Malawi - Second Lilongwe Water Supply Project

Document of The World Bank FOR OFFICLAL USE ONLY Report No. 13744 PROJECT COMPLETION REPORT MALAWI SECOND LILONGWE WATER SUPPLY PROJECT (CREDIT 1742-MAI) DECEMBER 2, 1994 Infrastructure Operations Division Southern Africa Department Africa Region This document has a restricted distribution and may be used bv recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EOUIVALENTS Currency Unit = Malawi Kwacha (MK) US$1.00 = MK 1.80 (July 1986) US$1.00 = MK 6.52 (March 1994) MK 1.00 = 100Tambala SDR = US$1.20 WEIGHTS. VOLUMES AND MEASURES 1 Cubic Meter (m3) = 220 Imperial Gallons, 264.2 US Gallons, or I Kiloliter: (ki) 1 Liter per Capital per Day = lcd IMeter = 3.28 Feet (39.37 inches) 1 Kilometer = 0.62 Mile I Square Kilometer (Km2) = 247 Acres (ac) ABBREVIATION AND ACRONYMS DLV = Department of Lands and Valuation DOW = Department of Lands and Water EIB = European Investment LDVW = Department of Lands, Valuations and Water LWB = Lilongwe Water Board MMWS = Ministry of Works and Supplies (now MOW - Ministry of Works) ODA = Overseas Development Administiation OPC = Office of the President and Cabinet PCR = Project Completion Report SIDA = Swedish International Development Authority WHO = World Health Organization FISCAL YEAR Government of Malawi & Lilongwe Water Board = April 1 to March 31  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation December 2, 1994 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Malawi Second Lilongwe Water Supply Project (Loan 1742-MAI Attached is the Project Completion Report on Malawi - Second Lilongwe Water Supply Project (Loan 1742) prepared by the Africa Regional Office, with Part II contributed by the Borrower. Largely prepared under the previous loan the project focused primarily on physical capacity expansion for water supply improvements in Malawi's capital, Lilongwe. Secondary objectives included management strengthening, and inputs for present and future sanitation improvements. The project met its objective in physical terms and provided economic benefits earlier than planned. Rapid in-migration associated with moving Malawi's capital to Lilongwe has spurred population growth and capacity additions financed by the project have already been utilized. Training and reorganization took place in the utility. Financial results have not matched appraisal expectations due to continuing government delays in tariff increases and greater-than-expected increases in operating expenses. The project outcome is rated as marginally satisfactory, and its inst,tutional development as modest. The sustainability of project benefits is uncertain because the utility is not recovering its capital costs. The PCR is of satisfactory quality. The project may be audited. Attachment 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 PROJECT COMPLETION REPORT MALAWI - SECOND LILONGWE WATER SUPPLY PROJECT CREDIT 1742-MAI TABLE OF CONTENTS Eage No. PREFACE . ................................................... i EVALUATION SUMM4ARY ....... .......... .. .................... H PART I: PROJECT REVIEW FROM ASSOCIATION'S PERSPECTIVE ....1..... A. Project Identity .......................1...................... B. Project Background ....... ............ ..1..................... C. Sector Development Objectives and Policy Context ...................... 1 D. Project Objectives and Description ................................. 2 E. Project Foundation and Design. 2 F. Project Implementation ........................................ 4 G. Major Results of Project ...................... 6 H. Project Sustainability ....................... 9 I. Association Performance ........................................ 9 J. Borrower and Executing Agency Performance ............... ........ 10 K. Project Relationship ......................................... 10 L. Consulting Services . ......................................... 10 M. Project Documentation and Data .................11................. 1 PART II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE .... ...... I I PART m: STATISTICAL INFORMATION ............................ 15 ANNEXES 1. Related Bank Loans and/or Credit ................................ 15 2. Project Timetable . .......................................... 16 3. Credit Disbursement ........................................ 17 4. Project Implementation ...................................... 18 5. Project Costs . ............................................ 19 6. Project Financing . ......................................... 20 7. Economic Impact . .......................................... 21 8. Financial Impact . .......................................... 22 9. Economic Rate of Return Data Streams ............................. 23 I This document has a restricted distribution and may be used by recipients only in the performance of their official dutis. Its contents may not otherwise be disclosed without World Bank authorization. I  Table of Contents (Continued) Page No. 10. Short and Long Term nstitutional Development ........................ 24 1 1. Short-Term Practical Experience Program Through Twinning ................ 25 12. Status of Covenants ........................................... 27 13. Use of Bank Resources . ...................................... 29  - l - PROJECT COMPLETION REPORT MALAWI - SECOND LILONGWE WATER SUPPLY PROJECT CREDIT 1742-MAI PREFACE 1. This is the Project Completion Report (PCR) on the Second Lilongwe Water Supply Project, for which Credit 1742-MAI in the amount of SDR 16.6 million (US$20.0 million equivalent) was approved on November 20, 1986 and declared effective on May 12, 1987. The Credit was closed, as scheduled, on June 30, 1993. The undisbursed balance of SDR 93,008.56 was canceled on November 15, 1993. 2. The European Investment Bank (EIB) and the Overseas Development Administration of the United Kingdom (ODA) financed the project in parallel with IDA and the Government of Malawi. An amount of US$19.37 million equivalent of IDA's credit was on-lent to Lilongwe Water Board (LWB) at an interest rate of 11 percent for 25 years, with a 6-year grace period and the remaining US$0.63 equivalent was passed on to LWB as grant. Of the US$0.67 million equivalent of the ODA grant to the Government, 60% was passed on to LWB as grant and 40% was on-lent on terms similar to those of the IDA Credit. The EIB contribution was on-lent to LWB at 11 9% interest rate. 3. The PCR was jointly prepared by the Infrastructure Operations Division of the Africa Region (Parts I and Im) and the Borrower, (Part 11). 4. Preparation of the PCR was started during the Bank's final supervision mission in July 1993, and is based on data provided by the Lilongwe Water Board, the Staff Appraisal Report, the Development Credit Agreement, supervision reports, correspondence between IDA and the Borrower, and internal Bank memoranda.   - ii - PROJECT COMPLETION PROJECT MALAWI - SECOND LILONGWE WATER SUPPLY PROJECT CREDIT 1742-MAI EVALUATION SUMMARY Objective 1. The principal objectives of the project were to: (a) augment and secure the water supply to Lilongwe; (b) strengthen Lilongwe Water Board's management through management staff support; (c) continue with the construction of the on-site sanitation program started under a preceding Engineering Credit (Cr. 1272-MAI); and (d) study future options for Lilongwe's waste water disposal and sanitation systems. Implementation Experience 2. The Second Lilongwe Water Supply Project was a follow-on from the Lilongwe Water Supply Engineering Project which assisted LWB: (i) to formulate long range water supply, sewerage, and sanitation programs; and (ii) assess the manpower and training needs, including recommendations for a training program. The project, which covered the first phase of LWB's long-term investment program, was identified from the Master Plan study prepared under the engineering project. The feasibility studies, detailed design, and contract documents were prepared under the Engineering Project. The advance preparation of project design and procurement, and contract documents contributed to the timely execution of the project components. Furthermore, the formulation of the long range water supply, sewerage, and sanitation programs, under the engineering project, give evidence of a well-conceived approach to project financing. The initial implementation schedule, i. e. a six year implementation period, reflected the Bank's concern regarding Government's ability to fund the investment as proposed initially. However in early 1988, as a result of availability of funds and the anticipated cost savings by a shorter implementation period, it was agreed that the project could be completed earlier. Therefore, implementation of all components was completed one year earlier. Although some difficulties and setbacks were experienced, project implementation had progressed smoothly. Results 3. The project was successful in meeting its stated objectives (para. 1). The above objectives were achieved through the: (a) construction of a new earthfill dam; (b) expansion of the water treatment plant; (c) extension of the water transmission and distribution systems; and (d) management staff support and training. 4. The project also provided financing for the preparation of a feasibility study for Lilongwe's  - iii - sanitation and sewerage and for implementation of on-site sanitation systems. The feasibility study examined: (a) existing and prospective urban development in Lilongwe to the year 2005; and (b) sanitation and water supply systems. Also, alternative sanitation projects for waste water treatment and disposal were evaluated and emergency programs with priority components to be built for immediate rehabilitation of the existing sanitation facilities were identified and designed. The On-Site Sanitation Program strengthened the management and implementation capacities of Lilongwe, Blantyre, Mzuzu, and Zomba City Councils. Sustainability 5. The major achievements under the Project were an adequate water supply for Lilongwe to meet the anticipated demand and an effective and efficient institution that is financially self-sufficient. While Government agreed to periodic requests throughout project implementation, a more automatic mechanism for periodic adjustments of tariffs is required to guarantee financial viability. Furthermore, to further improve and sustain what was achieved under the project, institutional development effort needs to be continued. These issues are being addressed by a proposed National Water Development Project, currently under preparation. Findings and Lessons Learned 6. The main lessons learnt from the preparation and implementation of the Second Lilongwe Water Supply Project are: (a) The value of a rigorous assessment of institutional capacity and the design of appropriate arrangements for technical assistance and training. Under the project, this was reflected in: 'i) The appointment of able senior management staff who have the experience and the training to build and strengthen the capacity of the implementing agencies is very essential. Under the preceding project, a technical assistance component aimed at improving management was included and was executed successfully. The technical assistance component continued under this project allowed for experienced expatriate engineers to occupy the Engineer/Manager, Financial Controller, and divisional engineering posts of distribution, operations, and projects and for relatively inexperienced local engineers to be appointed to deputy divisional engineering posts. The program provided for the gradual increase in responsibility to be given to the deputy engineers and for them to take over the engineers' posts before completion of the project, which they did. The location program was, therefore, executed successfully; and (ii) A thorough assessment of staffing, recruitment and training needs. Under the Engineering Credit an assessment of staffing, recruitment and training needs were made and under the Second Lilongwe Water Supply Project, the program was well executed. The strength of LWB's management and the improvement in the financial viability is the result of the above efforts. (b) Determination of the basis for a successful resettlement and compensation program at appraisal is essential. The need and basis for resettlement and compensation was not adequately defined during preparation/appraisal of the project and, therefore, the net effect of its impact on households could not be assessed.  - iv - (c) Government commitment is essential in order to develop a self-supporting organization. While Govermment was somewhat supportive of LWB through tariff increase approvals, the increases were often too little, too late. Government will therefore need to examine its role and procedures for the future approval of tariff increases. (d) Flexibility in changing targeted covenants to make them consistent with implementation experience is important. For example, although LWB's financial performance was generally sound, it failed to meet the covenanted rates of return. This issue was raised with LWB by each supervision mission, but no further action was taken by the Bank. In retrospect, the rates of return on the average current net value of LWB's fixed assets in operation (operating income/one-half of the sum of the current net value of LWB's fixed assets in operation at the beginning and at the end of that fiscal year) should have been renegotiated at the completion of the revaluation of the fixed assets as the actual value of the revalued net fixed assets were much higher than the SAR estimates. (e) Early preparation of key requirements for effective implementation. The advance preparation of design and procurement documents, as well as draft contracts avoided potential bottlenecks and resulted in project completion within the original estimated time frame.   PROJECT COMPLETION REPORT MALAWI - SECOND LILONGWE WATER SUPPLY PROJECT CREDIT 1742-MAI PART 1: PROJECT REVIEW FROM ASSOCIATION'S PERSPECTIVE A. Project Identity Project Name Second Lilongwe Water Supply Project Credit No. 1742-MAI RVP Unit Southern Africa Department Infrastructure Operations Division Country Malawi Sector Infrastructure Sub-sector Water B. Project Background 1. The transfer of the administrative capital from Zomba to Lilongwe made Lilongwe a focal point for agricultural, commercial, and industrial activities. As a result of this transfer, Lilongwe experienced a very rapid population growth; the population grew from 169,000 in 1986 to about 360,000 in 1992, and it is now expected to continue to grow at 8 percent (6.5 percent per annum was estimated during appraisal). This rapid increase in population put a strain on the City's water supply in the early 1980s which resulted in water restrictions and rationing. Since the tributaries flowing into the Lilongwe River, the sole source of water to the City, had been considerably reduced and occasionally ceased during the dry season, the Kamuzu Dam was built in 1966. In 1983, during project preparation, the water treatment plant had a capacity of 20,000 m3/day (reliable working capacity of the plant was 15,000 m3/day) while the peak demand was estimated at 28,000 m3/day. However, in 1984, an Emergency Augmentation Scheme, with an additional treatment capacity of 10,000 m3/d, was constructed to provide additional treated water. The booster stations and transmission lines worked at or above their design capacities, particularly during the high demand period of the year. Water supply, although servicing almost 100% of the population, within the City boundary was rationed by cutting off supply to certain areas during peak demand months for up to 12 hours per day. In order to meet the unmet as well as expected increase in demand, piped water supply needed to increase from 22,000 m3 per day in 1985 to about 67,000 m3 per day in 1997. Therefore, the Government requested that the Bank finance the construction of a new dam/reservoir and expand the water supply system to cater for the increased demand. C. Sector Development Objectives and Policy Context 2. The Government's policy reflects those of the International Drinking Water Supply and Sanitation Decade and aims at providing clean potable water to all people to reduce the incidence of water borne diseases and to reduce the time devoted by individuals to water collection. Furthermore, as a result of the financial and management deterioration of key parastatals in the early 1980s, which added a financial burden on the Government, the Government formulated appropriate policies to make these parastatals financially self-sufficient. In the water sector, to strengthen the mandate of the two existing Water Boards and make them financially self-supporting, supplemental legislation to the Waterworks Act of 1926, was passed in 1984. The supplemental legislation were the Blantyre Waterworks Act and the Lilongwe Waterworks Proclamation. Furthermore, to avoid duplication of effort by various agencies involved in  providing water to other than Lilongwe and Blantyre, a comprehensive reorganization in the water sector was undertaken in 1979. This reorganization led to the creation of the Department of Lands, Valuations and Water (DLVW) under the Office of the President and Cabinet (OPC). DLVW was later split into the Department of Water (DOW) and the Department of Lands and Valuation (DLV). During the project, the Department of Water had the sole responsibility, except in Lilongwe and Blantyre which had water boards, for developing and managing water resources in Malawi. D. Project Objectives and Decrptio 3. Project Objectives. The objectives of the project were to augment and secure the water supply to Lilongwe, Malawi's rapidly developing capital, through the construction of a new dam/reservoir and to expand Lilongwe's water supply system to cater for the anticipated increase in demand. The additional capacity was to help meet Lilongwe's water demands up to about 1997. The project was also designed to strengthen LWB's management and to help determine an appropriate strategy for Lilongwe's long-term sewerage and sanitation requirements. 4. Project Components. The Project comprised the following: (a) construction of a new earthfill dam on the Lilongwe River near Masula, which would be 24 meters high and 750 meters long, including site clearance, earthworks, concrete works, instrumentation, pipework, and fencing, and road and water pipeline diversions; (b) remedial works, including the raising of the crest level by one meter, on an existing dam serving Lilongwe; (c) extension of water treatment plant capacity by 27,000 m3/day; (d) extension of the water transmission and distribution system, including the laying of transmission mains, and construction of new service reservoirs, elevated water tanks and booster pumping stations; (e) supply of vehicles and equipment to LWB; (f) consultancy services for engineering and construction supervision; (g) strengthening of LWB through management staff support and training; and (h) implementation of the on-site sanitation scheme and preparation of the feasibility study of Lilongwe's long-term sanitation and sewerage options. E. Project Foundation and Design 5. Foundation. Although the need for additional water supplies for Lilongwe was first identified in 1978 through the Bank's Cooperative Program with the World Health Organization (WHO) and a Cooperative Project of the Swedish International Development Authority (SIDA), preparation of the project did not start until after the Government's long-range water supply, sewerage and sanitation programs were identified. In 1980, UNDP provided financing for prefeasibility studies for the long range development of Lilongwe's water supply system and to prepare draft terms of reference for the associated water supply master plan and feasibility studies. In September 1981, the Government requested IDA's  - 3 - assistance for the financing of the master plan and feasibility studies. In addition to the master plan and feasibility studies, the preceding Project, the Lilongwe Water Supply Engineering Project, financed: (a) detailed engineering design and preparation of contract documents for the first phase; and (b) management and staff support and training to strengthen the newly-created LWB. The Master Plan recommended that the water supply development program should be implemented in two stages. The first stage to be operational by 1990 and reach its capacity by 1997 and the second stage to be operational in 1997 and reach its operational capacity by 2005. The first stage was financed under the Second Lilongwe Water Supply Project. 6. Design. Although the initial project design was appropriate for the projected demand, in order to cut down initial construction costs, the feasibility studies for the first phase of the water supply scheme recommended the following amendments to the proposals in the Master Plan: (a) scaling down population and demand projections to reduce the design capacity of the proposed treatment works from 45 mIl/d to 32 ml/d; (b) building the dam in two stages with the first stage to serve until 1997 and raising the dam by 5 meters to double its storage capacity; (c) eliminating a technically desirable western ring main; and (d) reducing treated water storage volumes by 30%. 7. Based on revised population and demand projections (Table 1) for the City of Lilongwe, which resulted mainly from the outbreak of cholera, LWB was required to provide service to the peri-urban area of the city. This resulted in a substantial increase in demand on their facilities and therefore the design capacity, which was anticipated to be adequate until 1997, has now been utilized. The SAR population projections were based on 6.5 percent growth per annum starting with an estimated figure of 124,900 in 1983, while the revised estimates are based on 8 percent growth per annum starting with an actual figure of 245,756 in 1987 for the expanded Lilongwe area of service. Table 1: Population Proiections YEAR SAR Estimates Revised Estimates (Original Area) (Expanded Area) 1990 213,000 309,582 1995 272,000 454,877 2000 348,000 668,384 2005 444,000 982,048 ources LWB In retrospect, implementing the original project design would have been appropriate. This is specifically true for the construction of the dam: The cost of raising the dam by 5 meters now will cost much more than the estimated cost during project preparation. The need for commissioning of a Third Lilongwe Water Supply Project (Proposed Project to be financed by ADB) in 1996, one year earlier than anticipated, and the justification for the project, mainly an increase in demand resulting from more than anticipated population growth, prove that the original design was appropriate.  8. Advance identification and preparation of the Project design under a well focused study contributed to its smooth implementation. The engineering designs, bid documents, and contract documents were prepared by the time the project was appraised. F. Project Implementation 9. Background. The Second Lilongwe Water Supply Project was prepared through the Lilongwe Water Supply Engineering Project (Credit 1272-MAI) (para 5). The Master Plan prepared under the Engineering Project was revised because the initial estimated cost of the project (US$47 million) exceeded the Government's investment capability. Furthermore, it was felt that the projected annual tariff increase required to support the proposed investments (40 % increase in 1985 and decreasing to 11 % in 1991) would be beyond the affordability of a large segment of the population. The final engineering design, prepared under the Engineering Project, reflected the changes (para 6). 10. The project Implementation Schedule is presented in Annex 4. The main issue raised and extensively discussed during project preparation and appraisal was LWB's ability to fund the investment as proposed under the Master Plan. The initial implementation schedule reflected this concern. Although a shorter implementation period would have been cost effective, implementation was phased over six years - one year for final preparation, two years for construction of the dam, and three years for construction of the treatment plant to reduce the budgetary demands on LWB. However in early 1988, after the finalization of FY88/89 budget and the ten-year budget projections, LWB confirmed its ability to provide the local counterpart funding earlier than anticipated during appraisal. Therefore, the implementation schedule was revised. The revised Implementation Program is also presented in Annex 4. As a result of cost savings from the shorter implementation period, construction of the Headquarters building was included under the project. This work was included in the project as originally conceived, but was deleted during appraisal in order to reduce project costs and counterpart funds required. Although some difficulties and setbacks were experienced, project implementation progressed smoothly. Furthermore, capacity development efforts that were initiated in advance, allowed LWB to implement the staff development program at an early stage of project implementation. 11. Construction of Kamuzu Dam No. 2 and Repair to Kamuzu Dam No. I (Contract) A : While remedial works on Kamuzu Dam No. I were completed on time, the construction of the new dam started in February 1987 as originally scheduled, but was completed in June 1990, seven months after the scheduled completion date. The causes of completion delays were: (a) late arrival of the heavy construction plant and equipment; (b) plant breakdowns; (c) difficulties in obtaining adequate quantities of cement; (d) low production rate of washed sand; and (e) unusually heavy seasonal rainfall which caused high back water levels. 12. Construction of Water Treatment Works. Pulling Stations and Reservoirs & Water Towers (Contracts B & E). The two contracts were awarded to the same contractor. Start of construction work was advanced by eight months and completed ten months ahead of appraisal estimate. Although the contracts were completed earlier than anticipated during appraisal, they were completed five months after the revised completion date. The reasons for the completion delay were: (a) poor quality of plant that experienced frequent break downs; (b) shortage of essential equipment, which required sharing of some equipment for both contracts; (c) late ordering of materials, specifically pipes, (d) lack of adequate supervisory staff; and (e) high rate of staff turnover on the part of the Contractor. 13. SuDply and Installation of Water Treatment Equipment (Contract C). This contract was advanced by eleven months and was completed ten months ahead of appraisal estimate. The contract experienced minor technical and administration difficulties. In order to minimize difficulties and ensure a more reliable and less expensive source of chlorine for water treatment, LWB requested and the Bank approved the financing of the  - 5 - On Site Electrolytic Chlorination (OSEC) plant, instead of using imported chlorine gas for disinfection. Local generation of chlorine from salt contributed to foreign exchange savings and improved reliability of supply of chlorine. 14. Supply and Installation of Pumping & Electrical Eguipment (Contract D). Work started in April 1989, ten months ahead of appraisal estimates, and was completed in June 1991, two months after the revised completion date. Completion of this component experienced some problems with transformers which were believed to have been damaged in transit, and with raw water pumps which did not meet the specifications. The latter problem was resolved quickly by bringing a skilled fitter to re-fit the pumps. However, the former took longer as the transformer tanks needed to be reordered. 15. New Office Building. In December 1987, the Water Board requested that the construction of a new office building (that would cost about Kwacha 2.6 million) be added under the dam construction contract and financed under the project. The justification for the request was based on the fact that to have these facilities ready for use during project construction would save some money as it would eliminate the need to provide office accommodation for construction and supervisory personnel. Construction of the new office building started in August 1989 and was completed in June 1990. The new office building generally improved performance as it brought operation and supervisory staff to one location and reduced communication as well as transport costs. It also resulted in savings from high rental accommodations. 16. Contracts F. G and H, relating to the design and supervision of construction of the distribution system, were handled in-house with the Bank's approval in order to provide experience for the LWB's own staff. This was discussed and agreed with the Bank in March 1988. This was possible because of the addition to LWB's staff of a project engineer and an operations engineer through technical assistance. The work was done with the assistance of, and under the direction of, the TA team attached to LWB as part of the capacity building effort under the project. The successful completion of these contracts provided valuable experience in contract management to LWB staff. 17. Resettlement. The number of families initially estimated to be resettled and compensated for the loss of land, crops, trees, and houses, as a result of the project was 50. Nearly 200 families were actually compensated. The total compensation payments by LWB out of its own funds prior to dam construction were about MK 300,000, 50% more than estimated (MK 200,000) at appraisal. As the standard of living of the families displaced was not known during appraisal and there has not been effective follow-up since the compensation was paid out, it is not possible to determine the net impact of the project on the displaced persons. 18. Institution Building. Implementation of the institution building and localization (replacing expatriate engineers by trained Malawians) program, still on-going in some areas, was successfully implemented. A twinning arrangement with South Staffordshire Water Company of the UK provided LWB with about 50 person- months of training both in Malawi and in the UK through short-term attachments. Additionally, LWB's staff had the opportunity to review on the job training in the installation of mechanical/electrical equipment during construction of project works. Furthermore, staff in all key management areas were trained and all posts, except the Financial Controller's post, have now been localized. The Financial Controller's post is in the process of being localized with a suitable candidate having being identified. 19. Some of the main organizational improvements achieved were: Planning (a) A planning division was created to co-ordinate the previously diverse planning functions;  - 6 - (b) LWB's first Corporate Plan was produced; (c) An annual planning cycle was formulated; Organization (a) The Engineering Department was split up into an operations Department and a Technical Department. (b) A commercial division was established within the Finance Department Communicati A Management Information System (MIS) was introduced. 20. Sanitation ComRonent. The feasibility study for sewerage and sanitation for Lilongwe was successfully completed. The low-cost sanitation component was completed, although it experienced some delays as a result of shortage of local funds. 21. Project Costs and Financing. The appraisal estimate of the total project cost was Kwacha 55.0 million, including price and physical contingencies. While actual cost of the Project in terms of Malawi Kwacha was much higher than the appraisal estimates (Kwacha 77.7 million), the actual cost in terms of US Dollars was much lower than the appraisal estimates. The IDA Credit of SDR 16.6 million (US$ 20.0 million equivalent) increased to USS 21.6 million equivalent when the Project Closed. The actual and estimated costs in Kwacha and US dollars are given in Annex 5. As indicated in the table, the scope of the project was expanded to utilize the cost savings. The actual cost includes the financing of the construction of new headquarters building. 22. Procurement and Disbursements. Procurement of Bank-financed items proceeded without delays. The estimated and actual disbursements of the Credit are given in Annex 3. As can be seen from this table, actual disbursements were slightly ahead of the estimated disbursements. G. Major Results of Project 23. Project OQ jetive. Overall, the project was successful in meeting its stated objectives of expanding Lilongwe's water supply system, strengthening LWB's management and determining an appropriate strategy for Lilongwe's long-term sewerage and sanitation requirements. Through the: (a) construction of a new earthfall dam source, security was ensured; (b) extension of water treatment plant; including new service reservoirs, transmission and distribution mains, a reliable water supply was ensured; and (c) management staff support and training program, LWB's management capacity,necessary to operate and maintain the systems, was developed. 24. The Second Lilongwe Water Supply Project also provided financing for a feasibility study of Liongwe's sanitation and sewerage and implementation of an on-site sanitation system. The feasibility study examined the existing and prospective urban development in Liongwe to the year 2005 and sanitation and water supply systems. Also, an alternative sanitation projects for waste water treatment and disposal were evaluated and an  - 7 - emergency programs with priority components to be build for immediate rehabilitation of the existing sanitation facilities were identified and designed. 25. Physical Results. The physical targets of the project were achieved with less than anticipated cost and much earlier than the time anticipated at appraisal. This was done because of LWB's ability to come up with the counterpart funding earlier than anticipated during appraisal. 26. Financial Performance. LWB's income statements, balance sheets and sources and application of funds comparing SAR forecasts with actual results for the period 1988 - 1993 are presented in Anne14. LWB's operating results are summarized below. The actual net revenues were higher than the appraisal estimates because implementation schedule of the project was moved ahead of SAR estimate. This resulted in higher production and sales. The actual operating expenses were also higher than appraisal estimates. Table 2. LWB's Operating Results, 1990 - 1993 Comparison of SAR Estimates with Actual Results MK '000 - '1 *t- i -4 Not Revenue 7,476 6,332 9,063 7,634 I',46 8,422 15,188 11,396 Opeatfing Expene.a 6,820 5,294 7,714 5,844 9,372 6,385 13,746 10,017 Operating Income 656 1,039 1,349 1,790 2,474 2,037 1,442 1,330 Operating Ratio 91 33 85 76 79 75 90 87 Onratins Ratio: Opemting expene/net revenue - give. the relationship between opeating expenses, including depreciation, cad met revenue. 27. The two major factors that contributed to the less than anticipated financial performance are the rate of inflation and the rate of increase of average water tariffs. The higher rate of inflation caused operating expenses to be higher, and the less than inflation tariff increases resulted in lower than anticipated revenues. As shown below, except in FY90, FY91, and FY92, the tariff increases granted to LWB were lower than the rate of inflation. Although the increases granted in FY90 and FY92 were above the rate of inflation, they were granted in two stages, with the second increase granted about 6 months into the financial year, thus reducing its beneficial impact. In FY93, the second increase of 24% was requested due to major increases in costs caused by the two devaluations and wage increases, but only 15% was granted.  - 8 - Table 3: Tariff Increase Requested and Granted -1980 - 1989 t1- 9911 19921 199. 11994 |Inf. Rate SAR Estimates 10 9 9 9 9 9 9 Actual 26 31 16 11 12 22 26 Tariff Increase Requested 12 25 27 17 18 22 & 24 28 Granted (nominal) 12 18 27 15 16 18 & 15 18 Granted (real) (14) (13) (11) (4) (4) 11 (8) 28. At the time of project preparation in 1986, the average tariff in Lilongwe was 42 tamnbalas/m3 which was below Blantyre's average of tambalas 63/m3 although both cities had about the same per capita income levels. Household expenditure for water in Lilongwe constituted between 1.2% and 2.0% of per capita income. At project closing, water charges averaged 2.2% of per capita income and on the basis of convertional practice which considers 5% or less of income for water and sanitation as affordable, LWB has a margin of flexibility within which to adjust tariffs. Although tariffs incremented at a negative real rate (See Table 3) per capita income growth was even lower, hence the lack of political will on the part of Government to increase tariffs. 29. Rate of Return on net Fixed Assets. As shown in Table 4, the actual rates of return are lower than the appraisal estimates. The highest rate of return achieved during the period up to March 31, 1993 was 2.0% (in FY92), as compared to a projected level of 5.8% for the samne year. The reasons for the lower than anticipated rates of return are: (a) net revenues increased at a lower rate than operating expenses which resulted in lower surplus, (b) larger increases in the actual value of net fixed assets than projected, about 3 times greater during 1990-92, and about 1.8 times higher in 1993, the last year of the project, and (c) even after a sound review of asset lives and through revaluation processes, the assessed depreciation provision was abated by 15 % up to FY91 and by 20 % thereafter, resulting in the tariff being set too low and failure to meet the rate of return objective. The actual value of the revalued net fixed assets were much higher than the SAR projections because: (a) Implementation schedule of the project was moved ahead by one year resulting in a larger accumulation of fixed assets in operation earlier than anticipated in the SAR. (b) The Malawi kwacha was devalued by 15 % in 1988, and by a further 40 % in 1992. This had a direct impact on the kwacha value of Lilongwe Water Board's fixed assets. (c) As a result of macro-economic factors, including the devaluation, the actual rates of inflation during this period were significantly higher than the SAR projections. The net fixed assets were, therefore, revalued using much higher factors than originally anticipated.  - 9 - Table 4: Rate of Return HW n t 191. ' - 1990. 199.1 19 :, : 19.3 SAR Estimates 2.2 3.3 5.4 5.8 1.9 5.8l Actual 0.05 0.9 1.2 2.0 0.8 30. Economic Re-Valuation. The assumptions and the data for the economic re-valuation are shown in Annexes 7 and 9. At appraisal, the economic evaluation of benefits in quantitative terms was limited to revenue from incremental water sales which accounted for 98 percent of the project cost. At appraisal, the economic rate of return for the project was estimated at 5.5 percent. The actual ERR is 4.65% and this is lower than the 5.5% estimated in the SAR. This is mainly due to the decline in real average tariffs which contributed to less than projected net benefits. H. Project Sustainability 31. The major benefits of the project are an improved water service delivery to the City of Lilongwe and the building of efficient and effective institution. To ensure continued financial viability of LWB, there is a need to establish a mechanism for periodic adjustments of tariffs to reflect the economic cost of providing the services. From the experience gained under the project, it is evident that LWB would not be able to generate adequate funds internally to finance its future investment needs without adequate and timely tariff adjustments. Furthermore, to further improve the LWB's management there is a need to continue the institutional development effort that was started under the Project. 1. General 32. A component to protect the raw water quality from inflows of impurities from the catchment was not included under the project. LWB has requested the Bank to consider including catchment protection under the National Water Development Project. J. Association Performance 32. Although the need for additional water supplies for Lilongwe was first identified in 1978 and a number of short-term water supply extensions were implemented, the Bank's involvement awaited the development of LWB's long-range water supply and sewerage and sanitation program. The long-range water supply, sewerage and sanitation programs for Lilongwe were formulated under the preceding Lilongwe Water Supply Engineering Project. The Bank performed well in reviewing the Master plans, feasibility study, and providing assistance in developing an affordable and financially viable project. The Bank's strategy for the project, first identifying the long-range water supply needs and preparing a priority investment program, was well-founded. 33. The supervision missions, although only once a year on average, facilitated project implementation by reviewing procurement documents and discussing the financial and economic justification for any additional work LWB proposed to carry out under the project. During each supervision mission, the failure of LWB to meet the covenanted rate of return was raised, but no further action was taken by the Bank as LWB seemed to be performing well in all other respects. In retrospect, the covenanted rates of return should have been renegotiated at the completion of the revaluation of the fixed assets. When the revaluation exercise was completed the revalued fixed assets were between 100%-  - 10 - 200% higher than those used in the SAR. Therefore, the covenanted financial rates of return were unrealistically high. K. Borrower and Executing Agency Performance 34. The implementing agency for the Project, LWB, performed very well during project implementation. Procurement documents were prepared on tune and financial statements were audited and submitted to the Bank before the due dates. As shown in Annex 12, the Borrower was in compliance with the Credit and Project Agreement covenants except the rate of return. L. Project Relationship 35. The good relationship between the Bank and LWB, the implementing agency, and the Government of Malawi was instrumental to the successful implementation of the project. M. Consulting Services 36. The engineering and construction supervision consultants' performance was good. Quarterly reports, providing details of implementation and progress in general, were prepared on time. The consultants who prepared the feasibility study for Lilongwe's sanitation and sewerage completed the study on schedule and within budget. N. Project Documentation and Data 37. The Loan Agreement and Appraisal reports were adequate and clear and were instrumental to the smooth implementation of the Project. Other relevant information, progress reports, Master Plan, etc., on the project were adequately documented.  PROJECT COMPLETION REPORT PART II. -PROJECT REVIEW SECOND LILONGWE WATER SUPPLY PROJECT A. ADEQUACY AND ACCURACY OF FACTUAL INFORMATION IN PART III The information in Part III was prepared by the Financing Agency in close collaboration with the Implementing Agency, Lilongwe Water Board. The data thus reflects the status of the Second Lilongwe Water Supply Project from progress reports and other available data. B. EVALUATION OF BANK'S PERFORMANCE At appraisal, the covenants that were agreed upon with the borrower were in some cases unrealistic, e.g. the stipulated expected rate of return on investment as a measure of the financial performance of the Lilongwe Water Board. The tariff levels that were agreed upon at appraisal were not honoured by the borrower. The Bank's stance should have been to insist on the borrower honouring its commitments as regards levels of increases in tariffs to ensure continued sound financial standing of the Implementing Agency. Lessons learned: The absence of an effective monitoring system for conformity with conditions agreed upon at appraisal during the life of the project has resulted in some covenants, mostly on tariff levels, not being honoured. As a result, the Implementing Agency's financial status has deteriorated, and the ability to borrow addstional funds for further expansion projects is jeopardised. At appraisal, the Bank should have stipulated specific actions it would take 'for failing to honour the covenants agreed upon at appraisal stage. C. EVALUATION OF BORROWER'S OWN PERFORMANCE During the implementation stage of the Second Lilongwe Water Supply Project, the organisation's establishment lacked properly experienced managers, hence the need to include a management support component for a -.General Manager, Financial Controller and Training through Twinning with the South Staffordshire Water Company.  -12- This assisted the development of properly qualified personnel, especially since part of the project was implemented by direct labour. In contrast with this situation, the supervision consultant worked in almost total isolation, and there was little transfer of skills to the Malawian professionals as regards project management. Major Lessons Learned During the implementation stage there was need to appoint a project liaison officer, project accountant, procurement officer, and quality assurance officer, to look into the interests of the Board and have an indepth knowledge of all issues pertaining to the project, including a smooth handing over of completed parts of the project. Save for the dam, there is little information on operation and maintenance as regards some of the project components. D. ADEQUACY OF PROJECT COMPONENTS IN MEETING OBJECTIVES The project, when completed, partly met the original objectives since capacities of various components are almost reaching full capacity within the two years of commissioning. The main reason for -this is that the population figures which were used for planning purposes turned out to be unrealistic, e.g. the projected population for -the year 2000, was already exceeded by 1992. An assumed growth rate of 6% was used and the actual turns out to be much higher at 8%. The actual demand is about 30% above that projected, hence current raw water storage capacity, which was assumed to be adequate till the year 2000, will be inadequate by 1997. Hence the decisions to: (a) cut back on dam height by 5m; (b) reduce treatment capacity at TW2 to 27 Ml/Tay; and (c) remove sections og the transmission main from the Second Lilongwe,Water Supply Project; have left the Board with sections of the system not able to meet the demand in the near future, and hence the need for the Board to embark on a project to increase on the capacities of those elements of the project that are fully stretched.  -13- Major Lessons Learned The design horizon for some of the main components, e.g. dam and treatment works, should have been reviewed in light of recent information on population growth and demand. This would have eliminated the need to embark on a further expansion project soon after completion of the current project. The decision to cut the dam embankment to save MKI.O million at the time has already translated into MK25.0 million to construction the dam embankment to maintain the reliability of the supply for 1996 onwards. Similarly, the decision to cut back on treatment capacity has resulted in the available unit operating to full capacity as of now. The absence of a project co-ordinator and quality assurance officer to independently criticise some aspects of the project has landed the Board with high maintenance costs within a short period. This is more evident with electro/mechanical equipment. E. EVALUATION OF CONSULTANTS' PERFORMANCE The consultant was technically very competent. However, his performance was adversely affected by lack of interaction with the essential staff of the executing * agency who would take up responsibility for operation and maintenance of all aspects of the project when handed over. As a result, a little information on the project as regards operation and maintenance was handed over to the client. There was very little transfer of skills on specific aspects of the project, and information required for further expansion is always hard to get. There was very poor relationship between the consultants and some of the contractors, - This resulted in a