Compliance with Operational Manual Statements .26 Table 12: Bank Resources: Staff Inputs .26 Table 13: Bank Resources: Missions .27 Table 14: ISKI's Income Statements (1987-1997) .28 Table 15: ISKI Balance Sheets (1987-1997) ...................... 29 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Table 16A: ISKI Key Indicators for Project Performance and Operation (1987-1994) ................ 31 Table 16B: ISKI Key Indicators for Project Performance and Operation (1994-1997) ................ 32 APPENDICES: A. Mission's Aide-Memoire B. Borrower Contribution to the ICR C. Borrower Comments to the Draft ICR D. Maps: IBRD 29702 & 29703  IMPLEMENTATION COMPLETION REPORT REPUBLIC OF TURKEY ISTANBUL WATER SUPPLY AND SEWERAGE PROJECT (Loan No. 2888-TU) PREFACE This is the Implementation Completion Report (ICR) for the Istanbul Water and Sewerage Project in Turkey, for which Loan No. 2888-TU in the amount of US$218.0 million equivalent was approved on December 1, 1987 and made effective on February 23, 1988. The loan was closed on January 31, 1998, compared to the original closing date of December 31, 1995 after two extensions of one year each and a final extension of one month. The last disbursement took place on July 31, 1998. US$215.5 million of the loan was disbursed and US$2.5 million cancelled. The ICR was prepared by Daniel Coyaud (ECSIN), Alptekin Orhon (ECCTR) and Herman Nissenbaum (Consultant). It is based on material in the project file. The Borrower contributed to the ICR by preparing its own evaluation of the project implementation, and by commenting on the draft ICR. The Guarantor also commented on the draft ICR.  IMPLEMENTATION COMPLETION REPORT REPUBLIC OF TURKEY ISTANBUL WATER SUPPLY AND SEWERAGE PROJECT (Loan No. 2888-TU) EVALUATION SUMMARY 1. Introduction. In the 1 980s, the Republic of Turkey set improvement targets for water supply and sewerage, for which the Government raised sector investment from 5 percent of the total public investment program in 1984 to 9 percent over 1985-89. Water and sewerage targets established by the Government were: (i) by 1990, the total urban and rural population should have adequate water supplies; (ii) by 1990, the total rural population should have satisfactory excreta disposal facilities; and (iii) by 2000, the total urban population should have satisfactory sewerage and sewerage disposal facilities. Accordingly, the Istanbul Water Supply and Sewerage General Directorate (ISKI) undertook a program to meet the city's rapid growth and remedy service deficiencies. T'he sewerage components of the Project would continue the pollution abatement program initiated under the Istanbul Sewerage Project (Loan 2159-TU). 2. Project Objectives. Project objectives were to: (a) improve and extend sewerage services to 75 percent of Istanbul by 1996; (b) enhance sewage disposal in order to protect the waters and shorelines of the Sea of Marmara and the Bosphorus; (c) reduce unaccounted-for water; and (d) strengthen ISKI's operations, maintenance and industrial waste management capacities. Its project components included: (a) Water distribution - replacement of some 1,200 km of water pipes; rehabilitation of some 20 pumping stations; installation of a supervisory control system; (b) Sewage collection and treatment - construction/rehabilitation of pumping stations, collectors, treatment plants in five drainage zones; construction of about 2,500 km of street sewers; and installation of about 330,000 service connections; and (c) Technical Assistance for institutional development. 3. The Project, with a total cost of US$557 million, was only part of ISKI's over- ambitious 1987-94 investment program amounting to US$3.1 billion, including physical investments and working capital requirements. The Project was ambitious as it called for building four sewerage treatment plants while ISKI was only building one when the Project began (and it was also still executing the Sewerage Project (Loan 2159-TU)). In addition, only a year after Loan 2888-TU was signed, the Project was enlarged to assume substantial unfunded works from the previous Bank-financed project. The combination of these investments proved to comprise an unduly heavy workload, straining ISKI's financial and implementation capacity. Also, ISKI was then still maturing after its establishment in 1981.  - iii - 4. Implementation Experience. After several important components started and proceeded satisfactorily, the works slowed after local elections and changes in ISKI's management team in 1989. More active project execution proceeded in the next year. At the same time, however, ISKI's finances deteriorated and it launched an aggressive policy in 1991 to increase its customer base. This succeeded, but the new practice also triggered deteriorating collections, illegal connections and invalid billings which began to crimp revenues. This led to increased short term borrowings and arrears on supplier and contractor payments and affected the availability of counterpart funding for the Project. Severe droughts in the early 1990s also depressed water sales and increased operating costs. In 1993, matters became even more disrupted with the imprisonment of ISKI's General Director on corruption charges, after which most project activity stopped for almost 1 1/2 years. 5. In mid-1994, sewage treatment was considerably behind schedule and it was clear that only part of the planned facilities could be completed by the scheduled end CY95 Closing Date. Unaccounted-for water had not been reduced and financial problems had grown. Operating costs had more than doubled since 1987 and there had been significant tumover in ISKI's Board and in its managerial and professional staff. These deteriorating conditions led to a mid-term Bank review. 6. This review concluded that the Project was too ambitious as ISKI was unable to mobilize the necessary counterpart funds and also lacked implementation capacity. Agreements were made to scale down the scope of the Project. A new action plan recast ISKI's financial and operational performance targets, calling for phased: (a) reductions in operation ratios and unaccounted-for water; (b) increases in tariffs, in the percentage of completion of remaining works, and in self-financing; and (c) reductions of accounts receivable and increases in the debt service coverage ratio. At the same time, the Loan's Closing Date was extended to end CY96 and the Bank assumed a larger share of project costs. After this, ISKI provided sufficient counterpart funds to carry out the scaled-down Project scope, and project implementation generally improved. 7. Project Accomplishments. The Project fell short of achieving its originally intended goals for major civil works, and institutional and financial strengthening. Three new sewage treatment plants were built out of four originally planned and partial improvements in two drainage zones were carried out. Replacement of water supply pipes and pumps occurred two years late with longer delays on works in three drainage zones. Progress on other aspects is indicated in the following figures for 1987 and 1997 respectively: * The sewerage network's length was increased some 45 percent, from 5,650 km to 8,191, slightly more than the amount estimated during appraisal. The provision of sewerage services increased from coverage of 55 percent of the population to 65 percent, appreciably less than the targeted 75 percent. The volume of  -iv - appropriately treated sewage disposed of increased from 57Mm3/year in 1989 to 133 Mm3/year in 1997. * ISKI's customers increased 150 percent (from 1.0 to 2.5 million). The number of employees/thousand water connections fell from 6.3 to 3.0. The number of firms monitored for wastewater discharges more than doubled, rising from 3,245 to 8,341, a multiple of the SAR target. At the beginning, there were no industrial waste discharge permits in force; at the conclusion there were 582. 8. Relatively little was accomplished on the planned institutional development although ISKI did successfully absorb the Supervisory Control and Data Acquisition System (SCADA). The entity also carried out useful studies on the Yenikapi effluent discharge, on expansion of the Istanbul sector master plan, and on alternative treatment plants. The first mentioned responded to the reservations raised about sewage disposal into the Bosphorus. It successfully supported the design of the outfalls (built under the prior project) and the Bank's reluctance to finance secondary treatment facilities, due to lack of economic justification. 9. The Project's internal rate of return (together with ISKI's other investments during the period) was 6.4 percent, half of the forecast 13.6 percent result. This rate of return, however, does not take into account the full economic benefits of the project as it does not reflect positive externalities such as property value increases, and health and productivity gains. 10. Factors Affecting Outcome. ISKI was intended to be fully autonomous and administratively self-sufficient--but instead it became part of the Istanbul political structure. It is governed by the Metropolitan Municipal Council of which, Istanbul's Mayor is the Chairman. The Council appoints all Board members, approves ISKI's investment plans, budgets, staffing, tariffs and borrowings, while the Board approves operating rules and senior staff appointments. This governance structure successfully engaged local interests into sector operations. Unfortunately, it also exposed the entity to partisan, often short-term, political influences which stood against establishing a clear commercial orientation for the utility and caused discontinuities in ISKI's staff and repeated project delays. 11. Another aspect which adversely affected the project outcome concerned ISKI's finances, as the provisions in the project financing plan (ISKI's 1987-94 financing plan), proved to be unduly optimistic, particularly regarding ISKI's internal cash generation. In fact, ISKI's gross internal cash generation was less than one third the amount estimated at appraisal (US$71 1.1 million actual vis-a-vis 2,520.0 million estimated at appraisal). This shortfall was mainly the result of actual water sales (1.9 million m3 ) substantially below appraisal estimates (3.4 billion m3). Other estimates were also on the "optimistic side", although by a lesser margin; actual revenues per m3 sold: $0.84 actual vis-a-vis US$0.91 estimated at appraisal, and total operating costs during the period: US$1.0 billion actual - vis-a-vis US$0.9 billion estimated at appraisal.  12. Assessment of Outcome. The Project's results were mixed. The positive outcomes were significant: increasing wastewater collection coverage; increasing the volume of treated wastewater discharged in the Bosphorous and Sea of Marmara which helped the environment; reducing the number of employees per water connection; and more effectively enforcing the compliance with regulations on wastewater discharge. However, the Project did not fully achieve its objectives as not all the works planned were satisfactorily completed or on time. In addition, the agreed institutional strengthening measures fell short of what was required and their poor implementation failed to stem the problems that ISKI confronted. On balance, what was accomplished was positive, especially under the circumstances of an over-dimensioned project and over-estimated executing agency's implementation capacity. Ideally, the project should be rated as "marginally satisfactory." However, since such a category does not exist in the evaluation process, in a conservative judgement the project is rated "unsatisfactory." 13. ISKI's limited sewerage accomplishments had some explanation, as ISKI focused more on water supply vis-a-vis sewerage improvements in the wake of the extended droughts through 1994, and in view of rapid population growth. 14. Borrower Performance. The Borrower's performance under the Project was uneven. Difficulties in project implementation during the initial years (1989-93) included mounting operating costs, poor bill collection and undesirable project design changes. ISKI had also poor project planning and cost accounting. After the 1994 project restructuring, however, ISKI's performance improved. This was evident in the better conduct and pace of project implementation, and ISKI's rigorous pursuit of stronger finances, e.g., abolition of the free water program, temporary forgiveness on late bills' interest, containing operating costs, substantially increased water billings and reductions in unaccounted-for water. 15. Bank Performance. The Bank's performance was similarly mixed. It correctly identified the critical development requirements at the outset but failed to identify and address adequately the main institutional issues that stood in the way of a sound commercial operation. The overestimation of water sales was a major deficiency of the project appraisal, as it led to the endorsement of an investment program that was not financially sustainable. The staff made suitable judgments on location of the large sewage interceptors and sewage treatment plants, as well as preliminary designs of treatment processes. The mid-term mission produced valuable stocktaking and course redirection (although its occurrence in the last year of a scheduled 8-year project execution period was tardy). During 1988-1992, the Bank could have held more supervision missions to address problems such as low tariffs, inadequate collection of bills and accounting deficiencies. 16. Sustainability. Project sustainability is uncertain. ISKI, on the one hand, has gained significant experience in the development and operation of water and sewerage  - vi - infrastructure. It also has shown the capability to increase water and sewerage service coverage in a rapidly growing city like Istanbul. ISKI has also accumulated a valuable knowledge of the environmental issues affecting the city, and of the most appropriate solutions to these issues. ISKI, on the other hand, continues confronting serious institutional and financial weaknesses, which raise concerns about the likelihood of the entity being able to adequately operate and expand the existing infrastructure. A fundamental concern relates to the continued vulnerability of ISKI to the impact of political changes in city government. Ultimately, a sustainable sound operation will require fundamental changes in ISKI's governance structure. 17. Future Operations. There are no plans for a Bank-financed follow-up project. The Bank did offer to consider a project in connection with significant institutional changes, e.g., private participation in service operation and delivery. This did not evoke a positive response. 18. Lessons Learned. (i) Utilities that are only nominally autonomous, and are subject to partisan, often short-term political interests are unlikely to provide good quality services on a sustainable basis. These fundamental issues need to be fully resolved before Bank lending is provided. Alternatives to unsatisfactory governance should be sought and implemented. The selected alternative should ensure that the utility's board is directly accountable to broad segments of the society, and not only to short-term elected or appointed officials, and that the Board and management have adequate incentives, job security, and autonomy. Available alternatives range from full privatization of the services to public utilities with boards of directors selected by a variety of municipal stakeholders such as non-governmental organizations, business associations, universities, research centers and municipal authorities; (ii) Projects where the planning assumptions regarding the values of significant variables such as revenues and costs are on the "optimistic side" are likely not only to experience financial difficulties during implementation but also to yield lower than expected economic and financial rates of return. Project appraisal should ensure that the assumptions are realistic and that contingency plans are drawn for coping with deviations from the most likely range of outcomes. The range of outcomes with their respective probabilities, however subjective, should be listed in the appraisal document; (iii) It is highly risky to attempt to compensate for past under-investment in and neglect of water and sewerage services by undertaking investment programs which are overly ambitious in relation to the utility's proven managerial and financial capabilities. Project appraisal should make sure  - vii - that required investments are adequately staged and that Bank loans generally finance only those investments that can be realistically implemented during the grace period of the loan; (iv) The decision process regarding the design and location of sewage treatment plants should include early consultations with affected populations, and all issues regarding final site selection should be fully sorted out before Board presentation; and (v) A timely mid-term review could have assisted in addressing some of the major project issues and project outcomes. At the time of project preparation, mid-term reviews were not Bank requirements and one was not planned in the initial project design.  IMPLEMENTATION COMPLETION REPORT REPUBLIC OF TURKEY ISTANBUL WATER SUPPLY AND SEWERAGE PROJECT (Loan No. 2888-TU) PART I. PROJECT REVIEW FROM BANK'S PERSPECTIVE A. INTRODUCTION 1. In the 1980s, the Republic of Tirkey set improvorment targets for water supply and sewerage, for which the Government raised sector investment from 5 percent of the total public investment program in 1984 to 9 percent over 1985-89. Water and sewerage targets established by the Government were: (i) by 1990, the total urban and rural population should have adequate water supplies; (ii) by 1990, the total rural population should have satisfactory excreta disposal facilities; and (iii) by 2000, the total urban population should have satisfactory sewerage and sewerage disposal facilities. 2. As a part of this effort, the Istanbul Water Supply and Sewerage General Directorate (ISKI) undertook a large investment program to meet the challenges presented by the city's rapid growth and to remedy serious service deficiencies. Istanbul's population increase (4.3 percent per year average in 1980-85) had exceeded new piped water connections since 1'983. ISKI's figures showed that only about 55 percent of the 5.7 million city residents was connected to public sewerage facilities. In addition, metering of water production and usage was deficient and 50-60 percent of estimated production was not billed (in part due to a generous program of "exemptions" to a variety of consumers). 3. ISKI's task moreover was complicated by the metropolis' terrain, which made it difficult to dispose of sewage in ways protecting the waters and shorelines of the adjacent Sea of Marmara and the Bosphorus. Istanbul's growth had overtaxed sector services, resulting in unsatisfactory maintenance of old, frequently poorly built sewer lines and septic tanks. Sewage was discharged into the nearest receiving waters, usually at the shore lines, causing heavy pollution. About 90 percent of the sewage effluent still then went untreated into the Golden Horn (a narrow inlet on the European side of Istanbul) and the Marmara Sea; the remainder into the Bosphorus. These conditions caused such problems as the contamination of the popular Marmara beaches which a long discussed, repeatedly revised Istanbul Sewerage Plan sought to alleviate. It was decided to remedy these difficulties by discharging treated sewage effluent into the Bosphorus and/or the Sea through submarine outfalls.  -2 - 4. This solution was however the subject of persistent popular doubts and debate in the city, as were the design, site and output of various planned sewerage facilities. There was then heightened environmental sensitivity in Istanbul, as evidenced by the interventions of Turkish environmental authorities and scientists in the issues. A key, often contentious subject was concern over whether the currents of the Bosphorus and the Marmnara could dilute sewage effluent without environmental damage. The scientific and engineering concept of the 1983 Istanbul Master Plan held that they would. But this was challenged by ISKI's incoming administration after the 1989 elections on the basis of some beliefs that the lower and upper Bosphorus currents mixed, thereby bringing significant sewage volumes back into the Marmara. ISKI's environmental problems moreover were worsened by the fact that its planned works sometimes affected congested areas, often close to historic buildings. B. PROJECT OBJECTIVES 5. Project objectives were to: (a) improve and extend sewerage services to 75 percent of Istanbul by 1995 from 55 percent in 1987; (b) to provide for appropriate disposal of sewage in order to protect the waters and shorelines of the Sea of Marmara and the Bosphorus; (c) reduce unaccounted-for water; and (d) strengthen ISKI's operations, maintenance and industrial waste management capacities. The sewerage components of the Project would continue the pollution abatement program initiated under the Istanbul Sewerage Project (loan 2159-TU). 6. The project components included: (a) Water distribution-replacement of some 1,200 km of water pipes in the Asian distribution network; rehabilitation of some 20 pumping stations; installation of a supervisory control and data acquisition system; (b) Sewage collection and treatment-construction/rehabilitation of pumping stations, collectors, pre- and secondary treatment plants in five drainage zones (Atakoy, Kucukcekmece, Tuzla, Kadikoy-Pendik and Uskudar); construction of about 2,500 kIn of street sewers; and installation of about 330,000 service connection; and (c) Staff training, other technical assistance and equipment for ISKI's institutional development. 7. These constituted an ambitious volume of activity for ISKI, as reflected in the Project's estimated US$557 million cost. Moreover, this was only a relatively modest portion of the entity's overall US$3.1 billion investment program, including working capital requirements, for the 1987-94 period. These investments were almost eight times greater than its 1982-87 program. Similarly, Loan 2888-TU called for ISKI building four sewage treatment plants while it was only building one when the Project began. Also, it was still busily executing the Istanbul Sewerage Project (Loan 2159-TU), on which  -3 - disbursements continued for almost two years after the new Loan became effective. Moreover, only a year after Loan 2888-TU was signed, the Project was enlarged to encompass the construction of some 15 km of sewage interceptors and tunnels from Kabatas to Baltalimani, and a sewage treatment plant and sea outfall at the latter. (This was a carryover from Loan 2159-TU, after a re-estimation of project costs turned out to be erroneous, and newly added components lacked funding.) Unhappily, the combination of these investrnents proved to comprise an unduly heavy workload, straining ISKI's financial and implementation capacity. 8. They also posed a formidable challenge for a relatively new, still maturing institution. ISKI had only been established in 1981, in which process it had to merge 24 municipal systems besides Istanbul's. The appraisal team judged that its creation successfully contributed to strengthening sector organization and commended the institution for strengthening its management (which the SAR termed a "model" for the Turkish sector), as well as its industrial waste control and training programs. The SAR discussed other aspects of ISKI's operations without noting any weaknesses either, although the project scheme did call for upgrading its operational capabilities. In keeping with this favorable assessment, the SAR forecasts that ISKI's strength could help contain the Project's financial risks of inadequate tariff hikes and cost overruns. The actual project experience, however, demonstrated that these assessments were unduly favorable. C. PROJECT FiNANCING PLAN 9. Another aspect of project design which adversely affected the outcome concerned ISKI's finances. The Loan was partly justified by the premise that it would reaffirm "support for Istanbul's strong cost recovery policy for the sector." This followed from the Government's intention (with Bank endorsement) in decentralizing water and sewerage activity in 1980 to have locally generated funds cover an increasing part of investments. This was founded in part on the belief that Istanbul's water sector had substantially untapped financial capacity, which the Project would be a good vehicle for testing. The change also reflected the concems of Central Government authorities about containing Turkey's fiscal deficit. 10. Consequently, ISKI's overall 1987-94 financial plan called for ISKI to internally generate cash amounting to US$2.7 billion or 88 percent of total investments (US$3.1 billion, including working capital). To this end, the entity indicated that it intended to maintain the average 1987 tariffs and other charges in real terms. The financial plan called for ISKI financing 62 percent of the total cost of the Bank-financed project. The funding plan thereby contrasted sharply with that in the Sewerage Project (Loan 2159- TU), which originally called for ISKI's only covering operation and maintenance, and debt service expenses. Loan 2888-TU's provisions accordingly did not involve any Government contribution for project financing. 11. Unfortunately, the provisions in ISKI's 1987-94 financing plan regarding internal cash generation proved to be unduly optimistic as seen in the table below.  -4- Actual SAR (1987-94) (1987-94) (US$ million) Volume of Water Sold (million mi3) 1,913.0 3,426.0 Total Operating Revenues 1,607.3 3,130.2 Total Operating Cost 991.0 874.0 Net Income from Operations 616.3 2,256.2 Depreciation 95.1 263.9 Gross Internal Cash Generation 711.4 2,520.1 Total Operating Revenue per m3 sold (US$) 0.84 0.91 Total Operating Cost per m-3 sold (US$) 0.52 0.26 12. The shortfall in gross internal cash generation was mainly the result of actual water sales substantially below appraisal estimates. The appraisal called for water sales amounting to 3.4 billion m3 in 1987-94 while actual sales amounted only to 1.9 billion m3 or 56 percent of the appraisal estimate. This large difference in volume of water sold, together with revenues per m3 sold slightly lower than expected (actual US$0.84 vis-a-vis US$0.91 at appraisal) resulted in total operating revenues equivalent to about half of the appraisal estimate (US$1.6 billion actual vis-a-vis US$3.1 estimated at appraisal). Actual total operating costs (US$1.0 billion) were somewhat higher than appraisal estimates (US$0.9 billion). The combined result of the shortfall in water sales and revenues per rn3 and the higher than expected operating cost was a reduction in ISKI's cash generation for debt service and contribution to investments from US$2.5 billion estimated at appraisal to US$0.7 billion actual. D. IMPLEMENTATION EXPERIENCE 13. The implementation of the Istanbul Water Supply Project was uneven, with substantially varied degrees of work activity. Several important components, e.g., the Uskudar zone works and several large diameter sewers, the installation of a new supervisory control and data acquisition systems (SCADA), started promptly on schedule and proceeded satisfactorily. However, the works slowed after local elections and changes in ISKI's management team in 1989. The new officials questioned the design for disposal of pre-treated sewage in the lower current of the Bosphorus and the level of treatmnent for the plants located on the Marmara Sea's shores (for the reasons indicated in paras. 3-4). Fortunately though, more active project execution proceeded in the next year when work on most of the main drainage areas resumed. This excluded the Kadikoy and Baltalimani treatment plants whose proposed locations sparked other local disputes and suggestions of different sites. 14. At the same time, ISKI's financial situation became so weak that the entity's management launched an aggressive policy starting in 1992 to increase its customer base  -5 - and thus its revenues. The number of customers increased by 13 percent per annum through 1993 compared to 3 percent earlier. However, the new practice also soon triggered deteriorating bill collection, illegal connections and invalid billings which began to crimp ISKI's revenues. This in turn led to the expediencies of increased short term borrowings, arrears on supplier and contractor payments, and non-debt service. This is reflected in ISKI's negative working capital (current assets minus current liabilities) every year starting in 1990. Total current and overdue liabilities, which had increased from US$72 million at the end of 1988 to US$207 million at the end of 1991 peaked at US$469 million at the end of 1993. ISKI's current and overdue liabilities with Treasury increased from US$8.5 million at the end of 1990 to US$45.4 million at the end of 1993. The cash shortfalls also significantly weakened ISKI's ability to provide the requisite counterpart funding for the Project, as well as increased its debt service burden. 15. In the early 1990s, severe droughts depressed water sales and increased operating costs. Furthermore, they lowered the levels of the raw water reservoirs on the European side of the Bosphorus even more than usual, necessitating costly water imports and cloud seeding. The situation caused ISKI to re-focus priorities and to allocate more resources to water supply than sewerage. 16. During 1990-93, the Bank and ISKI had additional disagreements on the design of project works, which resulted in delays in project implementation. ISKI's operations were even more disrupted in mid-1993 with the imprisonment of ISKI's General Director, on corruption charges. The investigations which followed found improprieties in certain contracts (none of which were related to Bank-supported procurement). During the process, the entity's accounts were frozen for almost a year by the courts. The main project execution activity stopped for almost a year and a half under an interim management, while ISKI's creditworthiness also deteriorated. 17. After several years of start-and-stop cycles and numerous deficiencies (especially during 1989-93), the Bank classified the operation as a "problem project." Sewage treatment was considerably behind schedule. It was clear that only part of the planned facilities could be completed by the scheduled end CY95 Closing Date. Unaccounted-for water had not been reduced and financial problems had grown. Operating costs had more than doubled since 1987 (from US$0.34 to US$0.79 per m3 of water sold). For financial reasons, the entity had terminated the contracts of the engineering firms supervising the sewerage activities. It instead enlisted local universities to bolster ISKI's staff, which did not provide an adequate replacement. ISKI was then only building the Tuzla sea outfall, and the loan substantially lagged in its scheduled disbursements. All these worsening conditions brought the Project's future course into examination, for which a late 1994 Bank mission conducted a mid-term review. 18. The mission concluded that, as originally designed, the Project was too ambitious, as ISKI was unable to mobilize the necessary counterpart funds and also lacked implementation capacity to carry out the full scope of planned activity. It was agreed that the Project should be scaled down and measures taken to improve ISKI's financial  - 6 - situation while it reduced its debt. More realistic investment priorities were designed. The number of sewage treatment plants to be built was cut from four to three (Uskudar, Baltalimani and Tuzla), and all new work on two additional plants (Kadikoy and K.Cekrmece) was eliminated. A 1994-97 Action Plan also recast ISKI's financial and operational performance targets to bring improved functioning over the next three years. 19. The 1994-97 Action Plan envisaged the completion of the Bank-financed project with investments amounting to US$187.1 million, and investments outside the project amounting to US$1.2 billion. The investment plan was to be financed largely (81 percent) through loans and to a lesser extent (19%) through internal cash generation. Actual investments amounted to only US$483.6 million. The overall estimates of internal cash generation were more realistic, although with substantial differences in individual items between the Action Plan and actual results. For example, total operating revenues were substantially overestimated, costs were also overestimated (this may, in part, be explained by the evolution of inflation and exchange rates). The net result was gross internal cash generation amounting to US$592.5 million vis-a-vis US$705.0 million estimated in the Action Plan. The table below shows the comparison between the Action Plan and actual results: Actual | Action Plan (1994-97) (1994-97) (US$ million) Volume of Water Sold (million m3) 1,179.0 1,214.0 Total Operating Revenues 1,037.6 1,602.0 Total Operating Cost 519.1 958.0 Net Income from Operations 518.5 644.0 Depreciation 74.0 61.0 Gross Intemnal Cash Generation 592.5 705.0 Total Operating Revenue per m3 sold (US$) 0.88 1.32 Total Operating Cost per m3 sold (US$) 0.44 0.79 20. ISKI's 1994-97 investment and financing plan was accompanied by a set of financial and operational performance parameters to bring about its improved performance. The parameters involved items such as: (a) reductions in operation ratio, in operating cost per m3 of water sold, and in unaccounted-for water; (b) increases in tariffs, and in the percentage of completion of remaining works under the Bank-financed project; and (c) increases in self-financing, collection ratio and debt service coverage ratios. 21. An imaginative feature of the Action Plan defined rating criteria for these parameters in terms of deviations against set goals and assigned different weights for each parameter. ISKI and the Bank also agreed on the levels by which the resulting scores would be considered adequate for purposes of gauging performance, against the backdrop of pursuing 25-50 percent average improvements quarterly in the target areas.  This performance target system replaced the Loan's major financial covenants which ISKI had been unable to fulfill (minimum self-financing of capital expenditures and limits on accounts receivable). Table 16B includes the comparison between targets and actual results for the parameters in the Action Plan. 22. The Loan's Closing Date was extended to end CY96 with the possibility of a subsequent additional year's financing with better project performance. The Bank likewise agreed to assume a larger share of project costs. For its part, ISKI agreed to augment the project supervision force by having the universities employ consulting firms for complementary expertise, and they performed satisfactorily. After the restructuring, ISKI also provided sufficient counterpart funds to carry out the reduced scope of the Project. On the basis of the changes, project implementation generally improved thereafter through the closing of the Loan on January 31, 1998. E. PROJECT ACCOMPLISHMENTS 23. The Project fell short of achieving its originally intended goals for major civil works and for institutional and financial strengthening. Three new sewage treatment plants were built out of four originally planned. The Atakoy works were not implemented; the Zeytinbumu pumping station was rehabilitated outside of the Project; Kucukcekmece was not implemented due to the substantial design and location changes and it was agreed to exclude it from the Project since there was not enough funds available to finance it. Kadikoy-Pendik drainage zones were only partly finished. Replacement of water supply pipes and pumps occurred two years late, with even longer delays on works in the Uskudar, K. Cekmece and other drainage zones. Some components implemented (e.g., North Halic and Baltalimani zones) were added to the original scheme while some equally new Tuzla works extended into 1998. But overall, less was done than planned. Progress on other aspects is indicated in the following figures on key yardsticks, which are stated with regard to 1987 and 1997 respectively (ref. Table 6 for fuller details): (a) Water and Sewerage Operations -- The sewerage network's length was increased some 45 percent, from 5,650 km to 8,191, slightly more than the amount estimated during appraisal. -- The number of sewerage connections rose from 521,000 (55 percent of the populace) to 762,000 (65 percent), below the desired 75 percent.  - 8 - -- The level of unaccounted-for water was reduced by half (60 percent to 30 percent). This fell short of the original project target of 23 percent but surpassed the 38 percent goal set in the mid-term revisions.' -- The volume of appropriately treated sewage disposed of in the Bosphorus and the Sea of Marmara increased from 57/Mm3/year in 1989 to 133 Mm3/year in 1997 (no target for this objective was set at the outset). -- ISKI's customers increased by 150 percent (from 1.0 to 2.5 million). -- ISKI's operational and maintenance efficiency improved significantly as the number of its employees/thousand water connections fell from 6.3 to 3.0. -- The number of firns monitored for wastewater discharges more than doubled, rising from 3,245 to 8,341, a multiple of the SAR target. -- At the beginning of the project, there were no industrial waste discharge permits in force; at the conclusion there were 582. (b) Finances -- ISKI's self-financing percentage rose from 22 percent to 70 percent, in contrast with the revised 31 percent target for 1997. But customer contributions provided only eight percent of project financing instead of the projected 15 percent. -- The debt service coverage ratio rose from 1.60 to 2.38, better than the 2.27 goal set under the midterm revisions. -- The operating margin (operating revenues per m3 sold minus operating costs per m3 sold) rose from the equivalent of US$0.11 to US$0.78 per m3 sold, better than the target of US$0.71 for 1997. 24. Relatively little was accomplished on the planned institutional development although ISKI did successfully absorb the recommended supervisory control and data acquisition system (SCADA), which now practically covers its entire water production, 1 Although some reduction in the unaccounted-for water level most likely took place during project implementation, there is no readily available evidence to support remarkable achievements in this area. This is particularly true for the claimed reduction between 1994 and 1997 (from 50 to 30 percent). The substantial upward and downward variations during project implementation also suggest caution in using these figures. A telling example of these variations is that the Staff Appraisal Report is based on 30 percent unaccounted-for water in 1987 with a target of 23 percent by the end of the Project in 1994. This estimate was later revised to 60%, without satisfactory explanation.  - 9 - transmission and distribution functions. Also, ISKI did avail itself of the technical assistance resources to carry out useful studies on the Yenikapi effluent discharge, expansion of the Istanbul sector master plan and pre-feasibility study of alternative sewage treatment plants on the Asian side (ref. Table 7). The first mentioned study responded to the reservations raised by ISKI's post-1989 management regarding sewage disposal into the Bosphorus (ref. para. 13). The findings successfully confirmed the design of the outfalls (built under the prior project) and supported the Bank's reluctance to finance proposed secondary treatment facilities (although it was agreed to design the Baltalimani plant for this possibility later). 25. The internal rate of return of the Project combined with ISKI's other investments during the project period was 6.4 percent or less than half of the forecast 13.6 percent result. The rate of return was reestimated using the same methodology as in the SAR, where incremental sales resulting from ISKI's overall investment program were taken as an indicator of benefits for rate of return calculation. The methodology is weak for several reasons. First, by relying on actual tariffs it fails to capture consumer surplus benefits. Hence, it cannot be regarded as a measure of economic benefits. In addition, the full economic benefits should take into account such important factors as resultant property value increases, health improvements and productivity gains (which were not calculated). However, all these qualifications cannot contest the finding that benefits were significantly lower than expected, largely because sales volumes fell significantly short of the projected level. F. FACTORS AFFECTING OUTCOME 26. One significant element in determining the Project's fate was ISKI's management structure. It was created to be separate, fully autonomous and administratively self- sufficient---but instead became enmeshed in the Istanbul political structure. It is governed by the Metropolitan Municipal Council of which, Istanbul's Mayor is the Chairman. The Council appoints all Board members, approves ISKI's investment plans and budgets, staffing, tariffs and borrowings, while the Board approves operating regulations and senior staff appointments. This management structure successfully engaged local participatory interests into sector operations but, unfortunately, it also exposed the company to the harmful effects of strong partisan, often short-term political influences at the municipal level. These caused discontinuities in ISKI's staff (even down to middle management levels) and disruption in its investment program. There were repeated project implementation delays on the occasions of municipal elections, and ISKI had four changes of administration during the 10-year period. 27. Project implementation was also disrupted when ISKI's new 1989 management team questioned the design of the planned facilities for sewage disposal, the degree of treatment by the plant on the shores of the Marmara, and the locations of Kadikoy and K. Cekmece sewage plants. This problem was compounded by ISKI's management's failure during project preparation/execution to win the agreement of residents on the location of the sewage treatment plants. There were also arguments over whether surface  - 10- runoff sewage from the City was substantially retuniing to the Sea of Marmara, which kindled demands for additional tests and analyses. 28. The counterpart funding problems which plagued project execution suggest that expecting ISKI to achieve financial self-sufficiency in a short period of time was unrealistic, as it failed to consider social and political resistance to higher tariffs. The Sewerage Project PPAR (Loan 2159-TU) reported on this aspect quite comprehensively, examining ISKI's financial performance during roughly the same years as Loan 2888's. Its conclusions are equally applicable here. It stressed that the institution early on assumed a very large development program, and went through a rapid pace of buildup in the 1980s. At the end of 1990, its financial statements showed growth rates in double digits for all major components of its balance sheet and income statements in real terms (save for share capital). However, these rates slowed markedly thereafter. 29. Further, it may appear from statistics on ISKI's self-financing of capital expenditures that the entity achieved the Central Government's objective of municipal self-sufficiency. The figures are deceptive since they mask the underlying frequently low level of investments in execution. In addition, the Central Government stepped in to service most of ISKI's debt. 30. Moreover, the PPAR for Loan 2159-TU demonstrated that ISKI's financial policy in the late 1980s was to rely on borrowings to finance its investments (90.5 percent of ISKI's 1986-90 investments were debt financed). It failed to adjust tariffs quarterly to compensate for the large inflation at the time, and its annual hikes were generally below the costs of services. ISKI did not make the required increases in 1987-89; managed to do so in 1990-93 (save for 1992); but later failed to raise tariffs to the levels targeted for 1997 (both in real terms). On this, there is some belief that ISKI had no incentive to raise its tariff levels at the same time that the Central Government was subsidizing large water production operations in Istanbul without seeking any recovery of capital investments. It might also have been difficult to do so simultaneously with the politically inspired free water program that benefited a wide segment of the population and not only the poor as originally intended. Other factors hindering cash generation included the overestimated water sales projections and poor collections. In addition, ISKI's incomplete sewerage improvements limited the cash contributions which would have come from real estate developers' fees for the networks. 31. ISKI's financial management did not progress to the same standards as its technical side. It had persistent difficulties with billings as well as collections during the Project's execution. Regrettably as well, its employment of a semi-private collection agency midway through the project to improve matters was equally unsuccessful. ISKI also was unable to achieve the covenanted confinement of accounts receivable to two months billings.  - 11 - 32. ISKI followed Bank's Procurement Guidelines in a satisfactory manner. ISKI, however, was not satisfied with the Bank's requirements as it considers that the Bank is not flexible enough in allowing the use of local procurement procedures (see ISKI's assessment of project implementation - Appendix B). According to ISKI this lack of flexibility results in higher contracting costs, as the contractors bid higher prices to compensate for risks emerging from competing under a system (international bids) which is not familiar to them. This assertion is not substantiated by ISKI. 33. It is important to note that the Loan Agreement includes provisions for procurement using local procedures and that these were followed for small contracts. In those cases the Bank allowed provisions to facilitate the participation of local contractors. These provisions include the use of Turkish in the contract when it is awarded to a local contractor, and the use of the "discount rate system" under which the bidders quote on the basis of a discount on the construction prices determined by ISKI. G. ASSESSMENT OF OUTCOME 34. Despite the favorable project outcomes reported in paras. 23-24, we rate the Project as unsatisfactory. Clearly, the Project fell short of many of its original objectives; not all the works planned were satisfactorily completed or done on time; and the financial rate of return was lower than expected. In addition, the entity's management failed to stem the problems that ISKI confronted during the project period. The rating does not deny, however, that the entity did accomplish significant project goals, particularly environmental improvements through increased volumes of sewage collection and treatment, as well as improved monitoring of industrial waste discharges. In addition, the Project scored some important related attainments, e.g., more than doubling ISKI's customer base. On balance therefore, what was accomplished was quite important, especially under the circumstances of an over-dimensioned project and over-estimated strength of the executing agency. It also warrants noting that ISKI's efforts to develop simultaneously water supply and sewerage (including secondary treatment facilities) was bolder than many other developing countries' customary sector undertakings. Ideally, the project should be rated as "marginally satisfactory." However, since such a category does not exist in the evaluation process, in a conservative judgement the project could be called "unsatisfactory." 35. ISKI's partial sewerage accomplishments were not fundamentally caused by poor performance but by a justifiable shift in priorities from sewerage to water supply to respond to unforeseen circumstances--extended droughts, as well as higher than anticipated population growth. After the prolongation of the droughts from the early 1990s through 1994, ISKI's management rightfully concentrated on mitigating its perceived "water crisis." This situation also prompted the Central Government to step in to finance large water supply production facilities (US$1.2 billion Melen scheme, US$275 million for Yesilcay) during project execution. At the same time though, ISKI  - 12- did err in undertaking operations for drainage rehabilitation and free water distribution, and this may be traced to the flaws in its governance that are discussed in para. 26. H. BORROWER PERFORMANCE 36. The Borrower's performance was uneven. During 1989-93, its difficulties included mounting operating costs, poor bill collection, unsatisfactory management of the planned technical assistance, and undesirable design changes. Generally throughout as well, ISKI had poor project planning and cost accounting. After the 1994 project restructuring, ISKI's performance improved. This was evident in the better conduct and pace of project implementation, e.g., the Baltalimani sewage treatment plant was built in practically 18 months. Improvements were also notable in ISKI's relatively rigorous effort to become more financially viable, e.g., the abolition of the free water program, temporary forgiveness on late bills' interest, containing operating costs, substantially increased water billings and the sharp reduction in unaccounted-for water. I. BANK PERFORMANCE 37. The Bank's performance was similarly of mixed quality. It correctly identified the critical need to improve the provision of water supply and sewerage services in Istanbul at the outset. The same applies to the judgments made on location of the large sewage interceptors and sewage treatment plants, as well as the preliminary designs of treatment processes. 38. However, there were significant deficiencies too. The Bank erred during project appraisal in accepting very unrealistic projections of water demand and consequently a non-viable financing plan, particularly when the bulk of financing for ISKI's overall investment plan was to be provided from internal cash generation. In addition, the appraisal failed to realistically assess the failings in ISKI's governance, and its vulnerability to short-term political concerns. Hence, our overall assessment of quality at entry is poor. 39. The optimism in the projections was debated within the Bank during project processing but ultimately no adjustments were made to the water demand projections and to the investment plan. The water demand projections were considered justified by the arguments that there was substantial suppressed demand, real incomes were expected to grow, and the proposed tariffs represented relatively modest fractions of disposable income. In addition, the increased 1987 charges were accompanied by rate restructuring, which was believed to ensure tariff affordability. In fact however, the project design did not adequately provide for the contingency that eventuated: customers' refusal to pay the higher tariffs. 40. Another shortcoming of the appraisal included the failure to carry out public consultations regarding the location of the wastewater treatment plant, and to ensure that all issues related to this were fully sorted out before the start of project implementation.  - 13 - 41. In addition, it would have been prudent to help see the sector decentralization put into effect with suitable safeguards. Especially pertinent ones would have been greater precautions against financial shortfalls and overtaxing the fledgling entity. In the latter connection, the decision to add the unfulfilled components of the Istanbul Sewerage operation to this Project reflected questionable judgment as to ISKI's absorptive capacity. These points moreover comprise an especially sorry record considering that the Bank had been active in Turkey's water sector since the late 1960s. A few of these weaknesses even transpired in the prior Sewerage Project. 42. Implementation progress and problems were appropriately reported, and the quality, skills mix and continuity of the Bank staff were good. The mid-term mission provided a good example of how valuable this mechanism can be for stocktaking and course redirection (although its occurrence in the last year of a scheduled 8-year project execution period was tardy). In addition, the Bank directly provided constructive technical assistance by developing a spreadsheet model for project monitoring, and assisting ISKI's staff in maintaining it as a useful tool for project implementation. The 1994 action plan's rating scale for assessing project performance was likewise valuable. During 1988-1992, the Bank could have held more supervision missions to address problems such as low tariffs, inadequate collection of bills and accounting deficiencies. J. SUSTAINABILITY 43. Project sustainability is uncertain. ISKI, on the one hand, has gained significant experience in the development and operation of water and sewerage infrastructure. It also has shown the capability to increase water and sewerage service coverage in a rapidly growing city like Istanbul. ISKI has also accumulated a valuable knowledge of the environmental issues affecting the city, and of the most appropriate solutions to these issues. ISKI, on the other hand, continues confronting serious institutional and financial weaknesses, which raise concerns about the likelihood of the entity being able to adequately operate and expand the existing infrastructure. A fundamental concern relates to the continued vulnerability of ISKI to the impact of political changes in city government. Ultimately, a sustainable sound operation will require fundamental changes in ISKI's governance structure. K. FUTURE OPERATIONS 44. There is not now any plan for a follow-up project. The Bank did advise ISKI and local authorities that one could be considered in connection with significant changes in the institutional structure for Istanbul's water and sewerage activity, such as private participation in service operation and delivery. There has not been a positive response to this proposal.  - 14- L. LESSONS LEARNED 45. The lessons learned from this experience are: (i) Utilities that are only nominally autonomous and are subject to partisan, often short-term political interests are unlikely to be able to provide good quality services on a sustainable basis. These fundamental issues need to be fully resolved before Bank lending is provided. Alternatives to unsatisfactory governance arrangement should be sought and implemented. The selected alternative should ensure that the utility's board is directly accountable to broad segments of the society, and not only to short-term elected or appointed officials, and that the Board and management have adequate incentives, job security, and autonomy. Available alternatives range from full privatization of the services to public utilities with boards of directors selected by a variety of municipal stakeholders such as non-governmental organizations, business associations, universities, research centers and municipal authorities; (ii) Projects where the planning assumptions regarding the values of significant variables such as revenues and costs are on the "optimistic side" are likely not only to experience financial difficulties during implementation but also to yield lower than expected economic and financial rates of return. Project appraisal should ensure that the assumptions are realistic and that contingency plans are drawn for coping with deviations from the most likely range of outcomes. The range of outcomes with their respective probabilities, however subjective, should be listed in the appraisal document; (iii) It is highly risky to attempt to compensate for past under-investment in and neglect of water and sewerage services by undertaking investment programs which are overly ambitious in relation to the utility's proven managerial and financial capabilities. Project appraisal should make sure that required investments are adequately staged and that Bank loans generally finance only those investments that can be realistically implemented during the grace period of the loan; (iv) The decision process regarding the design and location of sewage treatment plants should include early consultations with affected populations, and all issues regarding site selection should be fully sorted out before Board presentation; and mid-term reviews should be held in a timely manner. In 1987, however, mid-term reviews were not Bank requirement and one was not planned in the initial project design. In September, 1994, however, a mid-term review was carried out and assisted in restructuring the project design. (v) A timely mid-term review could have assisted in addressing some of the major project issues and project outcomes. At the time of project preparation, mid-term reviews were not Bank requirements and one was not planned in the initial project design.  - 15 - PART II. STATISTICAL TABLES Table 1: Summary of Assessments Table 2: Related Bank Loans/Credits Table 3: Project Timetable Table 4: Loan Disbursements: Cumulative Estimated and Actual Table 5A: Key Indicators for Project Implementation Table 5B: Status of Completion and Operation Table 6: Key Indicators for Project Operation Table 7: Studies Included in Project Table 8A: Project Costs Table 8B: Project Financing Table 9: Project Costs and Benefits Table 10: Status of Legal Covenants Table 11: Compliance with Operational Manual Statements Table 12: Bank Resources: Staff Inputs Table 13: Bank Resources: Missions Table 14: ISKI's Income Statements (1987-1997) Table 15: Balance Sheets (1987-1997 as of December 31) Table 16A: Key Indicators for Project Performance and Operation, 1987-1994 Table 16B: Key Indicators for Project Performance and Operation, 1994-1997  -16- Table 1: Summary of Assessments A. Achievement of Objectives Substantial Partial Negligible Not applicable () () () () Macro Policies D D1 D:1 x Sector Policies D D D [} Financial Objectives D Dg D- Institutional Development D E Z] Physical Objectives x X D : Poverty Reduction EZ EE El Gender Issues El l E [ Other Social Objectives El E Cl Environmental Objectives El FII El Public Sector Management E E E] x Private Sector Development E E El B. Project Sustainability Likely Unlikely Uncertain () () () C. Bank Performance Highly Satisfactory Deficient satisfactory Identification l El Preparation l E X Appraisal E E l Supervision E E l  - 17 - D. Borrower Performance Highly Satisfactory Deficient satisfactory Preparation C D Implementation D D Covenant Compliance D DI X Operation F El E. Assessment of Outcome Highly Satisfactory Unsatisfactory Highly satisfactory satisfactory FEl El Fx1 El  - 18- Table 2: Related Bank Loans/Credits Loan/Credit Title Purpose Year Status Preceding Operations Credit 324-TU Istanbul Urban To provide assistance for the 1972 Closed Development Project preparation of an urban development program and urban transportation/land model 844-TU Expansion of water supply system for 1972 Closed in 1981 Istanbul Water Supply Istanbul and set up a regional water PAR No. 4853.1983 supply & sewerage authority 2159-TU Expansion of sewerage system, 1982 Closed in Dec./1988 Istanbul Sewerage I including sewerage treatment, for the old area of Istanbul and strengthen the PCR No 10114, Nov/91 water supply & sewerage authority (ISKI) 2537-TU Preparation studies for the Cukurova 1985 Closed Cukurova Region Urban Urban Development Project Engineering Project 2818-TU Water supply, sewerage and sewage 1987 Closed in Dec.1995 Izmir Water Supply and Sewerage treatment facilities for Izmir ICR No.15958, Project 8/1996 2819-TU Urban infrastructure, improvement of 1987 Closed Cukurova Urban Development policies, institutional and financing Project arrangements for the Cukurova region Following Operations 3151-TU Sewerage and sewage treatment works 1989 Active Ankara Sewerage Project for Ankara 3565-TU Water supply, sewerage including 1993 Active Bursa Water and Sanitation sewage treatment (Water supply and sewerage component) 3893-TU Expansion of water supply, wastewater 1995 Active Antalya Water Supply and and drainage system, institutional Sanitation Project strengthening, private operator participation, and changes in municipal financial policies for Antalya 4315-TU Water supply, sewerage including 1998 Active Cesme -Alacati Water pre-treatment plant for Supply and Sewerage Project wastewater  - 19- Table 3: Project Timetable Steps in Project Cycle Date Planned Date Actual/ Latest Estimate Identification Early 1985 Early 1985 Preparation 1985 1985-1986 Appraisal 3/1987 5/1987 Negotiations 9/1987 10/1987 Board Presentation 12/1/1987 12/1/1987 Signing 12/10/1987 12/10/1987 Effectiveness 2/23/1988 2/23/1988 Midterm Review 9/1/1994 Project Completion 12/31/1994 3/31/1998 Loan Closing 12/31/1995 01/31/1998 Table 4: Loan/Credit Disbursements: Cumulative Estimated and Actual (US$ millions) Fiscal Year Appraisal Estimate Actual Actual as % of Estimate FY88 24.40 21.00 86% FY89 46.20 37.50 81% FY90 81.00 47.90 59% FY91 120.20 76.10 63% FY92 159.40 104.60 66% FY93 189.80 135.60 71% FY94 206.00 158.90 77% FY95 214.80 174.80 81% FY96 218.00 184.50 85% FY97 218.00 196.90 90% FY98 218.00 218.00 100% Date expected for last Disbursement: July 31, 1998  - 20 - Table 5A: Key Indicators for Project Implementation I. Key Implementation Indicators in SAR Implementation Period Project Component Estimated in SAR Actual TA & Training for Reduction of Water 1988-90 1993-94 Losses Replacement of Pipes 1988-94 1990-97 Pumping Stations 1989-94 1992-93 SCADA 1987-89 1987-94 Large Diameter Sewer Pipes 1988-91 1988-97 Atakoy Drainage Zone 1988-89 Component not implemented because of change in design K.Cekmece Drainage Zone 1988-92 Component started to be implemented, then stopped. Treatment plant excluded frorm the project after project restructuring Kadikoy-Pendik Drainage Zone 1988-92 Component partially completed. Treatment plant excluded from the project after project restructuring Tuzla Drainage Zone 1988-93 1990-98. Estimated to be completed by end of 1998 Uskudar Drainage Zone 1988-90 1988-94 Street Sewers and Connections 1987-94 1988-92 Industrial Waste Management 1988-89 1991 Training 0 & M Equipment Purchase 1988-92 1990-94 II. ModifiedIndicators Project Component Targeted Actual Completion North Halic and Baltalimani Drainage By the end of 1997 Component added to the project in 1988. Zone Completed by mid 1997 Tuzla Drainage Zones By the end of 1997 Estimated by mid-1998  - 21 - Table 5B: Status of Completion and Operation Project components already completed, commissioned and satisfactorily operated: (i) Uskudar: Sewage Collectors, Sewage Pre-treatment Plant (0.1 Mm3/day capacity) and Sea Outfall; at North of the Golden Horn: Sewage Tunnels and Interceptors between Kabatas and Baltalimani (8.8 km long with diameter from 3.2 to 3.6 m); the Baltalimani Sewage Pre-treatment Plant (0.4 Mm3/day capacity) and Sea Outfall (two 1.7 m diameter, 430 m long pipes reach a depth of 70 m at the bottom of the Bosphorus); (ii) Supervisory Control and Data Acquisition (SCADA) System, which covers practically the entire ISKI Water Production, Transmission and Distribution System; (iii) Replacement and laying of new water supply pipes of 100 mm and 150 mm diameter totaling a length of 1,380 km in the water distribution networks; and (iv) Street sewers and customer connections to the sewerage system. Components already completed but not commissioned because other works need first to be completed: (i) Sea Outfall (2.2 diameter, 2.2 km long, and 46 m deep) at Tuzla which will be commissioned by September 30, 1998 once the Sewage Treatment Plant at Tuzla; and (ii) First stage of the Dragos- Tuzla collector contract are completed. Components still under construction: Branch sewerage collectors North of the Golden Horn. The contract is about 95% completed and is expected to be fully completed and operated by September 30, 1998. (ii) Tuzla Sewage Treatment Plant (0.225 Mm3/day capacity) which is more than 97% completed and expected to be fully completed and commissioned by September 30, 1998; and (iii) Sewage Collectors and Tunnels at Orhantepe and Tuzla (40% completed). The first stage is expected to be completed and operated by September 30, 1998, which would allow commissioning of the Tuzla Treatment Plant. The second stage would be completed and operated by December 1998. Note: The Moda Tunnel at Moda and Caddebostan-Bostanci Interceptor will be used only after the Sewage Pre-treatment Plant at Kadikoy is financed and constructed, as agreed during restructuring of the project.  - 22 - Table 6: Key Indicators for Project Operation 1987 1994 1997 Key Indicators Units Estimated Actual Estimated Actual SAR Restruc- Actual SAR turing I. Water & Sewerage Operation Water produced Mm3/y 447 693 389 789 540 570 Water sales Mm3/y 180 533 194 623 333 400 Unaccounted-for water percentage %prod. 60% 23% 50% 21% 38% 30% Water customers (000) 993 1,120 1,609 1,235 2,485 Water connections (000) 586 776 831 Total length sewer system km 5650 6,100 7,038 8,191 Total Population (000) 6,420 7,828 8,511 8,763 9400 Nbr.Population connected to sewerage (000) 3,349 5,447 6,608 6110 % Population connected to sewerage % 52% 70% 64% 75% 65% Sewerage pre-treated Mm3/y 0 75 133 a/ Sewerage pre-treated per water sales % % 39% 33% Year end employees (000) 6,217 7,200 7,078 7,542 Employees per 1000 water customers No 6.3 6.4 4.4 3.0 I. Industrial Waste Management Year end discharges permits in force No 0 2,430 315 582 Firms monitored No 3,245 2,900 5,956 8,341 Year end employees No 37 85 95 135 III. Financial Indicators Total Operating Water & Sewerage $/m3 sold 0.45 0.98 1.08 1.72 0.76 Average Cost $/m3 sold 0.34 0.32 0.56 1.01 0.30 Cost Coverage Ratio No 1.30 3.04 1.94 1.71 2.55 Collection Percentage % 61 86 81 84 Debt Service Coverage Ratio 1.60 6.00 0.58 2.27 2.38 Self-financing Percentage % 324 106 (54.6) 32 70 a/ of which 89 Mm3/y are due to Yenikapi STP Table 7: Studies Included in Project Study Purpose as defined Status Impact of Study At appraisal/redefined Monitoring of Not foreseen at Implemented by Monitoring confirmed that the Yenikapi effluent appraisal Middle East Technical assumptions used in the design of the discharge at the University in 1992-93 effluents disposal system at the bottom bottom of the of the Bosphorus were basically correct. Bosphorus _ Master Plan for Not foreseen at Implemented by The study assisted in defining Tuzla Water appraisal International Collectors and Connections of Sewer Supply & Sewerage Consortium. Started Collectors to North Halic Interceptors. for Istanbul 1993, expected to be However, most of the benefits of the completed by May 98 study are directed towards future investments. Pre-feasibility Study Not foreseen at Implemented by Economical and technical comparison for Alternative Asian appraisal TBP/UBM in 1990-93 of different sites to locate sewage Side Sewage treatment plant on the Asian side, as Treatment Plants alternative to the Kadikoy site.  - 23 - Table 8A: Project Costs Project Project Actual Appraisal Amended Actual Overrun Project Components May 1987 Dec. 1988 over Amended US$ million % A. Water Distribution 1. Program reducing water losses 1.40 1.40 0.50 -64% 2. Replacement of pipes 60.40 60.40 36.83 -39% 3. Pumping stations 11.10 11.10 1.99 -82% 4. SCADA 13.50 13.50 25.40 88% B. Sewerage System Baltalimani 79.54 79.54 0% Large Diameter sewer pipes 26.40 26.40 26.62 1% Atakoy drainage zone 6.80 6.80 0.00 -100% K. cekmece drainage zone 70.90 70.90 12.01 -83% Kadikoy-Pendik drainage zone 124.90 124.90 53.40 -57% Tuzla drainage zone 97.00 97.00 97.40 0% Uskudar drainage zone 9.20 9.20 9.91 8% Street Sewers and Connections 118.60 118.60 201.19 70% Land Acquisition 2.70 2.70 20.36 654% Local Staff and Supervision 14.70 14.70 27.08 84% C. Training and Equipment Industrial waste management training 0.10 0.10 0.02 -80% O & M Equipment purchase 12.20 12.20 13.10 7% Total 569.90 649.44 605.36 -7% Table 8B: Project Financing Appraisal Estimates Actual/Latest Estimates Item Total Percent of Total Percent of (US$ million) Total (US$ million) Total 1. IBRD Loan 218.0 38 218 36 2. ISKI Revenues from Sales 265.9 47 336 56 3. Customer Contributions 86.0 15 50 8 Total 569.9 100 604 100  -24 - Table 9: Project Costs and Benefits Incremental Cost and Benefit Streams for Internal Rate of Return Project Other Sales Sales Incre- Incremental Other Average Capital Capital Incremental w/o with mental Water Operating Net water Operating Costs Costs Operating Program Program Sales Revenues Revenues Benefits Revenues Cost Year $Mn $Mn Cost$Mn Mm3 Mm3 Mm3 $Mn $Mn $Mn /rn3 sold $Mn 1987 2.8 -2.8 56.4 1988 47.2 12.8 4.7 240 240 0 -64.7 61.1 1989 31.4 24.4 16.3 240 268 28 11.8 -60.3 0.42 72.7 1990 100.7 3.0 93.5 234 234 0 0.0 0.0 -197.2 0.90 149.9 1991 84.2 53.1 111.1 225 297 72 81.4 2.0 -164.9 1.13 167.5 1992 118.9 72.7 135.3 216 270 54 48.1 70.0 -208.8 0.89 191.7 1993 60.8 190.8 141.4 208 266 58 69.6 42.0 -281.4 1.20 197.8 1994 16.3 73.0 37.5 194 194 0 0.0 6.0 -120.8 0.95 93.9 1995 16.8 58.6 70.5 186 256 70 53.9 57.0 -35.0 0.77 126.9 1996 38.3 118.1 73.6 179 329 150 120.0 44.0 -66.0 0.80 130.0 1997 26.3 136.2 37.9 172 400 228 152.8 37.0 -10.6 0.67 94.3 1998 46.9 197.1 69.5 165 460 295 206.5 28.0 -79.0 0.70 125.9 1999 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2000 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2001 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2002 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2003 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2004 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2005 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2006 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2007 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2008 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2009 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2010 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2011 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2012 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2013 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2014 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 2015 20.0 69.5 158 460 302 241.6 28.0 180.1 0.80 125.9 Aamtn Re Capital costs do not include: Rate of Return: 0.06 (a) former projects completed in 1987-88. Opportunity cost capital: 0.10 (b) works excecuted by DSI, not charged to ISKI NPV of incremental costs: 1393.81 Million After 1998 capital costs include renewal costs Discounted volume of water: 1275.23 Mm3 to maintain the level of sales Average increm. cost of water: 1.09 $/m3 Incremental operating costs are cash operating cost increases above the level of 1987.  - 25 - Table 10: Status of Legal Covenants Turkey: Istanbul Water Supply and Sewerage Project Agreement Section Covenant Present Original Description of Covenant Comments Type Status Fulfillment D ate _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ LA 3.03 (a) 05 C 01/01/1995 Carry out an Action Plan, satisfactory to the For the fourth quarter of Bankfor the purpose of improving the 1997, ISKI scored a Borrower's implementation of the Project and weighted average of 3.3 out its overall operational and financial of a scale of 4 in terms of performance. achieving the agreed targets for eight performance parameters action plan. LA 3.03 (b) 09 C 01/31/1995 Fumish to the Bank for review and approval, Reports are produced on a report on the progress achieved during said time quarter against the performance targets of the Action Plan. LA 3 (c) (d) 02 CP 01/31/1995 On the basis of forecasts review whether the Measures are taken to Borrower would meet the Action Plan substantially meet all eight performance targets for the next following performance targets, except quarter. Identify any measures that may be tariff. ISKI can meet the required to achieve these targets. Promptly other fmancial performance take all necessary measures (including, targets because of without limitation, adjustments of the Government subsidies. structure or levels of its rates) in order to meet Central Government is the Action Plan performance targets. serving most of ISKI's debt service and is fmancing most of investments required for future water production facilities. LA 5.01 (b) 01 C 06/30/1989 Independent audit of accounts and financial The 1996 accounts of ISKI statements and audited financial statements have been audited. The with auditor's report should be sent to the auditors were no able to Bank. express an opinion because of deficiencies in the accounting, commercial and intemal control systems. ISKI is taking action to implement the auditors recommendations. LA 5.04 02 C Consult with the Bank prior to contracting any None long-term debt exceeding in the aggregate: (a)US$5million in any fiscal year; and (b) US$250 million after date of the Loan -Agreement. LA Sched. 5 09 CP 02/28/1989 Prepare semi-annual reports on (i) financial Reports were too succinct. & 5.04 status; (ii) progress of project implementation; (a) (iii) monitoring indicators; and (iv) updated description of future investments. Covenant types: 7. = Involuntary resettlement Present Status: 1. = Accounts/audits 8. = Indigenous people 2. = Financial performance/revenue generation from 9. = Monitoring, review, and reporting C = covenant complied with beneficiaries 10. = Project implementation not covered by CD = complied with after delay 3. = Flow and utilization of project funds categories 1-9 CP = complied with partially 4. = Counterpart funding I1. = Sectoral or cross-sectoral budgetary or other NC not complied with 5. = Management aspects of the project or executing agency resource allocation 6. = Environmental covenants 12. = Sectoral or cross-sectoral policy/ regulatory/ institutional action  -26 - Table 11: Compliance with Operational Manual Statements There is no evidence of non-compliance. Table 12: Bank Resources: Staff Inputs Planned Actual Stage of Project Cycle Weeks US$ (000) Weeks US$ (000) Preparation to Appraisal Not available 38 78 Appraisal Not available 16 35 Negotiations through Board Approval Not available 11 24 Supervision Not available 149 468 Completion 16 27 Total 215 526  - 27 - Table 13: Bank Resources: Missions Stage of Project Month/ No. of Days in Specializa- Performance Rating2 Types of Problems3 Cycle Year Persons Field tion' Implement Development ation Objectives Through appraisal 15 Appraisal tbru May - Dec 87 2 16 F..2E. Ec Board approval Supervision Feb-88 2 5 F E 1 1 Coll.. Supervision Apr-88 2 5 F E 1 1 Pcr. Supervision Jul-88 2 5 F E 1 1 Coll.UFW. Supervision Jun-89 2 6 E 2 1 Coll. UFW.Acc. Supervision Nov-89 2 6 E 2 1 Coll. UFW.Acc. Supervision Apr-90 2 6 F E 2 1 Des. Turn. Drght. Acc Supervision Jun-90 2 3 F E 2 1 Des. Acc Supervision May-91 2 6 E F 2 1 UFW . Coll.. Acc. Supervision Jun-92 2 14 E 2F 2 2 Tar. Coll. UFW, Acc. Supervision Dec-92 4 12 E F 2 2 Tar. Coll. UFW,. Acc. Supervision May-93 3 20 E-F-Env 2 2 Tar. Coil. UFW,. Acc. Supervision Nov-93 2 12 E F 3 3 UFW. Tar.Coll.Turn. Acc. Supervision Jul-94 2 4 E F U U UFW. Tar.Coll.Turn. Acc. Supervision Sep-94 2 15 F E U U UFW.Tar. Turn. Acc. Supervision Feb-95 2 7 F E S S UFW. Tar. Turn. Acc. Supervision Feb-96 2 7 E F S S UFW.Tar.. Acc. Supervision Sep-96 2 10 E F S S UFW.Tar.. Acc. Supervision Sep-97 2 7 2E F S S Tar.. Acc. Completion Mar-98 2 8 E F S S Tar.. Acc. I - Key to Specialized staff 2 - Key to Performance 3 - Key to Types of Problems: skills: Ratings: Des. = Design disagreement Drght.= Drought, Low sales E = Sanitary Engineer I = Minor Problems Turn. = Turnover of Management Acc. = Accounting Insufficiency F = Financial Analyst. 2 = Moderate Problems UFW = Too high unaccounted-for water Env = Environmentalist 3 = Major Problems CoIl .= Too low collection of bills Ec = Economist S = Satisfactory Tar. = Too low tariff U= Unsatisfactory Pcr = Procurement  Table 14: ISKI - Income Statements (1987-1997) (US$ million) _________________________________ 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 Total Operating Revenue 80.7 115.1 128.4 229.8 276.8 321.2 246.6 208.7 223.7 301.1 304.1 Water and Sewerage 80.2 104.5 107.8 229.4 274.4 - - - - - 266.7 Other 0.6 10.6 20.7 0.4 2.4 - - - - - 37.4 Staff Costs 22.0 24.9 31.4 81.5 99.2 117.3 124.4 54.1 73.1 67.3 59.6 Energy Costs - 22.4 23.5 24.7 28.5 33.0 27.8 18.1 25.2 30.9 27.4 Raw Materials 9.3 7.0 8.6 12.9 13.8 12.3 14.6 4.2 8.2 9.4 3.6 Depreciation 5.6 8.7 10.1 12.9 16.4 14.1 13.4 13.9 16.6 18.4 25.1 Provision for bad and doubtful debts - - - 19.2 38.6 44.4 26.1 31.3 0.3 2.2 10.0 Sub-contracted services 23.7 4.9 7.2 11.0 10.8 6.4 7.9 3.9 3.8 3.7 - EDP services - - - 9.1 11.2 16.0 16.9 5.3 1.7 0.4 Provision for Empl. Term. Benefits - - 22.3 21.6 14.1 26.8 15.0 - Rent Expenses - - - 6.0 0.3 - - - - - - Other Expenses 1.4 1.9 1.9 4.7 3.7 6.6 6.2 8.4 15.0 18.4 3.6 Total Operating Expenses 62.0 69.8 82.8 182.0 222.5 272.5 259.0 153.2 170.5 165.7 129.4 Net Income from Operations 18.7 45.4 45.7 47.8 54.2 48.7 (12.3) 55.5 53.1 135.5 174.7 Non-operating income 26.0 15.7 11.9 20.2 26.4 71.2 32.4 103.2 52.7 47.0 38.4 Non-operating expenses 37.6 3.0 2.0 8.0 5.5 8.7 61.4 8.9 62.1 71.6 55.9 Net Non-Operating income (11.7) 12.7 9.9 12.2 21.0 62.4 (28.9) 94.3 (9.3) (24.6) (17.5) Financial expenses 30.1 63.9 45.3 49.5 57.1 111.1 105.6 106.9 104.3 21.7 Exchange Losses - - - - 40.9 50.0 161.1 111.7 114.9 - Net Income for the Year 7.1 27.9 (8.3) 14.8 25.7 13.1 (202.4) (116.9) (174.9) (108.3) 135.5 Extra-ordinary item - - - - 63.8 Net Income (loss) after extraordinary item 7.1 27.9 (8.3) 14.8 25.7 76.9 (202.4) (116.9) (174.9) (108.3) 135.5 Note: 1990-1996 ISKI financial statements are audited by private independent external auditors. 1987-1989 and 1997 ISKI financial statements are unaudited.  Table 15: Balance Sheets (1987-1997 as of December 31) (US$ million) 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 Assets Current assets Cash and due from banks 2.24 2.12 4.13 19.43 9.17 0.42 20.83 13.85 16.64 10.69 9.97 Receivables from customers 43.58 63.45 70.63 99.88 113.36 140.27 87.05 139.85 74.36 102.51 111.70 Pipes and other stocks 5.67 19.59 18.40 32.14 36.20 10.47 6.80 50.38 32.91 28.99 41.80 Advances and other current assets 7.63 53.88 46.76 2.94 12.93 48.27 118.20 9.76 71.03 13.25 37.99 Total current assets 59.12 139.04 139.92 154.39 171.66 199.44 232.88 213.84 194.94 155.44 201.46 Long-term assets Tangible assets Fixed assets-net 117.90 173.48 234.82 311.61 361.15 304.16 269.44 283.08 374.98 472.03 378.00 Construction in progress 94.78 97.64 187.31 302.64 296.48 334.21 354.12 214.64 171.16 105.23 213.89 Investments - - - 0.24 0.24 0.14 0.08 1.28 0.81 0.72 0.44 Advances and long-term deposits 46.24 2.17 4.88 6.16 6.57 15.00 8.16 2.33 1.49 0.95 0.89 Total long-term assets 258.92 273.28 427.01 620.65 664.43 653.50 631.81 501.33 548.44 578.94 593.23 Total assets 318.04 412.32 566.93 775.04 836.09 852.94 864.70 715.17 743.39 734.38 794.69  Table 15: Balance Sheets (1987-1997 as of December 31) (US$ million) 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 Liabilities and Equity Current liabilities Current portion of long-term loans 0.00 0.00 0.00 18.67 39.92 89.18 93.06 60.08 81.38 66.24 52.37 Payables to government institutions 1.39 54.20 94.46 95.61 102.58 99.18 218.97 202.33 281.51 324.68 265.64 Payables to Contractors 0.00 0.00 0.00 31.00 25.72 57.47 81.80 23.03 12.18 15.21 6.62 Trade accounts payable 0.00 15.50 6.76 10.81 10.69 8.62 17.48 8.83 14.03 5.20 2.52 Accruals and other current liabilities 40.96 2.40 7.10 25.29 27.62 38.85 57.25 49.60 21.42 20.13 20.17 Total current liabilities 42.34 72.11 108.33 181.39 206.53 293.30 468.54 343.87 410.51 431.46 347.32 Long-term liabilities Long-term loans 167.45 184.24 239.72 269.84 256.26 233.08 260.63 300.66 299.05 267.47 229.82 Employee termination benefits 0.00 0.00 0.00 0.00 7.53 32.18 35.45 21.53 31.40 29.20 33.18 Customer deposits 11.66 7.74 6.32 24.61 39.22 42.29 38.70 30.25 19.57 11.54 7.84 Long term payables to Government A 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Other-long term liabilities 1.56 1.87 1.88 2.18 1.27 3.69 2.93 3.53 3.28 2.77 4.36 Total long-term liabilities 180.66 193.85 247.92 296.63 304.27 311.24 337.70 355.96 353.30 310.98 275.19 Equity 95.03 146.36 210.68 297.02 325.30 248.40 58.45 15.34 -20.43 -8.07 172.18 Total liabilities and equity 318.04 412.32 566.93 775.04 836.09 852.94 864.70 715.17 743.39 734.38 794.69 End of period FX rate(TL/$) 1,018 1,813 2,311 2,927 5,075 8,556 14,458 38,418 61,054 107,505 204,750 Note: 1990-1996 ISKI financial statements audited by private independent external auditors. 1987-1989, and 1997 ISKI financial statements unaudited. The US$ balances are calculated on the current reported balances taking the year end exchange rates.  Table 16A: Key Indicators for Project Performance and Operation. 1987-1994 (US$ million) 1987 1988 1989 1990 1991 1992 1993 1994 SAR Actual SAR Actual SAR Actual SAR Actual SAR Actual SAR Actual SAR Actual SAR Actual Performance Parameters Self-financing Ratio (%) 160.0 324.6 79.0 25.7 72.0 (35.5) 73.0 52.4 79.0 52.1 83.0 22.7 93.0 (43.2) 106.0 (54.6) Debt Service Coverage Ratio 10.20 1.60 8.10 1.35 6.20 0.74 5.70 1.68 5.80 1.60 4.90 1.18 5.40 0.28 6.00 0.58 Collection Ratio (%) - 61.0 - 61.0 - 85.0 - 75.0 - 61.0 - 74.0 - 75.0 - 86.0 Total Operating Revenues per m3 sold($ 0.85 0.45 0.89 0.46 0.89 0.57 0.89 0.98 0.90 0.93 0.92 1.19 0.95 0.93 0.98 1.08 Total Operating Cost per m3 sold(%) 0.20 0.34 0.20 0.28 0.21 0.37 0.23 0.70 0.25 0.62 0.28 0.76 0.30 0.79 0.32 0.56 Operating Ratio (%) 24 77 22 61 24 64 26 71 28 66 30 64 32 86 33 52 Unaccoounted-for water (%) 30.0 59.7 29.0 50.8 28.1 52.7 27.0 29.1 26.0 32.0 24.9 42.9 24.1 48.0 23.0 50.1 w Other Operational Indicators Water Produced (mil m3) 437.0 447.0 510.0 504.0 541.0 474.0 571.0 330.0 600.0 437.0 630.0 473.0 661.0 512.0 693.0 389.0 Water Sold (mil. m3) 306.0 180.0 362.0 248.0 389.0 224.0 417.0 234.0 444.0 297.0 473.0 270.0 502.0 266.0 533.0 194.0 Total Operating Revenues 259.3 80.7 322.9 115.1 346.1 128.4 369.6 229.8 398.7 276.8 437.0 321.2 476.2 246.6 520.4 208.7 Total Operating Cost 1/ 61.5 62.0 71.6 69.8 83.0 82.8 94.9 162.8 111.0 183.9 130.1 205.8 150.6 211.2 171.3 107.8 Total Investments 120.5 2.8 305.6 60.0 351.5 55.8 359.4 103.7 361.4 137.3 360.5 191.6 357.9 251.6 351.0 89.3 1/Total operating cost figures exclude provision for bad and doubfful debts and provision for employee termination benefits.  Table 16B: Key Indicators for Project Performance and Operation, 1994-1997 (US$ million) 1994 1995 1996 1997 Performance Parameters AP Actual AP Actual AP Actual AP Actual Physical Progress (% of completion) 48.0 46.0 69.0 69.0 88.0 82.0 100.0 86.0 Self-Financing Ratio (%) (29.0) (54.6) 11.0 (122.3) 25.0 (32.4) 31.0 70.0 Debt Service Coverage Ratio 0.73 0.58 1.23 0.4 1.83 0.73 2.27 2.38 Collection Ratio (%) 77.0 86.0 79.0 88.0 80.0 85.0 81.0 84.0 Total Operating Revenues per m3 sold ($) 0.91 1.08 1.09 0.87 1.43 0.92 1.72 0.76 Total Operating Cost per m3 sold (%) 0.58 0.56 0.67 0.56 0.84 0.45 1.01 0.30 Operating Ratio (%) 64 52 61 64 59 49 58 39 Unaccoounted-for water (%) 36.0 50.1 42.3 51.1 41.5 40.6 41.0 29.8 Other Operational Indicators Water Produced (mil m3) 400.0 389.0 520.0 524.0 540.0 554.0 580.0 570.0 Water Sold (mil. m3) 256.0 194.0 300.0 256.0 316.0 329.0 342.0 400.0 Total Operating Revenues 233.0 208.7 328.0 223.7 452.0 301.1 589.0 304.1 Total Operating Cost 1/ 149.0 107.8 200.0 143.5 265.0 148.4 344.0 119.4 Total Investments 145.6 89.3 304.3 75.4 410.9 156.4 576.0 162.5 AP: Action Plan agreed with ISKI in 1994. 1/ Total operating cost figures exclude provision for bad and doubfful debts and provision for employee terminaton benefits.  Appendix A: Mission's Aide-Memoire WORLD BANK February 28, 1998 Second Water and Sewerage Project, Loan 2888-TU Project Completion Mission AIDE-MEMOIRE 1 This aide-memoire summarizes the results of discussions held in Istanbul between the representatives of ISKI and the World Bank representatives Daniel Coyaud. and Alptekin Orhon on a mission from February 23 to 27, 1998. The mission thanks ISKI for the assistance provided during the mission. This aide-memoire was presented to Treasury in Ankara.on March 3, 1998. The mission's recommendations are subject to confirmation by the Bank's management, and a letter will be sent to ISKI after the mission returns to Washington. Project Implementation i Implementation of the project is about three years behind appraisal schedule and about one year behind the revised schedule expected at the time of the project restructuring in September 1994. The project components already completed, and already operated are indicated in the operational plan, together with components already completed but not commissioned because other works need first to be completed (Annex 5 page 1), 2 Project components still under construction are: i the branch sewerage collectors North of the Golden Hom. The contract is about 95%/* completed and is expected to be fully completed by June 30, 1998; ii the Tuzla Sewage Treatment Plant (0.225 Mm3/day capacity) which is more than 97% completed and expected to be fully completed by April 30 1998; and iii the Sewage Collectors and Tunnels at Orhantepe and Tuzla (23% completed). The first stage is expected to be completed by April 30, 1998, which would allow commissioning of the Tuzla Treatment Plant. The second stage would be completed by December 1998. 3 About $9.5 million of the Loan were used by a contract for ductile iron pipes financed at the level of 100% under the Loan. The pipes contracted mid-December 1997 were delivered at the end of January 1998. 4 On the whole the restructured project is about 92% completed. On February 25, 1998, the situation of project implementation according to different criteria is summarized as follows: Total Amount in $ As percentage million Contracts already a varded $604 million 100% of total project cost $604 million Paid or Certified $554 million 92% of total project cost S604 million Loan Disbursed $210 million 96% of total Loan of $218 million  Page 2 5. The detailed implementation follow-up table in Annex 1 shows all project contracts and for each contract already awarded, the amounts of works certified, the disbursements from the Loan 6. Project expenditures for priority works are shown in Annex 2, together with the graph of actual expenditures compared with the progress mutually agreed in the Action Plan of July 1994. 7. Master Pan For resumption of the study, a protocol was signed in December 1997. The completion of the study and issue of the final reports are expected by June 30, 1998. Financial Performance in 1997. 8. The following analysis is based on the provisional figures for 1997. 9. Water Produced and Water Sold. Water produced increased 2.9% in 1997(570 mil. m3) compared to 1996(554.1 mil. m3) Water sold increased 22% in 1997 (400.4 mil m3) compared to 1996 (329 mil. m3). 10. Unaccounted-for Water. For 1997, the level of UFW achieved (30%) is now quite reasonable and well below the level of 1995 (51%) and the average target of 39%. This improvement is achieved by continuing through replacement of pipes in high leakage areas,more control on the illegal water consumptioni, improvement of metering, and better administration of customer services. 11. Tariffs and Operating Revenues. The average water and sewerage charges per m3 sold in 1997 (US$0.67) which is substantially below the average target of US 1.15 for the year. 12. In 1997, total operating revenues decreased slightly to US$304 million compared to 1996 (US$306) due to ISKI's reluctance to increase tariffs every quarter to compensate for inflation. 13. However, operating ratio has decreased to 43% in 1997 compared to 52% in 1996 which is also lower than the average target of 58% for the year, because of the reduction in operating costs. 14. Operating Costs. In 1997, operating costs has decreased to US$129.4 million compared to 1996 (157.4 million). As a result, the operating cost per m3 sold is improved in 1997 (US$0.32 per m3 sold) compared to 1996 (US$0.48 per m3 sold). This is also much lower than the average target of US$0.78 per m3 sold. This improvement in operating cost per m3 sold is mainly due to decreasing of operating expenditures and the increase in water sales(22%). However, it should be noted that ISKI's operating cost accounts do not reflect adequately the depreciation. ISKI's depreciation accounting policy does not take into account the useful lives of its fixed assets. Because of the deficiencies in ISKI's fixed asset accounting procedures, like the absence of detailed fixed asset ledgers, and other deficiencies in the accounting system, ISKI can not determine the net book value of the fixed assets. Amount allowed for depreciation for 1996 (US$14.7 mil.) is low and it represents only 18% of ISKI's debt service(US$82.2 mil.) which was due in 1997. 15. Collection Ratio. In 1996, in spite of the efforts of ISKI, the collection ratio achieved was 84% which was lower than in 1995 (88%) and 1996(85%), but higher than the average targeted figure of 80.7% for the year. Although ISKI's collection ratio is slightly lower than in 1996, ISKI continued to implement measures to increase its collections. 16. Self-Financing Ratio. In 1997, ISKI's self-financing ratio improved in 1997(66%) due to substantial decrease in operating costs. This is also higher than the average target for 1997(20.3%). 17. Debt Service Coverage Ratio. In 1997, based on ISKI's debt service requirements, debt service coverage ratio was 2.3 which was higher than the target of 1.76 for the year.  Page 3 18. Employees per Thousand Water Connections. Employees per thousand water connections improved in 1997 (3.0) compared to 1996 (3.5). Status of Compliance on Action Plan in 1997 19. The compliance on the Action Plan formally agreed by the Bank through an Amendment to the Loan Agreement is presented in Annex 3 for the fourth quarter of 1997. In 1997, ISKI achieved an average score of 3.3 above the minimum score of 2 covenanted in the Legal Agreement. The overall improvement over the third quarter of 1994 is about 70% of the improvement targeted. In the fourth quarter of 1997, of the eight performance indicators, ISKI has achieved the maximum score of 4 for six parameters (unaccounted-for water, average costs per m3 billed, operating ratio,collection ratio,debt service coverage ratio, and self- financing ratio), a score of 3 for physical progress, and a score of 0 for average water and sewerage revenues. Details of the performance for the fourth quarter of 1997 is shown in Annex 3. Inadequacy of ISKI Water and Sewerage Tariffs. 20. Effective January 1, 1997 ISKI increased its water and sewerage charges about 66% in TL terms compared to 1996 and informed the Bank that the ISKI Board is authorized to make quarterly adjustrnents in line with inflation. However, effective August 15, 1997, ISKI only increased the first block (0-10 m3) for domestic of about 36% in TL terms without increasing the tariffs for other blocks and categories of customers. On average, it corresponds to an increase of about 19% lower than the inflation of about 35 % over the first semester of 1997. Although authorized to make quarterly adjustments, ISKI management, due to political influences, did not make other quarterly adjustments in 1997. Effective January 1, 1998, on the average ISKI increased its tariffs by about 90% compared to January 1997 and 70% considering the increase in August 1997 which are lower than inflation of about 100% for the whole year of 1997. Government Subsidies in 1997. 21. Although ISKI met the Action Plan Target for 1997, ISKI is not recovering the full cost of water because of inadequate depreciation of its assets(para 16) and large Central Govemment subsidies. A large part of (88%) its debt service regarding foreign loans was covered by Treasury. In addition, the Central Government is bearing the full cost of implementing large investments like Yesilcay ($275 million), and Melen Projects ($1.2 billion) needed to supply Istanbul in the future. In order to fully recover its costs, ISKI's tariff should be close to the long term incremental cost of providing water and sewerage services. Financial Projections for 1998 22. To meet the targets agreed in the action plan, ISKI needs to improve its financial situation in the future years. The projections for 1998 are presented in Annex 4. ISKI has to implement a tariff level which is adjusted quarterly to compensate for inflation, and based on this the equivalent average water and sewerage tariff should be about US$0.70 $/m3 in 1998 in order to cover debt service, and 60% of ISKI investments. ISKI needs to generate more revenues in order to service the debt for the loans borrowed and be able to meet the financial requirements of the large investments that ISKI plans to implement itself for future years like Ikitelli Water Treatment Plant (US$31 million), Pasakoy Wastewater Treatmnent (US$26 million), Kadikoy Wastewater Pretreatment Plant and Sea Outfall (US$46 million), Orhantepe-Pendik Collectors (US$21 million), Omerli-Sihli-Darica-Gebze force main (US$21 million),and Cavusogludere-Ikitelli force main (US$28 million).  Page 4 Audits 23. ISKI management informed the Bank that the 1997 financial statements, project account and the special account will be audited by ISKI's current private external auditor. ISKI management also agreed to submit the above mentioned three audit reports no later than June 30, 1998 Preparation of the Implementation Completion Report 24. The mission provided guidelines (Annex 5) for ISKI's own final evaluation report on the project that ISKI is required to prepare according to the General Conditions of the Loan (Section 9.07 (c). 25. During its visit the mission, assisted by ISKI, collected statistical data and information (Annex 6). 26. In the views of the mission, the Project's objectives were partially achieved. Projected and actual operating indicators listed in Table 6, show that: a) Unaccounted-for water was reduced from 60% in 1987 to 30% in 1997. The figure in 1997 is lower than the 21% expected at appraisal, but higher than the target of 38% mutually agreed in 1994 at the time of the project restructuring. b) Sewage treatment was extended.. Three new sewage treatment plants were built under the project (instead of four in the original project). The volume of appropriately treated sewage disposed in the Bosphorus and the Sea of Marmara increased from 57 Mm3/year in 1989 to 133 Mm3/year in 1997. Overall, the volume of sewage treated as a percentage of water billed increased from 23% in 1988 to 33% in 1997. c) The total length of sewerage network progressed from 7200 km in 1988 to 8200 km in 1997, i.e a progression of 1000 km, as compared to 2,500km estimated at appraisal. The number of sewerage connections progressed from 510,000 in 1987 to 840,000 (?) in 1997. As a result, provision of sewerage services were increased from 56% of the population in 1987 to 65% in 1997, instead of 75% as estimated at appraisal. d) ISKI's operation and maintenance capabilities was improved. The number of employees per thousand water connections decreased from 6.3 in 1987 to 3.0 in 1997. The number of firms monitored for wastewater discharges increased from 3,200 in 1987 to 8,300 in 1998. The cost coverage ratio (operating revenues over operating costs) improved from 1.30 in 1987 to 2.33 in 1997. 27. The actual costs of the restructured project amount to $604 million , i.e.6% higher than the appraisal estimate of $ 570 million. However, if we adjust the appraisal estimate by adding the cost of what was added to the project (North Halic Interceptors and Baltalimani Sewage Treatment Plant), the project actual costs are 7% lower than the adjusted appraisal estimate cost of $ 649 million (Table 8A). However, the restructured project includes only parts of the works related to the Kadikoy and Kucukcekmece drainage areas. 28. ISKI"s institutional capacity was very low in August 1993 when ISKI faced a major financial crisis and subsequent demand by ISKI's creditors for repayment of dues. Since the September 1994 Agreement on the eight performance parameters action plan, ISKI scored above the minimum acceptable, corresponding between 25% and 50% of agreed improvements. In 1997, ISKI achieved more than 70% of the agreed improvements. 29. ISKI's institutional capacity continues to be limited by its institutional structure which is closely  Page 5 linked to local politics since the Mayor of Istanbul is the Chairman of ISKI and appoints all the Board members. Any radical change in ISKI's institutional capabilities can come about only with the change in the current institutional structure. Such change was not envisaged in the project. In March 1994 a new mayor was elected and a new management was brought into ISKI. The new management, in its ability to contain operating costs, to reduce unaccounted-for water and to prioritize well for new investments, is a marked improvement over the previous management. However, the existing water and sewerage tariff level are insufficient to cover for the full economic cost of water supply to Istanbul (para 21). 1  SECOND ISTANBUL WATER SEWERAGE PROJECT Anne 1 Implementation P4p I $ million IBRD Appraisal Contract Contract Contract Contract Total lB-O Contract Bank Descrption No Company Signature Comp!etion Contract Disburser Works Value Financing Date Date Awarded ISKI Comphred Total Total Total A-Water Distribution $ $ $ equiv. $ equiv. S equiv. I Program Teductng water toSs 1.bo 1.bo 1.40 Heoduction ot Unacct TA Water tSEtIA-Austia Apr/19/93 Jun/30/94 0.50 0.43 2 Replacemr.r.td OT pipes Piping Water mains rehabilitation 24.20 24.20 6.10 T6949 Von Roll Dec/12/97 Jan/31/98 1.46 1.17 1.17 New Pipes + Fittings T6949 SADIP Dec/16/97 Jan/31/98 7.93 6.34 6.34 Pipe & PS Equipment Purcha 42.10 42.10 33.80 Pipes + Fittings T6410 PONT-A.NOUSSOt Sep/13/90 14.46 14.46 14.46 THYSSEN GUSS-A Oct/19190 8.52 8.52 8.52 INTER-IMPEX Oct/05/90 4.46 4.46 4,46 Pumps T5755 NIJHUIS Jan/07/92 0.78 0.78 0.QS T5755 KSB Feb/08/92 1.21 0.97 05.7 3 Pumping stations Pumping PS Rehabililation Civ. Works 4.80 4.80 1.20 4 SCADA 13.80 SCADA Supply SCADA System 12.30 9.80 Supply SCADA S1000 ABiTEKSER Oct/08/87 Nov/14/91 20.55 11.80 20.5a. SCADA Supn & Training 1.50 1.30 Design SCADA S1004 SOGREAi-USEYAS Mar/24/87 Jun/31/92 2.12 2.11 2.11 Darlik Pumping S4441 ELIOP Oct/12/90 Jul/05/92 2.46 2.02 2.02 Darlik Design S3047 BORTEK/SOGREA Oct/18191 Oct/3'192 0.27 0.'4 0.34 B-Sewerage system Norlh Halic Second- K23ti;8 YUKSEL INS.AS. May/l1696 Jan/05/98 8.84 4.