Ghana - Water Supply Projects

Document of The World Bank FOR OFCLl USE ONLY Report No. 13286 PERFORMANCE AUDIT REPORT GHANA ACCRA/TENA WATER SUPPLY PROJECT II (CREDIT 499-GH) AND WATER SUPPLY TECHNICAL ASSISTANCE AND REHABILITATION PROJECT (CREDIT 1342-GB) JUNE 29, 1994 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 EQUIVALENTS (Annual Averages) CWrrency Unit = Cedi (C) C 1.00 = 100 Peseva EXCHANGE RATE (CEDI TO US$) 1974 US$1.00 = 0.8696 1978 US$1.00 = 0.3636 1983 US$1.00 = 0.3636 1984 US$1.00 = 0.0244 1985 US$1.00 = 0.0175 1986 US$1.00 = 0.0111 1987 US$1.00 = 0.0061 1988 US$1.00 = 0.0049 1989 US$1.00 = 0.0037 1990 US$1.00 = 0.0031 MEASURES AND EQUIVALENTS in = inch ft = foot mi= mile mg = million Imperial gallons (1 mg = 4546./cu.m) mgd = mg per day mm = millimeter cu. m = cubic meter km = kilometer m = meter FOR OFFICIAL USE ONLY ABBREVIATIONS AND ACRONYMS AC Asbestos Cement AfDB African Development Bank ATMA Accra-Tema Metropolitan Area CIDA Canadian International Development Agency GDP Gross Domestic Product GOG Government of Ghana GWSC Ghana Water and Sewerage Corporation ID Institutional Development IDA International Development Association M & E Monitoring and Evaluation MIP Management Improvement Program OED Operations Evaluation Department O&M Operation and Maintenance PAR Performance Audit Report PCR Project Completion Report SAR Staff Appraisal Report TA Technical Assistance UFW Unaccounted-for-Water FISCAL YEAR January 1 - December 31 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation June 29, 1994 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on Ghana Accra/Tema Water Supply Project II and Water Supply Technical Assistance and Rehabilitation Project (Credit 499-GH and Credit 1342-GH) Attached is the "Performance Audit Report on Ghana - Accra/Tema Water Supply Project II and Water Supply Technical Assistance and Rehabilitation Project (Credit 499-GH and Credit 1342-GH)" prepared by the Operations Evaluation Department. The audited projects were prepared under emergency situations; the Accra/Tema Water Supply Project II after the collapse of the small dam at Weija, and the Water Supply Technical Assistance and Rehabilitation Project after a severe blockage of the Kpong - Accra pipeline, the water supply lifeline of the Accra metropolitan area. Furthermore, Ghana was facing a serious economic crisis at the time of the implementation of the latter project. This in turn caused a critical exodus of trained manpower to neighboring countries and overseas. Both projects achieved their physical objectives reasonably well. They also focused on enhancing the institutional development of the Ghana Water and Sewerage Corporation and its financial performance. The projects clearly manifest the complexities faced in, and the time and effort required for, achieving sustained results in these areas. The role of technical assistance proved to be essential in carrying out institutional development and policy reform programs. Through these two projects the sector agency gained significant experience and increased its management autonomy. The foundation was laid for further capacity building efforts which are now being implemented under an ongoing follow-on operation. The audit rates the outcome of the Accra/Tema Water Supply Project II as satisfactory, its institutional development as negligible, and the sustainability of benefits as uncertain. The outcome of the Water Supply Technical Assistance and Rehabilitation Project is rated as satisfactory and its institutional development as modest. The sustainability of benefits from this project is rated as likely, primarily thanks to much improved financial performance of the Ghana Water and Sewerage Corporation. These audit ratings agree with the previous PCR-based assessments. Robert Picciotto by H. Eberhard K6pp Attachment This document has a restricted distribution and my be used by recipients only in the performnce of their official duties. Its contents my not otherwise be disclosed without World Bank authorization.  PERFORMANCE AUDIT REPORT GHANA ACCRA/TEMA WATER SUPPLY PROJECT H (CREDIT 499-GH) AND WATER SUPPLY TECHNICAL ASSISTANCE AND REHABILITATION PROJECT (CREDIT 1342-GH) TABLE OF CONTENTS Page No. PREFACE ......................................................... i BASIC DATA SHEET ................................................ ii EVALUATION SUMMARY ............................................ vii 1. BACKGROUND ............................................... 1 A. Ghana's Macroeconomic Setting .................................... 1 B. Water Sector Setting ............................................ 2 2. THE PROJECT AND ITS IMPLEMENTATION EXPERIENCE ........... 4 A. Objectives and Description of the Project ............................. 4 B. Project Preparation ............................................. 6 C. Implementation Experience ....................................... 8 3. PROJECT RESULTS ............................................. 9 A. Physical Results ................................................ 9 B. Project Costs .................................................. 10 C. Institutional Development Results .................................. 12 D. Operational Results ............................................ 14 4. BORROWER AND BANK PERFORMANCE ............................ 16 A. Borrower Performance .......................................... 16 B. Bank Performance ............................................. 16 C. Compliance with Covenants and Directives ........................... 17 This report was prepared by Tauno Skytta (Task Manager) who audited the projects in October 1993, with administrative assistance from E. Tweddle. Table of Contents (Cont.) Page No. 5. PERFORMANCE ASSESSMENT AND ISSUES ...................... 18 A. Assessment of Project Performance ................................. 18 B . Issues ........................................................ 18 6. CONCLUSIONS AND LESSONS .................................. 19 A. Conclusions .... ........... ... **- - -*****- -***.- 19 B. Lessons of Experience ........................................... 19 i PERFORMANCE AUDIT REPORT GHANA ACCRAfTEMA WATER SUPPLY PROJECT H (CREDIT 499-GH) AND WATER SUPPLY TECHNICAL ASSISTANCE AND REHABILITATION PROJECT (CREDIT 1342-GH) PREFACE This is a Performance Audit Report (PAR) on the subject projects. The Accraffema Water Supply Project was approved by the Board of Directors on June 27, 1974 and financed by Credit 499- GH in the amount of US$10.4 million. The Credit was closed on June 30, 1981, after an extension of one year, and a small unused portion of it, about US$0.04 million was canceled. The Water Supply Technical Assistance and Rehabilitation Project was approved on March 29, 1983 and financed by Credit 1342-GH in the amount of US$13.0 million. The Credit was closed on June 30, 1990, after an extension of three years, and an outstanding balance of US$0.33 million was canceled. The PAR consists of an Evaluation Summary and a Performance Audit Report prepared by the Operations Evaluation Department (OED). The Project Completion Reports (PCR), prepared by the Water Supply Division of the then West Africa Regional Office for Credit 499-GH, and by the Infrastructure Operations Division of the Country Department IV (Western Africa) of the Africa Regional Office for Credit 1342-GH, were previously submitted to the Board as Report No. 4600, dated June 28, 1983, and No. 9732, dated June 28, 1991, respectively. The PCRs provide a detailed account of project experiences and a candid discussion on issues faced during project implementation. The PAR elaborates on particular aspects such as the sequence of actions from one project to another to achieve improvements in institutional development and financial performance. The PAR is based on the PCRs, the President's Reports, project legal documents, the SARs, review of relevant project files, and other background materials and study reports. The PAR also draws on field observations and on extensive interviews with officials of the Government of Ghana (GOG) and the Ghana Water and Sewerage Corporation (GWSC) in Accra during the audit mission in October 1993, as well as on discussions with Bank operational staff. The Audit endorses PCR findings in most respects. It particularly emphasizes the role of technical assistance in carrying out sustained institutional development and policy reform programs. Following standard OED procedures, the draft PAR was sent to the Borrower for comments; however no comments were received. ii PERFORMANCE AUDIT REPORT GHANA ACCRA/TEMA WATER SUPPLY PROJECT II (CREDIT 499-GH) BASIC DATA SHEET KEY PROJECT DATA Actual or Actual as % Appraisal Current of Appraisal Item Expectation Estimate Estimate Total Project Cost (US$ m) 44.70 51.70 115.70 Loan Amount (US$ m) 10.40 10.36 99.60 Economic Rate of Return % 6 negative Institutional Performance: Negligible CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS Appraisal Estimate (US$M) 10.40 Actual (US$M) 10.36 Actual as % of Appraisal 99.60 Date of Final Disbursement: March 31, 1981 PROJECT DATES Original Actual Identification 08/21/70 Preparation 09/ /72 Negotiations 06/04/74 Board Approval 06/27/74 Credit Signature 07/26/74 Credit Effectiveness 10/24/74 01/07/75 Credit Closing 06/30/80 06/30/81 iii STAFF INPUTS (staff-weeks) FYT2 FY73 FY74 FY75 FY76 FY77 FY78 FY79 FY80 FY81 FY82 FY83 Total PreappraisaL 1.8 .2 - - - - - - - - 2.0 Appraisat 30.3 - 53.0 - - - - - - - 127.9 Negotiation - - 7.4 - - - - - - - - . 9.4 Supervision - - - 18.4 26.5 21.5 12.5 14.3 4.6 8.4 4.0 16.8 127.4 Other - - 17.7 - - - - - - - - - 17.7 MISSION DATA Date No. of No. of Month/Year Days Persons Identification 10/69 1 10 2 Preparation 08/70 1 10 2 Preappraisal 05/71 1 10 2 Appraisal 11/71 1 15 3 Post Appraisal 12/72 5 2 Reappraisal 04/73 15 3 Reappraisal Update 02/74 15 2 Supervision 1 10/74 5 2 Supervision 2 04/75 5 1 Supervision 3 09/75 10 2 Supervision 4 02/76 5 1 Supervision 5 01/77 10 2 Supervision 6 06/77 5 1 Supervision 7 09/77 10 1 Supervision 8 04/78 5 1 Supervision 9 11/78 2 15 4 Supervision 10 02/79 2 5 1 Supervision 11 05/79 2 5 1 Supervision 12 05/80 2 5 1 Supervision 13 12/80 2 15 2 Completion 1 12/81 2 5 2 Completion 2 09/82 2 5 3 Total 190 OTHER PROJECT DATA Borrower: Republic of Ghana Executing Agencies: Ghana Water and Sewerage Corporation 1 Combined with supervision of Credit 160-GH 2 Combined with preparation of Credit 1342-GH iv PERFORMANCE AUDIT REPORT GHANA WATER SUPPLY TECHNICAL ASSISTANCE AND REHABILITATION PROJECT (CREDIT 1342-GH) BASIC DATA SHEET KEY PROJECT DATA Actual or Actual as Z Appraisal Current of Appraisal Item Expectation Estimate Estimate Total Project Cost (US$ m) 22.40 18.21 81.30 Loan Amount (US$ m) 13.00 12.67 97.50 Economic Rate of Return n/a n/a Institutional Performance: Modest CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS Appraisal Estimate (US$M) 13.00 Actual (US$M) 12.67 Actual as % of Appraisal 97.50 Date of Final Disbursement: October 31, 1990 PROJECT DATES Original Actual Identification 07/77 Preparation 04/78 Preappraisal 08/78 Appraisal Mission - 1 11/78 Appraisal Mission - 2 02/79 Reappraisal 11/80 Credit Negotiation 08/81 Credit Negotiation 05/79 01/83 Board Approval 11/81 03/83 Credit Signature 04/83 Credit Effectiveness 11/83 Credit Closing 06/87 06/90 Project Completion 10/87 10/90 v STAFF INPUTS (staffweeks) E17 fY80 FY81 M f 8 fflM f 8E FY88 FY89 FY90 FY91 Total PreappraisaL 12.8 - - - - - - - - - - 13.0 AppraisaL 8.4 .7 - - - - - - . - - - 9.1 Negotiation 1.3 6.2 13.4 6.6 7.7 - - - - - - - 35.2 Supervision - - - - 4.9 20.4 16.5 16.8 17.3 10.5 9.3 3.5 - 104.6 Other 5.0 6.5 2.4 3.4 5.4 - - - .1 - - - - 22.7 MISSION DATA Date No. of No. of Month/Year Days Persons Identification 07/77 15 2 Preparation 04/78 15 2 Preappraisal 08/78 20 2 Appraisal I 11/78 30 2 Appraisal I 02/79 10 2 Appraisal II 11/80 20 2 Supervision 1 06/83 10 2 Supervision 2 09/83 10 2 Supervision 3 02/84 10 2 Supervision 4 05/84 5 2 Supervision 5 09/84 10 2 Supervision 6 09/85 5 1 Supervision 7 11/85 10 2 Supervision 8 03/86 2.5 1 Supervision 9 10/86 5 2 Supervision 10 02/87 2.5 1 Supervision 11 5-6/87 10 2 Supervision 12 09/87 10 2 Supervision 13 11/87 5 1 Supervision 14 02/88 10 2 Supervision 15 06/88 10 2 Supervision 16 06/89 5 1 Supervision 17 11/89 5 1 Supervision 18 08/90 10 2 Total 245 vi OTHER PROJECT DATA Borrower: Republic of Ghana Executing Agencies: Ghana Water and Sewerage Corporation - vii - PERFORMANCE AUDIT REPORT GHANA ACCRA/TEMA WATER SUPPLY PROJECT H (CREDIT 499-GH) AND WATER SUPPLY TECHNICAL ASSISTANCE AND REHABILITATION PROJECT (CREDIT 1342-GH) EVALUATION SUMMARY Background 1. The economic crisis of the late 1970s GWSC is a decentralized organization with and early 1980s caused a severe setback in nine regional offices split into 41 district the economy of Ghana. The decline in the management units. ATMA is the regional standard of living led to an exodus of skilled office for the Accra/Tema capital area manpower. The per capita income was (paws. 1.03-1.04). further eroded by high population growth estimated at 2.6% between 1970 and 1984. 3. Since the 1960s, all major projects As a result, the capacity of the GOG to for water supply and sanitation have been adequately maintain essential services, instituted with substantial external aid; the including water supply, was severely Bank has assisted with four projects. In its constrained. The Bank assisted Ghana economic recovery program, the during these difficult times through several Government's strategy for the water sector operations to support the Government's has aimed at (i) rehabilitating and improving recovery program (paras. 1.01-1.02). utilization of existing assets, and (ii) improving management and capacity of the 2. In the early 1980s, the urban water GWSC (pars. 1.05-1.06). supply coverage (piped water) was estimated at approx. 85%. The coverage for rural 4. The Bank began its support of areas was estimated to be below 20% Ghana's water and sanitation program in (mostly wells and boreholes). In Ghana, the 1969. The first credit was to assist in the nation-wide responsibility for public water implementation of the Accra/Tema Water supply and sewerage is entrusted to the and Sewerage Project. The water supply Ghana Water and Sewerage Corporation component of the project exceeded its (GWSC). On-site sanitation is the targets. The sewerage component, however, responsibility of the local authorities. The was not a success; only the system was - viii - constructed and no consumers were 6. A change of government prolonged connected until much later. The Bank the preparation of the Stage 11 project. The continued its involvement in the water sector collapse in 1968 of the existing (small) through a systematic program of projects Weija dam further complicated the specially targeting water resource problems preparation causing the redesign of the main and institutional issues. The focus on component. During this period, devaluation capacity building was however soon of the Cedi and general price escalations overshadowed by urgent rehabilitation needs affected cost estimates; bids for the two of the Kpong transmission line (paras. 1.07- main components, the WeiJa Dam and the 1.10). rural water supply works exceeded the projected estimates by 30% and 49% Projects and Their Implementation respectively (paras. 2.04-2.06). Experience 7. The TAR project was appraised in 5. The Stage II Project was to increase 1978 with special focus on TA and heavy the water supply capacity in the ATMA area emphasis on strengthening the GWSC's (about 800,000 people at the time) by management performance and capacity for constructing the Weija Dam and extend planning and project implementation. A near service to some 170,000 people in adjacent disastrous deterioration of the Kpong - rural areas. The TAR Project was to ATMA transmission main in 1980, however, provide emergency maintenance and repairs changed the turn of events. The Bank was to the Kpong-ATMA transmission pipeline asked to consider the restructuring of the and primary distribution in the ATMA area. project to include urgent rehabilitation of the In general, both projects were to strengthen pipeline. The revised proposal was the management capacity of the GWSC, and approved in March 1983 (paras. 2.07-2.08). especially the TAR Project was to improve GWSC's operating and financial 8. From a purely technical point of performance to make it a viable and efficient view, the Audit assesses that the "quality at sector institution. It was also to prepare entry" of both projects was acceptable. The detailed plans for suitable investment same does not, however, apply to the projects for possible future funding by the institutional aspects. in general, the Bank. The Stage II Project was a parallel processing of both projects was delayed due intervention between the Bank (IDA), AfDB to Government changes, staff turn-over, and and CIDA; its total cost was estimated at eventually, critical technical events caused US$44.7 million of which the IDA credit the project proposals to be substantially was US$10.4 million or 23.3%. The TAR redesigned. The total elapsed time from Project was estimated to cost US$22.4 identification to appraisal was well over four million, of which the IDA credit was to years for the Stage H Project and over three cover the whole foreign exchange years for the TAR Project; in the latter case component of US$13.0 million or 58% it took another two and a half years before (paras. 2.01-2.03). the project was finally approved (paras. 2.09-2.10). -ix - 9. With substantial parallel funding, the additional work on it has continued since Stage II Project required close coordination. under funding from the Italian Government This proved to be difficult. For example, the and AfDB (paras. 3.01-3.03). completion of interlinked components such as the dam, the treatment plant, and the 12. Although some components of the transmission mains were not properly Stage H Project had maJor cost increases coordinated. As to institutional (about 50% for the WeiJa Dam), the overall improvements, the Stage II Project achieved cost increase was only 16%, including the very little in terms of set management and costs of the components funded in parallel operational goals. Another special concern by other donors. The total costs of the TAR was the poor performance in the installation Project decreased by about 18% mainly due of sewer connections in Accra; no to the devaluation of the Cedi. Costs of connection program was prepared and some components increased, however, sewerage operations have remained significantly because of changes in the inefficient ever since (paras. 2.11, 2.12, scope. These increases in costs were and 2.15). covered through a high provision for contingencies which was considered justified 10. The negligible results in institutional at appraisal because of the complex nature improvements prompted another attempt, of the project (paras. 3.04-3.07). under the TAR Project, to focus on the strengthening of the GWSC to become a 13. The TAR Project was a new attempt viable and efficient sector institution. Again, to achieve the institutional improvements the results were less than anticipated. The that had failed under the Stage II Project. expatriate TA manpower which was to assist The MIP was prepared, but its results as to in the implementation of the MIP and to the implementation of its recommendations strengthen the GWSC management was were minimal. The TAR Project was more forced instead, under the circumstances, to successful in improving the GWSC's concentrate on the day-to-day operation of financial performance; the tariffs were the GWSC. Also, the GWSC failed to increased rather regularly after a drastic complete the revaluation of its fixed assets correction of tariffs in 1986. As a result, (paras. 2.13, 2.14, and 2.16). the GWSC has been able to run its operations without subsidies ever since. Project Results Reasonable progress has also been made in metering water connections, although the 11. Both projects managed to number of metered connections in ATMA is substantially achieve their physical targets. still rather low at 45% (paras. 3.08-3.10). With regard to the Stage II Project, however, components under parallel funding 14. The staffing level of the GWSC has by CIDA and AfDB were delayed, and thus decreased from 33 in 1976 to 13 staff per the full realization of benefits was delayed 1000 water connections in 1993, i.e. below by several years. The Kpong transmission the level common in developing countres. main, Accra's lifeline, was put back into Similarly, the share of personnel costs of the operation under the TAR Project but total operating expenses is clearly lower than in many other cities in the developing II Project, revaluation of assets and lack of countries. These indicators show progress counterpart funds under the TAR Project, made in operational efficiency. Under the and in general, limited achievements of TAR Project, plans were prepared for a institutional goals under both projects follow-on project, i.e. the ongoing Sector (paras. 4.01-4.03). Rehabilitation Project which continues the implementation of many institution building 17. The Bank provided effective and tasks left unfinished earlier (paras. 3.11- timely support in the implementation of both 3.12). projects. This was essential considering the difficult country situation. Supervision 15. Both projects were to carry out intervals were around five to six months and training of O&M staff but the results were the quality is reported to have been good meager in part due to the complex nature of (para. 4.04). the projects' technical aspects that drew the attention away from operational 18. Although the compliance with improvements. Therefore, very little has covenants was reasonable, some problems been achieved in improving the were encountered in meeting the unaccounted-for-water ratio which remains requirements of both technical and financial at over 50% today. Surveillance of raw covenants under the Stage II Project. These water reservoirs and dams is limited, and aspects were made part of the TAR Project improvements to treatment plants remain and the performance has now improved. unfinished although such plans were The most serious non-compliance was in the prepared. The GWSC does not have a clear revaluation of the GWSC's fixed assets policy on providing service to the poorer which is being completed only now under segments of the city's population who the follow-on project. Similarly, the currently buy their water from vendors requirements of OMSs on the cost recovery (kiosk operators) at a higher cost than would (2.25) for the Stage 11 Project and be the case if the supply was provided by preparation of institutional components the GWSC. Improvements to some of these (2.28) for the TAR Project were not issues are, however, envisaged in the most adequately met (paras. 4.04-4.06). recent Corporate Plan that is under implementation (paras. 3.13-3.17). Performance Asessment and lksues Borrower and Bank Performianc 19. The Audit finds that the Stage II Project achieved all of its physical 16. Both Projects were implemented objectives, but its benefits were realized under very difficult circumstances as the much later due to the delayed completion of GWSC was loosing skilled manpower. With the supplementary components. This project the help of the TA (twinning) personnel, the achieved very little of its institutional performance of the GWSC was reasonably objectives. On balance, the Audit rates the satisfactory. The main problems were project's outcome as satisfactory, its delays in components funded in parallel and institutional development as negligible, and poor financial performance under the Stage the sustainability of its benefits uncertain. (para.4.04) - Xi - Also the TAR Project completed its physical operations the sector agency has gained objectives and it achieved marked significant experience and increased its improvements in the GWSC's financial management autonomy and is today in much performance. On this basis, the Audit rates better health financially. The key lessons the project's outcome as satisfactory, its learned from these operations include institutional development as modest, and the (paras. 6.01-6.02): sustainability of its benefits as likely. Ratings of both projects agree with the PCR- a) To achieve institutional based ratings (paras. 5.01-5.02). improvements in projects that are demanding technically requires 20. These two projects dealt with many careful planning prior to project operational deficiencies of the GWSC but launch, and concerted effort, firm fell short in achieving sustained commitment, and resource allocation improvements. There are still problems in during implementation. The best providing adequate training, necessary tools, approach would be a gradual spare parts, and vehicles. The GWSC is program through several operations contemplating to solve some of these that entail realistic targets for each operational issues through privatization. The step; issue of equitable service to the urban poor in Accra is awaiting a sound policy decision b) The approach of using twinning as a by the GWSC. The Stage II Project handled means of institutional strengthening the resettlement of communities displaced requires careful design involving due to the Weija dam impoundment in a both parties concerned in the satisfactory manner. Improved housing and selection of staff, -both staff for piped water supply was provided. The providing the services to ensure resettled families have adopted fishing in the adequate quality, and local staff for reservoir as a new and productive form of receiving training to ensure sufficient livelihood (paras. 5.03-5.05). time and commitment. Similarly, the arrangements for the attachment Conclusions and Lessons of local staff to the twinning organization need to have specific 21. These projects clearly illustrate the and achievable objectives; time and effort needed in order to achieve sustainable results under difficult country c) To ensure the sustainability of the conditions. The eroding institutional institutional and operational capacity of the implementing sector agency development efforts, appropriate further complicated matters. The Bank was equipment and manpower provision ready to face the challenge and invested the need to be made on a long-term necessary support through frequent and basis; and quality supervision to first deal with the emergency construction and rehabilitation d) To ensure the sustainability of tasks, and eventually with far reaching improved operational and financial institutional objectives. Through these two performance, an adequate level of - xii - authonomy is required to make crucial decisions on tariffs, manpower development, and investments. PERFORMANCE AUDIT REPORT GHANA ACCRA/TEMA WATER SUPPLY PROJECT H (CREDIT 499-GH) AND WATER SUPPLY TECHNICAL ASSISTANCE AND REHABILITATION PROJECT (CREDIT 1342-GH) 1. BACKGROUND A. Ghana's Macroeconomic Setting 1.01 Ghana had traditionally enjoyed a fairly high standard of living compared to most West African nations, but the economic crisis of the late 1970s and early 1980s caused a severe setback. The causes were a prolonged draught in 1975-77; falling prices of Ghana's main export commodity cocoa on the world market with subsequent decline in foreign exchange earnings; a greatly overvalued exchange rate leading to flourishing illegal economic activities such as smuggling; large public sector deficits; a difficult balance of payments situation; low productivity; low domestic savings; and declining investments. In comparison with many African countries Ghana had a large reservoir of skilled and trained manpower, but the low salaries in the public sector and the declining standard of living led to a brain drain and an exodus of this manpower to neighboring countries and overseas. The real per capita income was further eroded by a high population growth estimated at 2.6% between 1970 and 1984. 1.02 As a result, the capacity of the GOG to adequately maintain essential services, including water supply, was severely constrained. The Government launched an "Economic Recovery Program" in 1982, which is said to have been one of the most successful adjustment programs in Africa attaining about 5.5% annual net GDP growth. The Bank assisted Ghana during these difficult times through several operations to support the Government's recovery program.2 The relative success of the program in Ghana can specifically be attributed to realistic projections of This information is summarized from "I. The Economy" of the Report and Recommendation of the President for Water Supply Technical Assistance and Rehabilitation Project (Cr. 1342-GH), dated March 7, 1983, and "I. Introduction and Background" of the SAR for Water Sector Rehabilitation Project (Cr. 2039-GH), dated May 18, 1989. 2 See, "Adjustment in Sub-Saharan Africa; Selected Findings from OED Evaluations", Report No. 12155, June 30, 1993. The Bank support included two "Reconstruction Imports" Credits, in 1983 and 1985, an "Export Rehabilitation" Credit in 1984, an "Industrial Sector Adjustment" Credit in 1986, and two SAL Credits, in 1987 and 1988. The total amount of these credits was US$520.5 million. 2 the growth in exports and production, and the consistency with which the Government applied (economic) reform policies. One specific lesson from the African SAL, as well as from the Ghanaian experience, is noteworthy; "more attention should be given upfront to developing the institutional capacity and borrower ownership needed to sustain reform. Until this is in place, it would be prudent to accept a lower level of adjustment assistance, with more funds targeted to specific sectoral programs or investment projects (e.g. for basic infrastructure, social programs)." B. Water Sector Setting Sector Development and Strategy 1.03 For many years beginning in the late 1960s, water and sanitation sector operations and development had suffered from the Government's constrained ability to finance the local costs of the projects, and the lack of foreign exchange to obtain imported materials and spare parts essential for O&M of the systems. Furthermore, the mass exodus of qualified managerial and technical staff hit the water sector hard. At the same time, overstaffing of the lower echelons was commonly a serious problem. In the early 1980s, the urban water supply coverage (access to piped water supply) was estimated at approx. 85%. The coverage for rural areas was estimated to be below 20%, mostly from wells and boreholes equipped with hand pumps. In the ATMA area, the Kpong Water Works at Volta River, about 60 km north of Accra, is the main source of water, about 85% of the total supply. The capacity' of the Kpong plant was expanded to 40 mgd in 1966. The Stage II Project, see para. I La below, extended the capacity of the Weija dam, currently the other main source of supply for ATMA, to about 60 mgd. The capacity of the attached water treatment plant is 30 mgd and it became operational in 1985, at the completion of the CIDA funded components. 1.04 In Ghana, the nation-wide responsibility for the overall planning, development, and operations of public water supply and sewerage is entrusted to the Ghana Water and Sewerage Corporation (GWSC) established in 1966. On-site sanitation is the responsibility of the local authorities. The GWSC operates under the general direction of the Ministry of Works and Housing (MWH) and is a decentralized organization which has nine semi-autonomous regional offices. Regional offices are divided into 41 district management units. In Accra/Tema, the national capital area, the regional office is called ATMA. ATMA's operations are as large as those of the rest of the country combined. The GWSC has long suffered from overstaffing in lower staff categories, and inadequate salaries compared with parastatals in other sectors resulting in the attrition of experienced senior and middle level managerial staff. 1.05 Since the 1960s, all major projects for water supply and sanitation development have been instituted with substantial external aid. The Bank has assisted with four projects one of Production capacities presented here are based on estimates given in various reports. 3 which is under implementation. One new community water supply and sanitation project is being negotiated. The other donors are AfDB, CIDA, GtZ, and Italy. In the past, bilateral donors have been the main supporters of the rural programs. 1.06 In accordance with the objectives of the economic recovery program, the Government has made serious efforts to improve essential services. The strategy for the water sector has primarily aimed at (i) rehabilitating and improving capacity utilization of the existing assets, and (ii) expanding the essential infrastructure including water supply and sanitation. In order to achieve these objectives, the Government recognized the need to improve the management and organization of the GWSC through staff training and technical assistance for planning, project preparation and implementation, and O&M of the assets. Bank's Role 1.07 The Bank began its financial support of Ghana's water and sanitation program in 1969 with a credit of US$3.5 million (Cr. 160-GH). The credit was to assist in the implementation of the Accra/Tema Water and Sewerage Project.' The project, completed in 1973, achieved its physical objectives in providing extended water supply coverage in the ATMA area. An indication of its success was the increase in connections from about 17,000 in 1968 to approx. 40,000 in 1976, a figure that exceeded the appraisal estimate by 18%. The sewerage component was to lay the foundation for a new sewerage system in Accra including about 20 miles of collector, interceptor, and trunk sewers, three pumping stations, and an ocean outfall. The sewerage component was not a success as only the system was constructed, and no consumers were connected to it until much later. The number of connections was only slightly over 100 in 1976 and it is estimated that the number of connections today is in the order of 500. After this first intervention, sewerage has received very little, if any, attention and on-site solutions such as septic tanks and pit latrines are common. 1.08 In addition, institutional issues and actions were identified as an urgent area of support. Therefore, the Bank envisaged a systematic program of projects to follow. This logical, and well justified, approach focusing on capacity building was, however, soon to be overshadowed by urgent rehabilitation needs such as the Kpong transmission pipeline, see paras. 2.07 and 2.08 below. 1.09 Several lessons had been learned from the first operation and specific actions had been defined. They included: (i) realistic analysis of the central government's involvement in the GWSC's policy and decision making process is essential at appraisal of new operations; (ii) analysis of poor institutional achievements needs to be carried out; (iii) analysis of demand and technological options for sanitation development needs to be prepared; and (iv) detailed review of the GWSC's financial performance and action plan for meeting its obligations needs to be developed. This project was among the first generation of water supply and sewerage projects prepared and financed by the Bank. Its achievements are discussed in a PAR No. 1683, dated July 29, 1977. 4 1.10 The GWSC suffered from manpower shortages, especially at the senior and middle management level because of the large scale brain drain in Ghana in the 1970s and even more severely in the 1980s. In addition to the urgent rehabilitation of the Kpong pipeline, the TAR Project (see para. 2.01,b below) continued to address the above institutional issues, some of which were left unfinished under the Stage II Project. 2. THE PROJECTS AND THEIR IMPLEMENTATION EXPERIENCE A. Objectives and Description of the Projects 2.01 The Staff Appraisal Reports (SAR), and legal documents, specify the objectives of the two projects under Audit as follows: a) Accra/Tema Water Supply Project II, Cr. 499-GH (Stage II Project) Objectives: the project was designed to (i) increase the water supply capacity in the ATMA area to meet the demand of about 800,000 people; (ii) extend water supply service to some 170,000 people in adjacent rural areas; and (iii) in general, improve the management capacity of the GWSC. Components: * the IDA funded components included (i) construction of a 17 meter high earth dam at Weija (Densu River) including intake structures and the pumping station; (ii) supply of pipes and fittings for expansion of secondary and tertiary distribution systems including the adjacent rural areas; (iii) supply of water meters for new connections; (iv) supply of equipment for a leak detection program; (v) preparation of a connection program for the sewerage system; (vi) training program and related facilities and equipment; and (vii) consultants services for design of all components and construction supervision. * In addition to the above, the project included (i) components funded by the AfDB such as civil works for a treatment plant at Weija and major service reservoirs, supply of material (pipes and fittings) for the transmission main from the Weija Works, and equipment and technical assistance for installation of primary system pipelines; (ii) components funded by the CIDA such as pumping equipment, water treatment facilities at Weija, and supervision of their installation; and (iii) components directly funded by the GWSC such as civil works for laying of small diameter pipes in adjacent rural areas, construction of small service reservoirs (6 nos), construction of maintenance yards (7 nos), installation of secondary, primary, and tertiary distribution mains (material supplied under IDA, AfDB, and 5 CIDA funding above), land acquisition and village (4 villages) resettlement in the Weija reservoir catchment area. b) Water Supply Technical Assistance and Rehabilitation Project, Cr. 1342-GH (TAR Project) Objectives: the project was designed to (i) strengthen the management capacity of the GWSC; (ii) improve GWSC's operating and financial performance so that it could function as a viable and efficient sector institution; (iii) provide emergency maintenance and repairs to the Kpong - ATMA transmission pipeline and the primary distribution in the ATMA area; and (iv) prepare detailed plans for suitable investment projects for possible future funding by the Bank. Components: to meet the objectives, the components included (i) preparation of a "management improvement program" (MIP) to recommend organizational, and policy improvements in the management, operation, training, and financial functions of the GWSC; (ii) TA services (about 30 man-years) to strengthen the GWSC's higher level staff and to assist in the implementation of the recommendations of the MIP; (iii) repair and rehabilitation works of the Kpong - ATMA steel pipeline by replacing corroded sections, recoating of damaged sections, replacement of defective cathodic protection, repair/replacement of defective valves, and provision of maintenance equipment; (iv) supply to ATMA of spares, tools, water meters, valves and fittings, and vehicles for emergency repair and maintenance of the distribution system including improved metering of connections, and 250 existing and 100 new public (metered) standpipes; (v) supply of teaching aids for the GWSC training schools; and (vi) provision of consultant services for final design and contract document preparation for works under (iii) and (iv) above, updating of plans and documents for a future water supply and sanitation project, and for assistance in implementation of a surveillance program for the Weija reservoir and Densu river. 2.02 The Stage II Project was a parallel funding intervention between the Bank (IDA), AfDB, and CIDA. Its total cost was estimated at US$44.7 million of which the IDA credit was US$10.4 million or 23.3%. The funding shares of AfDB, CIDA and GWSC were 10.7%, 12.3%, and 49.2%, respectively. The gap of US$2.0 million was provided by CIDA as a grant and was used for engaging a consultant for supervision of the Weija treatment plant installations. As such, the project became quite complex and required a lot of coordination during implementation. 2.03 The TAR Project was estimated to cost US$22.4 million, of which the foreign exchange component was US$13.0 million (58%) and the local cost component US$9.4 million (42%). Thus the IDA credit of US$13.0 million was to cover the whole foreign exchange part of the project cost. In the cost estimate, the most critical element was the estimate for the repair and rehabilitation of the Kpong pipeline as the full extent and the nature of the damages were 6 difficult to determine before the mobilization of the actual implementation.' This estimate was prepared by the consultant retained prior to project appraisal under Bank PPF funding. The project was a rather straightforward operation although its institutional objectives were far- reaching but not well defined at the appraisal stage. B. Project Preparation 2.04 The Stage H1 Project was a logical follow-on project to the first project, see para x above. Its preparation was initiated as part of the Water Supply and Sewerage Master Plan for the ATMA area which was completed in 1965. The second phase of the master plan proposal was presented to the Bank for funding consideration when the first project was still under implementation and the appraisal took place in 1971. After some delays, caused among others by the change of government in Ghana in 1972, the project was finally approved in June 1974. It may be noted here that the information on the preparation and implementation of the Stage II Project is limited due to many changes in the management and staff of the GWSC after the project was completed in June 1981. Therefore, the discussion in this report is to a large extent based on the PCR and is primarily intended to provide the linkage between the first project (Cr, 160-GH) and the TAR Project. The outcome of the project is, however, analyzed later in this report. In this regard, the Audit's discussions in the field with the current GWSC management were very informative. 2.05 During the prolonged preparation time, the cost estimates of the project escalated quite rapidly due to general price escalations in Ghana and the devaluation of the Cedi. It can be assumed also that, as the project design was initially drawn from a master plan level proposal, the accuracy of the cost estimates base was questionable, especially for such a critical component as the construction of a major dam. In 1973 during the process, the GWSC decided to invite bids for two main components; one for the dam and one for the construction of rural water supply works. The bids that were received were 30% to 40% higher than the engineer's estimates, assumed to be caused by the first international oil crisis. The Bank, realizing that more funding was required, encouraged AfDB and CIDA to participate as cofinanciers. CIDA's parallel funding was tied to Canadian procurement and consultant services. 2.06 The preparation of the project was further complicated by the fact that the then existing (small) Weija dam was destroyed by floods in 1968. This caused the redesign of the main component of the project and updating of other aspects of the proposal. 2.07 The TAR Project was initially prepared and appraised in 1978 as a specially focused TA project with heavy emphasis on the strengthening of the GWSC's general management performance and improving its capacity for planning and project implementation. This approach was logical for a follow-on operation as the institution building objectives of the two previous projects were not fully achieved due, in part, to the high turnover of the GWSC's senior level In parallel with the Bank funded repair work of the Kpong pipeline, the Italian Government financed another phase of the repair. 7 officers, general overstaffing of the organization, and the lack of managerial autonomy from the Government. The Audit finds, however, that the strategy and necessary actions were never developed to the extent that sustainable results could be expected; the target still in this project was to train a certain number of managerial and other staff without really paying attention to the development of regulatory and policy frameworks, essential for a sustained institution building effort. 2.08 As in the case of the Stage II Project, matters took a dramatic turn after a near disastrous collapse of the Kpong-ATMA transmission main' in 1980. The processing of the proposed project had already been delayed because of the delayed tariff increase that was a condition set by the Bank for credit negotiations. In 1980, the Bank was asked to consider the restructuring of the proposed project to include urgent rehabilitation of the pipeline as it was still, at the time, the only reliable source of supply for the whole metropolitan area (although the Weija dam was already constructed, the CIDA funded treatment plant was still under implementation, see para. 2.11 below). Eventually, the Bank agreed to include the requested rehabilitation work in the project as well as some distribution facilities that were observed to be deteriorating rapidly since the first appraisal. The revised proposal was appraised in November 1980, and finally approved in March 1983. An additional delay was caused by another change of government in Ghana at the end of 1981, and subsequent necessary updating of all the cost estimates and renegotiations in January 1983. 2.09 From a purely technical point of view, the Audit assesses that the "quality at entry" of both projects was acceptable. The same does not, however, apply to the institutional development components of the audited projects. The projects' objectives placed the focus on institutional aspects; they were also priority areas identified in the Government's sector strategy to be supported by new projects. Project design on these aspects was still vague at appraisal, and as such not adequate. As the construction of the new dam in the Stage II Project, and the rehabilitation of the pipeline in the TAR Project became urgent and dominant items, the detailed design of institutional components received much lower priority. Furthermore, neither project SAR required systematic M&E of progress and project outcome beyond the regular reporting on the progress of physical implementation. It is no surprise then that the ID results were less than satisfactory as discussed in detail below. 2.10 In general, the processing of both projects was delayed due to Government changes, staff weaknesses and turn-over, and eventually, some critical technical events (dam and pipeline) caused the project proposals to be substantially redesigned. The total elapsed processing time from identification to the final appraisal was well over four years for the Stage H Project and over three years for the TAR Project; in the latter case it took another two and a half years before the project was finally approved. Under such circumstances, the focus of the initial project concept changed drastically. It may also be assumed that the ownership by the 6 The pipeline is a about 75 km long, 42" dia, steel pipe internally coated with cemrent mortar. The cement moitar coating was steadily deteriorating and eventually, sometime in 1980, caused a total blockage of the pipeline. 8 Government and the GWSC's commitment toward such components was substantially weakened as a result of the urgent and last minute changes in scope. C. Implementation Experience 2.11 The Stage II Project included components funded in parallel by three agencies (AfDB, CIDA, and IDA). Therefore, close coordination was essential. This became a difficult task and the completion of interlinked components such as the dam, the treatment plant, and transmission mains were not properly coordinated. For example, the dam was ready to provide raw water for treatment, about 3 to 4 years before the treatment plant was ready and operational. In the TAR Project, the design included two phases of the Kpong pipeline rehabilitation. Phase I was under Bank funding and Phase II was financed under a grant from the Italian Government. Implementation of both phases was carried out in parallel (completion in 1987), and there were no coordination problems reported. Due to the nature of the work, it became evident that more rehabilitation was required after completion of Phases I and II. Italian aid came to the rescue with an additional grant and the work was nearing completion at the time of the Audit. 2.12 The implementation of the major institutional improvements was heavily based on the use of TA. The Stage II Project was expected to provide support to improvements of the then existing training facilities (at Weija and Kumasi) of the GWSC. At the time of project appraisal these components were left rather open ended in want of an overall training program which was under preparation at the time. The GWSC failed to complete the training program and consequently, no funds for training support was expended. In brief, the Stage H Project achieved very little in terms of set management and operational goals. 2.13 Because of these negligible results, another attempt was made in the TAR Project which was expected to provide especially dedicated support to strengthen the GWSC. The MIP was to give the framework and the heavy TA manpower provision was to assist in the implementation of the recommendations so that the GWSC would become (see para. 2.01, b above) a viable and efficient sector institution. Again, the results were less than anticipated. The MIP was completed, although later than initially expected, but the implementation never got off the ground. The project was implemented under very difficult circumstances; first, the turnover of senior level officers was high (due to the exodus), there was a lack of managerial autonomy from the GOG, and heavy overstaffing at the lower levels. Second, the expatriate TA manpower, under a twinning contract with a UK water authority, which was brought in to assist in the implementation of the MIP, and strengthen the GWSC management was forced, under the circumstances, to concentrate on the day-to-day operations of the GWSC instead. This was crucial for avoiding a major disaster but it did not serve the original institutional objective of the project. 9 2.14 The main time delays were caused by the late signing of the twinning contract with a UK water authority for institutional strengthening and training. In order to complete the twinning contract, the closing date was extended by another two years. This delay allowed for including one important training component i.e. improvements to the existing training school at Weija. Similarly, as the institutional aspects were given more time, preparation of the feasibility study to form the basis for a follow-on project was appropriately completed. 2.15 The Stage II Project was especially concerned about the poor performance in installing sewer connections in the central parts of Accra where a limited system was constructed under the Accra/Tema Water and Sewerage Project. A specialist was provided in the TA program of the Stage II Project, but the individual who came stayed only a few months and was never replaced. As a result, no connection program was prepared and the sewerage operations of the GWSC have remained inefficient ever since. The TAR Project did not include any sewerage activities. There is more on these aspects later in this report, see paras. 3.10 and 4.05,a. 2.16 When the TAR Project was prepared, it was determined that the valuation of the GWSC's asset base was long outdated. Therefore, the GWSC was required to identify and value its fixed assets starting with the ATMA area. This revaluation of assets turned out to be a rather difficult task. It was delayed by about 3 years (was completed for 1987 accounts), and was first based on indices to adjust the book value of the assets. Although this was not a normally acceptable method, it helped to bring the assets to a more realistic level and it more than doubled the total value of the assets. At the time of the Audit, the detailed asset by asset revaluation process was under way. 3. PROJECT RESULTS A. Physical Results 3.01 Physical results are summarized below project by project as follows: a) Stage 11 Project All the components funded by IDA were completed as scheduled. As the project is fairly old, the records, however, do not make it possible to review some of the components in full detail. The Audit was not informed of any major deficiencies or omissions. One exception was the supply of pipes for the secondary and tertiary reticulation systems, as the GWSC preferred to procure these pipes from a local AC-pipe manufacturer. Therefore, IDA funding was required only for valves and fittings. The main problem was caused by the fact that the components under parallel funding by CIDA and AfDB This was about one year after the original credit closing date. 10 were badly delayed. Thus the full realization of the expected benefits was delayed by several years. b) TAR Project The physical targets were substantially met. The main component was the rehabilitation of the Kpong pipeline which covered the most critical sections of the emergency repair work. This was completed in parallel with other sections of the pipeline under funding from the Italian Government, and the pipeline was put back into operation. The above represented only the first steps in getting this critical transmission main, Accra's "lifeline", into functional condition; work has continued since under financing from the Italian Government and AfDB. 3.02 In general, the Stage II Project resulted in an increase in raw water production at the constructed Weija Dam. It was unfortunate, however, that the expansion and improvements to the treatment capacity were completed much behind the established schedule thus delaying the flow of benefits from the improved water supply to consumers and consequent increase in revenue to the GWSC. After completing the extension of the treatment capacity at Weija, the GWSC completed the extension of the distribution systems (at various levels) thus improving the access to increased supply of potable water to various sections of the Accra/Tema population including lower income groups in the city and adjacent (rural) villages. The increased supply of treated water from Weija became extremely critical after the collapse of the Kpong pipeline in 1980. The repair of the pipeline was carried out under the TAR Project and was a task that required specialized skills. The pipeline was kept in operation to the maximum extent during the repair but curtailment of the demand over a long period of time could not be avoided. The rehabilitation/replacement work has continued until recently but is now reaching its final stages. 3.03 The TAR Project provided a sizable allocation for vehicles, public standpipes, spare parts, water meters, and tools for improved O&M; this provision was nearly quadrupled during project implementation. The Stage II Project's contribution to these aspects came mainly through a provision for leak detection equipment. For example, in the early 1980s when the TAR Project was being prepared, the vehicle fleet of the GWSC was in bad shape with some 80% of the vehicles out of operation. Due to the emergency situation throughout the 1980s the share of "non-revenue water" had increased from below 40% to well over 50%. Small progress has been achieved recently in this regard. B. Project Costs 3.04 The overall cost data for the Stage II Project are presented in Table 1 below. The table compares the costs at appraisal and actual costs at project completion. It should be noted that, due to unavailability of records for the total costs, the table presents only its foreign currency disbursements (for the IDA funded components only) as recorded in the PCR. 11 Table 1. STAGE U PROJECT: COMPONENTS AND THEIR COSTS Components Appraisal Estimate Actual Costs Change US$ % US$ % % Million Million a. Weija Dam with Intake and Pumping Station 4.90 47.1 7.32 70.7 + 49.4 b. Improvements to Existing Pumping Station, Water Meter. 0.45 4.3 0.39 3.8 - 13.3 and Lak Detection Equipment c. Pipes and Fittings for Secondary and Tertiary Distribution 1.20 11.5 0.26' 2.5 - 78.3 Mains d. Pipes and Fittings for Adjacent Rural Areas 0.65 6.3 0.67 6.4 + 3.0 e. Consultant Services 1.10 10.6 1.49 14.4 + 35.5 f. Technical Assistance 0.25 2.4 0.23 2.2 - 8.0 g. Unallocated 1.85 17.8 0.00 0.0 TOTAL 10.40 100.0 10.369 100.0 - 0.4 3.05 The cost of the Weija Dam increased by about 50% chiefly because of the oil price hike of 1973 but also due to other factors such as inflation and changes of over 15% in the scope of the work. A large reduction of IDA disbursements occurred in the case of pipes and fittings for (secondary and tertiary) water mains. This was due to the fact that the GWSC wanted to procure pipe material for these mains from a local manufacturer of AC pipes and the IDA credit was used to finance the supply of fittings and valves only. 3.06 The overall cost breakdown for the TAR Project is presented in Table 2 below, comparing the costs at appraisal and actual costs at project completion. * Pipes procured under local funding. 9 The total actual cost of the project, including parallel fanded components, was US$ 51.7 million, about 16% higher than estimated at project appraisal. 12 Table 2. TAR PROJECT: COMPONENTS AND THEIR COSTS Components Appraisal Estimate Actual Costs Change US$ % US$ % % Million Million a. Kpong Pipeline Rehabilitation 6.31 28.2 5.81 31.9 - 7.9 b. Standpipes, Vehicles and Equipment 1.15 5.1 4.19 23.0 + 264.3 c. Technical Assistance 3.99 17.8 2.87 15.8 - 8.0 d. Management Improvement Program 2.35 10.5 2.26 12.4 - 3.8 e. Consultant Services 1.27 5.7 2.23 12.2 + 75.6 f. Weija Training School 0.85 4.7 g. Contingencies 7.33 32.7 TOTAL 22.40 100.0 18.21 100.0 - 18.7 3.07 A major change in the cost share of vehicles and equipment is the most conspicuous factor in the above cost table. It was mainly due to the major increase, justifiably though, in the number of vehicles provided under the project. The value of the consultant services also increased significantly because of the need to engage an individual consultant for carrying out management duties during critical periods of shortage of senior staff. These substantial increases were covered through an unusually large provision for contingencies, perhaps justified for a component such as the major and complex pipeline repair work. The rest of the items were within -10% of the original estimate. C. Institutional Development Results 3.08 As mentioned earlier, both projects incorporated institutional development objectives for training, and the TAR Project was to assist in the preparation of a Management Improvement Program (MIP) along with the implementation of its recommendations. The results under the Stage II Project were negligible, limited to on-the-job training of local engineering staff by the TA personnel. A case in point is the Project Manager who was engaged under IDA funding from 1974 through 1979. As the TAR Project, at least initially, was to make a comprehensive effort in GWSC's capacity building and strengthening its operating performance, a heavy TA provision was earmarked including 30 man-years of expatriate specialists' inputs. The MIP was prepared but its results as to the implementation of the recommendations were minimal. Due to the circumstances throughout the 1980s, most of the TA (twinning) team's time was diverted to dealing with the day-to-day operations rather than assisting the GWSC in the implementation of the MIP's recommendations such as improving sector policies, the functioning of the GWSC, and training of operations' and 13 finance staff. This was an unavoidable (and maybe the only possible) arrangement at the time, and it prevented an even more severe deterioration of the GWSC and its operations. The major institutional improvements and reforms were not, however, achieved. This is now being carried out under the ongoing Water Sector Rehabilitation Project (Cr. 2039-GH), see para. 3.13 below. 3.09 The financial results of the GWSC were below acceptable level throughout the 1970s and continued to deteriorate until the first major tariff increase (500%) in 1986. Thereafter, tariffs were increased in 1987 and 1988 by 25%. After a gap of two years, the tariff was increased by 25% in 1991, and again by 60% in 1992. Basically this has helped the GWSC remain financially sound and manage without any operational subsidies since 1986. The Audit was informed that the current policy is to operate on a "cash flow approach" i.e. the revenue should cover O&M, debt service, and public investment program (PIP) commitments. In 1993, the most recent tariff increase to meet this criteria was 35%, and the latest projections"o indicate tariff increases of 49%, 26%, and 25% for 1994, 1995, and 1996, respectively. Overall accounts receivable increased during the 1980s, and is now in the order of 6 months billing which is still high. Collection efficiency is acceptable for private consumers, but the high level of Government arrears has led to liquidity/cash flow problems from time to time. Since 1989, a clearing house arrangement has helped in dealing with Government accounts. 3.10 In general, the GWSC has made reasonable progress in metering the water connections. The overall metering ratio is in the order of 50% (as of June 1993), and that of ATMA is about 45%." The development of the sewerage system in Accra was initiated during the first project but the sewer connection program progressed slowly, see para. 1.07 above. The Stage II Project was expected to make a special effort to improve this situation. The TA specialist engaged for this particular purpose left after only a few months in the post, and was never replaced. The funds were used for the recruiting of the project manager, see para. 2.15 above. Only recently, the GWSC has initiated activities under AfDB funding to improve the sewerage situation in Accra and develop plans for that purpose. 3.11 The institutional objectives, especially of the TAR Project, included improvements to the operating and financial performance of the GWSC. In the ATMA part of the organization, the staffing levels have improved dramatically over the last two decades. The staff of ATMA has generally been at about 25% of the total staff of the GWSC. Table 3 below shows the progress made in this respect.'2 10 Given in the GWSC's "Corporate Plan, 1993 - 1996." 11 The number of connections in ATMA is about 54% of the total number of consumer connections in GWSC's operations. 12 The ATMA staffing and connection data for the early years is collated from various documents which may not be totally consistent but indicate the trends. The latest figures for 1993 are from the most recent Corporate Plan. 14 Table 3. GWSC-ATMA: STAFFING LEVELS Year Total Water Staff/i 000 Personnel Costs/Total Staff Connections Connections Operating Costs (%) 1968 1,100 17,000 65 n/a 1976 1,300 40,000 33 n/a 1993 1,100 85,000 13 45 The current staff ratio of 13 per 1000 water connections is below the average level for Asian and Pacific cities" but is still well above ratios common in industrialized countries, say 2.9 for USA (average), 3.2 for Brussels and 4.5 for C. Banlieue, Paris. According to the GWSC's latest Corporate Plan, personnel costs are about 45% of the total operating expenses. This is clearly above some ratios collected from industrialized countries (say 30%), but fares well among ratios collected from some developing countries (say 60%). GWSC has clearly improved the efficiency of its operations through the period of these projects, and is expected to continue on this path judging on the basis of its Corporate Plan projections. 3.12 Plans and designs for the TAR Project were prepared under the Stage II Project. Similarly, the consultants provided under the TAR Project were to prepare plans for a follow-on project i.e. the ongoing Sector Rehabilitation Project. This project is implementing many of the institution building tasks left unfinished under the TAR Project The quality of these plans was reported to be satisfactory. D. Operational Results 3.13 The Stage II Project was to start a systematic O&M training by preparing a program for that purpose. No such program was prepared, and the next significant steps were taken only when the twinning TA team provided on-the-job training under the TAR Project. Although the capacity of the GWSC to carry out effective O&M of its facilities was improved, much remains to be done. A few results have already been achieved under the Sector Rehabilitation Project; a comprehensive "manpower improvement program" has been prepared, training facilities at Weija and in Kumasi are in full use, and some outside organizations are being hired to provide specialized training in management and finance areas. 3.14 The PCR on the Stage I Project cites that the unaccounted-for-water (UFW), or non- revenue water, was at the level of nearly 45%. The new estimates in 1978, indicated small improvements as the UFW had fallen slightly below 40%. Serious operational problems on the Kpong pipeline in the early 1980s drew the attention away from improvements to the distribution 13 Infrastructure Notes: Personnel Costs as an Indicator for Water Supply Utility Perfornance in Developing Countries, TWU, W&S No. WS-12, Januay 1994. 15 network that were initiated effectively during the Stage II Project including metering of connections. This event combined with the lack of vehicles for leak detection and repair crews, and the problems in billing and collection (illegal connections) increased the UFW again, even beyond 60% in the 1980s. The Corporate Plan reports that the UFW was 63% in 1988 and has since fallen to its current level of 57%. The target is set to reach the level of 45% at the end of the plan period, i.e. by 1996. The projects aimed at proper metering of the plant outputs, implementation of comprehensive connection metering, and improvements to billing but failed to accomplish such objectives. The Corporate Plan cites these same strategies again for achieving better efficiency and performance in the future; some improvements are being implemented under the ongoing project (Credit 2039-GH). 3.15 The objective of improving water supply to low income segments of the population was expected to be met by increasing the number of public standpipes overall and repairing the existing, malfunctioning standpipes. A special component was designed under the TAR Project to repair 250 existing standpipes and install 100 new ones; all were expected to be metered in order to control the operation of, and consumption through, the standpipes in an appropriate manner. The documentation available to the Audit provides very little information on the actual achievements by the individual projects. The GWSC now states that their objective is to avoid public standpipes in the inner city."' Because of this, some private connection holders in low-income areas have started to operate "kiosk" type of water vending facilities and are charging considerably higher rates than would be the case if the supply was provided by the GWSC. Policies are not clearly spelled out even in the Corporate Plan as to how to meet the demand of the lower income populations in an equitable manner. 3.16 The Stage II Project was to prepare a program for systematic collection of hydrological data in the catchment area of the Weija reservoir, and to arrange for periodic inspections of the dam by a qualified expert. None of this materialized; the lack of funds was stated as the main reason. The TAR Project was again expected to improve this situation. It did so partially as the programs for the dam inspections and raw water quality surveillance were initiated. Full scale hydrological observations are, however, not being carried out. In general, the lack of maintenance of the dam structures is evident. Slight signs of deterioration are already visible. 3.17 The Stage II Project was also to prepare a program for improving the efficiency of the Kpong treatment plant. The study was completed but improvements to the plant's operation are yet to be realized. Similarly, a leak detection program was initiated under the Stage II Project, but needed to be scaled down due to lack of vehicles." This explains why the UFW ratio remains at an unacceptably high level as discussed above. The Kpong pipeline remains the most critical water supply facility in the ATMA area. Much effort has been invested in its repair but the cathodic protection still remains the weak link in the system. The Audit was informed that a comprehensive survey of the whole pipeline is scheduled to be carried out with financial help from AfDB. 14 Public standpipes ar still being used in the adjacent rural village, and the resettlemfent villages (Weija Dam) as a means of water supply to lower incone population. Effective useful life of vehicles is generally low because of poor preventive maintenance. 16 4. BORROWER AND BANK PERFORMANCE A. Borrower Performance 4.01 The GWSC staff were hard pressed in the processing and implementing of these two projects, mainly because of their emergency nature. The situation got even worse in the implementation of the TAR Project during the 1980s; the GWSC was loosing its senior managerial staff as trained and well qualified Ghanaians drifted to the private sector and overseas. Despite this, the performance of the GWSC, as generally demonstrated by the compliance with the covenants, see para. 4.05 below, was reasonably satisfactory. This is partly due to the fact that an expatriate project manager was assisting in technical aspects of project implementation. The main problems that were encountered are discussed below. 4.02 The most serious problem during the Stage H Project were the delays in the implementation of the components financed in parallel i.e. the water treatment plant at Weija and the distribution mains. The GWSC was unable to coordinate the various contracts appropriately thus delaying the realization of the benefits from the increased raw water supply capacity from the newly constructed dam. The tariff increases that were required to produce the covenanted IRR of 7% throughout the project period were not adequate and therefore, the resulting IRR was negative at the time of the credit closing in 1981. ATMA's billing and collection performance also deteriorated badly during the project period. As a result, the GWSC was unable to meet its financing share of the project and the GOG had to cover the difference. 4.03 With regard to the TAR Project, the GWSC's manpower problems were even greater than during the previous project. Thanks to the TA (twinning) team," the GWSC was able to handle the day-to-day operations reasonably well although the team in fact was expected to assist the GWSC in implementing the recommendations of the MIP. Starting from 1985, adequate tariff increases were implemented, and as a result the financial performance was as expected in the legal agreements. There were initially some delays in submitting audited accounts but this was corrected later. Revaluation of assets proved to be a difficult task and had to be delayed to be implemented under the follow-on project. The GOG was slow in providing counterpart funding which hampered the timely completion of some local contracts. There were, however, no other delays in procurement of goods and implementation of major civil work contracts. B. Bank Performance 4.04 The Bank provided effective and timely support in the implementation of both projects, and the Audit was informed that the GWSC was satisfied with the Bank's effort. During the Stage II Project the Bank undertook 13 supervision missions i.e. an interval of less than 6 months; the longest interval between two missions was 12 months. The mission interval in the TAR Project was even less, about 5 months with the longest being 12 months. This frequency and the high quality of the missions (based on a statement by a senior officer of the GWSC) made it possible to achieve 16 Although the arrangement was found satisfactory in general terms, the GWSC felt that they did not have an adequate role in the selection of the members of the twinning team. 17 reasonably positive results under very difficult country conditions. The GWSC made a small point of Bank's "complex" procurement procedures which, in their view, caused some procurement delays. C. Compliance with Covenmts and Directives 4.05 In general, the compliance with key covenants was not a problem. There were, however, some areas under both projects that warrant a special note as follows: a) Stage H Project. * The legal documents of this project included covenants on a number of technical issues that were not resolved prior to Credit approval. These covenants covered issues such as (i) safety of the dam and its periodic inspections, (ii) hydrological surveillance program, (iii) sewerage connection program, (iv) leak-detection and mains laying programs, and (v) Kpong treatment plant improvement program. Although many of the programs were prepared, they led to very little concrete action. These aspects were again included as covenants under the TAR Project, see below. * The usual financial covenants, such as IRR, debt coverage, and audited accounts, were met in the early years of project implementation. Later on the GWSC, however, failed to comply with these covenants; in the case of IRR, the results were always below target, audited accounts were submitted but late, and in the case of the debt coverage, the required level of coverage was achieved throughout the 1970s, but it fell below the required level of 1.5 x the debt service in FY1980. b) TAR Project. * Most of the covenanted technical aspects that were delayed under the previous project and incorporated in this project, were satisfactorily complied with. The sewerage connection program and the plans for Kpong treatment plant improvements, however, never materialized (for details, see discussion in paras. 3.10 and 3.17 above). * The financial covenants on IRR (not less than 3%) and the debt coverage (not less than 80% of operating costs) were not met in the beginning of the project period. GWSC's financial performance improved after 1985 and 1986, respectively, mainly due to adequate tariff increases. It may be noted, however, that the indicators were based on the partially revalued fixed assets only, and may not give a fully true picture of GWSC's financial operations. Assets revaluation is receiving serious consideration by the GWSC today, and the situation is expected to improve in the near future, see para. 2.16 above. 4.06 The Audit focused on the most critical OMSs and OPNs only, and finds no areas of serious non-compliance. It may be, however, worth noting that the cost recovery requirements (OMS 2.25) were not achieved under the Stage II Project. This situation was of course significantly improved under the TAR Project. Similarly, the preparation of the institutional aspects of the TAR Project (OMS 2.28) were not adequately detailed, and the result were less than expected. 18 5. PERFORMANCE ASSESSMENT AND ISSUES A. Assessment of Project Performance Stage I Project. 5.01 The Audit finds that the project achieved all of its physical objectives. Most of the benefits were, however, realized only much later due to delays in the completion of the supplementary components financed by other donors. The results in institutional aspects were minimal as only on- the-job training was provided. The GWSC's financial performance was lower than expected during the project period, but it was able to improve its efficiency later during the follow-on project. On balance, the Audit rates the project's outcome as satisfactory, its institutional development as negligible, and the sustainability of its benefits uncertain. The PCR-based rating of the project outcome is satisfactory." TAR Project. 5.02 This project completed its physical components under very difficult circumstances. The emergency situation that was becoming critical in terms of ensuring reasonably reliable and safe water supply to the city was handled well. The GWSC was also able to prepare the required Management Improvement Program although the implementation of its recommendation fell short of expectations. The most commendable achievement during this project was the marked improvement of the financial performance of the GWSC and its ability to meet the covenanted financial targets through regular tariff increases and improved efficiency. On the above basis, the Audit rates the project's outcome as satisfactory, its institutional development as modest, and the sustainability of its benefits as likely. This agrees with the PCR-based ratings. B. Issues 5.03 The installation of new standpipes and rehabilitation of existing ones were the most important poverty focus aspects of the TAR Project. As discussed in para. 3.15 above, there seems to be some confusion on this issue. The PCR on Cr. 1342-GH mentioned that "a program relating to the improvement of public taps" was prepared as per the respective covenant in the Project Agreement. Yet the GWSC informs the Audit that the aim is to reduce the number of public standpipes, at least in the inner city of Accra. Neither does the Corporate Plan cast any more light on this issue. It is important to establish a sound policy on this aspect to ensure equitable provision of reliable water supply to the lower income segments of Accra. 5.04 Documents prepared in connection with the implementation of these two projects make references to the lack of vehicles, spare parts, tools, and similar items that made it difficult to achieve certain operational improvements such as collection of hydrological data, revaluation of assets, and leak detection and repairs, to mention a few. A serious effort was made in the TAR Project to improve the situation as the funding provision for these critical items was substantially increased during implementation. This, however, was again done without paying enough attention to capacity At the time of this PCR, no ratings were given on institutional development and sustainability of project benefits. 19 building aspects e.g. training in, and facilities for, maintenance of vehicles. Such an effort thus becomes a one time exercise and its benefits are short lived. The GWSC is now contemplating to privatize certain operational activities to make them more cost effective and sustained. 5.05 The construction of the Weija dam involved the resettlement of villages within the impoundment area. The University of Ghana was engaged to prepare necessary studies. Only six villages with some 100 families needed to be resettled. The villages were traditional farming communities with tribally held lands. This led to some land acquisition problems which took several years to resolve. The new villages were constructed in the catchment area so that the villagers did not need to move out of the area they were used to live in. Piped water supply was arranged and the new housing was considered to be an improvement. An unexpected benefit to the villagers has been fishing which they quickly adopted as a new and productive form of livelihood. 6. CONCLUSIONS AND LESSONS A. Conclusions 6.01 These two projects clearly illustrate the time and effort needed in order to achieve sustainable results under difficult country conditions. In this case, the eroding institutional capacity of the implementing sector agency further complicated matters. The Bank was ready to face the challenge and invested the necessary support through frequent and quality supervision to first deal with the emergency construction and rehabilitation tasks. Institutional objectives were too optimistic given the challenge faced in the technical aspects of both projects. Through these two operations the sector agency has gained significant experience and increased its management autonomy and is today in much better health financially. B. Lessons of Experience 6.02 The key lessons learned from these operations include: a) To achieve institutional improvements in projects that are demanding technically, requires careful planning prior to project launch, and concerted effort, firm commitment, and resource allocation during implementation. The best approach would be a gradual program through several operations that entail realistic targets for each step; b) The approach by using twinning as a means of institutional strengthening requires careful design involving both parties concerned in the selection of the staff, -- both staff for providing the services to ensure adequate quality, and local staff for receiving training to ensure sufficient time and commitment. Similarly, the arrangements for the attachment of local staff to the twinning organization need to have specific and achievable objectives; c) To ensure the sustainability of the institutional and operational development efforts, appropriate equipment and manpower provision need to be made on a long-term basis; and 20 d) To ensure the sustainability of improved operational and financial performance, an adequate level of managerial autonomy is required to make crucial decisions on tariffs, manpower development, and investments.