Uganda - Second Water Supply Project

Document of The World Bank FOR OFFICIAL USE ONLY ReportNo: 18728 IMPLEMENTATION COMPLETION REPORT REPUBLIC OF UGANDA SECOND WATER SUPPLY PROJECT (Cr. 2124-UG) December 22, 1998 Urban and Water Unit Eastern and Southern Africa 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 EQUIVALENT Currency Unit = Uganda Shilling (USh) US$1.00 = UShl,200 (July 1998) US$1.00 = USh370 (December 1989) GOVERNMENT FISCAL YEAR July 1 - June 30 ABBREVIATIONS AND ACRONYMS DCA Development Credit Ag:reement EEC European Economic Community GoU Government of Uganda GTZ German Agency for Technical Cooperation PICR Project Implementation Completion Report IDA International Development Agency MoWLE Ministry of Water, Lands and Environment NWSC National Water and Sewerage Corporation PSP Private Sector Participation SAR Staff Appraisal Report TA Technical Assistance TORs Terms of Reference UfW Unaccounted for Water Vice President Callisto E. Madavo Country Director James W. Adams Sector Manager Jeffrey S. Racki Task Team Leader Alain R. Locussol  FOR OFFiCIAL USE ONLY UGANDA SECOND WATER SUPPLY PROJECT (Cr. 2124-UG) IMPLEMENTATION COMPLETION REPORT TABLE OF CONTENTS PR EFA C E ......................................................i EVALUATION SUMMARY ..................................................... ii PART I - PROJECT REVIEW FROM BANK'S PERSPECTIVE ............................................ 1 A. Project Identity ................................. .................... I B. Background .....................................................1I C. Statement and Evaluation of Project Objectives and Components ......................... . 2 D. Achievement of Objectives ......................................................4 E. Major Factors Affecting the Project ......................................................, .6 F. Project Sustainability ......................................................7 G. IDA Perfonnance ...................................................... 7 H. Borrower Performance ......................................................8 I. Assessment of Outcome ........................... .. 8 J. Future Operations ..........................8 K. Key Lessons Learned ........................9 PART II - STATISTICAL ANNEXES .............................. , 11 Table 1 - Summary of Assessments .............................. 11 Table 2 - Related Bank Loans/Credits. .............................. 12 Table 3 - Project Timetable .............................. 12 Table 4 - Loan/Credi t Di sbursements .13.......................................... ........ 13 Table 5 - Key Indicators for Project ImplementatiorL .. 14 Table 6 - Key Indicators for Project Operations .15 Table 7 - Studies Included in Project .......................................... 15 Table 8A - Project Costs .16 Table 8B - Analysis of Five Major IDA-Financed Civil Works Contracts ... . 17 Table 8C - Project Financing .18 Table 9 - Economic Rate of Retur .19 Table 10 - Status of Legal Covenants .20 Table 11 - Compliance with Operational Manual Statements .21 Table 12 - Bank Resources: Staff Inputs .22 Table 13 - Bank Resources: Missions ..........................,,.,.,.,,.,,. 23 ANNEXES A. Final Supervision Report B. Borrower's Contribution to the ICR C. Project Operational Plan Map - IBRD No. 21985 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.  IMPLEMENTATION COMPLETION REPORT UGANDA SECOND WATER SUPPLY PROJECT (Cr. 2124-UG) PREFACE This is the Project Implementation Completion Report (PICR) for the Second Water Supply Project (Cr. 2124-UG) in the Republic of Uganda, for which Credit 2124-UG in the arnount of SDR 45.1 million (US$ 60.0 million equivalent) was approved on April 12, 1990, the Development Credit Agreement was signed on April 24, 1990, and the Credit made effective on January 23, 1991. The Credit was closed on June 30, 1998 after two extensions from the original closing date of June 30, 1996. The Credit was fully disbursed and final disbursement took place on October 14, 1998. The European Economic Community, the Austrian Government, and GTZ of the Federal Republic of Germany provided co-financing for the project totaling US$31.4 million equivalent. This ICR was prepared by Mr. Solomon Alemu, Water Supply Engineer (AFTU 1), with the assistance of Messrs. Alain R. Locussol, Principal Water and Sanitation Specialist (AFTU I) and Tejbir Singh Phool, consultant. The ICR was reviewed by Messrs. Jeffrey Racki, Sector Manager, AFTU I and James W. Adams, Country Director for Uganda and Tanzania. Preparation of the ICR was begun during an IDA supervision mission in February 1998 and a completion mission in July 1998, to which participated Messrs./Mmes.: A. Grudzinska, Financial Analyst (AFTU1), A. Brown and R. Lidonde (RWSG-Nairobi). The PICR is based on materials in the Project file and data obtained during the completion mission. The Borrower contributed to the preparation of the PICR by stating its views as reflected in the mission's Aide Memoire (Appendix A), and by providing an extensive separate evaluation of the Project's preparation and implementation, that is attached to the PICR (Appendix B).  ii UGANDA SECOND WATER SUPPLY PROJECT (Cr. 2124-UG) IMPLEMENTATION COMPLETION REPORT EVALUATION SUMMARY Introduction I. The Second Water Supply Project for which Cr. 2124-UG was approved in 1990 was supposed to be the first phase of a Program prepared in the late 1980s with IDA assistance to improve water supply and sanitation in seven major towns of Uganda, including its capital Kampala. Until the mid 1980s, neglect had resulted in sub-standard performance of the National Water and Sewerage Corporation (NWSC) responsible for the provision of the water supply and sewerage service. Previous Bank group involvement in the sector in Uganda consisted of a Water Supply Engineering Project (Cr. 1 11O-UG approved in 1981) and Water Supply and Sanitation Rehabilitation Project (Cr. 151 0-UG approved in 1984), which were both successfully implemented despite the political strife that plagued the country before the mid 1980s. Project Objectives and Components 2. Project Objectives. The objectives of the Project were to: improve water supply and sewerage services in seven towns; establish delivery mechanisms for subsequent large scale on- site sanitation programs; develop NWSC into a financially viable authority; strengthen NWSC in all its statutory functions; and prepare NWSC for taking over water supply and sanitation operations in other towns. 3. Project Components, Costs and Financing Plan. The Project included three main components: (a) substantial extensions of water supply and waste water systems in Kampala, Mbarara, Masaka and Jinja and more limited improvement of water supply systems in Entebbe, Mbale and Tororo; (b) strengthening of NWSC capacity including technical assistance, training to improve technical, commercial and financial operations and the supply of spares and chemical; and (c) Project coordination and construction supervision. Two of the objectives, i.e., the promotion of on-site sanitation and the extension of NWSC operations to other towns had no specific components or budget. The total cost of the Project estimated at US$118 million at appraisal was to be co-financed by IDA, the EEC, Italy, Austria and GTZ of Germany. However, co-financing for the Jinja and Entebbe components of the Project did not materialize and construction activities where limited to the five other towns. Implementation Experience and Results 4. Project Achievements. All construction activities in the five towns were substantially completed before closing; quality of construction is good. The Project increased production, transmission and storage capacity but did not significantly increase access to the water and sewerage service because of a combination of limited extension of distribution facilities and NWSC's inadequate commercial policy. Also, despite the enactment of the Water Statutes, technical assistance, training, metering and leak detection repair programs supported by the Project, NWSC still exhibits many weaknesses.  5. Project Cost and Timetable. The final Project cost was US$121 million equivalent, excluding unsettled claims, as compared with the appraisal estimate of US$118 million. 6. Project Implementation. The closing date of the Project was extended twice to June 30, 1998 from its original date of June 30, 1996. Limited components were added to the original Project description. Also, the Austrian financing initially of about US$5 million was increased to US$18 million to support the Kampala leak detection and repair program. At closing, whilst the IDA Credit was fully disbursed, GoU and NWSC were still investigating options to finance about US$6 million worth of claims presented by contractors. 7. Key Factors Affecting Implementation of the Project. Outside of Government Control: The Project was implemented during a period of political stability. The depreciation of the Ugandan Shilling vis a vis the US Dollar resulted in high escalation of the local currency portion of construction contracts financed by counterpart funds. Towards the end of its implementation, the Project suffered from adverse climatic conditions related to the El Ninio phenomena. Subject to Government Control: GoU approved tariff increases regularly, but its performance as one of NWSC's major customers was far from being satisfactory; at closing, 40% of NWSC's accounts receivable were with Government agencies. GoU did not enforce an effective monitoring of NWSC and did not act decisively to ensure a steady improvement of its performance indicators. Land acquisition was an obstacle to construction activities because arrangements were not made prior to mobilization of contractors. Delays in construction activities also resulted from late releases of counterpart financing. The introduction of the VAT in July 1996 further aggravated the situation, since no provision in the Project budget was made to finance it. Subject to Implementing Agency Control: NWSC implemented the physical components of the Project, but frequent changes in designs resulted in significant variation orders and price escalation. Also, weak contract management by NWSC and its consultants allowed contractors to present large claims which they justified by changes in construction conditions and late payment of the local portion of the contracts. An assessment of the merits of these clairns could be made by NWSC when construction contracts were completed a few months before closing, at which time the IDA Credit was almost fully disbursed. 8. The Project is unlikely to be sustainable. It did not significantly extend the water supply service in the five towns it covered; realizing its full benefits would require additional investment to increase the capacity of the distribution systems to match that of the production facilities. The Project also did not improve environmental conditions in the towns and receiving bodies; additional investment would also be required to match waste water collection capacity with that of the treatmnent facilities. The Project did not test methodologies to irmplement a large scale on-site sanitation program. The Project did not transform NWSC into a financially viable water utility company and its large investment has not resulted in additional revenues. At closing, NWSC's recurrent and development costs were far too high, its commercial operations were highly inefficient, its commercial policy was inadequate and acted as a deterrent to both existing and new customers and its long term debt was mounting; NWSC is unlikely to be able to fund proper maintenance of facilities built by the Project. The Project did not assist NWSC in extending its operations in a significant number of new towns, which now have to operate their piped water systems by themselves. 9. Overall, IDA performance was unsatisfactory. During preparation and appraisal: During identification, IDA did not identify NWSC's high recurrent and development costs, inadequate commercial policy and inappropriate incentive framework as major issues and proposed to address NWSC inefficiencies simply by providing technical assistarnce and training. Although  iv two of the key objectives, i.e., extending the water supply and sanitation service and strengthening NWSC statutory functions, were supported by appropriate components, two objectives were not supported by specific components or budget and the fifth objective of transforming the NWSC into a financially viable entity, was supported by a weak component. Also, detailed designs were not sufficiently prepared before appraisal. During supervision: Even if IDA supervision focused mostly on technical issues, it did not foresee upcoming claims; it also did not collect data on monitoring indicators agreed at appraisal, and despite NWSC's poor performance, it often rated the Project satisfactory on Development Objectives 10. Overall the Borrower's performance was also unsatisfactory. Even if GoU approved tariff increases, it put NWSC in a difficult financial position by regularly defaulting on the timely payment of its water bills and by changing the on-lending conditions of the Austrian loan that co- financed the Project. GoU also did not enforce an effective monitoring of NWSC performance. During preparation, NWSC took an active role in defining the technical assistance component and the implementation arrangements of the Project. But overall contract management was weak, and poor communication with contractors and consultants who supervised construction led to a situation where contractors could not fully be paid from the available budget. Although NWSC generally complied with the financial covenant of contributing to the investment program from cash generated from its operations, it defaulted on other covenants such as the reduction of unaccounted for water and accounts receivable. 11. Project Outcome. Whilst most facilities initially included in the Project description have been built, the Project did not achieve its two main objectives of extending the water supply and sanitation service and of developing the NWSC into a financially viable water supply utility company. The Project did not establish delivery systems for large scale on-site sanitation projects and did not prepare NWSC to take over operations in other towns. The outcome of the Project is assessed as unsatisfactory. 12. Future Operations. At closing, NWSC was in serious financial trouble but needed to preserve the significant investment, human resources and information system developed under the Project. NWSC has prepared an Operation Plan, but to achieve sustainability, GoU would have to implement a series of reforms to place its national water utility company within a framework that provides the necessary incentives to extend the service and reduce the cost of its provision. Comprehensive private sector participation in the delivery of service is likely to be the only option left to salvage the sector. GoU recently agreed to investigate such options as part of a review of its urban water supply policy, and IDA has accepted to finance the necessary studies from an on-going project. Summary of Findings, Future Operations, and Lessons Learned 13. The main lesson learned from the Project are: * the provision of simple technical assistance and training is inadequate to address institutional issues if the operating framework of the water utility company does not provide incentives to improve the service and reduce the cost of its provision; * financial sustainability of a water utility company depends as much on an adequate tariff level, as on the development of distribution capacities to reach new customers, cost control and appropriate tariff structure; and * physical components of projects have to be carefully designed before awarding construction contract and a strong contract management capacity must be developed within the  v implementing agency, especially if the Project is implemented under a limited number of large contracts.  1 UGANDA SECOND WATER SUPPLY PROJECT (Cr. 2124-UG) IMPLEMENTATION COMPLETION REPORT PART I - PROJECT REVIEW FROM BANK'S PERSPECTIVE A. Project Identity Name : Second Water Supply Project Credit Number 2124-UG RVP Unit Africa Country Uganda Sector Water Supply and Sanitation B. Background 1. Uganda is a landlocked country of East Africa, whose total and urban populations were estimated to be about 16.5 million and 1.3 million respectively in the late 1980s. In the late 1990s, these figures have reached 20.5 and 1.7 million. Uganda is well endowed with water resources, and about 20% of its surface is actually taken up by waterways and lakes. Agriculture is the country's main contributor to the GDP. The country encountered prolonged political and economic difficulties until the mid-1980s. 2. Major towns were mostly equipped with piped water and waste water collection and disposal facilities in the 1950s and 1960s. In the 1970s, political and economic instability resulted in decay of the facilities, sub-standard technical, commercial and financial performance of the National Water and Sewerage Corporation (NWSC) responsible for the provision of service, and a general decrease in the quality of service. Since the mid 1980s, substantial involvement from donors such as the African Development Bank, the European Economic Community, Italy, Germany in the urban sector and UNICEF, UNCDF, Canada and Denmark in the rural sector has helped improve the situation somewhat. 3. Previous Bank Involvement. The Bank Group has financed two water supply projects since the early 1980s: (a) a Water Supply Engineering Project (Cr. 11 10-UG approved in 1981); and (b) the Water Supply and Sanitation Rehabilitation Project (Cr. 151 0-UG approved in 1984), aimed at rehabilitating badly deteriorated water and sanitation infrastructure in Kampala, Jinja, Entebbe, Tororo, Mbale, Masaka, and Mbarara. Despite the political strife that plagued the country until the mid 1980s, both projects were successfully implemented. But in the late 1980s, the service coverage was still low in these seven towns, and NWSC performance was still poor. The main lessons learned from these two projects were the need for implementing frequent tariff increases to cope with high inflation and the need for developing adequate procedures for payment to contractors. 4. Urban Water Supply and Sanitation Program. In the late 1980s, the Govemment of Uganda (GoU) prepared with IDA assistance, an investment "Program" for the above seven towns, estimated to cost US$194 million. IDA agreed to support the Second Water Supply Project (WSP-11) as a first  2 phase of this Program. IDA also agreed to finance in 1991 the First Urban Project (Cr. 2206-UG) to develop other urban infrastructures and services. C. Statement and Evaluation of Project Objectives and Components 5. Project Objectives. As stated in the SAR, the main objectives of the WSP-II, were to: (a) improve public health, labor productivity, and economic production potential; (b) alleviate the traditional burden on women in the provision of water; and (c) promote sound environmental and institutional policies by expanding clean water supply and waste water treatment, by building strong institutions and by introducing effective legislation. The Project's specific objectives were to: * follow-up on previous activities by providing acceptable water supply services to a larger proportion of the seven towns' population and by providing sewerage for areas that cannot feasibly be covered by other systems; * establish methodologies and delivery mechanisms for subsequent larger scale implementation of on-site sanitation; - develop NWSC into a financially viable authority through the introduction of appropriate tariff levels and structures and implementation of billing, revenue collection and accounting systems and procedures; * strengthen NWSC in all its statutory functions through continued training and technical assistance that will gradually be phased out; and - prepare NWSC for also taking over other urban areas. 6. Project Components. To achieve the above objectives, the following three main components were designed: (a) Main Civil Works * Kampala water supply: construction of a new production facility (Gaba II - capacity 80,000 m3 /day), 100 km of 100-500 mm transmission and distribution mains, 20,000 m3 of storage capacity, and rehabilitation of two booster stations; * Kampala sewerage: reinforcement and rehabilitation of 35 km of existing sewers, rehabilitation of the Bugolobi waste water treatment plant and construction of stabilization ponds; * Jinja water supply and sewerage: extension of the existing water production, transmission and distribution capacity to reach about 50,000 m3/day; replacement of existing sewers, construction of new sewers, rehabilitation of existing waste water stabilization ponds, and construction of on- site sanitation facilities in low income areas: * Mbarara water supply: construction of a river intake, a 7,500 m3 /day treatment plant, 33 km of transmission and distribution mains, 4,100 m3 of storage capac:ity, booster stations and office and residential buildings; * Mbale water supply: improvement of the existing intake and rising main; * Masaka water supply and sewerage: construction of a river intalce, a 5,000 m3 /day treatment plant, 24 km of transmission and distribution mains, 3,700 m3 of storage capacity, booster stations; construction of 2.5 km sewers, waste water stabilization ponds, office and residential buildings; * Entebbe water supply and sewerage: increase of the water production and distribution capacity to reach 12,000 m3/day and extension of the waste water collection and treatment capacity; and * Tororo water supply and sewerage: improvement of the water distribution system and extension of the existing sewerage network.  3 (b) Organizational Strengthening and Minor Works and Supplies * provision of technical assistance and training to improve NWSC's financial, commercial and technical operations, water quality control, block mapping, leak detection and repair, sewer cleansing in Kampala; and * provision of meters and essential spares, chemical and materials. (c) Project Supervision and Coordination * support to the Project Coordination Unit (PIU); and * technical assistance for construction supervision. 7. Project Cost and Financing Plan. The total Project cost was estimated at US$118 million, including physical and price contingencies, local taxes, and duties, of which US$101 million (86%) in foreign exchange. This amount was to be financed as follows: * the EEC approved US$23.8 million for the construction of the Gaba II treatment plant and associated works, water quality control, and technical assistance to NWSC; * the Govemment of Italy initially committed US$14.5 million to finance the Jinja water supply and sewerage component: * the Government of Austria initially approved US$4.7 million to finance the Kampala leak detection and repair program, but ended up providing altogether US$18.2 million equivalent; * GTZ of Germany agreed to contribute US$2.9 million to finance the block mapping and survey exercise, the computerization of billing and substantial technical assistance to NWSC; * IDA was the single largest contributor and approved US$60 million for the construction of water supply and sewerage facilities in Kampala, Mbarara, Mbale, Masaka and Tororo, the training of NWSC staff; IDA also contributed to the computerization of billing, water quality control and block mapping exercises, the cleaning of Kampala sewerage system, the supply and installation of meters, and the supply of chemicals and spares; * external financing for the Entebbe water supply and sewerage component could not be identified at the time of effectiveness of the IDA Credit; and * GoU and NWSC agreed to provide US$11.9 million as counterpart financing to cover local expenditures expenses and taxes. 8. Additional Components. Because the appreciation of the SDR vis a vis the US Dollar resulted in additional resources, IDA later agreed to finance: (a) in 1994, the rehabilitation of four post-stressed concrete tanks (three in Kampala and one in Jinja); (b) in 1996, the expansion of water supply system to the south industrial area of Kampala (16 km of distribution and transmission main, a booster station and 570 m3 capacity elevated tanks); and (c) in 1997, a standpost program in Masaka and Mbarara to expand coverage to unserved areas. 9. Project Implementation Timetable. The Project was to be implemented in six years between 1990 and 1996, with an initial closing date of June 30, 1996. Due to delays in start-up of the construction activities, the closing date of the Credit was extended twice to June 30, 1998. 10. Evaluation of Project Objectives and Design. In view of the institutional and managerial weaknesses identified at appraisal and the low access to the water supply and sanitation services in the seven towns, the overall objective of building the necessary institutional and legal capacity for improving and expanding clean water and sanitation service was appropriate. Two of the five  4 objectives, i.e., the development of a framework for large scale on-site sanitation program and the preparation of NWSC to take over operations in other towns had neither clearly defined components nor allocated budgets. 11. The design of the physical systems was heavily biased towards the extension of water production capacities and not sufficiently targeted on reaching unserved consumers through expansion of distribution networks. The same applies, to a lesser extent, to the extension of waste water treatment capacities and sewage collection capacities. The design of the institutional support was restricted to the provision of technical assistance and training for reducing leakage, improving metering, billing and collection. The objective of developing the NWSC into a financially viable utility company was translated primarily into the need for regular tariff increases, reduction of unaccounted for water and accounts receivable. D. Achievement of Objectives 12. Overall Assessment. All constnrction activities of the original Project description for which co-financing was secured at approval, as well as the components which were added later, were substantially completed before closing. Quality of construction is good. At the same time, the Project did not significantly increase access to water supply and sewerage or to improve NWSC's performance. 13. Objective 1: Improve and Expand Water Supp4' and Sewerage Services in the Seven Towns. The Project resulted in good quality construction, but water has yet to reach all beneficiaries. The Project initially included construction activities in seven towns, but because the financing for the Entebbe component was not identified when the IDA Credit became effective and the commitment of the Italian Government to finance the Jinja component did not materialize, substantial construction activities were limited to Kampala, Masaka and Mbarara. More limited construction took place in Mbale and Tororo. A Jinja water supply and sewerage component was later included in the IDA financed Small Towns Water and Sanitation Project (Cr. 2583-UG), currently under implementation. 14. The final Project cost is US$121 million equivalent, excluding US$6 million worth of claims that were still unsettled at closing, as compared with US$118 at appraisal. Despite a total of US$95 million invested in construction activities, the Project achieved only limited additional coverage mostly because of the limited extensions of the distribution networks it financed, and of NWSC's inadequate commercial policy (para. 17). For example, in Masaka and Mbarara, where about US$25 million were invested, only about 600 new water connections, serving no more than an additional 10,000 people, were built between June 1995 and June 1997. A standpost expansion program was introduced in these two towns in 1997, but at closing it was not possible to evaluate its effect. In Masaka and Mbarara, the number of sewerage connections stagnated at a total of 500. Similarly in Kampala where the Project invested US$16 million for water supply and US$9 million for waste water, only 2,650 new water connections serving an additional 40,000 people, were built between June 1995 and June 1997 and the number of active sewerage connections actually decreased. 15. As of March 1998, NWSC had a total of 49,250 registered water connections in the 11 towns served, but only about 31,500 were classified as "active". Altogether, 35% of the connections were "inactive", following the interruption of service for non-payment. 'Thus, it is likely that, based on an average rate of 15 persons per connection, no more than 30% of the total population in the 11 towns, estimated at about 1.5 million, have direct access to piped water. In addition there were a total of about 900 public standposts possibly serving an additional 25% of tie population (based on an average population per standpost of 500). NWSC also serves 10,700 sewerage connections.  5 16. Objective 2: Establish Methodologies and Delivery Systems for Subsequent Larger Scale Implementation of On-Site Sanitation. The SAR did not include specific components and budget allocation to achieve this objective. On-site sanitation issues were therefore not addressed at all during the Project. 17. Objective 3: Develop NWSC into a Financially Viable Authority. Activities such as leakage detection and repairs, block mapping, bulk and individual metering, and improvement of billing and collection procedures were identified to help transform NWSC into a financially viable utility company. At closing, and despite the heavy technical assistance, training, metering, and leak detection programs supported by the Project, NWSC still exhibits many weaknesses. NWSC: * employs a total of 1,870 staff: at about 40 staff per 1,000 connection, NWSC's performance is way behind that of good African water utility companies (10 to 12 staff/l,000 connections, with the best performance at 4 staff/1,000 connections); * bills only about 45% of the water it produces, up from 35% at the beginning of the Project: at about 55%, NWSC's unaccounted for water (UfW) is significantly higher than the 25% achieved by good African water companies (15% for the best performing one); and * has seen its accounts receivable increase seven folds in volume during the last seven years: as of March 1998, they reached 16 months of billing; while 40% of the arrears are with Government agencies, the volume of private arrears increased by 65% between June 1994 and March 1998. 18. Since the Project did not succeed in improving commercial operations, NWSC hired in 1998 a consulting firm to implement the Kampala Revenue Improvement Program (KRIP), designed to overhaul billing and collection procedures. The contract, awarded on a sole source basis and financed exclusively from NWSC operating revenues, includes a bonus if the agreed minimum collection performance is exceeded. The KRIP is implemented by a large team of expatriate staff and adds significantly to NWSC already high operating costs. 19. NWSC tariff was regularly increased between 1989 and 1994, but not since; at closing, it was about 2.5 times that of 1989, in real terms. NWSC now charges one of the highest tariff in Africa, averaging US$ 1.0/m3 equivalent. The tariff structure is still complex and includes cross- subsidies among categories of consumers but has no protection of small domestic consumers. At an average US$300, the fee for a new connection is obviously unaffordable to most households; at US$40, the reconnection fee is also too high to reactivate the 35% connections that have been cut-off for non- payment. In 1998, NWSC completed a tariff study aimed at justifying a 25% increase, limiting distortions created by the current structure, protecting small domestic customers, and rescheduling NWSC's growing long term debt. 20. Objective 4: Strengthen NWSC in all its Statutory Functions. The Water Statutes and the NWSC Statutes prepared in 1995 under the Project were eventually passed into laws, after much delay in preparation, review, and approval. 21. Objective S: Prepare NWSC to take over operations in other Urban Areas. The SAR did not include specific components and budget allocation to achieve this objective, which therefore was not achieved. On the contrary, GoU adopted since approval of the Project, a decentralization policy aimed at transferring responsibility for operation and maintenance of small piped systems to local govemments and that of point systems to users groups. This policy is supported by several donors, including IDA under the Small Towns Water Supply Project (Cr. 2583-UG). At closing, about 15 towns had a small piped water supply system either in operation or under construction. A review by  6 various donors of the technical, commercial and financial performance of these decentralized operations is planned for 1999. 22. Environmental Assessment and Action Plan. The environmental assessment carried out before appraisal focused on the proper disposal of additional quantities of waste water generated by improved water supplies. The Project was successful in improving the water legislation and water quality monitoring procedures. Also, the NWSC was able to complete the rehabilitation and equipment of water quality laboratories in the Project towns. Consequently, the monitoring of drinking water quality at head works, in the distribution networks and of effluent at waste water treatment plants is being carried out systematically. But very little was achieved by the Project in extending the sewerage service; for example, at closing almost no waste water reached the stabilization ponds built in Masaka. Also, the extension of the stabilization ponds in Kampala was postponed because little additional waste water was generated. Operational problems plague sewerage systems rehabilitated or constructed under the Project, anid quality of effluents leaves much to be desired. 23. Land Acquisition and Resettlement. Evidence that the resettlement process had been carried out in compliance with Bank guidelines was produced during negotiations for the Gaba waterworks site in Kampala. Land acquisition contributed to delays at most construction sites in particular in Masaka and Mbarara, but resettlement was carried out according to Bank guidelines. 24. Studies financed by the Project. The Project financed the preparation of the Water Statutes and NWSC Statutes which helped clarify the legal environment of the water supply and waste water sector. Late in its implementation, the Project also financed the preparation of a water tariff study. Although it presents some shortcomings, its recommendations should help NWSC improve its pricing policy. 25. Economic Rate ofReturn. The ERR of the Project was recalculated to be negative, as compared with 12.5% in the SAR (Table 9); calculations carried out at appraisal and at completion ignore the consumer surplus and externalities and thus underestimate actual benefits. The negative ERR estimated at completion can mostly be explained by higher investment and the poor commercial performance of NWSC. The calculation carried out at completion ignores potential benefits of the KRIP, an activity that was not part of the Project. E. Major Factors Affecting the Project 26. Factors Outside of Government Control. The Project was implemented during a period of political stability. The significant depreciation of the Ugandan Shilling vis a vis the US Dollar since the-early 1990s resulted in higher than expected price escalation of the local currency portion of the construction contracts, financed by counterpart funds. Towards the end of its implementation, the Project suffered from adverse climatic conditions related to the El Nino phenomena. The virtual destruction of the Mombasa highway in Kenya, through which most goods are imported to Uganda resulted in late deliveries of material and equipment and additional costs. Also, the newly constructed intake in Mbale was silted up and the river bank protection works were destroyed by torrential floods and had to be rebuilt. 27. Factors Subject to Government Control. GoU approved water tariff increases regularly, at least until 1994, but its performance as one of NWSC's major customners was far from being satisfactory; at closing, 40% of NWSC's accounts receivable were with Government agencies. GoU did not enforce an effective monitoring of the NWSC and did not act decisively to ensure a steady  7 improvement of its performance indicators. Land acquisition was a constant obstacle to construction activities because arrangements were not made prior to mobilization of contractors; disagreement on compensation resulted in construction delays and also less than satisfactory protection perimeters on source works or access to facilities; local governmnents did not often assist NWSC in solving its problems. Although GoU/NWSC actual contribution to project cost compares with that projected at appraisal (about US$12 million), late releases in counterpart financing resulted in delays in construction activities and claims for interest on late payments. The introduction of the VAT in July 1996 further aggravated the situation, since no adequate provision was made in the Project budget to finance it. 28. Factors Subject to ImplementingAgency Control. NWSC implemented the physical components of the Project, albeit with delays and cost overruns; quality of construction is good. The Project was appraised on the basis of designs prepared in the late 1980s and adjustment to detailed designs were not made prior to tendering in 1993; frequent changes in designs and delays in submitting them to contractors resulted in significant variation orders and price escalation. Weak contract management by NWSC and its consultants also allowed the few contractors who implemented the Project to present large claims that they justified by changes in construction conditions. An assessment of the merits of these claims could be made by NWSC a few months before closing, when construction contracts were completed; at that time the IDA Credit was almost fully disbursed. At closing, NWSC and GoU were still investigating options to finance about US$6 million worth of unsettled invoices to four contractors for claims deemed legitimate. F. ProJect Sustainability 29. The Project is unlikely to be sustainable for the following reasons. * the Project did not significantly extend the water supply service; realizing its full benefit would require additional investment to increase the capacity of the distribution systems to match that of the production facilities; - the Project did not improve environmental conditions in the towns and receiving bodies; additional investment would also be required to match waste water collection capacity with that of the treatment facilities; also, methodologies to implement large scale on-site sanitation were not tested; * the Project did not transform NWSC into a financially viable water utility company and its large investment did not result in additional revenues; at closing, NWSC's recurrent and development costs were far too high, its commercial operations were highly inefficient, its commercial policy was inadequate and acted as a deterrent to both existing and new customers, and its long term debt was mounting; it is unlikely that NWSC will be able to fund proper maintenance of facilities built by the Project; and * NWSC has not been able to extend its operations in a significant number of new towns, which now have to operate their piped water systems by themselves. G. IDA Performance 30. Overall, IDA performance was unsatisfactory. During identification, IDA failed to identify NWSC's high recurrent and development costs, inadequate commercial policy and inappropriate incentive framnework as major issues and proposed to address NWSC's inefficiencies simply by providing technical assistance and training. IDA performance was also unsatisfactory during preparation and appraisal. Two of the key objectives, i.e., extending the water supply and sanitation  8 service and strengthening NWSC's statutory functions, were supported by appropriate components. Two other objectives, i.e., promoting on-site sanitation and extending NWSC's operations to other towns, were not supported by specific components or budget. The fifth objective of transforming NWSC into a financially viable water utility company was supported by a weak component. Also, detailed designs were not sufficiently prepared before appraisal. The SAR is also often confusing, referring to both a medium term development Program and its first phase Project, without providing a clear description of the Project components. IDA performance during supervision was also unsatisfactory. Even if supervision focused mostly on technical issues, it failed to foresee the large claims that were still not financed at closing. The supervision did not collect data on the monitoring indicators agreed at appraisal, and despite NWSC's poor performance often rated the Project satisfactory on Development Objectives; as a result, supel vision dici not propose and help implement timely corrective actions. H. Borrower Performance 31. Overall the Borrower's performance was also unsatisfactory. Though GoU approved tariff increases, it put NWSC in a difficult financial position by defaulting on the timely payment of its water bills and by changing the on-lending conditions of the Austrian loan that co-financed the Project. GoU did not play an active supervisory role of NWSC and did not act decisively to ensure a steady improvement of its performance indicators. 32. During preparation, NWSC took an active role in defining the technical assistance component and the implementation arrangements of the Project. But overall contract management was weak, and poor communication with the contractors and the consultants who supervised construction led to a situation where contractors could not be fully paid from the available budget. Also poor coordination between the NWSC and local authorities on land acquisition issues iresulted in construction delays. Although NWSC generally complied with the financial covenant of contributing 20% of the investment program from cash generated from its operations, it defaulted on other covenants such as reduction of unaccounted for water and accounts receivable. I. Assessment of Outcome 33. Whilst most facilities initially included in the Project description have been built, the Project did not achieve its two main objectives of extending the water supply and sanitation service and developing the NWSC into a financially viable water supply utility company. It also did not establish delivery systems for large scale on-site sanitation projects and did not prepare NWSC in taking over operations in other towns. The outcome of the Project is assessed as: unsatisfactory. J. Future Operations 34. At closing, NWSC was in serious financial trouble: its customier base was shlinking, revenues were stagnating, already high operating costs were further increased by the KRIP, arrears were mounting, and the debt service was significantly inflated by the new on-lending conditions imposed by the Ministry of Finance for the Austrian loan that co-financed the Project. At the same time, NWSC needed to preserve the large investment, human resources and information system developed under the Project. NWSC has prepared a draft Operational Plan, which is attached as Appendix C. But to achieve sustainability, GoU which is concerned with the overall performance of the sector, would have to implement a series of reforms to place its national water utility company within a  9 framework that would provide the necessary incentives to rapidly extend the service and reduce the cost of its provision. 35. Based on experience in African countries that had to address similar problems, comprehensive private sector participation in the delivery of service is likely to be the only option left to salvage the sector. While a "lease contract" is prima facie the minimum option to be sought, it may be possible to rapidly develop a "concession arrangement" whereby the responsibility for selecting and financing new projects would be transferred to a private operator. As indicated in the Operation Plan, GoU has recently agreed to investigate such options as part of a review of its urban water supply policy. IDA has agreed to finance the corresponding studies from Cr. 2583-UG (Small Towns Water Supply Project). Recommendations should be available in 1999. K. Key Lessons Learned 36. Institutional Development. The main lesson is that the provision of simple technical assistance and training is inadequate to address institutional issues if the water utility company operating framework does not provide incentives to improve the service and reduce the cost of its provision. Other lessons are: - carefully monitoring a water utility company performance against technical, commercial, and financial indicators, both at appraisal and during supervision is necessary to identify issues to be addressed; and * addressing major institutional deficiencies needs a comprehensive program; a piecemeal approaches resulting from an incomplete analysis may well have no effect on the bottom line. 37. Financial Development. The main lesson learned is that financial sustainability of a water utility company depends as much on an adequate tariff level, as on the development of distribution capacities to reach new customers, cost control, and appropriate tariff structure. Components of this lesson are: * the tariff structure should be designed to allow access of lower income groups to the water supply service; high connection fee, high re-connection penalties and absence of protection of minimum consumption act as deterrent to both new and existing customers; * IDA must carry out, at least once a year, a full supervision of the financial performance of the water utility company it lends to, in order to review evolution of key performance indicators, prepare short-term financial forecast and propose action plans; and * financial covenants linked to cash generation are appropriate for utility companies with poor collection performance, as evidenced by NWSC compliance with such covenant. 38. Physical Implementation. The main lessons learned are that physical components of projects have to be carefully designed before awarding construction contracts and that a strong contract management capacity must be developed within the implementing agency. Components of these lessons are: * detailed definition of physical components to be implemented is necessary before contract award to avoid delays and cost overruns; * advanced land acquisition and timely compensation are key to avoid delays and cost overruns;  10