Turkey - Istanbul Sewerage Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 10852 PROJECT PERFORMANCE AUDIT REPORT TURKEY ISTANBUL SEWERAGE PROJECT (LOAN 2159-TU) JUNE 29, 1992 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EgULENT Currency Unit - Turkish Lira (TL) December 1981 - TL 130 - US$1.00 (SAR) December 1988 - TL 1790 - US$1.00 GLOSSARY OF ACRONYMS ISKI - Istanbul Water Supply and Sewerage General Dire- DSI - State Hydraulic Works IB - Iller Bankasi (Bank of the Provinces) ISI - Istanbul Water Authority FISCAL eer January 1 -December 31 FOR OFFICIAL USE ONLY THE WORLD BANK Washington. D.C. 20433 U.S A. of Director-General --tinns Evaluation June 29, 1992 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Performance Audit Report on Turkey Istanbul Sewerage Project (Loan 2159-TU) Attached, for information, is a copy of a report entitled "Project Performance Audit Report on Turkey - Istanbul Sewerage Project (Loan 2159-TU)" prepared by the Operations Evaluation Department. Attachment This document has a restricted distribution and may be used by recipients only in the performance of their officiit duties. Its contents may not otherwise be disclosed without World Bank authorization. FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT TURKEY ISTANBUL SEWERAGE PROJECT (LOAN 2159-TU) TABLE OF CONTENTS Page No. Preface ............................................................. i Basic Data Sheet....................... .. ........... .............. ii Evaluation Summary ...................... iv I. PROJECT qUMMARy.. ............................ ... 1 A. Background.............................................. 1 B. Bank Involvement........................................... 3 C. The Projects and its Objectives..o......................... 3 D. Project Implementation..................................... 7 E. Project Costs and Disbursements............................ 8 F. Procurement................................................ 11 G. Performance of Contractors................................. 12 H. Institutional Developments................................ 12 I. Operational Aspects....................................... 12 J. Sustainability............................................. 13 II. ISKI's FINANCIAL PERFORMANCE............... .................. 13 A. Compounded Growth Rates of Financial Indicators......o.... 13 B. Financial Plan of ISKI................................... 14 C. Self-financing of Capital Investments................ 16 D. Foreign Exchange Exposure............................... 17 E. Government Assistance..................................... 18 F. Accounts Receivable..................... . ..... ...... 19 G. Financial Covenant..................... ......... .20 H. Current Ratioo.......................................... 21 I. Return on Assets.......................................... 21 J. Return on Equity................................ . 22 K. Debt-equity Ratio...*o**......... .. ......... . 22 L. ISKI's Financial Policy......... ooooo...o..*. ooooo. .. 22 M. Accounting and Audit.....*................. . . . ...... . 23 N. Maturity, Grace Period and Exchange Risk of the Bank Loan. 23 0. Environmental Aspects.............. .............. 24 P. Economic Assessment........... ..... ............ ..... 25 Q. The Role of the Bank ................ 26 III. CONCLUSIONS AND LESSONS.......... ............... . ........... 27 Annex 1 Comments from the Borrower .................................. 30 This document has a restricted distribution and may be used by recipients only in the performance of their otficial duties. Its contents may not otherwise be disclosed without World Bank authorization. PROJECT PERFORMANCE AUDIT REPORT TURKEY ISTANBUL SEWERAGE PROJECT (LOAN 2159-TU) PREFACE 1. This report presents the results of the Performance Audit of the Istanbul Sewerage Project for which Loan 2159-TU for US$ 88.1 million was approved on May 27, 1982. The Loan was made to the Istanbul Water Supply and Sewerage General Directorate (ISKI), the Guarantor being the Government of Turkey. 2. The Loan Agreement was signed on May 28, 1982 and the loan became effective on February 23, 1983 some six month after the originally planned date. The loan was closed on sch-dule on December 31, 1988 but the account was kept open until September 30, 1989 at which time the amount of US$ 54,000 was canceled. 3. The Project Performance Audit Report (PPAR) was prepared by the Operations Evaluation Department. The Parts I and III of the Project Completion Report (PCR) was prepared by the Europe, Middle East and North Africa Region and Part II by the Borrower. The PCR was forwarded to the Board for information on November 8, 1991. OED has reviewed the PCR, the Staff Appraisal and Presidents Report, the legal documents and Bank files and the transcript of the meeting of the Executive Directors at which the loan was approved and discussed the project with the Bank staff involved. An OED mission visited Turkey in November 1991 and discussed the project experience with officials of the borrower and other involved bodies. The cooperation and assistance of these officials is gratefully acknowledged. 4. The Audit found that the PCR gave a concise and adequate account of the project experience and has drawn appropriate conclusions. The PPAR further analyzed certain aspects of the PCR's findings and provides independent comments and conclusions on various aspecto of the project. 5. The draft PPAR was sent to the government and the borrower for comments. The comments received from the Government are reproduced as Annex 1 to the PPAR. ii PROJECT PERFORMANCE AUDIT REPORT TURKEY ISTANBUL SEWERAGE PROJECT (LOAN 2159-TU) BASIC DATA SHEET KEY PROJECT DATA Actual or Actual as % Appraisal Current of Appraisal Item Expectation Estimate Estimate Total Project Cost (US$ m) 222.1 200.7 90.36 Loan Amount (US$ m) 88.1 88.05 99.94 Economic Rate of Return N/A N/A N/A Institutional Performance: Good CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS Appraisal Estimate (US$M) 88.1 Actual (US$M) 88.05 Actual as % of Appraisal (%) 99.94 Date of Final Disbursement: September 19, 1989 PROJECT DATES Orisinal Actual Identification 1977 Preparation 1978 Appraisal 12/81 Negotiations 04/82 Board Approval 05/27/82 Loan Signature 05/28/82 Loan Effectiveness 08/82 02/83 Project Completion 12/87 12/90 Loan Closing 12/88 12/88 iii STAFF INPUTS (staffweeks) FY72 Ft4 FY75 FY76 FY77 FY78 FY7. FY80 FY81 FY82 FY83 FY84 FY85 FY86 FY87 Preappraisat 1.2 1.36 27.43 5.76 5.62 2.99 8.25 28.87 40.12 6.45 AppraisaL - - - - - 26.24 - Negotiation - - - - - - - 5.8 Supervision - - - - - 1.81 16.94 19.03 12.19 24.o' 13.81 Other - 2.42 2.43 1.28 6.78 1.63 0.11 0.74 2.3 8.77 - - - - FY88 FY89 FY90 FY91 TntaL Preappraisat - - - - 128.05 AppraisaL - - - - 26.24 Negotiation - - - - 5.8 Supervision 3.03 1.9 2.43 7.65 103.75 Other - - 26.56 MISSION DATA Date No. of No. of Month/Year Days Persons Supervision 07/82 5 2 Supervision 12/82 9 2 Supervision 06/83 9 2 Supervision 09/83 9 2 Supervision 03/84 5 2 Supervision 05/84 10 3 Supervision 12/84 8 2 Supervision 02/85 5 1 Supervision 07/85 5 1 Supervision 09-10/85 8 2 Supervision 05/86 5 2 Supervision 07/86 4 2 Supervision 02/88 5 2 Supervision 04/88 5 2 Supervision 07/88 3 2 Supervision 09/88 6 2 Supervision 12/88 3 2 Supervision 06/89 9 2 iv OTHER PROJECT DATA Borrower Government of Turkey Executing Agencies: IstAnbul Water Suppi 7 and Sewerage General Directorate (ISKI). V PROJECT PERFORMANCE AUDIT REPORT TURKEY ISTANBUL SEWERAGE PROJECT (LOAN 2159-TU) EVALUATION SUMMARY Background water born diseases were annual occurrences with high incidenct- of i. At the time of appra..al (1981) death (PPAR pars. 4-5). Turkey's population was about 45 million of which some 56% lived it iii. A major national effort to urban areas. The average growth rate improve service levels was launched was 2.5% but much higher at 4.1% in in 1978 with the "Conference on urban areas. Istanbul and Ankara, Planning and Development oC Community the two largest cities had population Water Supply and Sanitation." of 3.2 and 1.9 million respectively. Ambitious '.argets were set for The provision of water supply and achieving 100% water supply coverage sewerage services traditionally was by 1990 and sewerage by year 2000. the responsibility of municipalities The impressive progress toward these and village councils. As these targets were shown by the statistics usually lacked adequate human and of 1989, indicating water supply financial resources the actual design service levels at 97.7% and sewerage and implementation of service at 56%. The Conference also proposed facilities were gradually absorbed by improved sector organization three major government agencies arrangements. The first major step namely the State Hydraulic Works in this direction came with the (DSI), Iller Bankasi (Bank of the creation, in 1981, of the greater Provinces) and the Directorate for municipal areas and with it the Roads, Water and Electricity (YSA) creation of the autonomous entity of (PPAR paras. 2-3). the Istanbul Water Supply and Sewerage General Directorate (ISKI). ii. The 1979 sector statistics The new agency (ies) were to become indicated that about 63% of the total financially self supporting, through population had adequate water the charging of appropriate tariffs supplies. Remarkably, service levels and replace the old system of sector were about equal in urban and rural investments financed by government areas. Sever systems existed in the grant and low interest loans (PPAR larger cities only serving only about paras 5-7). 4.4 million people but none of these systems had treatment facilities. Bank Involvement The indiscriminate discharge of sewage into local receiving waters iv. Bank involvement started in the created unacceptable levels of late 1960a leading to the approval of pollution and sanitary conditions, Loan 844-TU in June 1972 for IS$37 particularly in the densely million for the augmentation of populated, low income areas were Istanbul's water sources. The project highly unsatisfactory. Outbreaks of was completed with a six years delay vi and substantial cost increases, vii. Preparation of the project had largely due to design changes, a long histor7 from the early 1970s procurement delays and rapid to appraisal in 1981. Much of this inflation. OED's PPAR (Report No. was taken up by the revision and 4853) was issued in December 1983. refinement of the Master Plan but it The project achieved its physical has also taken several years for the objectives 'ut it failed to improve Turkish government to create an the institutional and financial institutional form, ISKI, which was situation. Nevertheless the efforts acceptable to the Bank. made under this project contributed to the eventual creation of ISKI. viii. Physically the project The revision of Istanbul's sewerage consisted of construction of lateral Master Plan was assisted under IDA sewers, collectors and interceptors, Credit 324-TU (US$2.3 million) pre-treatment works, pumping granted for studies for urban facilities and a submarine outfall. developments in Istanbul (PPAR paras. Consulting services, laboratory and 8-9). maintenance equipment and rraining facilities were also provided for. The Proiect and its Obiectives It covered the most densely populated area of Istanbul and its principal v. The rapid growth of Istanbul impact was to be the prevention of after the post-war years resulted in discharge of waste waters on the large squatter settlements with south shore of the Golden Horn inadequate and ill maintained water and the North-East shore of the Sea supply and, particularly, sewera&e of Marmara (PPAR paras. 14-18). services. All waste waters were discharged, directly or indirectly ix. The fundamental premise of the into the main receiving waters, the disposal aspect of the Master Plan Golden Horn, the Sea of Marmara and was the heavily researched concept the Bosphorus. No waste water that the strong bottom currents of treatment was provided. The shallow the Bosphorus would carry pre-treated and relatively static waters of the sewage to the anaerobic depth of the Golden Horn became anaerobic at the Black Sea without environmental upper end and the popular basing damage. Effluent discharged into the beaches of the Sea of Marmara were Sea of Marmara was to receive full highly contaminated. The Bosphorus treatment. In the late stages of with its strong currents was less project implementation the affected but its north to south effectiveness of the "Bosphorus surface currents carried additional current" theory was brought into pollution to the Sea of Marmara. seri,us question, on the grounds that th? lower and upper currents do mix vi. The desire to improve these and significant volumes of sewage conditions essentially set the would be carried back into the Sea of objectives of the project which aImed Marmara. An intensive study to at the rehabilitation and extension examine the issue was lauichizd and of secondary sewer system, this is expected to be completed in interception of all discharges into 1993. However, this issue may be the Golden Horn, provision of rendered somewhat academic by the suitable disposal systems and the decision of the Turkish environmental improvement of the institutional and authorities to prohibit the discharge financial capacities of ISKI (PPAR of waste waters into any receiving paras. 10-13). waters without full treatment. This vii decision has major implications for US Dollar, the refinement of design the future sewerage developments in during construction and the efticient Istanbul and is discussed later in project management. The financing this report (PPAR paras. 19-21). plan of the project also changed significantly. Reduction of Implementation government contributions required ISKI to increase its internally x. The project had a late start generated contribution from the due to delayed effectiveness of the estimated 1% to 46%. Disbursements Bank loan and further, early, delays of the Bank loan lagged behind in the occurred due to the initial slow early years due to the delays award of major contracts. Once the mentioned, caught up by the fifth works got under way, implementation year and the loan was closed on was rapid and efficient resulting in schedule with the eventual high quality finish. This is cancellation of US$ 54,000 (PPAR particularly commendable as paras. 26-28). substantial design and alignment changes were required during Procurement and Contractors construction largely due to the congested areas on the route of the xii. The procurement process large interceptors and the need to presented no problems. While avoid damage to historical buildings. contract awards were delayed in tie Further changes were brought on by initial stages,, with the growth of the addition of the North Halic ISKI's autonomy matters improved interceptor which required the considerably. All consultants and amending of the Loan Agreement in contractors performed satisfactorily 1985. The originally planned (PPAR paras. 29-30). components of the project were completed about one year behind Institutional Aspects schedule and the loan was closed, as scheduled, on December 31, 1988 (PPAR xiii. The lengthy process leading to paras. 22-25). the creatiun of ISKI was followed by a remarkable rate of development Project Costs and Disbursements notwithstanding the magnitude of its task. At the same time as ISKI took xi. The original estimated cost of over the management of Istanbul's the project was US$222 million complex water supply and existing compared to the final cost of sewer system it also undertook the US$231.7 million, a 4% apparent implementation of the largest increase. This small difference infrastructure project in the city. hides considerable variations in the Technical progress was eminently cost of individual components as successful but the management of detailed in the PCR and this report. financial affairs was much harder to While the cost of some components get under control. Particular were substantially overestimated at problems wL-- faced in billing and appraisal, major, overall, savings collection for which several permitted considerable additional alternative arrangements were tried works to be carried out. The main including a semi-private collection contributing factors were the agency, ultimately, successfully competitive bids of Turkish (PPAR paras. 31-33). contractors, the devaluation of the Turkish lira, the fluctuation of the viii Operations and SustaLnability whose outcome is uncertain (PPAR para. 57). xiv. Operations and maintenance of the water supply facilities is xviii. The Bank Loan contributed to satisfactory. ISKI's operating the exchange risk build up as well as capacity for the sewer system, to the mismatch of maturities between particularly of the proposed advanced borrowings and the economic life of treatment facilities is yet to be fixed assets (PPAR paras. 59-60). proved but its present performance gives no reason for doubt. A much Environmental Aspects more severe test will face ISKI's financial management in the funding xix. Although by nature and design of the planned major investments and the project was a straight forward the continued reliable maintenance of urban infra3tructure investment, the expanded facilities. However, in there is no doubt that the principal the light of the findinge of this motive force for it was environmental review. tht %udit sez no reason for protection and improvement. While concern regarding the sustainability concern for public health was also a of either the physical facilities or strong incentive, the formidable the benefits of the project (PPAR public and government support was paras. 34-35). "bought" by the promise of clearing up the Golden Horn and the shores of Financial Aspects the Sea of Marmara. xv. ISKI has shown a very rapid xx. The technical solutions for the growth of its balance sheet, partly collection and interception of sewage due to the large investments carried flows were standard and involve only out in the 1980s and partly to the construction difficulties in the high rate oi inflation. The congested historical parts of the revaluation of its fixed assets in city. Prevailing scientific and operation allowed ISKI to justify engineering views also proposed l.Arge tariff increases in recent apparently satisfactory solutions for years but collection is lagging sewage disposal based on the behind billings and the ensuing transporting and assimilative arrears are threatening ISKI's credit capacities of the Bosphorus. In the worthiness at the time when another late stages of the project not only ambitious investment program is were these thsories brought into planned for the 1990s (PPAR paras. question but the evolving 36-47). environmental awareness of the relevant Turkish authorities raised xvi. Although ISKI is not facing the required disposal standards into liquidity problems the financial all receiving waters to fully treated indicators based on cash flows have effluent. been negative since 1987. The debt- equity ratio has been between 1 and xxi. The implications of the latter 2:1 but the Revaluation Fund which decision are very large. It not only has no cash equivalent makes up most requires the construction of large, of the equity (PPAR paras. 53-56). complex, mechanical and/or biological treatment plants with associated xvii. ISKI's financial policy has operational difficulties but, the changed since 1987 which shows the problems of locating suitable sites utility being in a transitory period for these plants, due to Istanbul's ix topography, would require lonp and The Role of the Bank axpensive pipelines and tunnels. The technical feasibility of these xxiv. There is every indication that solutions is not in question but the Bank's more than twenty year their affordable phasing and funding sustained relationship with Istanbul will severely test ISKI's capacities and, increasingly, the rest of Turkey as well as the Bank's guiding hand. was effective and beneficial. In the earlier years the primary targets of xxii. Another of ISKI's policies Bank advice were the technical and which deserves particular note is the institutional issues with particular decision to develop, in the future, credit due for helping to create water supply -and sewerage services ISKI. Consistently competent Bank completely in parallel. The findings staff in adequate strength was, of a recent OED study of some 130 undoubtedly a major factor in the Bank funded water supply and sewerage effectiveness of the assistance. projects confirm that this is, While sound technical advice will probably, a unique policy in continue to be required in view of developing countries. IF fully the ever increasing complexity of implementec. it wi'_l not only ensure future developments, the Bank should the full public health impact of increase its emphasis on further providing safe water supply but, in improving ISKI's financial management view of the service level target and and performance if the monumental achievements mentioned earlier, it planned investments are to become an will also ensure equitable service to affordable reality (PPAR paras.72- the poor (PPAR paras.61-68) 73). Economic Assessment Conclusions and Lessons xxiii. The benefits resulting xxv. The project achieved all its from the project to public health, physical objectives. Achievement of increased property values, the the financial viability objecti.ve was fishing industry, public beaches and made less convincing by the tourism could not be quantified. The fluctuation evident in some years. SAR calculated the Average (long run) Incremental Cost (ACI) of the project xxvi. After initial delays, due to investments, at the time, as a guide highly bureaucratized procedures for a future tariff study. The PCR project implementation proceeded recalculated these values using the efficiently. Substantial cost same methodology but with revised savings materialized on the original cost and benefit streams. The components partly due to the resulting ACI, for combined water and competitive performance of Turkish sewerage services, of 0.65 US$/m3 of contractors and partly to water consumed is less than the 1990 overestimation of the cost of some average tariff charged by ISKI at components. These savings permitted 0.80 US$/m3 confirming that ISKI the execution of substantial charges economic prices for its additional works. The quality of services. Because of the progressive finished works was high. There were tariff structure the tariffs are no procurement problems. affordable to the lowest income group at an estimated 3% of household xxvii. The beneficial impact of the income (PPAR paras. 69-71). project is already visible on the quality of water in the Golden Horn x and the environmental improvements are enhanced by the municipality's city beautification works in the same area, benefiting both the local and tourist population. The extensive public relation work carried out by ISKI and the Municipality undoubtedly contributed to the acceptance by the population of Istanbul of the project and the corresponding tariff increases (PPAR paras. 74-76). xxviii. The principal lessons to be drawn from this project are the following: (i) sewerage/waste disposal projects need to be planned with not just the consideration of individual service improvements but with the full demonstration of the overall environmental benefits to the community as a whole and the informing of the beneficiaries accordingly if the public is to accept the cost; (ii) a well organized institution with adequate autonomy is essential for successful implementation and the Bank is well justified in waiting for this to materialize before proceeding with a project. PROJECT PERFORMANCE AUDIT REPORT TURKEY ISTANBUL SEWERAGE PROJECT (LOAN 2159-TU) I. PROJECT SUMMARY A. Background 1. Some 97% of Turkey's 780,000 km2 territory lies in Asia and 3% in Europe. The country's most developed region is its coastal area which also receives the highest level of precipitation exceeding 600 mm per annum. The arid plateau of central Anatolia receives around 300 mm and it is less developed. The mountainous Eastern Anatolia with its harsh climate is the least settled and developed region of the country. 2. The country's population in 1980, (appraisal 1981) was about 45 million of which about 24 million (56%) lived in urban areas. In the preceding decade the average population growth rate was 2.5% - 4.1% in urban and 0.3% in rural areas. Istanbul and the capital city, Ankara, had populations of 3.2 and 1.9 million respectively, together representing about 21% of the total urban population. 3. Traditionally the provision of water supply and sanitation services has been the responsibility of municipalities and village councils. However, with the exception of a few large cities, the local bodies generally lacked the human and financial resources to provide these services efficiently and, over time, the planning, design and construction of new facilities were gradually absorbed by three central government agencies which were, originally, created as advisory and financing agencies. These were: The State Hydraulic Works (DSI) of the Ministry of Power and Natural resources was responsible for national water resources and water supply development in cities exceeding 100,000 population; Iller Bankasi (Bank of the Provinces), an autonomous corporation responsible for water supply in the remaining towns and sewerage developments in all urban areas except Istanbul. Iller Bankasi's shareholders were the municipalities; Directorate for Roads, Water and Electricity (YSA) in the Ministry of Rural Affairs and Cooperatives was responsible for rural programs excluding rural sanitation for which no agency carried formal responsibility. 4. Available statistics for 1979 indicated that some 63% of the total population had adequate water supplies, 18% were inadequately served and 19% had no access to public water supplies. Remarkably, the service levels were about 2 equal in urban and rural areas. Sewer systems existed only in the major cities, serving only about 4.4 million (18%) people and none of the systems had treatment facilities. Indiscriminate discharge of liquid and solid wastes resulted in the serious degradation of rivers and coastal waters. Sanitary conditions, particularly in densely populated, low income, urban areas were highly unsatisfactory. These conditions inevitably contributed to the frequent contamination of drinking water supplies; of water samples tested in 1977, 38% were found bacteriologically unsatisfactory. The outbreak of water born diseases was an annual occurrence. In 1977 the recorded cases excaaded 15,000 with 143 deaths. 5. A major effort to improve the service levels in the country was launched with the organizing, in 1978, the "Conference on Planning and Development of Community Water Supply and Sanitation." The Conference set the following service level targets: i) by 1990 the total urban and rural population would be served with adequate water supply; ii) by 1990 the rural population would be provided with satisfactory excreta disposal facilities; and iii) by year 2000 all urban populations would be served by satisfactory sewage disposal facilities. These were ambitious targets yet, as of 1989 (latest statistics) the following remarkable achievements are reported: Urban Population served with water = 97.7% Rural Population served with water - 85.0% Total Population served with sewers - 56.0% Total Population served with septic tanks - 29.0% 6. The Conference also proposed alternative sector organizational arrangements to eliminate the existing gaps and overlaps in responsibilities and increase overall efficiency. However, major changes did not take place until 1981 following the creation of the greater municipal areas. Under the relevant Executive Order, the water supply and sewerage services of small municipalities were merged with the nearest large urban center. In the case of Istanbul, 24 surrounding municipalities were merged with the services of ISI, Istanbul's semi- autonomous Water Supply Authority. In the same year, ISI was "converted" into the Istanbul Water Supply and Sewerage General Directorate (ISKI), an autonomous entity responsible for the planning, design, construction and operation of all water supply and sewerage services in Grater Istanbul with the right to set water and sewerage charges without outside approval. 7. In the past, sector investments were generally financed by government grants or low interest loans (which were seldom repaid). Under the new arrangements the Government's intentions were to enable the sector entities to become financially self supporting through the charging of appropriate tariffs and they were expected to finance a gradually increasing proportion of their investments by internally generated funds. 3 B. Bank Involvement 8. The Bank's first involvement in the sector was the approval of Loan 844-TU on June 30, 1972 for US$ 37 million to help finance the Istanbul Water Supply Project. The primary objectives of this project were the augmentation of Istanbul's water supply which, at the time, could only meet 60% of the area's water demand and major improvements in the organization, management and financial performance of ISI. The project suffered somes 6 years delay in completion due largely to design changes, procurement delays and institutional inertia. The project was completed in December 1982 and OED's PPAR was issued in December 1983 (Report No 4853). Project costs have increased from the appraisal estimates of US$ 85 million to US$ 159 million due to rapid inflation, delays and design changes. The physical targets were achieved in that ISI's production capacities were increased from 115Mm3 to 315Mm3 by 1981. However, water sales lagged far behind, mainly due to the slow expansion of the distribution system and the increase rather than the projected decrease of the level of unaccounted for water (43% v. 26%). The project also failed to meet its institutional and financial objectives. However, the overlapping efforts of this project and the preparatory work for the Istanbul Sewerage Project did contribute to convincing the Government of the need for major changes in the sector which, eventually, brought about the creation of ISKI in 1981 and thereby laid the groundwork for the subsequent significant improvements in all areas of the provision of water and sewerage services in Greater Istanbul. 9. Simultaneously with the water supply project, IDA Credit 324-TU was approved for US$2.3 million for the preparation of master plans and related studies for urban development programs in Istanbul. Under this credit, assistance was provided for the review and revision of the Master Plan for Istanbul Sewerage. While this small project was also much delayed and was largely unproductive, the revised Sewerage Master Plan provided the foundation for the Istanbul Sewerage Project (Loan 2159-TU) the performance of which is the subject of this review. C. The Proiect and its Obiectives 10. Istanbul, the capital of the former Byzantine and Ottoman Empires and one of the oldest historical cities in the world, is the foremost cultural, industrial and economic center of Turkey. The population of Greater Istanbul, at the time of appraisal, was estimated at 3.73 million, projected to increase to 7.84 million by the year 2000. The city' industrial growth drew a heavy immigration of workers and resulted in the proliferation of large squatter settlements. The overall growth rate of the city, in 1980, was recorded as 4.1% projected to decline to about 2.6% by 2000. 11. The rapid growth of the city overtaxed both the water supply and waste disposal services which, particularly in the densely populated, low income, areas resulted in unsatisfactory maintenance of the old, and frequently poorly constructed, sewer lines and holding or septic tanks. All sewage was discharged into the nearest receiving waters, usually at the shore line, causing heavy pollution. No waste water treatment was provided. 4 12. The topography of the Greater Istanbul area naturally divides it into 14 drainage zones. About 90% of the effluent went into the Golden Horn and the Sea of Marmara and 10% into the Bosphorus. The water flows from the Aegean Sea through the Dardanelles into the Sea of Marmara and the Bosphorus to the Black Sea as a bottom current. A surface current of lower salinity water flows back from the Black Sea to the Aegean Sea. While the deep, strong currents of the Bosphorus provides effective dispersion of effluent, the weaker surface currents of the Golden Horn (particularly in its inner portions) and the Bosphorus carry a heavy pollution load to the Sea of Marmara, resulting in unacceptable shore pollution inhibiting leisure use and tourist potentials. Water samples at the Sea of Marmara basing beaches frequently showed unacceptable levels of coliform bacteria content. Testing of ocean bottom sediments showed high concentration of lead and other heavy metals in the upper Golden Horn and chromium in the Marmara near the outlets from tanneries. 13. The conditions described above essentially set the objectives of the project which were straight forward and appropriate. They are set out in the Appraisal Report as follows: (i) to extend and improve the coverage of sewerage services by rehabilit&tion and extension of the senondary sewer system to cover 60% of the population of which some 70% would be low income people; (ii) to improve environmental conditions by reducing the amount of sewage discharged at the shorelines of the Golden Horn and the Sea of Marmara by the construction of an interceptor system; (iii) to provide satisfactory sewage disposal by construction of pre-treatment facilities and a deep submarine outfall main; (iv) to improve the institutional capacity in project preparation and implementation, operation and maintenance and financial management by the development of a suitable organization; and (v) to assure financial self-sufficiency by the introduction of adequate water supply and sewerage charges. A broader objective was the development of country wide procedures for pollution control. 14. The preparation of the project had a long history. The first, comprehensive Master Plan for solving Istanbul's sewerage problems was prepared in 1971. Over subsequent years this was repeatedly reviewed and revised and, to this day, continues to be changed in the light of additional information and sharpened environmental considerations. Focused Bank interest -Ln a sewerage project is first recorded in 1972 with financing of related studies included in Credit 324-TU, approved for funding the preparation of an urban development master plan for Greater Istanbul. 15. All variations of the sewerage Master Plan were based on the concept of discharging sewage effluent, treated to various degrees, into the Bosphorus 5 and/or the Sea of Marmara through deep or shallow submarine outfalls. The 1983 version of the Master Plan recommended 12 separate sewer systems, eight discharging into the Bosphorus and four into the Sea of Marmara. Prevailing scientific and engineering opinions held that pretreated sewage (removal of grit, oils and floatables) can be discharged into the bottom currents of the Bosphorus through deep outfalls where high degree of initial dilution (1:100) will be achieved and all residuals will be safely carried by these currents to the Black Sea, the "world's largest anaerobic digester". As the assimilative capacity of the Sea of Marmara was considered much more limited, discharge, deep or shallow, was considered to require higher degree of treatment to achieve environmentally acceptable disposal of effluent. The Golden Horn was to receive no discharge in any form. The above principles and the topographic delineation of the drainage basins effectively defined the proposed layout and locations of the interceptor and outfall network to be constructed (in stages) for the Greater Istanbul area. 16. Although the PCR gives the date of "identification" for the project as 1977, file records of activities go back to 1972. With appraisal taking place in December 1981, the project had an exceedingly long "preparation" period. Much of this was occupied by the Bank waiting for revision of studies, decision with regard to priority of components to be constructed and, above all, for the Turkish government to enact legislations creating the institutional format which satisfied the Bank. 17. From this process emerged ISKI, the borrowing (with government guarantee) and implementing agency and the Project, as eventually appraised. In this form it covered Istanbul's most densely populated area from the south shore of the Golden Horn to the western end of the medieval walled city encompassing much of the Yenikapi-Kabatas drainage basin. Its principal environmental impact was to be removing sewage discharges from the south shore of the Golden Horn and the north-eastern shore of the Sea of Marmara. This impact was further enhanced with the parallel "beautification" efforts by the city in these areas with the relocation of the proliferating small and medium size industries which also represented major point pollution along the shore lines. 18. The project consisted of the following major components: (i) laterals and collectors (285 km) (ii) interceptors (26 km) (iii) tunnel (2.4 km) (iv) pre-treatment works and pumping stations (v) pumping mains (2.3 km (vi) submarine outfall (2.0 km) (vii) consulting services, industrial waste water and oceanographic surveys, laboratory facilities, operation and maintenance equipment and establishment of a regional training center. As Istanbul is rich in historical buildings, the proposed alignment, design and method of construction of each facility was subject to review by the Turkish Archeological Commission. The design and method of construction of the large interceptors was further influenced by their potential effect on traffic flows within the already congested city. As these criteria essentially excluded the possibility of future duplication of the interceptors and the outfall, their 6 hydraulic design was for maximum capacity as dictated by the projected saturation development of the areas served. On this basis the design flows for the interceptors along the Golden Horn and the Sea of Marmara were 5.5 and 6.5m3/sec respectively and for the pretreatment plant and the sea outfall 12.0 m3/sec. 19. As mentioned above, the fundamental assumptions guiding the formulation of the Master Plan and, consequently the formulation of the project, were the scientific and engineering understanding of the flow of currents in the Bosphorus and the Sea of Marmara and the capacity of these waters to dilute, absorb and transfer sewage effluent without adverse environmental effects anywhere. The characteristics of the Bosphorus currents have been studied extensively. Mediterranean water increases in density and salinity from the excess of evaporation over precipitation as it circulates from Gibraltar along the African coast to the Aegean Sea. From the Aegean this heavy, highly saline water flows through the Turkish straights to the anaerobic lower portions of the Black Sea with depth below 2000m. Less dense, brackish, surface waters carry runoff from the tributaries of the Black Sea (mainly the Danube and the Dnieper) and create a surface current in the opposite direction. Various surveys attributed a range of values to the volumes and velocities of these currents. Generally accepted figures range from 3,000 to 6,100 m3/sec for the bottom and from 12,600 to 30,000 m3/sec for the upper current. There is a stable boundary layer between the upper and lower currents at depth of about 10 m at the south entrance but it lies below 35 m at Kilometer 20 in the Bosphorus. These characteristics indicate that sewage discharged into the bottom current via deep outfalls with or without diffusers would mix rapidly with sea water and reach the Black Sea in about 18 hours. Studies show that neither significant oxygen depletion nor excessive sedimentation would result from this process in the Bosphorus. 20. While the general characteristics of the process described above are not being disputed, in the late stages of the project, heavy criticism was leveled by the new administration of Istanbul against another important design assumption namely that the lower and upper currents do not mix. It was claimed that such mixing does occur and that partially treated sewage will enter the upper current and will be carried back to the Sea of Marmara perpetuating its pollution. To eliminate this danger it was proposed that full biological treatment be provided at the Yenikapi (and all other proposed) outfall and an extensive oceanographic sampling study of the Bosphorus currents has been commissioned (to be completed in 1993). A further step in this direction which, in effect, overrides the specific issue of the mixing of the currents, came in mid 1991 when the Turkish environmental authorities issued a directive that, in future, all waste waters must receive full biological treatment before discharge into receiving waters. ISKI has immediately adopted this policy. 21. The project under review has been virtual completed when the issue of the mixing of the currents surfaced and the Yenikapi outfall has been functioning, technically in violation of national environmental policies and will do so (in the absence of possible alternative solution) until such time when a full treatment plant can be built. The effect of these environmental policy changes are potential much more serious for the ongoing second project, Istanbul Water Supply and Sewerage (Loan 2888-TU). This project, for which a loan US$218 million was approved in 1987, includes, inter alia, provisions for the construction of pre-treatment plants and submarine outfalls at Uskudar, Kadikoy 7 and Baltalimani. On the basis of the new policies the Government is currently denying permission ior ISKI to construct the Uskudar outfall (contract due for bidding) and similar action is likely for Baltalimani. The longer term implications of the policy are a massive increase in costs in ISKI's future sewerage investment plans. The Audit field mission discussed these matters with ISKI's management and the overall impression is that unless a satisfactory program of staged construction of the future facilities can be arrived at, it is doubtful whether ISKI would have the financing capacity to carry out the program and, perhaps more important, the costs would be affordable to Istanbul's population. These issues are under intensive discussions between the Turkish authorities and the Bank. D. Project Implementation 22. The very lengthy preparation period mentioned above was followed by an impressively rapid loan processing. Appraisal was in December 1981, negotiations took place in April 1982, and Board approval and loan signing followed on the 27th and 28th of May 1982. At that point bureaucracy intervened and effectiveness only took place on February 23, 1983 (there were no special, project related, effectiveness conditions). The loan was closed on the originally planned date of December 31, 1988. 23. The originally scheduled components of the project were virtually complete by December 1988, about one year behind schedule. Data in the PCR (Implementation Schedule) clearly shows that all major contracts were delayed at the start, largely due to a laborious approval process which, however, later much improved. Once the contracts were approved, execution times were generally shorter than estimated. The contract for thG supply of large pipes for the interceptors was delayed by about on year due to the dispute whether a new pipe factory should be provided or an existing one should be utilized. The latter solution prevailed. 24. As mentioned above the construction of the interceptors which had to be laid in congested areas and, often, in the proximity of historical buildings. These conditions, inevitably, resulted in changes in the routing and, in some cases the type (tunnel versus pipe) of these large conduits. The comparison of planned and actual construction is shown in Table 1. below. A major change which required an amendment of the Loan Agreement (October 23, 1985) took place about mid way through the project implementation. At the end of 1984 ISKI reviewed the project cost estimates and concluded that substantial savings were likely due to the devaluation of the Lira and lower than expected bid prices. To utilize these savings the Bank was requested to agree to the inclusion of additional components to be constructed on the north side of the Golden Horn namely the Rabatas Interceptors, treatment plant and outfall. Subsequently it turned out that the reestimation of the project costs were erroneous and only the North Halic (Golden Horn) interceptors could be accommodated and all other additional components were to be completed under the second project. 8 Table 1. PROJECT COMPONENTS Description Appraisal Estimate Actual South Halic Interceptor 26.0 km 10.5 km Fatih Tunnel 2.4 km 6.5 km Yenikapi Treatment Peak Capacity=12 m3/sec As designed Plant Pumping Mains 2.3 km 2.3 km Submarine Outfall 2.0 km Twin Pipelines 1.1 km Laterals/collectors 285.0 km Substantially more North Halic Interceptor Not foreseen 8.4 km 25. The changes in the project components, while entirely within the project description, were significant though not outside the expectations for this type of work. To have completed the works within the estimated construction schedule in the congested conditions of Istanbul and with th-- high quality observed by the Audit mission is a credit to all parties involved and augurs well for ISKI's future program. E. Proiect Costs and Disbursements 26. Actual total cost of the project amounted to US$231.7 million equivalent compared to the appraisal estimate of US$222.0 million, an increase of 4%. The closeness of these figures hide significant differences in component costs as would be expected from Table 1. above. Furthermore, in addition to the North Halic Interceptor substantial amounts of water supply pipes and water meters were financed, on an emergency basis, under the project resulting in a total "not foreseen" expenditure of US$46.92 million equivalent. According to the PCR, the actual cost of the secondary sewer system exceeded appraisal estimates by 168%, due entirely to the expanded scope of this component. If this is correct, it implies that an additional US$60.68 million equivalent extra cost was also met from overall savings under the original project components giving a total "savings" of US$107.6 million. A detailed compa-ison of the appraisal and actual costs is shown in Table 2. below. All costs are expressed in US Dollars and the Turkish Lira costs were converted at the exchange rate ruling at the time the expenditure was incurred. The line items of the Appraisal cost estimates include the relevant physical and price contingency elements. Some of the cost differences represent clear and substantial overestimation at the time of appraisal such as the treatment plant (75%) and the Fatih tunnel where the actual average unit cost per meter length was only 50% of that estimated at appraisal. On the other hand similar unit cost for the interceptors show a mere 7% 9 difference. Precise analysis of the reasons for these differences is hadly ever possible but fil#z search and the Audits field discussions seem to indicate that the following factors may have had significant influence: (i) the unanticipated competitiveness and competence of local contractors and suppliers; (ii) the thirteen fold devaluation of the Turkish Lira during the implementation period; (iii) the fluctuation of the value of the US Dollar; (iv) the continuous refinement of the design of the works during construction and (v) the efficient supervision of construction by ISKI and its consultants. Table 2. (US$ Million Equivalent) DESCRIPTION APPRAISAL ACTUAL CHANGE Z Interceptors 42.78 18.56 -57 Fatih Tunnel 11.21 15.03 +37 Pretreatment Works -- civil works 40.15 10.34 -74 -- equipment 25.52 6.10 -76 Submarine Outfall 44.41 13.63 -69 Secondary Sewers 36.19 96.97 +168 0 & M Equipment 10.73 14.09 +31 Land Acquisition 1.16 0.90 -22 Proj. Mgt. & SPN, 8.05 7.4 -8 TA & Training PPF 0.50 0.5 0 Front-end Fee 1.30 1.30 0 North Halic Not Included 15.46 New Item Interceptor Water Pipes & Not Included 31.46 New Item Meters TOTALS 222.00 231.75 +4 10 27. The financing arrangements, as agreed at appraisal and as actually carried out are shown in Table 3. below: Table 3. PROJECT FINANCING SOURCE APPRAISAL ACTUAL US$ 2 US$ % Million Million IBRD 88.1 40 88.0 38 Government Contribution 131.2 59 37.6 16 Internal Cash Generation 2.8 1 106.1 46 TOTAL 222.1 100 231.7 100 The most dramatic change in the financing plan was the increase in ISKI's self f*nancing percentage. At the time of appraisal the Bank clearly had low expectations regarding the revenue generating capacity of ISKI and the financial performance covenant of the Loan Agreement (Section 5.04 (a)) only required ISKI to generate ..."revenues sufficient to cover its operating and maintenance costs and its debt service in excess of provisions for depreciation". For the financing of project expenditures the Guarantee Agreement obligated the Government to... "provide the Borrower or cause the Borrower to be provided with such funds as are needed to meet such expenditures". ISKI's overall investment funding requirements over the "project period" were estimated at appraisal as US$ 375.2 million of which US$201.6 million were to be provided by the Government "in the form of contribution (Guarantee Agreement)". The PCR reports that the Government's contributions to ISKI's investments started to decline already in 1983 and, by 1987, the project financed by Loan 2888-TU shows no Government contribution. The withdrawing of government funds forced ISKI in the right direction of financial self sufficiency and the evident success of this is further discussed in the financial section of this report. 11 28. Disbursements were scheduled to take place over a period of six and a half years, about one year less than the country profile at the time of appraisal. Due largely to a six month delay in effectiveness and the initial slow award of major contracts, disbursements lagged significantly in the first years of the project but caught up with the forecasts in the fifth year. The Loan was closed as scheduled on December 31, 1988 although disbursements continued until September 1989 at which time US$54,000 was canceled. The comparison of projected and actual disbursements is illustrated in Graph 1, below. GPAPH 1 g0 110 80 100 z 6D0 -80 m) L -70 50- 0 -60 r40- -50 ~ -40 20 -30 10 x1 0L = -. 110 1983 1984 1985 1986 1987 1988 1989 1990 FISCAL YEAPS Appraisal Estimate A Actual -'- Percentage F. Procurement 29. In respect of following the Bank's procurement guidelines or generally managing the procurement process in a satisfactory manner, neither the PCR nor the file search highlighted any significant problems. However, while the process was satisfactory, efficiency, particularly in the early stages, was far from laudatory. Most major contracts suffered nearly a years delay in the award process due to the cumbersome approval process. Subsequently with ISKI's autonomy increasing this process is also improved significantly. While all major contracts were open to international competitive bidding, little foreign interest 12 was shown, a testimony to the known competitiveness of the Turkish construction industry. G. Performance of Consultants and Contractors 30. The conclusion of the PCR that all consultants, contractors and suppliers generally performed well was confirmed in the discussions of the field Audit. The Audit mission also visited a number completed and in progress facilities and noted the high quality workmanship in construction and installation of equipment. H. Institutional Developments 31. The evolution of ISKI has been described in earlier sections of this report. It was along and tortuous process due both to the complexity of the existing sectoral responsibilities and the political implication of major changes. Over the years it took to arrive at the final solution the Bank followed a correct and single minded approach which eventually bore fruit. The present organization had a monumental task at its birth namely to take over and mange the existing responsibilities of four organizations, the DSI, Iller Bankasi, ISI and the Istanbul Municipal Sewerage Department. At the same time, the new agency had to take in hand, "on the run", the implementation of one of the largest infrastructure projects ever launched in Istanbul. The resources for this were limited. General shortage of skilled manpower, staff drawn from different organizations and limited experience in managing an organization of the size and complexity of ISKI. Particularly severe were the shortages in competent financial personnel. 32. .lthough progress was far from trouble free, ISKI's development was remarkable by any standard. It is rare to see, in similar Bank financed projects, that an implementing agency which is created at the very start of the project (regretfully a common occurrence) execute a large and complex project with the efficiency that ISKI achieved. This project, once again, amply demonstrate, that adequate autonomy is an absolute prerequisite to successful institutional and managerial development. 33. While the technical side of ISKI's development was steady and satisfactory, the progress of financial management was more uneven. Admittedly there was a longer and, in many ways, harder road to travel and the issues involved, such as new accounting system, massive tariff increases and improved billing and collection, were more intractable than the technical aspects with, as is usually the case, even grater manpower constraints. I. Operational Aspects 34. There is limited information on the operating performance of ISKI, particularly on the sewerage side. The project provided the first proper waste disposal facility in Istanbul, the Yenikapi Pre-Treatment Plant and Outfall and there is no long term performance record as yet. The Audit mission's visit to the plant found fully satisfactory operating and maintenance practices. Another visit to ISKI's computerized water supply coatrol center confirmed that system 13 control, in terms of sophistication and quality of operating staff had left little to be desired. J. Sustainability 35. At present, ISKI appear to have the technical, managerial and financial (see following section) capacity to fully sustain the performance and benefits of its water supply and sewerage facilities. However, ISKI also has a development plan, particularly in sewerage, which, in its scale, dwarfs its present facilities. The future introduction of secondary and (perhaps) tertiary waste treatment facilities will demand a new level of operating skills. This, combined with the projected intensive (and necessary) pollution monitoring and control, on land nd at sea, will demand a quantum increase in both management and staff training/recruitment. While the task is more than formidable, the findings of this Audit give no reason to doubt that ISKI will cope with the problems provided its autonomy is not impaired and realistic staging of its future developments facilitates the maintenance of a sound financial base. II. ISKI's FINANCIAL PERFORMANCE A. Compounded Growth Rates of Financial Indicators 36. At the end of 1990, ISKI was still a growing utility with a balance sheet totaling TL2.39 billion or US$816.2 million, but it had been through a rapid growth in the 1980s. In nominal or current TL terms, Table I shows growth rates well above 50% compounded per year for all major components of its income statement and balance sheet. In real terms, however, the same growth rates are much lower albeit still in the double digit range except for the share capital. Also noteworthy is the marked slowdown of these real terms growth rates in the most recent period which will have a bearing on the implementation of the strategic plan for the 1990s. The apparent favorable trend of real terms profitability is entirely due to exceptional profits recorded in 1990 following a loss in 1989 and which may not be indicative of future profitability. Overall, the inconsistency of growth rates over the last three years (range from 0% to 187%) indicates an entity which has not yet absorbed the impact of the large investments undertaken in the early 1980s. 14 Table 1: ISRI's Nominal Growth Rates in Z compounded p.a. in current terms in real terms' 1990/1982 1990/1987 1990/1982 1990/1987 gross fix. assets in op +101.5 + 94.9 + 52.3 + 28.7 works in progress +122.6 +113.2 + 73.4 + 47.0 sales of water + 91.1 +106.9 + 41.9 + 40.7 net profits + 72.7 +186.8 + 23.5 +120.6 depreciation + 92.2 + 86.9 + 43.0 + 20.7 cash flows + 76.1 +153.6 + 26.9 + 87.4 reserves +361.2 +109.2 +312.0 + 43.0 paid-up capital + 19.9 + 0.0 - 29.3 n.a. long-term loans +129.8 + 76.0 + 80.6 + 9.8 37. Total fixed assets increased 368 times between 1982 and 1990 in current TL terms. In real terms (i.e., before revaluation of fixed assets in operations), the growth is still 52 times. In US dollar terms, it is 24 times as shown in Table 2. At the end of 1990, total fixed assets amounted to TL1,912 billion or US$653 million, of which 51% were in operation. Table 2: Fixed Assets in TL billion/% 82 83 84 85 86 87 88 89 90 Fix.assets in operation 3.6 14.1 41.3 61.? 91.9 132.3 339.1 588.8 978.6 Works in progress 1.6 14.6 3.3 11.3 39.7 96.5 177.0 433.0 935.0 Total fixed assets 5.2 28.7 44.6 72.5 131.6 228.8 516.1 1022 1912 in real terms (ISKI) 3.6 6.0 22.2 26.1 40.9 52.5 130.4 140.9 186.4 in US$ terms (million) 27.6 101.4 100.5 125.7 :73.7 224.1 284.4 441.6 653.1 WIP/total fix. assets 30.8 50.9 7.4 15.6 30.2 33.4 34.3 42.4 48.9 B. Financing Plan of ISKI 38. In four years, -otal resources have increased almost sevenfold in current TL while they more than doubled in constant US dollars. Table 3 provides the financing plan for the period 1987-1990 which corresponds to a change in ISKI's financial policy (PPAR, para. 57). ISKI has relied almost equally on long-term borrowings from abroad (mostly from the Bank) and on its Revaluation Fund to finance its expenditures. Cash flows have not been as steady a source of funds as the build-up in short-term liabilities. On the use side, a growing share of resources has been allocated to fixed assets although this is partly due to their 1 Deflated by the consumer price index (IMF, "International Financial Statistics", January 1992). The compounded rates of inflation were 49.2% p.a. from 1982 to 1990, and 66.2% since 1987. 15 annual indexation onto the inflation. Indeed, works in progress (WIP), which are not revalued, have been more erratic in terms of share of total funding. 1987 marked a change in ISKI's financial policy when all the share capital was repaid (US$9.8 million) and 68.4% of retained earnings were transferred out of its balance sheet. Table 3: Financina Plans for 1988-19) 1987 1988 1989 1990 Sources (in Z of total) Cash flows 6.1 11.5 3.1 15.0 Long-term Borrowings 63.3 37.2 31.9 36.9 (of which) Foreign loans 58.5 36.2 29.6 31.9 Revaluation Fund 16.2 28.6 34.5 29.2 Increased Current liabilities 14.5 20.9 19.0 16.9 Decreased current assets 0 1.5 11.4 1.4 Uses (in % of total) Increased fixed assets 54.5 63.9 74.0 76.0 (of w.) Work in progress 31.9 17.9 36.9 42.7 Increased current assets 27.5 34.6 21.7 20.8 (of w.) Accounts receivable 9.8 15.6 17.5 17.6 Decrease of net worth 18.1 0 0 0 Total funding (TL billion) 178.02 450.56 693.65 1176.18 same in US$ million 207.67 316.80 326.93 450.88 39. Table 4 summarizes the recent trends in financing ISKI's various assets. The sum of cash flows, of long-term borrowings, and of fixed assets revaluation has financed easily the increase in fixed assets during the last five years. The large share contributed by the Revaluation Fund may, however, give the impression that ISKI has not relied heavily on long-term borrowings (49% in 1990) to finance its capital investments. This is an accounting presentation required to offset the upward revision of fixed assets in operation. Actually long-term borrowings do finance a large share of the physical investments (i.e., total fixed assets less the revaluation for the year). The last line of Table 4 shows that, although the trend has been downward, long-term borrowings financed 90.5% of cumulative investments between 1986 and 1990. Since foreign loans represented an average 86.9% of total long-term borrowings over that period,2 it implies that a large share of local costs has been financed with foreign currency resources. Nominal cash flows have financed a small though uneven share of total fixed assets, but this is misleading because of the large accounts receivable which decrease the free cash flows (PPAR, paras. 11-14). 2 In part due to the exchange risk added at the end of year onto the outstanding balance of foreign borrowings. 16 Table 4: Financing Ratios 1986 1987 1988 1989 1990 Lg-term borrowings/tot.fix. assets 77.6 113.3 58.2 43.1 48.6 Foreign loans/total fixed assets 66.7 107.3 56.7 40.0 42.0 Cash flows/total fixed assets 16.1 11.1 18.0 4.2 19.7 Revaluation Fund/fixed assets 26.9 29.6 44.8 46.6 38.4 Cash flows + borrowings + reval/FI 120.6 154.0 121.0 93.9 106.7 Cash flows + revaluation/WIP 89.3 69.6 224.0 101.9 103.5 Cash flows + borrowings/WIP 200.4 212.8 272.1 94.9 121.6 Incr. curr.liab./incr. acc. receiv. 222.5 147.9 133.1 108.6 96.0 Long-term borrow./FI not revalued 110.0 164.8 105.7 83.2 78.8 40. While ISKI was planning to issues certificates in 1991 to mobilize domestic savings (for US$29.9 million), this novel way had not been tested during the period under review for this Audit despite the relatively high return on assets recorded in two of the last three years. Short-term loans from commercial banks have been used only for liquidity management purpose twice in the last nine years: in 1987 for TL6.5 billion, immediately repaid the following year, and in 1990 for TL36.5 billion. Until 1990, customer deposits have not been a factor in financing long-term assets, increasing from TL2 billion in 1982 to TL14.6 billion in 1989 which translated into a 53% decline in real terms. In 1990 the amount was increased fivefold to TL72 billion. While modest compared to other long-term resources, this provided 13.3% of incremental long-term funds in 1990 but 44.2% when excluding the exchange risk incorporated into foreign borrowings at the end of 1990. C. Self-financing of Capital Investments 41. ISKI's performance in self-financing its large investments depends on the indicator chosen. Cash flows have been always positive since 1982 despite a loss recorded in 1989. Table 5 shows that the trend has been too erratic to be predictable. The depreciation component of cash flows has shown a steady rise, however, both in nominal and dollar terms, the latter particularly since 1987. On the other hand, the continuous increase of accounts receivable (PPAR, para. 48) has reduced available cash flows to the point of being negative in the last four years. 17 Table 5: Structure of Cash Flows in TL million 82 83 84 85 86 87 88 89 90 Cash Flows 1903 2523 9247 26256 9503 10784 52014 3899 175919 Depreciation 107 2172 1363 1179 2320 4752 12351 21494 33676 Deprec./Cash flows (%) 5.6 86.1 14.7 4.5 24.4 44.1 23.7 551.4 19.1 Deprec. in US $ million .66 9.63 3.72 2.56 3.44 5.54 8.68 10.13 12.91 Cash flows adj. acc.rec. n.a. 2826 5729 9414 7197 -6642 -18489-44318-16076 42. Table 6 summarizes the contradicting picture of self-financing (SF). Nominal self-financing measured by the Corporate Self-financing ratio3 was between 10 and 20% whenever a profit was recorded (not in 1989). The same self- financing of actual fixed assets (i.e., excluding of revaluation) was above 30%. However, after correcting cash flows for the accounts receivable, SF has been negative since 1987. If, furthermore, debt-servicing in principal was deducted first, the resulting SF ratio was even more negative. Table 6: Self-financing in % 83 84 85 86 87 88 89 90 Corporate self-financing 10.7 57.7 94.4 16.1 11.1 8.1 .8 19.7 adjusted for Reval. Fund 16.3 186.2 222.2 59.7 37.5 32.8 1.5 32.1 same adj. for receivables 18.3 115.4 79.7 17.7 -9.7 -11.7 -16.7 -2.9 43. The ex post financing plan of the Bank project shows a far greater self- financing than anticipated at appraisal. It is partly due to the Government's decision after project launch of not fulfilling its commitment to contribute 59% to the project. ISKI did not mobilize additional loans to fill the financing gap, but instead relied on tariff increases to generate cash flows. The strategy was timely to take advantage of the grace period on the Bank loan, thus freeing any cash flows generated during project construction to finance investments. Now that debt servicing is under way, this policy is not as effective, but it is to the credit of ISKI to have taken a unique window of opportunity. Not only it allowed the project to be implemented without delay, but it demonstrated that tariff increases do not need to wait until the project is commissioned in order to service the debt, but should be implemented as soon as the project is started to enable self-financing. D. Foreign Exchange Exposure 44. Foreign debt has increased 755 times between 1982 and 1990. This is partly due to the exchange risk which is included in the outstanding debt amount. There are two types of revaluation working concurrently: 3 Defined as cash flows divided by the increase in total gross fixed assets (including works in progress) during the same year. 18 (1) revaluation of fixed assets in operation according to domestic inflation to approximate their replacement cost; and (2) revaluation of foreign borrowings to take into account the depreciation of the TL in terms of the currencies required for debt service. The former has generated TL791.7 billion since 1983 and represented 80.9% of fixed assets in operation at the end of 1990. The latter is not disclosed as a separate item in the balance sheet. Over 1989-1990,,the incremental exchange risk amounted to TL464.8 billion and accounted for 80.1% (79.6% in 1989, 80.3% in 1991). The Bank loans have contributed to this increased exchange risk (PPAR, para. 59). It is not clear from the financial annexes to the PCR whether ISKI is bearing the exchange risk when it is servicing its debt. Given the rapid decline of the Lira, it is important to display separately ISKI's foreign exposure which could be a major cause of financial under-performance in the future. E. Government Assistance 45. Tax Sharing Fund. Municipalities in Turkey obtain resources from the Government through the Tax Sharing Fund (FIA). Table 7 shows that this resource was allowed to decline in real terms until 1990 when the n-minal amount was doubled. Given that cash flows have been reduced by inc-aase in accounts receivable, the tax sharing is comparatively a reliable reuource to support ISKI's self-financing. Still ISKI has been assuming that this source would level off at only 1% of sales in its 1992-2000 projections while it was 6-7% during the last three actual years. This conservative stance is explained in part by the subsidy nature of the Tax sharing and the importance of not relying excessively on it to balance its accounts. Table 7: Tax Sharing Fund TL million 1987 1988 1939 1990. Tax sharing 12281 12109 18626 35437 in % of profit 203.6 30.5 loss 24.9 in real terms ($ mill.) 14.3 8.5 8.8 13.6 46. ISKI has borrowed from the Public Participation Fund but only three times in the last eight years (in 1986, 1987 and 1989). At the end of 1990 the outstanding balance was TL27.97 billion which represented only 3.0% of total long-term borrowings, partly due to the fact that all foreign loans are revalued whereas domestic borrowings are not. There was no other domestic borrowing which indicates the low leverage of ISKI in the local financial market despite good nominal returns. A recent attempt in 1991 at mobilizing funds on the domestic market has been successful albeit limited in scope US$4 million at 85% p.a. interest and maturity as short as six years. 19 F. Accounts Receivable 47. Accounts receivable are one of the two majors causes of utilities' financial distress (with inadequate borrowing maturities). ISKI is no exception to this observation. The most appropriate indicator is the comparison of increase in receivables with sales which are both flows. Table 8 shows that the arrears increased generally several times faster than sales. As a result, additional arrears represented a larger share of sales, up to almost four months in 1990. Although no age structure of arrears was available, the 1989 external audit disclosed that TL60.7 billion (or 34.9%) of total were over one year old. The situation must have worsened in 1990, as Table 9 seems to indicate, due to the relative pace of arrears and sales. Table 8: Sales and Arrears TL million 82 83 84 85 86 87 88 89 90 Sales 3438 8774 12366 36087 28575 69034 163662 272330 611303 % change 155 41 192 -21 41 42 60 45 Receiv. increase -304 3518 16842 2307 17426 70503 48216 191995 % change 379 -86 655 305 -32 298 Receiv./salea 28.5 46.7 8.1 25.2 43.1 17.7 31.4 same in days of sales 104 170 30 92 157 65 115 Table 9: Receivables in relation to other Indicators % 82 83 84 85 86 87 88 89 90 Receiv./Current assets 66.3 42.4 40.3 56.2 50.3 43.3 45.6 50.5 64.4 Cash & bank/Receiv. 15.0 49.8 99.7 51.3 18.7 5.1 3.3 5.9 3.3 Current liab./Receiv. 101.8 172.2 62.9 49.2 64.0 96.8 113.6 153.4 116.5 Incr. Receiv./Cashfl. n.a. n.a. 38.0 64.1 24.3 161.6 135.6 1,237 109.1 48. The above table shows two distinct periods. Prior to 1986, ISKI's efforts were to offset the increase in accounts receivable by retaining more cash and deposits in banks. After 1986, the offset is done by running up current liabilities which is typical of corporations which are not in complete control of their finances. The collection of accounts receivable remains the best avenue for ISKI. It is currently entrusted to a subsidiary which is staffed entirely from outside ISKI. Although the public is aware that Suser is ISKI's subsidiary, it is well accepted partly because it is responsible for bringing water to outskirt municipalities through a fleet of 160 trucks. As an incentive, collection staff is paid on commissions linked to actual collection. As a result of the vigorous effort sustained in the recent years, recovery has improved from 52% in 1988 to 68% as of September 1991. Since the billing cycle is around one month, the actual collection rate was above 70%. 20 Table 10: Recent Collection Performance Sept. 1991 in % of total billings Domestic customers 63 48 (of w.) billed monthly 86 (5.7) Hotels 97 n.a. Industries 88 13 Public institutions 36 14 Municipalities 84 n.a. 49. There are both room for improvement and limitations due to the affordability of tariffs. Customers billed monthly have a 40% better payment record than those billed quarterly, but the formers are only 30,000 out of 1.1 million. To convert more customers would improve ISKI's intra-year liquidity and reduce arrears. Similarly the number of collection locations was increased from only three until 1991 to 22, therefore it is too early to assess the result although the change is in the right direction. Conversely the unavailability of an age structure of arrears makes a little dubious the effort to manage arrears which are becoming bad debts. The latter can be provisioned only after all legal pursuits have been exhausted which delays the recognition of losses. Neither the PCR nor the external audit report for 1989 mention any provisions for unrecoverable arrears. 50. The affordability of water charges is achieved through subsidization. The first 10 m3 of water is free if the monthly consumption is below this figure; only a nominal sewerage surcharge is then billed. The same approach is taken when handling poverty groups in arrears. ISKI waives their water bills, but not the sewerage bills if they pay all outstanding arrears (often paid in installments) and commit to consume less than 10 m3 a month; if the consumption is above 10 m3 there is no waiver. This accommodating stance concerns only 2% of ISKI's individual consumers. G. Financial Covenant 51. The covenant in Loan 2159-TU required ISKI to adjust its tariffs whenever revenues are insufficient "to cover its operating and maintenance costs and its debt service in excess of provision for depreciation" (Loan Agreement, section 5.04 (a)). The validity of this covenant was tested on the first difference between operating revenues and operating expenses as shown in Table 11. Except for a loss in 1986, the operating margin has been comfortably positive. As shown in the same table, the second difference between depreciation and debt-servicing was negative only twice (in 1987 and 1988) but not enough to reduce significantly the operating margin. Overall the covenant was met. Yet, the above analysis on accounts receivable shows the Bank covenant to be inadequate. 4 As none was made available to the field mission for this Audit. 21 Table 11: Financial Covenant in TL billion 82 83 84 85 86 87 88 89 90 Operating revenues 3.4 8.8 12.4 36.1 28.6 69.0 163.7 272.3 611.3 Operating expenses 4.3 8.0 10.4 21.5 32.8 53.0 99.2 175.5 419.4 Operating margin -.8 .8 1.9 14.6 -4.2 16.0 64.5 96.9 191.9 Depreciation .1 2.2 1.4 1.2 2.3 4.8 12.4 21.5 33.7 Debt service n.a. n.a. .6 .8 1.0 7.5 12.5 18.7 23.0 52. Table 12 summarizes the trend of operating margin before and after deducting the yearly increase in accounts receivable from the operating revenues. Except for 1989, the adjusted margin was always below I in recent years. Although the impact of growing arrears was clear, the covenant in the follow-up project (Loan 2888-TU) was changed to require ISKI to self-finance 100% of the Project expenditures which are not covered by the Loan plus at least 45% of the other capital expenditures. This is not more realistic than the previous covenant given the impact of arrears on the actual self-financing (PPAR, para. 42). Table 12: Sensitivity of Operating Margin to Arrears % 82 83 84 85 86 87 88 89 90 Operating margin 80.3 110.3 118.6 167.6 87.2 130.2 165.0 155.2 145.8 adj. for acc. receiv. n.a. 114.1 84.8 89.4 80.2 97.3 93.9 127.7 99.9 H. Current Ratio 53. Table 13 shows a steady declining trend of the current ratio. While a ratio above one is usually a positive sign, in the case of ISKI the downward trend means that it has offset the growing arrears with current liabilities. While the latter have to be repaid eventually, the former do not provide an adequate collateral as long as 35% are more than one year old and the outstanding arrears increase faster than sales. Table 13: Current Ratio times current liab. 82 83 84 85 86 87 88 89 90 Projected (SAR) 1.0 1.2 1.0 .6 .6 .9 1.1 1.2 n.a. Current ratio 1.48 1.37 3.95 3.61 3.11 2.38 1.93 1.29 1.33 I. Return on Assets 54. Table 14 shows that the traditional return on fixed assets in operations (ROA) usually covenanted in Bank projects has declined in the second half of the 1980s (average of 13.8% since 1987) compared to the first half and that it has been more erratic. After correcting for the yearly increase in accounts receivable, the ROA has been negative since 1987, averaging -6.4% which represents a 20-point drop from the nominal average. The more meaningful return 22 on total assets (Corporate ROI) shows a similar pattern at lower levels though because of the larger assets base. The nominal rate averaged 8.1% since and - 3.9% after adjusting for receivables. Unless the arrears is being reversed, ISKI's creditworthiness may be jeopardized by negative return on assets regardless of the definition. Table 14: Rates of Return on Fixed Assets % 82 83 84 85 86 87 88 89 90 Return on Assets 52.9 28.5 33.4 51.2 12.4 9.6 22.1 .8 22.5 adj. for Receiv. n.a. 32.0 20.7 18.4 9.4 -5.9 -7.8 -9.6 -2.1 Corporate ROI 36.9 14.9 25.2 44.8 9.3 6.0 14.1 .5 12.0 adj. for Receiv. n.a. 16.7 15.6 16.1 7.1 -3.7 -5.0 -5.8 -1.1 J. Return on Equity 55. Since the initial capital was repaid in 1987, equity was defined here as capital plus the Revaluation Fund for purpose of computing the return (ROE). The trend shows large variance although it is downward overall. After correcting for cash flows which are tied up in arrears, the ROE has been negative since 1987 which is also a worrisome sign not mentioned in the PCR. Table 15: Rates of Return on Equity % 82 83 84 85 86 87 88 89 90 Capital + Reval.Fund n.a. 41.5 58.8 111.5 19.7 15.5 36.1 1.2 28.4 adj. for Receiv. n.a. 46.4 36.4 32.2 14.9 -9.6 -12.8 -13.5 -2.6 K. Debt-equity Ratio 56. The SAR projected that the debt-equity ratio (DE) would never rise above .69:1, but it did 5 out of the last nine years, remaining above 1.1:1 in the last four. Given the size of the investment program carried out by ISKI, this is not, however, excessive. The strategic plan until 2000 assumes more than a doubling of long-term borrowings in dollar terms which may not be compatible with the accounts receivable situation which deplete free cash flows for self- financing and debt-servicing. Table 16: Debt-equity Ratios % 82 83 84 85 86 87 88 89 90 Projected (SAR) .64 .69 .64 .52 .45 .47 .41 .37 n.a. Actual .60 1.28 .47 .27 .64 1.88 1.48 1.10 1.12 L. ISKI's Financial Policy 57. The above tables show a clear change of ISKI's financial policy since 1987. It has been more aggressive with some riskier results. Short-term loans from commercial banks have been used for liquidity purpose; consumer deposits have 23 been increased advantageously given their quasi-equity nature; depreciation has been increased in dollar terms; the initial capital has been repaid; a real liability such as the foreign exchange risk has been offset with a bookkeeping revaluation of fixed assets; the arrears from customers have allowed an increase in current liabilities. Despite nominal returns which are positive, ISKI does not have the real financial autonomy derived from positive returns adjusted for accounts receivable. It is therefore in weak position to undertake the large investment program outlined in its strategic plan without reversing the trend in arrears which reduce its effective cash flow generation ability. M. Accounting and Audit 58. ISKI has been audited by Government auditors during project implementation, but was in the process of shifting to private auditors for the accounting year 1990. The last external audit report available was, however, the one for 1989 prepared by the Auditors' Council of the Ministry of Finance (MOF). They have focused on compliance of recording with domestic regulations and with provisions of the Loan Agreement, but raised no issue with the unfavorable aspects of ISKI's financial situation, such as arrears, unaccounted-for-water, exchange risk, etc. The follow-up project required the appointment of private auditors starting with FY 1990, but invitations to bid were only recently sent and the audited 1990 accounts should be ready only early 1992, too late compared to the covenant requesting audited accounts within seven months. It is hoped that the new arrangement will eventually allow these documents to be circulated within a reasonable time after the end of the FY. N. Maturity, Grace Period and Exchange Risk of the Bank Loan 59. The Loan was granted to ISKI fcr 18 years of maturity, including 4 years and 7 months of grace, and with a commitment charge equal to .75% p.a. of the undisbursed principal. A front-end fee of .25% or US$1, 301,970 was also charged. The exchange risk is borne by ISKI and the portion due to the currency- pooling system amounted to US$11.26 million as of the end of 1991 or 21% of the nominal outstanding debt. Between May 1982 (signature) and September 1991 the Lira depreciated by 97% against the US dollar, thus adding far more to the debt- servicing burden. 60. From ISKI's viewpoint, the availability over time of borrowed funds can be measured by an indicator called the Average Loan Life (ALL). It is defined as: the ratio of (a) the sum, until maturity, of the loan balances (in US dollars) outstanding at the end of each year over (b) the loan amount net of cancellations. This indicator measures the number of years during which the entire loan prDceeds stay effectively at the borrower's disposal assuming up- front disbursement. The faster the disbursements and the slower the repayments, the longer the availability of loan funds. In the case of Loan 2159-TU, it had an expected average loan life of 7.4 years at the time of appraisal. The slower than expected disbursements reduced it to 6.5 years. The exchange risk due to the currency-pooling system realized on past repayments (but that accumulated until September 30, 1991)s cut it back by seven months to 5.9 years. Assuming 5 Deadline for data collection for this PPAR. 24 this accumulated risk does disappear but does not worsen, the ALL is further reduced by five months to 5.48 years. This indicates that the actual terms of the Bank loan were short to finance slow-recoverable investments such as sewerage and suggests that ISKI will be effectively relying on Istanbul Municipality's general budget for its debt-servicing. 0. Environmental Aspects 61. While in its design and implementation the project is, essentially, a straight forward urban infrastructure project, its real focus was entirely environmental improvement, in particular the cleaning up and permanent protection of the Golden Horn, the Sea of Marmara and the less effected Bosphorus. The earliest sewers in Istanbul were built during the Ottoman times and further developments took place in the 1920s and 1930s. The rapid and largely uncontrolled developments of the post World War II era eventually completely overwhelmed not only the sewage collector system but the indiscriminately used receiving waters, particularly the Golden Horn and the Sea of Marmara. It is difficult to pinpoint the exact time of what one might call an environmental awakening in Istanbul but take place it did. 62. The first, fully comprehensive plan to clean up the city and its surrounding waters was the so called "DAMOC Master Plan Revision" (a revision of an earlier Drainage Master Plan) in 1974. This plan remains valid to this day to the extent that it envisaged the complete prevention of discharge of untreated sewage at the shore line of any receiving waters by building a network of interceptors that would transport all sewage to specific disposal points. However, as mentioned earlier, this plan still envisaged discharge of only pre- treated sewage into the bottom current of the Bosphorus as an environmentally acceptable solution. The Bank has, in general, supported this plan and the project under review was its first major stage to be implemented. 63. The review of the project documentation and the Audit's discussions with Bank staff and Turkish officials clearly indicate that parallel with ISKI's general development there has been a steady sharpening of the environmental focus both within ISKI and the central environmental agencies. This process was both fueled and supported by the increased environmental awareness of the public and greater visibility and involvement of the Turkish scientific community. Members of the latter were the principal actore in challenging the full effectiveness of the "bottom current" theory resulting in the extensive study of this phenomenon which is due to be completed in 1993. 64. Although the findings of this study should be of major importance, its significance, in a sense, has already been diminished by a far reaching decision of the Turkish environmental authorities in the late 1980s. This decision effectively prohibits the discharge of any but fully treated effluent into any receiving waters. ISKI declared its full support for this decision and proceeded to revise its sewerage Master Plan accordingly by incorporating the provision of (at least) secondary sewage treatment plant in all drainage basin designs. As claimed, this, indeed, puts Istanbul in the forefront in the terms of environmental protection policies just about anywhere in the world. In addition, studies are under way for the recycling of sewage effluent for irrigation and cooling water purposes. 25 65. In the context of the ever increasing pressure, on a world wide basis, for better protection of the natural environment these policies can not be faulted. However, one of ISKI's many useful publications describing and justifying future plans, repeatedly uses a seemingly innocuous phrase which points to the problems to come. This states: "Our Administration's objective is to establish a chain of full treatment plants using the most appropriate technology available, whatever the cost,' in order to solve this environmental problem once and for all". 66. The new policy not only adds, at several disposal points, the cost of large scale secondary treatment plants to designs previously providing for pre- treatment only but, because of Istanbul's topography, suitable level sites for large treatment plants can only be found in remote areas. This, in turn, adds the substantial extra cost of long puning mains/tunnels (the proposed Kadikay - Riva tunnel is 28.5km). All in all, the implementation of the new policy represents a quantum increase in ISKI's projected investment costs planned for the next 10 to 30 years with little indication that much thought has been given as to where that level of funding might come from. 67. Over the years the Bank and ISKI have developed a close working relationship. This now needs to be put to good use in finding a solution that preserves the principles of the commendable environmental protection program of Turkey and ISKI without hasty commitment to an unaffordable level of expenditures that would jeopardize ISKI's even now shaky financial situation. Such chain of events could not only halt the progressive improvements of the past years but may put at risk the full sustainability (effective maintenance etc.) of the facilities already built and endanger the already spectacular achievements of the program such as the transformation of the Golden Horn. 68. It should also be noted that while the most visible effects of this impressive pollution control effort have and will be on the shores and waters of the Golden Horn and the Sea of Marmara, an equally important policy of ISKI is the commitment to develop water supply and sewerage in parallel in Istanbul. A current review of some 120 past Bank financed water supply/sewerage projects revealed that in virtually all countries/cities these developments were grossly uneven with water supply much in advance. The resulting mismanagement of waste waters not only resulted in a vast backlog of sewerage investments (which may never catch up in some places) but the polluted streets and receiving waters clearly diminished the health improvement effects of the safe(er) water supplies provided. P. Economic Assessment 69. The SAR presents (paras. 6.01 - 6.04) a brief qualitative assessment of the expected project bsnefits on public health, property values, fishing in the Sea of Marmara, the use of public beaches and tourism and rightly points out that none of these are realistically quantifiable. The SAR also calculated the Average (Long Term) Incremental Cost (AIC) of the sewer system extensions projected to the year 2010 at discount rates of 8% and 10%. The benefit side of 6 Author's italics for emphasis 26 these calculations is represented by water consumption per year as the charges for sewer services would be in the form of a surcharge on water consumed. The resulting values were 0.072 and 0.079 US$/m3. These were essentially theoretical figures as no sewer charges were in effect at the time but they were to be considered in the recommendations of the forthcoming tariff study. These calculations were flawed in that the incremental benefits were derived from total instead of incremental water sales. 70. The PCR (Para. 6.04) recalculated the ACI using actual costs and incremental water sales for the project period and revised projections based on current estimates. The changes are particularly dramatic in the projected water sales which take into account a much higher growth rate (5.4%) in Istanbul than foreseen in the SAR. The recalculated ACIs, at the same 8% and 10% discount rate, are 0.126 and 0.138 US$/m3. The PCR further states that the combined water/sewerage ACI is "in the order of 0.65-0.75 US$/m3. As the average combined water/sewerage tariff in 1990 was 0.80 US$/m3 equivalent, ISKI has, indeed, appear to be charging economic prices for its services. 71. The Audit also attempted to assess the direct effect of these investment on the poor. The SAR provided a brief statistical assessment of the pr-jected benefits of the project for the poor stating that the drainage areas c 4red by the project house some 50% of Istanbul's low income population and that the sewerage service levels in the area would be increased from 73% to 95%, in other words, virtually full service coverage will be achieved providing service to an additional 300,000 lower income people. The Audit has not been able to obtain specific statistics relating to low income service provisions but sees no reason to argue with the conclusion that the 95% service level confirmed to have been achieved in the area does provide the projected coverage for the poor. As to the affordability of the services by the poor, ISKI's tariff structure is highly progressive with industrial/commercial rates up to 130% of domestic tariffs, permitting substantial cross subsidization. Water, for consumers using less than 10 m3, is free but the 100% sewerage surcharge is levied. These tariffs, for households below the poverty level, result in charges of less than 3% of household income and appear to be well accepted by the public. Q. The Role of the Bank 72. Bank involvement in water supply and sewerage in Istanbul now spans an unbroken 20 years and this has spread to Izmir and Ankara in the late 1980s. Over this 20 years sector developments in Istanbul have been immense. From the acute water shortages and primitive sewer services of the late 1960s, handled by ISI with only 34 engineers to today's 90% service levels and a strong and competent ISKI with a multi billion dollar investment program is remarkable progress by any standard. In development work it is never possible to state precisely who should get the credit for what. However, there can be little doubt of the Bank's major role in the Istanbul developments. For more than 20 years it provided sustained and consistent advice, patiently working through years of inertia, flexible when appropriate but it also stood firm when justified (a not too frequent attitude). Above all, the Bank should get substantial credit for the steady effort exerted to create ISKI. The records appear to show that the Bank - Turkey/Istanbul relationship was always cordial, more of a partnership 27 than anything else. The individual staff were competent, the skill mix generally right. As a result, supervision work was well received and was effective. 73. As sector developments expand further, both in Istanbul and Turkey as a whole, the Bank should maintain and strengthen this relationship. Projects are getting increasingly complex technically, environmentally and economically. Selection of priorities an judicious financial planning will be the utmost importance and to play its rightful role, the Bank must ensure that it devotes adequate resources to it both in terms of quantity and, above all, quality. III. Conclusions and Lessons 74. The PCR gives a frank and concise assessment of the project experience. The efficient implementation of the project required little additional analysis and the Audit concentrated its work on the broader environmental aspects of the complex and sensitive waste disposal problems of Istanbul and ISKI's financial situation. The two are closely interrelated in that the massive proposed developments, resulting largely from the increased environmental awareness generated by the project, have financial.implications which are, perhaps, not fully reflected in the PCR. The detailed analysis of ISKI's finances, as presented in this ieport, highlight the critical issues which will need to be in view in all future Bank involvement in this complicated but rewarding area of Istanbul's development. 75. The SAR listed five specific objectives for the project (see paragraph 13 of this report) the first four of which were met in full. A question mark must remain against the fifth, the objective of financial self-sufficiency. While improvements in this are were also very significant, performance tends to fluctuate from year to year sufficiently to raise concern. This is further discussed below. 76. After initial, largely bureaucratic, delays the project was implemented efficiently, on schedule and well within cost estimates. Designs and quality of construction were of high standard and ISKI has proved .tself to be a competent organization well capable of maintaining and expanding its facilities. As mentioned above, the project was first and foremost an environmental improvement effort. While this was only the first stage of the Master Plan, the Audit has no reason to doubt, that the eventual completion of the Plan will further enhance and expand the already dramatic envirotmental impact visible on the Golden Horn. It should also be mentioned that the Istanbul Municipality has reinforced the effects of the cleaner waters by extensive beautification of the shore lines. ThZs, apart from being sound urban management, much increased the visibility of the results of the project to the people of IstanbAl and, importantly, to the tourists. 77. ISKI and the Municipality have been jointly conducting a sustained publicity campaign on the project and the program. Impressive publications keep the public informed of work in progress, results and future plans. There is little doubt that this is instrumental in the general acceptance (so far, at any rate)of the substantial tariff increases implemented by ISKI in recent years. 28 78. Over the years of project implementation, possibly as a result of the visible improvements in reducing pollution, environmental awareness within ISKI, the Turkish scientific community and the country at large, appear to have been steadily increasing. The demands of environmentalists for ever stricter standards of protection are increasing. This, apart from internal pressures, probably also (at least partly) due to Turkey being "sandwiched" between two very large environmental efforts, the Mediterranean Action Plan and the forthcoming 3lack Sea Convention. A direct result of all this has been the decision to require full treatment for all effluent to be discharged into the Istanbul waters. As already stated, the Audit can not dispute the principles of this decision but wishes to emphasize the need for prudent and economically and financially viable planning with the full consideration of affordability by the people, the industries and institutions of Istanbul and the country in general. Just to launch into such undertaking " whatever the cost" would be irresponsible. 79. Regarding financial issues ISKI is a case where the twin revaluation of foreign debt and fixed assets in operation creates some confusion about the true state of its finances. The exchange risk added to the original foreign debt is not really a resource available for financing capital investments as the formal funds flow statement suggests. Similarly, the yearly addition to the fixed assets through revaluation and the creation of the Revaluation Fund neither provides any new real funds to finance actual expenditures nor can it substitute for the original capital endowed to ISKI (repaid in 1987) and then being used exclusively to support the leveraging (although moderate) of the utility. Furthermore, the current levels of accounts receivable curtail cash flows to the point of generating negative returns on assets and equity when the impact of these arrears is fully taken into account. Coupled with the more aggressive financial policies, implemented since 1987, ISKI is carrying greater risks due to the arrears and foreign exchange risk exposures. 80. The main problem likely to be faced by ISKI in the future will be passing or, through the tariffs, the heavy cost of the planned investments when the affordability to the consumers will be in question. Users could be reluctant to pay an ever rising share of their income for water and sewerage on the untested assumption that they also want a cleaner Bosphorus and Marmara. Furthermore, if the past is any indication ISKI's finances could suffer a set back if the arrears situation was to worsen further. For this reason, it is proposed that any future lending operation be subordinated to the reduction of arrears, particularly from public sector institutions, so as to produce a positive return on assets and, secondly, a minimum return on total fixed assets should be covenanted after discounting cash flows by any increase in accounts receivable. 81. Financial reporting could be improved by disclosing items which are relevant for assessing ISKI's exposure to risks as well as its real cash flow generatirg capacity. To achieve this the following items should be disclosed on the balance sheet: - the outstanding exchange risk on foreign borrowings; - the accumulated interest during construction capitalized in the fixed assets; - the operating costs capitalized, if any, in the fixed assets; - the debt falling due in the coming year. 29 82. Notwithstanding the cautionary notes above, the principal lesson from this project is that sewerage and urban waste disposal in general, should be treated as not just another form of necessary urban infrastructure, but planned, with full information of the beneficiaries, as an indispensable environmental protection act that will impact on every facet of the city's life. 83. The project also reinforces the often repeated lesson in this sector that a sound and well organized institution with adequate autonomy is a prerequisite to successful implementation. In Istanbul's case the Bank is to be commended for holding back until satisfactory actions have been taken to achieve this. It did take a long time and the Audit noted in its file search the management's consideration of terminating the project. Well done the staff that persevered. 30 Annex 1 Page I of I Comments from the Borrower THE REPUBLIC OF TURKEY PRIME MINISTRY THE UNDERSECRETARIAT OF TREASURY AND FOREIGN TRADE lef: DEt-IV-3-t42 0 92 26934 Ankara, Mr. Yves Albouv Chiet Infrastructure and Energy Division Operations Evaluation Department Ref: tstanbul Sewerage Project Draft Project Performance Audit Report. Dear Mr. Albuuy We are very pleased to receive the Draft Project Performance Audit Report. We are of the opinion that this Report will be very contributive and useful reference material to evaluate the performance of the other projects. We have examined the Report in depth with all the related institutions and appreciated the efforts for the preparation of this Report. It is a pleasure for us co express our fully agreement on the context of the Draft Report. With our best regards. lead of ppartmelnb-