Ghana - Water Sector Rehabilitation Project

Document of The World Bank FOR OFFICLAL USE ONLY Report No. 18753 IMPLEMENTATION COMPLETION REPORT REPUBLIC OF GHANA WATER SECTOR REHABILITATION PROJECT (CR 2039-GH) December 22, 1998 Water and Urban 2 Country Department 10 Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency Unit = Cedi (C) US$1.00 = C 228 (annual average exchange rate 1988) = C 270 (1989) = C 330 (1990) = C 368 (1991) = C 437 (1992) = C 652 (1993) = C 957 (1994) = C 1,200 (1995) = C 1,635 (1996) =C 2,050 (1997) = C 2,544 (1998) FISCAL YEAR OF BORROWER January 1 - December 31 WEIGHTS AND MEASURES 1 meter (m) = 3.28 feet (ft) 1 kilometer (kin) = 0.62 miles (mi) 1 square kdlometer (km2) = 0.386 square miles (mi2) 1 liter (1) = 0.22 Imp. galls. or 0.264 US galls. 1 cubic meter (m3) = 1,000 liters or 220 Imp. galls. or 264 US galls. 1 million gallons per day (mgd) = 4,500 cubic meters (m3) per day ABBREVIATIONS & ACRONYMS ADC Austrian Development Corporation AFD Agence Francaise de Developpement AfDB African Development Bank ATMA Accra-Tema Metropolitan Area CIDA Canadian International Development Agency DFID Department for International Development ECG Electlicity Corporation of Ghana GOG Government of the Republic of Ghana GWSC Ghana Water and Sewerage Corporation ERR Economic Rate of Return IDA International Development Association ODA Overseas Development Administration (UK) PCMIU Project and Construction Management Unit PMC Project Management Consultant VRA Volta River Authority Vice President: Jean-Louis Sarbib, AFR Country Director: Peter Harrold, AFCIO Sector Manager: Letitia A. Obeng, AFTU2 Team Leader: David Henley, AFTU2  FOR OFFICIAL USE ONLY REPUBLIC OF GHANA WATER SECTOR REHABILITATION PROJECT (CR 2039-GH) IMPLEMENTATION COMPLETION REPORT TABLE OF CONTENTS PREFACE ...............i EVALUATION SUMMARY .................. , ii PART I - PROJECT IMPLEMENTATION ASSESSMENT ........................................ 1 A. Background .........................................1 B. Statement/Evaluation of Objectives .........................................1 C. Achievement of Objectives ........................3......3 D. Major Factors Affecting the Project .............................7 E. Project Sustainability .............................8 F. Bank Performance ............................. 8 G. Borrower Performance .............................8 H. Assessment of Outcome ...............................9 I. Future Operation .............................9 J. Key Lessons Learned .............................9 Part II. STATISTICAL TABLES ............................ 12 Table 1: Summary of Assessments ............................ 13 Tabl e 2: Related Bank Loans/Credits .............................. 14 Table 3 - Project Timetable ............................ 14 Table 4: Credit Disbursements: Cumulative Estimated & Actual ............................................ 14 Table 5: Key Indicators for Project Implementation ............................................ 15 Table 6A: Targeted and Achieved Output ............................................ 16 Table 6B: Planned & Actual Implementation Schedule ............................................ 17 Table 7: Studies Included in Project* ............................................ 18 Table 8A: Project Costs* ............................................. 19 Table 8B: Actual Project Financing* ............................................ 20 Table 9: Economic Costs and Benefits ............................................. 21 Table 10: Status of Legal Covenants ............................................ 21 Table 11: Compliance with Operational Manual Statements ............................................ 22 Table 12: Financial/Operational Indicators ............................................ 22 Table 13: Bank Resources: Staff Inputs ........................................... 24 Table 14: Bank Resources: Missions ........................................... 24 Appendix A Mission's Final Aide Memoire Appendix B Borrower Contribution to ICR Appendix C Map 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.  REPUBLIC OF GHANA WATER SECTOR REHABILITATION PROJECT (CR 2039-GH) IMPLEMENTATION COMPLETION REPORT PREFACE This is the Implementation Completion Report (ICR) for the Water Sector Rehabilitation Project in Ghana, for which Credit 2039-GH, in the amount of SDR 19.3 million (US$25.0 million equivalent), was approved on June 14, 1989, and made effective on June 18, 1990. The credit was closed on June 30, 1998, compared with the original closing date of June 30, 1997. Final disbursement took place on November 16, 1998 at which time the balance of SDR 2.38 million was to be cancelled. Cofinancing for the project was to be provided by Overseas Development Agency (ODA-UK), Overseas Economic Corporation Fund (OECF - Japan), Austrian Development Corporation (ADC), African Development Bank (AfDB), and the French Development Agency (AFD), however, the latter donor withdrew from the project due to the continued lack of creditworthiness of the implementing agency, Ghana Water and Sewerage Corporation (GWSC). The ICR was prepared by Eleanor Wamer (Fin. Analyst), Water & Urb.n 2, Eugene Okongwu (San. Engr) Nigeria Resident Mission, and reviewed at a meeting dated December 7 chaired by Ms. Theresa Jones, Acting Country Director for Ghana. The beneficiary and implementing agency, GWSC, contributed to the preparation of the ICR by commenting on the Bank's draft and preparing an evaluation which is attached in Appendix B. Preparation of this ICR began during the Bank's June 1998 final supervision/completion mission. The ICR is based on material in the project file and on information provided by GWSC. The draft was sent to cofinanciers for review; comments were received from the ADC representative.  ii IMPLEMENTATION COMPLETION REPORT REPUBLIC OF GHANA WATER SECTOR REHABILITATION PROJECT (CR 2039-GH) EVALUATION SUMMARY INTRODUCTION i. The Water Sector Rehabilitation Project (WSRP) was the fourth Bank-financed operation intended to focus on priority rehabilitation needs of the water supply sector and to continue strengthening institutional capacity of the country's water supply agency, GWSC. Two earlier projects, approved in 1969 and 1974 respectively, focussed on expansion of the urban water systems and construction of a sewerage system in the Accra-Tema Metropolitan Area (ATMA). The third project approved in 1983 aimed to strengthen the management of GWSC and rehabilitate two major transmission pipelines in ATMA. PROJECT OBJECTIVES ii. The project's objectives were to: (i) strengthen the managerial, financial and technical capabilities of GWSC both at headquarters and in the regions to enable it to effectively plan and implement a meaningful development program; (ii) improve GWSC's operating and maintenance ability; and (iii) improve GWSC's financial performance so that it might function as a viable institution. The project was also intended to assist GWSC to: (i) reduce the levels of non-revenue earning water; (ii) increase the output of existing water systems by replacement of plant and equipment and by general system rehabilitation; and (iii) provide for the expansion of some systems and completion of others to further increase water availability. iii. The covenants were aimed to ensure achievement of the project's financial objectives, including the requirement for GWSC to produce funds from internal sources equivalent to GWSC's annual capital expenditures incurred. The objectives and related covenants were clear, and reasonable measures were taken at appraisal to put GWSC on a path that would enable it to meet its objectives, including a recapitalization program in which foreign debt was converted into equity and a clearinghouse arrangement established to help ensure that government's liabilities to GWSC were met. The major pitfalls with the project were: (i) it was not well designed in terms of selection of components to meet the objectives (paras. 7, 30); and (ii) it grossly underestimated the timeframe and the inputs necessary for GWSC to meet its financial and institutional objectives. IMPLEMENTATION EXPERIENCE AND RESULTS iv. The project partially achieved its objectives. The financial objectives were partially met with the recruitment of professional finance and accounting staff, pilot computerization of accounts and billing in a few regional offices, and training and technical assistance. The physical objectives were also partially met. There was modest improvement in GWSC's operating standards as a result of technical assistance provided and the supply of operational facilities. However, the major works, i.e. completion of rehabilitation and limited expansion of systems, will not be fully completed until September 1999. Implementation of a program to reduce unaccounted-for-water (u-f-w), which was intended to be a major measure for turning around GWSC's financial performance, was not done. As financial performance is closely linked with operating performance, improvement in GWSC's financial performance could be expected only after completion of the works in 1999  (compared with SAR target of 1996), in addition to adequate pricing and cost containment measures. v. The key factors that affected achievement of the major objectives were: (i) management instability in the sector -- there were frequent changes at the ministerial level in the parent ministly, MWH; similar changes occurred at the top management level within GWSC including the position of Managing Director which was filled on a permanent basis in only two and half out of seven years of the project; (ii) lengthy start-up delays; and (iii) inadequate and untimely tariff increases. vi. Sustainability of the investments is likely as GWSC has agreed to carry out an Operational Plan for operating and maintaining the project systems. Additionally, the sector is being prepared for a major restructuring to be launched by mid-2000, supported by investments from the Bank and bilateral donors. Under the restructured sector, operations. and management will be the responsibility of the private sector while policy and regulation will be the responsibility of a downsized GWSC together with a newly established regulatory body, Public Utilities Regulatory Commission (PURC). vii. The total project costs estimated at US$120.4 million did not deviate much from the SAR estimate of US$124.9 million even though the scope was reduced. At appraisal, GWSC was expected to finance some US$18.8 million (16% of total cost); however, only about US$6.8 million was financed by GWSC (6% of total cost). It is the view of GWSC that donors should have financed 100% of total cost (para 2.4.7 of Appendix B). Due to GWSC's persistent cash shortages cofinanciers agreed to cover most of the local costs of their components, but throughout the project, the lack of counterpart funding continued to be a cause of delays in works' completion. viii. The Bank's performance was unsatisfactory. The project design was deficient. The scope of works necessary to achieve a reduced level of u-f-w was underestimated at project preparation. U-f-w reduction was inappropriately included as a study under the project as well as a measure to contribute to enhanced revenue generation. ix. The quality of supervision was affected by lack of continuity in task management (6 Task Managers over the period 1991 to 1998). This notwithstanding, the Bank was proactive in steering the Government of Ghana to sector restructuring when it became clear that the institutional environment needed to be improved to introduce efficiency into the sector. It is noteworthy that in spite of the frequent changes in task management, the Bank has sustained good relations with the cofinanciers and donors who are in support of the sector restructuring. x. The Borrower's performance was deficient. The project was executed during a period of management instability in the sector. Frequent changes in ministers at the MWH affected GWSC's ability to obtain adequate and timely tariff increases. Changes at the top management levels at GWSC interrupted continuity in sector dialogue. Frequent changes in leadership in the technical experts recruited to advice GWSC further contributed to handicapping the sector's abilitv to move forward. xi. The project's outcome is unsatisfactory. The immediate action programs designed to speedily (within 12 months) restore inoperative and partially operative systems were completed in 1996, four years later than planned. Capacity in the rehabilitated systems should be restored to design capacity by September 1999, some 3 years later than anticipated. Limited expansion should take place as planned with an equal delay. There was insufficient focus on the rehabilitation/expansion of the distribution systems and on consumer connections. The u-f-w  iv reduction program was not undertaken due to delays in designing an effective program, though GWSC has recently commenced a program using its own resources. The financial viability goals could not be achieved partly because of their link with physical works and also due to the inadequate pricing of water. KEY LESSONS LEARNED xii. Stable sector management is a necessary condition for project success. Throughout implementation there were frequent changes at all levels of sector management (MWH, GWSC and PCMU) which contributed to: (i) implementation delays; and (ii) delays in GWSC's ability to address its persistent cash flow difficulties. xiii. The assignment of local financing obligations for capital works to a utility should be informed by the utility's proven capability and not on projected performance. Prior to the project, GWSC was unable to cover its understated cash operating expenses from revenues and collections. Increased revenue generation expected during the project depended on several factors some of which were not within the control of GWSC such as periodic and timely approval of water tariff and the macro-economic environment. xiv. Project design: A positive innovation of the project was the successful establishment and implementation of a clearinghouse arrangement to settle Government's current liabilities to GWSC (para 6). This approach could be usefully implemented by other utilities facing problems with government arrears. xv. It is only by adequate customer identification and holistic packaging (commercial, financial and operational) that measurable results can be attained. Insufficient attention was paid to marketing, including distribution, connection, billing and collection which could have helped to improve revenue generation. Experience has indicated that the necessary commercial orientation is best addressed by private sector involvement. xvi. PCMU structure functioned as planned but had limited use and could not be easily integrated into GWSC. In sectors and countries with very low implementation capacity this structure is useful, but one should devise a way to integrate it into the agency so that (i) communication links are enhanced; and (ii) local professional capacity remains with the departure of the international consultants. xvii. More innovative contracting methods are needed for rehabilitation works on plants that must remain operational. Designs were not done at appraisal, possibly because of the difficulty in designing too far in advance of the actual rehabilitation of operating plants. Design and construct type of contracts with a high contingency and flexible budget may be more appropriate for rehabilitation works. This approach would have ensured that sufficient resources were mobilized for the project; eliminated the process of rescoping of contract packages to fit available financing; and ensured that the systems are fully rehabilitated. xviii. Reduction of u-f-w is a process involving significant works components as well as improved commercial management in addition to measures to stop commercial leaks. The scope of works necessary to achieve reduced level of u-f-w was underestimated at project preparation, hence the inability of this measure to contribute to enhanced revenue generation. U-f-w was insufficiently treated as a technical study.  REPUBLIC OF GHANA WATER SECTOR REHABILITATION PROJECT (CR 2039-GH) IMPLEMENTATION COMPLETION REPORT PART I - PROJECT IMPLEMENTATION ASSESSMENT A. BACKGROUND 1. Water supply systems deteriorated rapidly during the economic crisis of the late 1970s and early 1980s when Government's ability to adequately operate and maintain essential services was severely constrained. Essential commodities, such as fuel, lubricants and spare parts were in short supply resulting in failure of vital components of some of the water treatment and transmission systems. In some areas, entire systems were rendered inoperable causing complete failure of piped water supplies to smaller towns and severe reduction of supplies to larger urban centers. Increases in urban population continued unabated at about 5% annually over the last decade thus widening the substantial gap between water supply and demand. 2. The economic recovery program embarked on during the mid-1980s set about to, among others, reform public enterprises by increasing their autonomy with the objective to create effectively managed, financially viable institutions. Three water supply and sewerage projects were approved and completed between 1969 and 1990. They focussed on expansion and rehabilitation in the Accra-Tema Metropolitan Area (ATMA) in addition to technical assistance to the country's water supply agency, Ghana Water and Sewerage Corporation (GWSC). The projects were successful in completing the physical works, but the institutional development outcome was inadequate. The fourth Bank-financed operation, the Water Sector Rehabilitation Project (WSRP), was aimed to rehabilitate the water supply systems in the secondary towns while continuing to develop GWSC. B. STATEMENT/EVALUATION OF OBJECTIVES Project Objectives: 3. The overriding objective of the project as identified in the SAR was to strengthen the main water and sewerage sector entity, GWSC, to enable it to more effectively administer the sector through: (i) strengthening the managerial, financial and technical capabilities of GWSC both at headquarters and in the regions to enable it to effectively plan and implement a meaningful development program; (ii) improving GWSC's operating and maintenance ability; and (iii) improving GWSC's financial performance so that it would function as a viable institution. The primary goals were intended to be achieved by: (i) reducing the levels of non-revenue earning water; (ii) increasing the output of existing water systems by replacement of plant and equipment and by general system rehabilitation; and (iii) providing for the expansion of some systems and completion of others to further increase water availability.  2 Project Components: 4. The components of the project included: (a) Institutional development to complete the decentralization of GWSC, with assistance provided under an ongoing twinning arrangement with Thames Water Intemational UK, a Manpower Improvement Program, including recruitment of key professional staff and retrenchment of about 1,500 redundant employees, management and technical training and increased staff support to the Managing Director; (b) Emergency replacements of mechanical, electrical and water treatment plant and equipment which had deteriorated beyond economic repair; (c) Rehabilitation and completion of 37 water supplv systems including works designed to restore production levels to those for which the systems were originally designed and to complete partially constructed systems on which work was suspended because of lack of funds (river intakes, wells, transmission, treatment and distribution systems); (d) Limited expansion of 31 water supply systems including limited distribution system expansion in densely populated urban areas where there were no piped supplies and expansion of certain production facilities to meet extreme needs; (e) Improvement of workshops, stores and training centers and provision of staff housing including structural improvements and provision of equipment for workshops and stores at headquarters and in the regions for the two GWSC training schools; the component also includes staff housing as an incentive for recruitment and retention of key staff; (f) Supplv of essential equipment and materials including vehicles; replacement parts for existing plant and equipment; a one-year supply of chemicals; water meters; and laboratory equipment; and (g) Technical studies as follows: (i) hydrologic studies in 17 of the 37 systems to be rehabilitated or expanded; (ii) a feasibility study for the disposal of wastes from the Accra sewerage system; (iii) an engineering design for the expansion of Tamale water supply system; (iv) development of programs for the reduction of unaccounted-for-water in GWSC's largest water supply systems, including Sunyani, Cape Coast and Sekondi-Takoradi; (v) a study on the appropriateness of the tariff structure and levels; and (vi) revaluation of fixed assets. Covenants 5. The Project Agreement included covenants aimed at ensuring improved financial performance. GWSC was required to produce funds from internal sources equivalent to annual capital expenditures incurred, or expected to be incurred and its debt service requirements for each year after December 1988.  3 Evaluation of Objectives 6. The project's objectives and financial covenants were clear and important for the sector, but the major financial covenant was not realistic. While the project took specific measures to help fulfill the covenant, including a capital restructuring program and setting up of a clearing house arrangement' to settle the current liabilities of specific Government agencies, the timeframe for accomplishing the covenant was unrealistic. Prior to the project (1977 to 1986) GWSC was dependent upon government to cover over 50% of its recurrent expenses. Subsequently, it covered its operating costs but at a level constrained by cash and without regard to operational needs. The financial plan could have considered GWSC's very weak base and a more realistic covenant could have been designed to reflect gradual improvements over time in line with efficiency gains. 7. The risk identified at appraisal of GWSC not being able to generate sufficient internal cash flow to contribute to counterpart funds was expected to be mitigated through a program to reduce the level of unaccounted-for-water (u-f-w). However, the u-f-w program was designed to be implemented after the systems were rehabilitated and expanded by 1996. The rehabilitation and expansion program was considerably delayed, but even if it was not delayed, no measurable improvements could have been expected until 1996. Implementation of other measures to reduce u-f-w such as improving customer identification, billing and collection were similarly delayed. The covenant was intended to take effect from 1989. This risk reduction measure was therefore ineffectual and unrealistic. 8. Another risk was the inability of GWSC to manage a project of this size whichwas to be mitigated by substantially strengthening the Project and Construction Management Unit (PCMU) established to implement the project. It was necessary to have such a unit to help expedite the project and it performed fairly well. There were problems, however, which provide lessons in designing future units of this type: (i) there were lengthy delays in staffing the PCMU with counterpart staff due to shortage of technical staff at the time; (ii) because of the relative concentration of professionals (foreign and local) in the PCMU they were frequently called by management to do many unplanned ad hoc, albeit urgent, matters at the expense of the project; (iii) PCMU management (foreign) experienced a high tumover which disrupted the flow of the project, particularly in a context of management instability in the sector (paras. 26, 32); and finally (iv) it was difficult to integrate the PCMU into GWSC as intended and the PCMU evolved into a somewhat independent unit. C. ACHIEVEMENT OF OBJECTIVES 9. The objective of improving financial performance to allow GWSC to function as a viable institution was not achieved. Comparative indicators for the period 1990 to 1997 are presented in Table 12 and summarized below. As indicated in the table below, the targets for production, sales, and commercial efficiency were not attained, hence GWSC's inability to produce a positive rate of return on assets in use throughout the life of the project. Although I The clearinghouse was set up to clear on a current basis the mutual obligations of Electricity Corporation of Ghana (ECG), Volta River Authority (VRA), GWSC and the Government which helped to improve the predictability and reliability of GWSC's cash flows. Monthly payments are made by each agency based on estimates with adjustments for actual obligations done during a quarterly reconciliation process. The arrangement has worked well since its inception in 1989.  4 actual revenues grew at an average annual rate of about 16.5%, they did not keep pace with the projected revenues (after adjusting them to reflect cedi devaluation) and expenses; inflation averaged 35.9% annually. Comparative Key Indicators (Selected Years) 1992 1993 1994 1995 1996 1997 Proj Actual Proj Actual Proj Actual Proj Actual Proj Actual Proj Actual Production (bill. gallons) 32.9 40.9 35.6 39.8 39.1 41.4 41.9 41.1 42.7 42.1 42.7 42.1 Sales (bill. gallons) 19.1 18.2 21.3 18.5 23.4 17.4 25.1 17.7 25.6 17.9 25.6 20.2 Non-revenue water (%) 42% 56% 40% 54% 40% 58% 40% 570/o 40% 57/o 38% 53% Revenue (bill. cedis) 15.6 15.6 27.3 21.7 45.8 22.2 63.9 33.0 92.4 37.4 120.5 39.1 Collection (bill. cedis) 14.3 12.5 24.9 17.4 41.9 17.8 58.7 26.2 84.7 36.5 111.0 34.2 Collection/Revenue (%) 92% 80% 91% 80% 92% 80% 92% 79% 92% 98% 92% 87% Months revenue outst 1.0 6.9 1.0 5.8 1.0 5.7 1.0 6.1 1.0 6.4 1.0 9.6 Operating ratio (%) 81% 79% 79% 78% 79% 138% 85% 101% 85% 121% 85% 139% Rate of return (%) 3.4% - 4 - 3.7 - 2.8 - 2.8 - 2.8 10. The objective to improve GWSC's financial management (FM) capacity was partially achieved. New recruitment since 1992 increased the number of professional accountants and financial analysts to an appropriate level. Staff benefitted from in-house training which helped GWSC to produce and effectively use financial projections as a management tool. The project financed an FM specialist who helped GWSC computerize its commercial statistics, providing for the first time in 1995 detailed comparative information for each region on production, consumption, and billing and collection by consumer categories. Computerization has been introduced in some of the regions and at headquarters. Staff would need to continue upgrading FM capacity, but may be unable to do so because of GWSC's persistent cash constraint. The positive impact of the project on financial management capacity is further detailed in para 3 (b) of the Borrower's ICR in Appendix B. 11. The objective to strengthen the managerial capability of GWSC both at headquarters and in the regions to enable it to effectively plan and implement a meaningful development program was partially achieved. A twinning arrangement with Thames Water International (TWI), UK, early in the project provided useful professional exposure to senior staff who were seconded to TWI for brief periods. This arrangement came to a premature end in 1992 as GWSC was no longer able to secure the necessary Government approvals for staff travel. 12. Additionally, the project provided for the rehabilitation and equipping of two training centers to further strengthen GWSC. This has been partially achieved, and training sessions are now held year round primarily for junior and mid-level staff. The works have not been completed as envisaged at appraisal, but agreement was reached on a schedule for completion during the ICR mission (see Operational Plan, Attachment II). Two retrenchment exercises carried out in 1991 and 1993 reduced the number of junior staff and temporarily increased the professional and technical staff. Further recruitment of professional staff, however, has been constrained by weak cash flows and loss of staff to the private sector. 13. The creation of a project management cell (PCMU) at headquarters staffed with international experts and local counterpart staff helped to expedite all aspects of the project (accounting for multi-currency donor funds, engineering supervision, technical studies,  5 disbursements etc). There were constraints in this arrangement, one aspect of which was not within GWSC's control, that is, instability in the PCMU leadership (a total of 6 managers since 1990). There were also communication difficulties between the PCMU and other departments of GWSC, possibly exacerbated by unstable leadership. The result is that the PCMU evolved into a semi-independent unit. 14. The physical objectives were partially achieved. Table 6A shows the planned targets and achieved outputs for key activities; while planned and actual implementation schedules are compared in Table 6B. The objective is expected to be achieved to (i) increase the output of existing systems by replacement of plant and equipment and by general system rehabilitation; and (ii) provide for the expansion of some systems and completion of others to further increase water availability. The restoration of inoperative and partially operative systems was undertaken under the Immediate Action Program - IAP - (44 systems financed by IDA) and Priority Replacement Program - PRP - (6 systems financed by DFID). These programs were successfully completed in mid-1996, though four years later than planned due mainly to delays in completing the final designs and lack of counterpart funding. 15. At appraisal 37 priority systems were proposed for rehabilitation, completion and expansion to be financed by 4 bilateral donors (AFD-France, OECF-Japan, DFID-UK and ADC- Austria). Twelve of these systems were subsequently removed due to (i)AFD's withdrawal from the project and (ii) the transfer of 6 to other on-going projects. Co-financing agreements for works in 25 systems were signed and made effective in 1991 (for ADC and DFID-financed components) and in 1994 (for OECF-financed component). System planning, detailed designs and preparation of bidding documents were completed by 1994, while bidding and award of contracts continued until early 1996. At the Credit closing date, only one contract package (involving two systems - Kumasi and Sunyani) was substantially completed. The remaining systems are scheduled for completion at different dates through September 1999. 16. The objective of increasing the output of the existing water supply systems through rehabilitation, completion and expansion should be attained but at levels lower than the SAR estimates (247,700 m3/day) and lower than the revised target after system planning (310,323 m3/day) for the following reasons. The 45,000 m3/day proposed for Sekondi-Takoradi is the sum of capacities of two headworks at Daboasi (27,000 m3/day) and Inchaban (18,000 m3/day). The estimated reliable yield in the dry season of the impoundment at Anumkwesi (raw water source for Inchaban) is only 7,200 m3/day (if the impoundment is desilted). To match the capacity of the rehabilitated treatment plant at Inchaban, the yield of the reservoir needs to be increased by raising the Dam and desilting the impoundment (both of which were not part of this project). 17. Similarly, the capacity of the treatment plant for Cape Coast is 29,208 m3/day, but the dry season reliable yield of the raw water source (Brimsu reservoir) is only 13,750mV/day. The feasibility of increasing the reservoir by raising the Brimsu Dam was studied under the project but could not be implemented because of a funding constraint. At Kwanyaku, the impoundment on the Ayensu River needs to be desilted and the treatment plant uprated prior to attaining the production level of 23,000 m3/day. These works were not implemented under the project. Possible realizable capacity on completion of system rehabilitation is therefore about 220,000 ms/day. 18. In some of the systems where the headworks capacity was increased (such as Kumasi with 50% increase in headworks capacity) there was no matching expansion of the distribution system, and additional consumer connections were not financed under the project. Consequently,  6 output increases may not translate into proportionate increase in the population served and in increased sales. Rehabilitation and expansion of the distribution systems (including consumer connections) need therefore to be given priority in future investments. 19. The objective to improve GWSC's operation and maintenance (O&M) ability was partially achieved. Schedules of operational support equipment, tools and vehicles were prepared with the assistance of technical specialists retained through the PMC2 contract, and supplied under contracts financed by OECF. The delivery and distribution of vehicles greatly facilitated operation and maintenance. 20. With the assistance of technical specialists, the O&M department has been re-organized; procedures for accumulating operating data at the systems are now in place; and systems operating manuals produced. Pilot in-house training was conducted in six systems prior to commencement of systems rehabilitation contracts. This training which was combined with minor rehabilitation works could not be sustained because of funding constraints. 21. A pilot preventive maintenance program (PMP) was developed and implemented in 1994 in Volta and Ashanti regions. This program should have been extended and institutionalized at the completion of system rehabilitation works, backed by appropriate levels of spare parts or means for timely procurement of such spare parts, and training in the field or at the upgraded training centers for maintenance personnel. GWSC will now have to develop strategies for adequate maintenance of its M&E equipment after it is installed. 22. The objective to reduce the levels of u-f-w in three major systems in the project area as envisaged in the SAR was not undertaken in part because of delays in developing a viable proposal. GWSC has taken certain actions which will facilitate implementation of a u-f-w program in the future and has commenced a limited program with its own resources. These actions include: (i) procurement of production, zonal and consumer meters (financed by IDA and OECF) - installation of the meters is in progress but constrained by cash flows; (ii) procurement of leak detection equipment; (iii) training of distribution staff in four regions on leak detection and repair procedures; (iv) introduction of a computerized system of leak management; and (v) customer management survey of ATMA. 23. The economic performance related issues of the project are inevitably linked to the financial and technical problems presented in the previous paragraphs. In general, the analysis aims to capture the economic benefits deriving from improving the water supply, and refers to these measurements only as indicative of true economic returns. The SAR method aimed to measure these benefits using the willingness to pay approach and to confirm that the demand curve was reasonable. The estimates presented in the SAR were thus only meant to be illustrative of the general magnitude of the economic benefits of the project. 24. Regarding the quantifiable objectives presented in the ex-ante economic analysis, the project referred to four indicators: (i) production; (ii) physical leakage; (iii) net water for use; and (iv) new consumers. These objectives were not fully met. The target for estimated production capacity of treated water was 15.665 million gallons (90.4 Mm3) of water per year, but capacity will not reach this level as 6 systems that were intended to be financed by AFD were not done, and another 6 were dropped from the project for budgetary reasons. The analysis estimated the 2 Project Management Consultants (PMC) are the consultants retained to manage the project via the PCMU.  7 physical leakage of the system to remain 25%. The current u-f-w is estimated at 57%, much of which is administrative; physical leakage is now estimated at 25% to 30%. The net amount of water for use after the completion of the project was expected to reach 11.755 million gallons (53.3 Mm3). The number of new consumers added to the network at the end of the project was expected to reach 300,000. However, the number of customers connected to the network is stagnant as there was limited expansion of the distribution system. There has been very limited growth in the population served. 25. In estimating the economic returns of the project the main quantifiable costs and benefits are considered. The actual cost of the project is US$120.4 million spread over 7 years, 1992 to 1998; O&M is assumed to be 1% annually of total investment. The increased amount of water produced could serve as a proxy for estimating the quantifiable benefits of the project. Over the project period, water production for the 25 project towns increased from 49.5 Mm3 in 1988 to 70.7 Mm3 in 1997. The incremental amount produced is equivalent to 20.9 Mm3/year and falls short of the SAR estimates for the 25 towns of 27.5 Mm3/year. Assuming an average value of water equal to the present average tariff would give an economic benefit of US$7.2 million/year, which yields a marginally positive return of 2% compared with SAR projection of 15%. D. MAJoR FACTORS AFFECTING THE PROJECT 26. Factors not generally subject to government control: The implementation of system rehabilitation contracts was dependent on timely and successful conclusion of cofinancing agreements with external support agencies. Two of these agreements (with DFID and ADC) became effective in 1991. However, the inability to finalize agreements with AFD resulted in a contract package for six systems in the Eastem Region not being implemented. 27. Factors generally subject to government control: Management instability negatively affected pro.ject implementation. Over the period 1991 to 1998, there were 5 Managing Directors in GWSC who served in acting assignments for most of their respective terms. Similar frequent changes occurred in all of the key senior management positions within GWSC. Additionally, at the parent Ministry, MWH, there were frequent changes in Ministers which did not help with continuity and the ability to have policies approved as and when required. Devaluation of the Cedi at about 36% annually had a major negative impact on GWSC's cash flows and its ability to produce counterpart funds when needed. 28. Factors generally subject to implementing agency control: GWSC was required (in the financing plan for the project) to finance 47% of the local costs of the project (equivalent to US$18.8 m) from its internally generated revenue but was unable to meet this obligation. Persistent cash shortages resulted in protracted implementation periods and reduction in works under many contracts. Contract costs also increased as a result of claims on delayed payments. DFID provided supplementary financing of