Malawi - Lilongwe Water Supply Engineering Project and Second Lilongwe Water Supply Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 1643 0 PERFORMANCE AUDIT REPORT MALAWI LILONGWE WATER SUPPLY ENGINEERING PROJECT (CREDIT 1272-MAI) and SECOND LILONGWE WATER SUPPLY PROJECT (CREDIT 1742-MAI) April 3, 1997 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. Currency Equivalents (annual averages) Currency Unit = Malawi Kwacha (MK) MKl.00 100 Tambala 1982 US$1.00 MKl.17 1986 US$1.00 MKl.86 1987 US$1.00 MK2.21 1992 US$1.00 MK3.60 1994 US$1.00 MK6.82 (March) 1995 US$1.00 MKl5.00 (June) 1996 US$1.00 MKl5.05 (August) Abbreviations and Acronyms BWB Blantyre Water Board EDF European Development Fund GOM Government of Malawi IBI Industrial Bank of India LCC Lilongwe City Council LCD Liters Per Capita Per Day LWB Lilongwe Water Board MIWD Ministry of Irrigation and Water Development OD Operational Directive ODA Overseas Development Administration (UK) OED Operations Evaluation Department, O & M operation and maintenance OMS Operational Manual Statement PAR Performance Audit Report, PCR Project Completion Report, SAR Staff Appraisal Report, TA Technical Assistance WD Water Department of MIWD UFW Unaccounted-for Water Fiscal Year Government of Malawi and Lilongwe Water Board: April I to March 31 Director-General, Operations Evaluation : Mr. Robert Picciotto Director, Operations Evaluation Department : Mr. Francisco Aguirre-Sacasa Division Chief : Mr. Yves Albouy Task Manager : Mr. Tauno Skytta FOR OFFICIAL USE ONLY The World Bank Washington, D.C. 20433 U.S.A. April 3, 1997 Office of the Director-General Operations Evaluation MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on Malawi Lilongwe Water Supply Engineering Project (Credit 1272-MAI) and Second Lilongwe Water Supply Project (Credit 1742-MAI) Attached is the Performance Audit Report (PAR) on the Malawi Lilongwe Water Supply Engineering Project (Credit 1272-MAI, approved in FY82) and Second Lilongwe Water Supply Project (Credit 1742-MAI, approved in FY87), prepared by the Operations Evaluation Department. The main objectives of the Engineering Project were to: (i) enable the Government of Malawi (GOM) to formulate long-range water supply, sewerage and sanitation programs for Lilongwe, and (ii) strengthen the newly established Lilongwe Water Board (LWB) through improved management, financial control, and cost recovery. These objectives were to be achieved through the preparation of appropriate master plans, design and contract documents for the next investment phase, provision of advisors to LWB, and procurement of essential equipment. In addition, the Project was to develop on-site sanitation technologies through a pilot sanitation project. The principal objectives of the Second Project were to: (i) augment the water supply to Lilongwe, and (ii) continue the management support to LWB. The project's physical components were: construction of a dam, a water treatment plant, a distribution system and continuation of the on-site sanitation program. Training and management support was to be provided through a twinning program with an experienced water utility. All new water supply facilities were successfully constructed. Overall water availability was increased and access to water by the poor was improved as planned. The sanitation program was effectively established and continues to expand. The capacity building effort was a significant success and the LWB is now an effectively managed water utility. All managerial positions are occupied by capable local professionals. The GOM and LWB have agreed to a performance contract which is monitored through five-year Corporate Plans. Progress made so far is encouraging. The growing debt service, however, has become a significant burden for the LWB due to currency risks. The outcome of both projects is rated as satisfactory and the sustainability of project benefits as uncertain mainly due to LWB's weak finances. The institutional development impact is rated as substantial. These ratings agree with the PCR-based ratings. 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. 2 These operations demonstrate that a small utility operating under difficult circumstances can be managed efficiently given the right emphasis on timely and well designed capacity building efforts. In the future, however, it is essential that some foreign exchange risk relief be given to LWB to ensure the sustainable operation of water and sanitation services in Lilongwe. Attachment Contents Preface ............................................................ 3 Basic Data Sheet ..................................................... 5 Evaluation Summary ........................................ .......... 11 1. Introduction and Background ................................... ...... 15 Macroeconomic Setting............................................15 Water and Sanitation ....................... ......................15 Water Resources...............................................15 Sector Legislation and Institutions...................................15 Status of Sector Activities.........................................16 Sector Development and Bank Assistance...............................16 2. The Projects .................................................................................. 17 Project Objectives, Components, and Planned Outputs...................... 17 Relevance of Project Design and Quality at Entry......................... 17 3. Implementation Experience and Results ................................... 19 Project Highlights from the PCRs ............................... ..... 19 Physical Outputs ......................... .................... 19 Institutional Development ....................... ................. 19 Sustainability of Benefits .................................... ..... 19 Ratings .................... . ....................... ........ 20 Project Achievements; Audit's Analysis ............... ................ 20 Implementation Experience ..................... .................. 20 Institutional Development ................................ ......... 20 Water Availability ................................................. 21 Access to Water by the Poor .......................................... 22 Sanitation Program ......................... .................... 23 LWB's Financial Operations ................... ................... 24 Costs of Water Supply Operations ...................................... 25 This report was prepared by Mr. Tauno Skytta, Principal Evaluation Officer, who audited the projects in June 1995. Mr. Jean-Francois Landeau reviewed and analyzed the financial information collected during the audit mission. Ms. Helen Watkins provided administrative support. The author wishes to dedicate this report to the memory of Mr. Jozsef B. Buky, a dear friend and long-time colleague, who developed the approach applied in the two audited projects. 2 4. Issues for the Future ........................................... 27 Water Sources ................................................. 27 Institutional Development .................... ..................... 27 Sustainability ................................................ 28 5. Assessment of Performance ............................................. 29 Borrower Performance ...................... ..................... 29 Compliance with Covenants and ODs ................................. 29 Bank Performance .............................................. 29 Projects' Ratings .............................................. 29 6. Lessons and Recommendations ............................... ...... 31 Lessons ...................................................... 31 Recommendations .............................................. 31 Annexes A. Project Summaries ................................................. 33 B. Household Monthly Income Distribution in Lilongwe ......................... 37 Attachment Comments from the Borrower ...................................... ...... 39 3 Preface This is a Performance Audit Report (PAR) for the Lilongwe Water Supply Engineering Project (Credit 1272-MAI) and Second Lilongwe Water Supply Project (Credit 1742-MAI) approved on June 24, 1982, and November 20,.1986, respectively. The PAR consists of an Evaluation Summary and a Performance Audit Report prepared by the Operations Evaluation Department (OED). The Project Completion Reports (PCR) were prepared by the Africa Regional Office on May 7, 1990 (Report No. 8642), and on December 2, 1994 (Report No. 13744), respectively. The PCRs provide an accurate, but brief, account of the projects' achievements. The PAR builds on this information and elaborates on particular aspects such as institutional development, including financial performance of the project entity, access to water by the poor, efficiency of operations, and lessons from these operations. The PAR is based on the PCRs, the President's Reports, the projects' legal documents, and the Staff Appraisal Reports (SAR). In addition, a review of relevant project files and other background material collected during the audit mission in June 1995 helped in the analysis presented in the PAR. The draft PAR was sent to the Borrower for comments. The comments received from the Lilongwe Water Board are reflected in, and reproduced as an Attachment to the PAR.  5 Basic Data Sheet MALAWI LILONGWE WATER SUPPLY ENGINEERING PROJECT (CREDIT 1272-MAI) Key Project Data (Amounts in US$ million) Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 4.75 4.04 85 Credit amount 4.0 Cancellation 0.26 0.07 Economic rate of return N/A N/A Cumulative Estimated and Actual Disbursements FY83 FY84 FY85 FY86 FY87 FY88 Appraisal estimate (US$M) 400 1,500 3,200 4,000 Actual (US$M) 0 1,300 1,700 2,998 3,340 3,606 Actual as % of appraisal 0 86.7 53.1 75.0 83.5 90.2 Date of final disbursement: October 27, 1987 Project Dates Original Actual Initiating memorandum Negotiations 5/82 5/24-26/82 Board approval 5/82 6/24/82 Signing 6/82 9/27/82 Effectiveness 7/82 2/2/83 Closing date 6/30/86 6/30/87 6 Staff Inputs (staff weeks) Total Through appraisal 26.5 Appraisal through 4.5 Board Approval Board approval through 2.1 Effectiveness Supervision 41.1 Mission Data Date No. of Staff Specializations Performance Types ofproblems (month/year) persons days in represented rating field Identification through Effectiveness 02/79 1 5 - 06/79 1 1 Sanitary - Engineer 08/80 1 5 Sanitary - Engineer 02/81 1 3 Sanitary - Engineer 04/81 1 4 Sanitary - Engineer 07/81 1 5 Sanitary - Engineer 08/81 1 13 Financial - Analyst & Sanitary Engineer Supervision 1 1 3 Sanitary 1 Engineer Supervision 2 2 2 Sanitary I Management Engineer Supervision 3 2 7 Sanitary 1 Engineer Supervision 4 1 6 Sanitary 1 Engineer 1. 1 = No significant problems, 2 = Moderate problems 7 Date No. of Staff Specializations Performance Types ofproblems (month/year) persons days in represented rating field Supervision 5 1 6 Sanitary 1 - Engineer Supervision 6 1 6 Sanitary 2 - Engineer Supervision 7 1 3 Sanitary I - Engineer Supervision 8 1 2 Sanitary - Engineer Supervision 9 2 1 Sanitary 1 - Engineer Other Project Data Borrower/Executing Agency: Government of Malawi FOLLOW-ON OPERATIONS Operation Credit no. Amount Board date (US$ million) Second Lilongwe Water Supply Project 1742-MAI 20.0 11/20/86  9 Basic Data Sheet MALAWI SECOND LILONGWE WATER SUPPLY PROJECT (CREDIT 1742- MAI) Key Project Data (Amounts in US$ million) Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 55.0 77.7 141 Loan amount 20.0 20.0 100 Cancellation SDR 0.9 million Economic rate of return 5.5 4.7 Cumulative Estimated and Actual Disbursements FY87 FY88 FY89 FY90 FY91 FY92 FY93 FY94 Appraisal estimate (US$M) 0.6 2.8 6.8 11.2 15.2 18.0 19.4 20.0 Actual (US$M) 5.0 8.0 15.0 18.3 20.9 21.6 Actual as% of appraisal 179 118 134 120 116 111 Date of final disbursement: November 15, 1993 Project Dates Original Actual Identification 10/84 10/84 Negotiations 09/09/86 09/09/86 Board approval 10/20/86 11/20/86 Signing 02/11/87 02/13/87 Effectiveness 01/87 05/12/87 Closing date 06/30/93 06/30/93 10 Staff Inputs (staff weeks) Total Through Appraisal 26.6 Appraisal Through Board Approval 10.62 Board Approval Through Effectiveness 2.7 Supervision 34.6 Mission Data Date No. of Staff days in Specializations Performance (month/year) persons field represented' rating4 Identification 10/84 FA Preparation 8/85 Appraisal 02/86 5 4.0 EN,EC,FA,FA C Supervision 1 06/87 1 2.50 EN 1 Supervision 2 02/88 1 1.50 FA,EN 1 Supervision 3 09/88 1 0.20 EN 1 Supervision 4 06/889 1 1.00 EN 1 Supervision 5 06/90 2 2.20 EN 1. Supervision 6 02/91 1 0.40 EN 1 Supervision/Completion 7 07/93 2 3.00 FA,EN 1 Other Project Data Borrower/Executing Agency: Government of Malawi FOLLOW-ON OPERATIONS Operation Credit no. Amount Board date (US$ million) National Water Development Project 2753-MAI 79.2 6/28/95 2. The resources utilized from appraisal through board approvals is low because most of the work was done under the Engineering Project. 3. EN = Engineer, EC = Economist, FA = Financial Analyst, FAC = Financial Analyst - Consultant, TS = Training Specialist, FAD = Financial Advisor 4. 1 = No significant problems 11 Evaluation Summary Introduction and Background 1. Malawi's economy relies primarily on agriculture which makes it vulnerable to climatic changes. Recent droughts have cut food production and caused water shortages throughout the country. The gross national product fell by nine percent in 1994 with inflation at nearly 35 percent. Out of the total population (about 9.0 million in 1992), 88 percent live in rural areas. Population growth was 3.4 percent during the last decade. 2. Surface water resources vary greatly from the dry to the wet seasons and, depending on rainfall, year to year. Groundwater resources are widespread throughout the country but yields are generally sufficient only for small community supplies. 3. The Water Resources Act as amended in 1990 is the main statute dealing with management of water resources. Local government acts and public health and environmental legislation touch on sanitation and environment. The Blantyre Water Board (BWB) Act of 1971, and the Lilongwe Water Board (LWB) Act of 1986, are statutory and commercially operated bodies. The Water Department (WD) of the Ministry of Irrigation and Water Development (MIWD) is in charge of policy preparation, planning and monitoring, technical assistance in planning and design of sewerage, and control of water quality. Local government bodies are responsible for water-borne sanitation and septic tank emptying, as well as promotion of on-site sanitation programs. 4. Bank involvement was preceded by (i) distribution expansion funded by the European Development Fund (EDF) in 1981, and (ii) an emergency project for capacity augmentation funded by the Industrial Bank of India (IBI) in 1985. In addition to the two evaluated projects, the Bank has been involved in two other water sector operations; one with the BWB (1977), and one with WD (1989). A new project, the National Water Development Project, approved in June 1995 (Credit 2753-MAI), is being implemented by the MIWD. The Projects 5. The first of the two evaluated projects, the Lilongwe Water Supply Engineering Project (Engineering Project), Credit 1272-MAI of US$4.0 million, was approved on June 24, 1982, and was completed in October 1987, nearly two years behind schedule. The second project, the Second Lilongwe Water Supply project (Second Project), Credit 1742-MAI of US$20 million, was approved on November 20, 1986, and was completed in December 1992 as scheduled. 6. The Engineering Project set out to develop a long-range sector program for Lilongwe; support was provided to strengthen the LWB through improved management, financial control and cost recovery, design of a new investment phase and preparation of a sanitation strategy for Lilongwe, including testing of various technological solutions. The Second Project focused on implementing the above new investment phase and funding for the sanitation program. It also continued the institutional support began under the Engineering Project. 7. The Audit finds that the design of both projects was relevant and that their "quality at entry" was good. The Second Project clearly demonstrates the merit of a Technical Assistance 12 (TA) operation preceding an investment phase, i.e. provision of adequate time for a dialogue between the Bank and the Borrower and development of a firm commitment for necessary institution building and funding of follow-up investments. Implementation Experience and Results 8. The PCRs give a good picture of the implementation of the two projects in terms of timeliness, costs, and quality of works; details in Annex A. The Audit's analysis confirms the findings presented in the PCRs and presents specific findings as to the projects' impact on institutional development, availability of water, sanitation program, and LWB's operations. 9. The LWB has signed a performance contract with the GOM which sets out several performance indicators; the targets are presented in a "Corporate Plan". The Audit reviewed and analyzed the second Corporate Plan for the period 1995-2000 in detail. 10. Operational improvements emerged from both projects. Target rates for unaccounted- for water (UFW) are stringent but LWB has been able to meet them. The current level of about 16 percent is among the lowest recorded in any African city. The LWB improved the quality of its service, i.e. time to respond to new service applications and service complaints. 11. The twinning (training) program aimed at the "localization" of key managerial positions. The LWB took over the functions of the expatriate advisers effectively and all senior level management positions are filled by local staff. The staff productivity index also improved; the current ratio of 24 staff per 1,000 connections is still high but half of that in 1990. 12. The gross availability of water increased as a result of the Second Project and the capacity of the new dam (Kamuzu Dam II) was sufficient to meet the demand through 1995, less than estimated at appraisal due to long-term drought in Malawi. The gross availability of water rose to 160 liters per capita per day (lcd); it is expected to go down to 125 lcd by 1997 due to higher than expected population growth. The current data, however, show that actual production, reflecting the actual demand, were about of 110 lcd during the first half of the 1990s. 13. The distribution system expansion under the Second Project has kept up with the expansion of the city. In low-income areas, the supply was inadequate until recently, but improved during the Second Project; about 30 percent in 1994. A network of water kiosks in the traditional housing areas (and in surrounding villages) was developed to improve the access by the poor. The LWB kiosks are operated by private operators, all of whom are women. The LWB together with the Lilongwe City Council (LCC) has strongly promoted the operation of kiosks by consumer committees. 14. Costs of water supply have shown steady increase in line with the local inflation and the increase in the US dollar value of the Malawi Kwacha. Despite this, the overall cost profile of operations appears to be healthy and the costs per cubic meter produced, US$0.10 in 1994, compares favorably with the average of US$ 0.14 for a panel of 18 developing country utilities. 15. In spite of increases in the tariffs, the financial situation of the LWB has deteriorated markedly. The current ratio has been less than one in the last three fiscal years, in sharp contrast with the comfortable levels reached in FY89 and 90. The trend is clearly downward and, if continued, will lead to serious insolvency. Profitability was lost in 1994 mainly due to the 13 impact of the local currency devaluations on the foreign debt and the return on fixed assets was negative. On the positive side, accounts receivable were stable, even decreasing slightly. 16. The Engineering Project developed a sanitation demonstration program and technological options, as well as promotional and health education material. Some 2,000 latrines were constructed between 1986 and 1989. The Audit was informed that 4,000 additional latrines have been constructed since 1990 under the continuation of the program. The incidence of water borne diseases has decreased. Delivery of sanitation services is community based; LCC provides the coordination, and sells latrine platforms at cost to households, household provide the digging of the pit and erection of the superstructure. However, latrine emptying service is not easily available. The fees are high; about the same as the monthly income of half of Lilongwe's population. Issues for the Future 17. The quality of water in the two dams has been a concern due to inflowing impurities from the catchment area. An afforested buffer zone of about 100 meters wide around the reservoirs and along the river carrying raw water to the treatment plant has now been authorized. Another concern are the recent droughts. Annual rainfall has been at less than 50 percent of the average of 1,400 mm in the last five years and the dams are operated under drought conditions. The next expansion will raise the main dam and provide sufficient water to meet the demand until 2005. 18. The institutional model used in Lilongwe (and Blantyre) i.e., the Water Board, has become a national approach in handling the water sector in cities and larger towns. The Audit finds that the LWB appears to have adequate operational autonomy but still lacks the autonomy to take crucial tariff and investment decisions based on its actual financial position. The LWB has effectively started the privatization of parts of its operations by contracting out water kiosks. Also, property maintenance services are now handled by outside contractors. Future plans include outsourcing meter reading, vehicle maintenance, and rehabilitation works. 19. A twinning model of providing TA and management support to a starting and still weak institution proved to be effective, producing capable local managers. Many of the participants were selected internally thus providing and supporting the continuity within the organization. Unfortunately, the training effort continues on an ad-hoc basis; in the technical area, however, an arrangement has been made with the Umgani Water Authority of South Africa. 20. Tariff increases below inflation and heavy foreign exchange losses pose serious threats to the sustainability of LWB's operations. Both timely tariff increases in line with local inflation and a measure of foreign exchange risk relief are essential for shoring up this sustainability while keeping services affordable to the poor. Assessment of Performance 21. The performance of the Borrower is rated as satisfactory. It improved considerably during the latter part of the Engineering Project and especially during the Second Project. The financial statements and audit reports were submitted to the Bank before due dates. The LWB set acceptable managerial standards for carrying out its day-to-day operations. Performance was 14 poorer in meeting the financial ROR covenant; tariff increases were inadequate, and often sanctioned late. 22. The Bank's performance is rated as satisfactory. The approach it took to prepare the ground for a larger investment project through a well targeted TA operation worked well. The special assistance given in identifying training needs and preparing a comprehensive training program as a twinning arrangement was essential. Supervision missions handled the technical supervision and institutional aspects well but not quite so the financial ones brought about by the impact of the Malawi currency devaluation on the IDA debt. 23. The Audit rates the outcome of the Engineering Project and that of the Second Project as satisfactory. The institutional development impact of both projects is rated as substantial. The sustainability of project benefits for both projects is rated as uncertain mainly due to the financial uncertainties in the LWB's operations. These ratings agree with those based on the PCRs. Lessons 24. The Audit draws the following main lessons: * Given the weak institutional capacity base in Lilongwe in the early 1980s, the approach chosen to provide heavy and well targeted TA inputs prior to agreeing to enter into a major investment program proved to be successful. * The Second Project was effective in improving the access to water by the poor. Extension of the network of water kiosks operated by individuals as contractors to the LWB, or increasingly by community groups, proved to be a successful solution. * The projects got on-site sanitation programs effectively under way in Lilongwe. However, the O&M support still remains uncertain and too expensive for residents in low-income areas. Recommendations 25. The operation and development of water services can be sustainable only if the burden to cover debt service remains within the capacity of the project entity. For this to happen, some foreign exchange risk relief should be considered as a complement to tariff increases and future investments should be carefully screened and sequenced. More immediately, the accounting practices should be changed so that the unrealized foreign exchange risk is not entirely charged to the income statement; a practice that gives a distorted picture of the utility's operating finances. 15 1. Introduction and Background' Macroeconomic Setting 1.1 The economy of Malawi relies primarily on agriculture. This fact makes Malawi vulnerable to climatic changes. A drought in 1992 drastically cut food production and caused water shortages throughout the country. As a result, the gross national product (US$220 in 1993) fell by nine percent in 1994 and inflation increased to 34,7 percent. Out of the total population of Malawi (about 9.0 million in 1992) 88 percent live in rural areas. In 1980-91, the population growth was 3.4 percent. Water and Sanitation Sector Water Resources 1.2 The surface water resources of Malawi, apart from Lake Malawi, vary greatly from the dry to wet seasons and year to year depending on rainfall. Generally, the surface waters have high suspended solids content (especially during the wet season) thus raising treatment costs. The problem is aggravated by inadequate environmental watershed protection measures resulting in soil erosion and declining microbiological water quality. 1.3 Groundwater resources are widespread throughout the country. Depending on the type of aquifer, the yields vary but are generally sufficient only for small community supplies. The quality is normally acceptable, although some localized quality problems exist. Groundwater is presently exploited by 9,700 boreholes and 5,600 shallow wells equipped with handpumps. Malawi is also known for its extensive use of small scale, spring fed gravity schemes. Sector Legislation and Institutions 1.4 The Water Resources Act of 1969 is the main statute dealing with management of water resources in Malawi. The Act was amended in 1990; complementary statutes are the Blantyre Waterworks Act of 1971 and the Lilongwe Waterworks Act of 1986. Local government acts and public health and environmental legislation touch on sanitation and environmental aspects. 1.5 The Lilongwe Water Board (LWB) and the Blantyre Water Board (BWB) are statutory bodies responsible for water supply services to the two cities. They operate commercially without receiving any recurrent funding from the Government of Malawi (GOM). The Water Department (WD) of the Ministry of Irrigation and Water Development (MIWD) provides services to towns and rural areas. The WD's services include: policy preparation, planning and monitoring; control and monitoring of quality of water; provision of water to all of Malawi except areas covered by LWB and BWB; and technical assistance to institutions in planning and design of sewerage and effluent treatment. Urban areas are served by piped water supplies. 5. This section is based on information collected from the Staff Appraisal Report for the National Water Development Project (Report No. 13189-MAI), dated June 1, 1995. 16 Rural areas depend on gravity schemes, borehole/shallow wells with handpumps, and protected spring supply systems, all community managed. Local government bodies are responsible for water-borne sanitation, septic tank emptying, and sludge disposal, as well as promotion of on- site sanitation programs. Status of Sector Activities 1.6 In general, the water sector in Malawi is not financially self-sustaining and many water supply and water-borne sanitation systems suffer from a number of weaknesses; service coverage is low and the quality of the main water sources is declining, in some cases it is already poor. This is mainly, until recently, due to the absence of a formal policy framework including realistic pricing policies, and inadequate institutional arrangements to match tasks of delivering water services and managing water resources. 1.7 In early 1994, the GOM recognized that the priorities of the sector should be: a) to prepare an integrated policy framework; b) to put in place institutional arrangements for implementing this new policy and ensuring community participation; c) to develop a set of strategies to implement the new institutional arrangements; and d) to secure funding for investments in rehabilitation and extended coverage. The strategies seek to ensure that the policy framework is applied so as to achieve the sector objectives. They cover institutional development, capacity building, and the achievement of sustainable resource management and investment. The amended Water Act envisages the establishment of Regional Water Boards (RWB) and they are expected to make investment choices that are demand oriented. Sector Development and Bank Assistance 1.8 Bank involvement in Lilongwe was preceded by two important developments. First, a project of distribution expansion, costing US$2.9 million, was implemented with financing from the European Development Fund (EDF), often referred to as Phase I, with completion in 1981. Second, an emergency project was funded by the Industrial Bank of India (IBI) to increase the supply and treatment capacity, projected to meet the water demand in Lilongwe through 1985. The cost of this project (Phase II) was US$5.4 million. 1.9 The Bank has been involved in four water sector operations in Malawi; one with the BWB (1977), two with the LWB (1982 and 1987), the projects evaluated here (Phase III), and one with WD (1989). These operations assisted in building up the capacity of the Water Boards in Blantyre and Lilongwe. A new project, the National Water Development Project, approved in June 1995 (Credit 2753-MAI), is being implemented by the MIWD. 17 2. The Projects Project Objectives, Components, and Planned Outputs 2.1 The first of the two evaluated projects, the Lilongwe Water Supply Engineering Project (Engineering Project), Credit 1272-MAI of US$4.0 million, was approved on June 24, 1982, and was completed in October 1987, nearly two years behind schedule. The second project, the Second Lilongwe Water Supply Project (Second Project), Credit 1742-MAI of US$20 million, was approved on November 20, 1986, and was completed in December 1992 as scheduled. 2.2 The objectives, components (as per the respective SARs), and the actual outputs (as per the respective PCRs) are presented in detail in Annex A. In summary, the Engineering Project set out to develop a long-range sector program for Lilongwe; a new investment phase was designed, and support was provided to strengthen the LWB through improved management, financial control and cost recovery. The project was also targeted to develop a suitable sanitation strategy for Lilongwe together with testing of various technological solutions. The Second Project focused on implementing the above new investment phase and providing funding for the expansion of the sanitation program. On the institutional side, the project was to continue the support began under the Engineering Project. Relevance of Project Design and Quality at Entry 2.3 The TA operation included a carefully designed preparation of the Second Project and focused strongly on institutional capacity building. The Engineering Project was envisaged to: (i) build up the institutional capacity of LWB to the (minimum) level needed for the preparation and implementation of the next investment phase (Phase III, see para. 1.9), and (ii) operate the expanded facilities in an efficient way. The Audit finds this approach well justified when the initial implementation capacity is assessed to be especially weak as was the case in Lilongwe in the early 1980s. The project also piloted new approaches to sanitation through a learning process which helped in the development of a long-term sanitation program for LCC. The implementation of the sanitation program was further supported under the Second Project. 2.4 The Audit finds that the design of both projects was supportive of the objectives especially for the Second Project. This demonstrates the merit of the approach, i.e. the provision of adequate time for a positive dialogue between the Bank and the Borrower and development of a firm commitment for institution building actions and funding of follow up investments. As a result, the Second Project was implemented ahead of schedule, including important strengthening of the LWB, as discussed later in this report. 18 2.5 In the 1980s, the above approach was tested in other cities as well in the then Eastern Africa Region.6 In the OED water sector portfolio review, this approach was considered to be key to project success and became one of the four study recommendations. 6. Another successful example is found in Kenya; the case is reported in the OED Impact Evaluation Report, "Development of Housing, Water Supply, and Sanitation in Nairobi", Report No. 15586, April 25, 1996. 7. Water Supply and Sanitation Projects, The Bank's Experience; 1967 - 1989. Operations Evaluation Department; Report No. 10789, June 19, 1992. 19 3. Implementation Experience and Results Project Highlights from the PCRs Physical Outputs 3.1 The Engineering Project produced a two-staged water supply Master Plan covering 1990-1997 and 1997-2005, a feasibility study of the Phase III Lilongwe Water Supply Project, as well as engineering designs and contract documents for Phase III. Although technically (and economically) feasible, the Master Plan was not financially affordable. It was quickly revised to phase the construction of the proposed dam into two stages and approved in April 1984. Accordingly, the Phase III proposal was modified significantly to bring the total capital cost down from MK 37 million to MK 26 million. In general, the technical aspects of the Engineering Project appeared to have been well handled. 3.2 Phase III was funded under the Second Project including the construction of a new earthfill dam, water treatment plant, and water transmission mains. The distribution system was augmented at a less than anticipated cost (in US dollar terms) with completion at a much earlier date than set at appraisal. This was achieved because LWB was able to come up with counterpart funding earlier than anticipated during appraisal. 3.3 Additionally, the Second Project also provided financing for the preparation of a sanitation and sewerage feasibility study for Lilongwe and for implementation of an on-site sanitation program. The sanitation component experienced some delays due to shortage of local funds. Institutional Development 3.4 The Engineering Project supported the employment of (i) an engineer/manager, a financial controller, and a low-cost sanitation adviser; and (ii) the purchase of necessary computers and software to upgrade LWB's billing and inventory control systems. The Bank and ODA assisted in reviewing specific training needs of LWB. 3.5 Under the Second Project, LWB was strengthened in the areas of planning, organization and communication. However, its financial performance remains weak as a result of the higher- than-expected inflation rate and inadequate and late tariff increases. Sustainability of Benefits 3.6 The PCR of the Engineering Project does not rate the sustainability of project benefits. The PCR of the Second Project states that, LWB has to become efficient and financially self- sufficient if project benefits are to be sustained. The PCR also states that a mechanism for periodic adjustments of tariffs has yet to be effected, and institutional development efforts need to be continued. 20 Ratings 3.7 The outcome of the Engineering Project was rated as satisfactory, the sustainability of its 8 benefits as likely , and institutional development as substantial. The outcome of the Second Project is rated as marginally satisfactory, and its institutional development as modest (i.e. moderate). Sustainability was rated as uncertain because the LWB does not recover its capital costs. Project Achievements; Audit's Analysis Implementation Experience 3.8 The respective PCRs give a good picture of the implementation of the two projects in terms of timeliness, costs and quality of works. The Audit notes that for the Engineering Project, an extension of six months was well justified to carefully review the feasibility study and improve the affordability of the proposal for the next investment phase. Overall, the completion of the project was delayed by nearly two years to allow more time for the completion of the sanitation pilot component. The Second Project was completed within the original time schedule, thus confirming the quality of project preparation under the preceding project. 3.9 The Engineering Project was implemented within budget. The sector organization and training needs assessment was dropped as it was being carried out under a bilateral (ODA) funding program. This caused the cancellation of about US$285,000. The Second Project was implemented within its cost estimates in US$ terms; in fact, the construction of the LWB's new office building was incorporated in the project to utilize the dollar savings resulting from the devaluation of the Malawi Kwacha; in Kwacha terms, the total costs increased by about 40 percent. The quality of facilities constructed under the Second Project is satisfactory. Institutional Development 3.10 The LWB has signed a performance contract with the GOM. The contract sets out several performance indicators; the targets are presented in a "Corporate Plan". The main indicators cover operational effectiveness, efficiency, and finances of the LWB. In each area, several "internal" indicators are set to monitor the operations of the individual divisions of the LWB. The first Corporate Plan was prepared during the Second Project and covered the period 1990 - 1994. The Audit reviewed and discussed the details of the second Corporate Plan which covers the period of 1995 - 2000.9 Some of the key performance indicators are discussed below. 3.11 Unaccounted-for Water. A number of operational improvements emerged from both projects. The efficient operation of the system is crucial. For this purpose, the performance indicators for unaccounted-for water (UFW) are stringent and achieving this target is one of the most critical points in the Corporate Plan. Table 3.1 shows progress made since 1990; UFW has remained close to or below the Corporate Plan targets, except for 1992 (major primary main 8. The institutional strengthening objectives were met and the results formed a solid foundation for further improvements under the Second Project; hence OED's PCR review rated sustainability as "likely". 9. Lilongwe Water Board; Corporate Plan 1995 - 2000. September 1994. 21 breakages occurred during that year). The current level of about 16 percent is a commendable achievement and it is in fact among the lowest recorded in any African city and comparable to the UFW levels in Western European countrieso. Most importantly, results are reliable as both production and consumption are metered. Table 3.1: Unaccounted-for Water (%) 1990 1991 1992 1993 1994 Target* 23.0 21.5 18.9 18.9 16.7 Actual 25.0 20.6 26.8 21.5 15.6 *As per LWB Corporate Plan, 1995-2000 3.12 Staff Development. The Bank helped develop an effective management support and training program (see para. 5.2) which was carried out by a British water authority (ODA funding). After the twinning program was completed in 1994, the LWB managed to take over the functions of the expatriate advisers effectively. This "localization" was the key objective of the program. Several key positions were localized already in the early 1990s and with the appointment of a local Financial Controller in October 1993, all the senior level management positions are filled by local officers. Table 3.2 shows the progress made in the overall staffing; since 1992 the staff productivity index (staff per 1000 connections) has been better than Corporate Plan targets. Labor costs were at 37.0 percent of operating expenses in 1995 (down from 43.3 percent in 1990), but could be further lowered to say 30 percent since the LWB only operates a water utility. Table 3.2: Number of LWB Staff per 1000 Connections 1990 1991 1992 1993 1994 Target* 49.8 43.3 36.9 27.6 23.1 Actual 49.8 45.2 35.0 24.1 24.6 *As per LWB Corporate Plan, 1995-2000 3.13 Efficiency of Operations. During the first half of the 1990s, the LWB has met, or nearly met, the targets set for all main operational indicators. Promptness of customer service, i.e. time to respond to new service applications and service complaints, has been better than, or on target. However, critical problems have been encountered in meeting financial targets. The Corporate Plan specifies criteria for several cost components, such as administration, payroll, chemicals, electricity, and plant/vehicle operation (per cubic-meter of water produced), as indicators of operational effectiveness. In the early 1990s, the results were roughly on target; but, since 1992, operating costs have nearly doubled and put on extra burden on the LWB's financial position as discussed below (paras. 3.22 - 3.23). Water Availability 3.14 The gross availability of water increased as a result of the Second Project. Both the new dam (Kamuzu Dam II) and the treatment capacity, as completed in 1992, are expected to meet 10. Source: Water and Sanitation Utilities, Set 1, Operational Indicators; TWUWS, May 1996. 11. In 1992, the water kiosk operators previously employed by LWB, became individual contractors, see para. 3.16. 22 the demand until 1997.12 Table 3.3 shows the overall progress from 1985 to 1995. Because of the expansion, the gross availability of water stayed around 160 lcd; if the treatment capacity is not augmented, it is, however, expected to go down to some 120 led by 1997 due to a higher than expected population increase. The LWB currently uses 160 lcd and 20 lcd as average design criteria for house connections and supply through water kiosks, respectively, i.e., a weighted average of 125 lcd. However, actual service levels were in the order of 110 lcd during the early part of the 1990s (see Table 3.3). Table 3.3: Water Availability and Service Level 1985a 1992 1994 1995 1997 Population, thousand 160 360b 417 446 510 Treatment Plants; design capacity, 30 62 62 62 62/120c thousand m3/day Gross availability of water, led 188 172 149 139 122/235 Average production, thousand m 3/day 26.9 40.5 46.0 N/A N/A Actual service level, led 168 112 110 N/A N/A a. As per SAR for Cr. 1742-MAI. b. Population from 1992 onward as per LWB Corporate Plan, 1995-2000. c. LWB Corporate Plan, 1995-2000, indicates an increase of the treatment capacity. Access to Water by the Poor 3.15 Access to water has improved significantly. Lilongwe, being a new city, has so far had reasonably good coverage in middle and high income areas. The distribution system was further extended under the Second Project so as to keep up with the development of new housing areas and expansion of the city. In low-income areas, coverage has, however, been quite poor until recently (Table 3.4). After the completion of the Second Project, the service in unplanned areas has improved rapidly; the coverage was at about 30 percent in 1994. This has mainly been achieved through the expansion of the kiosk network. Table 3.4: Service Coverage (%) 1991/92 1992/93 1993/94 In planned areas 100 100 100 In unplanned areas 3.1 3.6 28 Total Lilongwe 67 67 75 3.16 The LWB has developed an effective network of water kiosks in the traditional housing areas of Lilongwe (as well as in some surrounding villages) to improve the access to water by the poor (Table 3.5). The number of kiosks has almost doubled over the last ten years. Furthermore, the LWB kiosks are operated by private operators (all of whom are women), thus enhancing efficiency. Since 1993, the LWB together with the LCC has strongly promoted the creation of kiosks run by consumer committees. Box 3.1 describes one such example (in Chinsapo, a village south-west of the city center) illustrating the management system of kiosks and how they can be 12. Due to long-term drought in Malawi, consumption restrictions were required already in 1995 to avoid a major failure of supplies. 13. LCC has six extension workers for the promotion of water kiosk and sanitation activities among low-income consumer groups. 23 sustainable if communities participate in their operation. The LWB appears to manage the kiosk operator contracts in a transparent and effective manner; no incidents of corruptive charging practices have been reported by consumers. Today, the majority of the kiosks is operated and managed by (village) consumer committees which also promote discipline in kiosk management. Table 3.5: Number of Water Kiosks in Lilongwe 1985 1990 1991 1992 1993 1994 1995 Under LWB control 113 97 85 85 80 80 80 Under LCC control 0 0 0 0 35 125 125 Total 113 97 85 85 115 205 205 Source: LWB Corporate Plan, 1995-2000 and Audit's notes BOX 3.1: Women Are Effective Water Kiosk Managers The provision of water for household needs is traditionally a women's job in Malawi. In Chinsapo, the need for clean water was, however, raised by women and men alike and brought to the attention of the community leadership. In 1993, the village was provided with piped water. Chinsapo is estimated to have a population of about 20,000, split into some 4,000 families averaging 5 persons. It has 44 Communal Water Points (CWPs). The number of families per CWP ranges between 102 and 69; an average of about 90 families per CWP. Each CWP is managed by a Committee comprised of 10 members elected by, and among, the users. Most officials are women; very few committees have male members. When elected to the committees, the role of the men has been to serve as arbitrators in conflicts and to perform heavier tasks. On the whole, the women have managed to run the CWPs without help from their male partners, and they have held on to the key positions of the Committees, such as Chairperson, Secretary and Treasurer. The Committee is responsible for all management, care and day-to-day operations of the CWP. It enforces the billing; users pay monthly in advance. It sees to it that the Lilongwe Water Board is paid for the service in due course. The committee furthermore ensures that the surroundings are kept clean and monitors excessive water use. It also encourages users to take part in the upkeep of the CWP. Source: Write-up by a LCC extension worker. Sanitation Program 3.17 The start of the sanitation program was based on the approach that the demand has to be established before embarking on a large scale investment program. A Sanitation Advisor was engaged (under the Engineering Project) to develop a demonstration program and to help develop technological options, and related construction and training manuals, as well as promotional and health education materials. Some 2000 latrines were constructed between 1986 and 1989 in the demonstration phase run by the WD. The Audit was informed that some 4000 additional latrines have been constructed since 1990 under the continuation of the program in the Second Project. The program is now run, and the construction supervised, by LCC personnel. 3.18 There are other programs under way that are using the same (training, promotional and health education) materials thus multiplying the effect of the sanitation initiative began by the 24 Bank-funded projects. The Audit was informed by LCC's health officials (although not documented) that the incidence of water-borne diseases has decreased; while diarrhea was a common problem prior to the ongoing sanitation program, only sporadic occurrences of diarrhea outbreaks are reported now, mainly during the rainy season. 3.19 Delivery of sanitation services is community based. LCC provides the coordination of the program, supervises the construction, and sells latrine platforms (called san-plat) at cost to households; the cost of K15014 is payable in three monthly installments. The household provides the digging of the pit and erection of the superstructure. Households face, however, problems in the maintenance of latrines as emptying services are not easily available. The LCC has only three vacuum tankers of which only one is currently in operating condition. There are two private contractors (three tankers in total) who provide this service, but latrine emptying has to compete with the demand for septic tank emptying. The rates are also quite high, K150 per load for the LCC service and K225 for the service of private contractors, i.e. at about the same level as the monthly income of about 50 percent of Lilongwe's population. LWB's Financial Operations 3.20 The financial situation of the Lilongwe Water Board (LWB) has deteriorated markedly according to the latest available financial data (fiscal year ending in March 1995). Solvency, as measured by the current ratio, is problematic. The current ratio has been below one in the last three fiscal years, dropping to .43 in FY95. This is in sharp contrast with the comfortable levels reached in FY89 and FY90 (table 3.6 below). Although data are incomplete, the trend is clearly downward and, if continued, will lead to serious insolvency. After four years of modest profits and despite water sales increasing by over 29 percent on average during the last 8 years, profitability disappeared in 1994 mainly due to the currency risk on the foreign debt. While there is still an operating surplus, it was down in FY95 (by 8.4 percent) despite water sales being up by 52 percent. Worse, it covered only 16 percent of interest and depreciation expenses. 3.21 Most damaging to the LWB's creditworthiness were the unrealized foreign exchange losses which were (apparently) entirely charged to the income statement.16 In FY95, they amounted to MK77.2 million or three times the water revenues. In two years (FY94-95), the foreign exchange losses erased almost 5 years of sales, which is nothing short of catastrophic. As a result, the return on fixed assets was negative. The debt increased by 55.2 percent in FY95, with IDA Credits accounting for 43.7 percent of the outstanding total. The debt-equity ratio has worsened from .86:1 in FY94 to 1.21:1 in FY95. On the positive side, accounts receivable are stable, even decreasing slightly; over the years, receivables, which plague the utilities in many developing countries, have been under control. 14. Covers the cost of raw material only. 15. Water sales increased, however, at a lower rate than operating expenses. 16. A better approach would be to charge the unrealized losses to the balance sheet and amortize them over a period, say 10 to 25 years, so as not to burden the income statement. 25 Table 3.6: Financial Indicators 1989 1990 1993 1994 1995 Current ratio (x/1) 2.73 2.65 .57 .92 .43 Debt-equity ratio (x:1) .57 .78 .83 .86 1.21 Water sales increase (%) 28.5 35.6 28.2 16.4 52.0 Accumulated accounts 22.2 24.2 38.5 32.0 15.7 receivable in % of sales Costs of Water Supply Operations 3.22 The Corporate Plans set targets for key operational indicators, including operational costs per cubic meter of water produced (para. 3.13). In the early 1990s, the LWB met all the targets quite well, but begun to fall below targets after 1993. In 1994 (Table 3.7), the cost for administration was more than three times over target, plant (including all capital costs) more than double, and chemicals more than 50 percent, payroll 13 percent and electricity 17 percent over the planned targets. As a result, the overall costs of operations were some 52 percent over the targeted level. Payroll and administration costs together increased by about 130 percent from 1990 to 1994, roughly in line with local inflation over the same period. The total of chemicals, electricity and plant costs, all of which are dependant on imports, increased also by about 130 percent (from 1990 to 1994) which corresponds to the US dollar value increase of the Malawi Kwacha. Table 3.7: Cost of Water Supply in Lilongwe, MK per m3 Produced 1990 1992 1994 % of % of % of % over Cost Total Cost Total Cost Total Targeta Target Payroll 0.18 43 0.15 31 0.38 39 0.34 13 Chemicals 0.05 12 0.05 10 0.11 11 0.07 55 Electricity 0.06 14 0.08 17 0.10 10 0.09 17 Plant 0.07 17 0.08 17 0.21 22 0.09 125 Administration 0.06 14 0.12 25 0.17 18 0.05 250 Total 0.42 100 0.48 100 0.97 100 0.64 52 Consumer Price Index 100.0 138.2 204.2 104b Exchange Rate, MK to US$ 1.00 2.73 3.60 7.00c 156 Notes: a. Target set in the Corporate Plan (1990-1994). b. Increase in percent (from 1990 to 1994). c. Average for second quarter of 1994. 3.23 Despite the above increases, the overall cost profile of operations appears to be healthy. Compared with average values from a number of (developing country) utilities,' LWB's staff costs at 39 percent of the total operating costs are well below the average of 46 percent. Similarly, the electricity costs at ten percent compares with the average of 18 percent. The total cost of water supply at MK 0.64 per cubic meter (equivalent to about US$ 0.10 per cubic meter) in 1994, also compares favorably with the average of US$ 0.14, based on data collected from 18 developing country utilities (ranging from US$ 0.05 in Karachi and Changchun to US$ 0.31 in Minas Gerais). In summary, the past records show that the LWB management is improving the 17. Water and Wastewater Utilities, Indicators; Set It Financial Indicators. TWUWS, Second Edition, May 1996. 26 effectiveness of its operations. The increasing foreign exchange losses are, however, becoming a major burden for LWB to cover, a factor totally beyond its control. 27 4. Issues for the Future Water Sources 4.1 The PCR for the Second Project raises the question of water quality in the two dams. The concern is the inflowing pollutants from the catchments area. The PCR concludes that the project should have included a component to afforest a buffer zone around the reservoirs and along the river carrying raw water to the treatment plant downstream. In addition to the water quality improvements, such a buffer zone would also help in lowering the silt load into the reservoirs thus prolonging their useful life. The Audit was informed that the LWB is taking appropriate measures to mitigate the effects of these omissions and the first major step will be to create a 100 meters wide buffer zone in the areas described above. 4.2 Since 1991, the Lilongwe area has experienced severe droughts. Annual rainfall has been drastically below the average of 1,400 mm; in the last five years, the average annual rainfall has been at about 615 mm. The LWB has been forced to operate the dams under drought conditions and is actively collecting data for the design of the next expansion. The plan currently includes the raising of the main dam (Kazumu II) by five meters which is estimated to provide sufficient supply of raw water to meet the demand up to 2005. Beyond that, a major expansion is likely to include a source further away; one option is to construct a dam at the Diamphwe River some 40 km south of Lilongwe. Institutional Development 4.3 The Bank projects have provided the urban water sector with a suitable institutional model, i.e. Water Boards. This approach in handling the water sector is now being replicated in other cities and larger towns. The Audit finds that the LWB seems to enjoy adequate operational autonomy but still lacks the autonomy to take crucial tariff and investment decisions based on its actual financial situation subject to performance criteria within the framework laid down in the Corporate Plan. 4.4 A program based on a twinning model of providing TA and management support to a starting and still weak institution proved to be effective. Although there were problems during implementation, the program produced capable local managers both at the senior and middle management level. Many of the participants were selected internally thus providing and supporting the continuity within the organization. The LWB is now continuing the training effort through local institutions, especially on the financial side, with positive and promising results. In the technical area, an arrangement has been made with the Umgani Water Authority (of South Africa) to strengthen the overall quality of engineering and operational staff as well. The LWB now grants loans to employees to cover the cost of tuition; when the trainee passes the final examination the loan is written off, thus serving as an incentive. 4.5 The LWB has effectively started the privatization of parts of its operations. First, the operation of water kiosks was contracted out to the former kiosk attendants (see para. 3.16). 18. These readings are from Chitedze Agricultural Station, north-east of the two dams. 28 Also, all security and property maintenance services are now handled by outside contractors. Future plans include outsourcing of meter reading, vehicle maintenance, and small scale capital (rehabilitation) works. It is too early, however, to determine the financial implications, i.e. the impact of these measures on the LWB's income statement. 4.6 Some important aspects in LWB's operations are, however, in need of urgent improvement. The Audit notes that the strengthening of O&M activities is currently hampered by a lack of funds to maintain an adequate supply of spares and fittings. The Audit was also informed that LWB's mechanical workshop would need to be upgraded so as to improve the quality and timeliness of pump and vehicle maintenance and repair. The floor space at the existing workshop is also limited thus making it difficult to install new equipment. Outsourcing of scheduled maintenance and major repairs, e.g. by using locally available dealers, would be another option. Sustainability 4.7 As seen in para. 3.22 above, the cost of water supply increased roughly at the rate of inflation during the first half of the 1990s. The tariff increases, however, were much below inflation resulting in a lower than expected ROR.19 Also, throughout the life of the Second Project, tariff increases were often staged in two steps thus delaying their beneficial impact. LWB's capacity to service its debts, which was given a serious blow by the local currency devaluation, was weakened further by such practices (see para. 3.21). Whenever the debt burden increases, it is appropriate to require from users that tariff increases do not lag behind the local consumer price index. 4.8 On the other hand, affordability of water to consumers could become an issue if the shock of currency devaluation were to be passed abruptly and entirely to the water user. Whenever affordability is in jeopardy, so are the socio-economic benefits and often the financial revenues which are to flow from water sales. The Audit estimates that for a low-income family (of six people) to buy 20 lcd water (LWB's criterion for minimum domestic consumption), the cost would be some K10/month; this represents at least 10 percent of income for some 25 percent of Lilongwe's population. At the other end of the income curve, for some 20 percent of the population, the consumption of 100 lcd would cost a family of five about K40 per month; i.e. less than 5 percent of their income20. While the often used five percent threshold is not always valid21, the above estimates confirm that some relief on LWB's foreign exchange risks is urgently needed as a complement to tariff increases for shoring up the sustainability of their operations. 19. SAR estimates for the ROR ranged from 2.2 percent in 1989 to 5.8 percent in 1992 whereas the actual values were 0.05 percent and 2.0 percent, respectively. 20. The income distribution curve for Lilongwe is presented in Annex B. 21. In its memo, dated November 13, 1996, AFTUI reports on studies conducted in several Southern African cities which indicate that the ability to pay by the poor often exceeds this threshold. 29 5. Assessment of Performance Borrower Performance 5.1 Generally, the Borrower performed satisfactorily. Initially there were some delays in the preparation and submission of progress reports to the Bank. The performance improved considerably during the latter part of the Engineering Project and especially during the Second Project. The same applies to the financial statements and audit reports which were generally submitted to the Bank before due dates. Both projects provided management support which helped the LWB perform its duties and set acceptable managerial standards for carrying out its day-to-day operations. One area where performance failed was meeting the financial ROR covenant mostly because tariff increases were too small, and often late. The most recent Corporate Plan through which the LWB's operations are routinely monitored is of good quality. Compliance with Covenants and ODs 5.2 The failure to comply with the ROR covenant, in violation of Operational Manual Statement (OMS) 2.25, was the only disappointment. The Audit finds no violation of other Bank ODs (or OMSs) in the preparation and implementation of the two projects. Bank Performance 5.3 The Bank has performed satisfactorily overall. The approach taken by the Bank to prepare the ground for an investment project through a well designed and targeted TA operation worked well. Bank missions provided good support to the preparation of studies and plans; of significant importance was the special assistance given in identifying training needs for the LWB and preparing a comprehensive training program22 as a twinning arrangement; the Bank helped in obtaining suitable funding for the program as well. Supervision missions handled the normal technical supervision and institutional aspects well, but not quite so the financial ones. The Borrower raises this question in its part of the PCR for the Second Project, and suggests that the ROR covenant was unrealistic given the high increase in the value of the LWB's fixed assets as a result of the project and the revaluation of assets in 1992. In any case, the Bank overlooked the likely impact of the Malawi currency devaluation on the IDA debt. Projects' Ratings 5.4 The Audit rates the outcome of the Engineering Project and that of the Second Project as satisfactory. For both projects, the institutional development impact is rated as substantial but the sustainability of benefits as uncertain, mainly due to LWB's weak finances. These ratings agree with the ratings given in the PCRs. 22. At the time, the Bank had a water sector training specialist who was available and greatly helped improve the project's "quality-at- entry".  31 6. Lessons and Recommendations Lessons 6.1 The Audit draws the following main lessons: * Given the weak institutional capacity base in Lilongwe in the early 1980s, the approach chosen to provide heavy and well targeted TA inputs prior to agreeing to enter into a major investment program proved to be successful. * The Second Project was effective in improving the access to water by the poor. Extension of the network of water kiosks operated by individuals as contractors to LWB, or increasingly by community groups, proved to be a successful solution. * The projects got on-site sanitation programs effectively under way in Lilongwe. However, the O&M support still remains uncertain and too expensive for residents in low-income areas. Recommendations 6.2 The operation and development of water services can be sustainable only if the burden to cover debt service remains within the capacity of the project entity. For this to happen, some foreign exchange risk relief should be considered as a complement to tariff increases and future investments should be carefully screened and sequenced. More immediately, the accounting practices should be changed so that the unrealized foreign exchange risk is not entirely charged to the income statement; a practice that gives a distorted picture of the utility's operating finances.  33 Annex A PROJECT SUMMARIES Lilongwe Water Supply Engineering Project (Credit 1272) Objectives The project's purpose was to enable the Government of Malawi and LWB to formulate long-range water supply, sewerage and sanitation programs for Lilongwe. LWB was also to be strengthened through improved management, financial control and cost recovery. The project was also intended to assist the Government of Malawi to develop appropriate technologies for sewerage and sanitation and to assess the sector's organizational, manpower and training needs. Project Components Planned (Cost: US$4,750,000) Actual (Cost: US$4,041,000) as per PCR a) Master plans for the long-range development of water supply a) Completed. Formed basis for and sewerage and sanitation in Lilongwe; Feasibility Study and for low-cost pilot sanitation program. b) A feasibility study, including field investigation, detailed b) Completed. Formed basis for engineering design and preparation of contract documents for Engineering Design. Phase III the next phase (designated as Phase III) of Lilongwe's water Project commenced--Credit 1742. supply development; c) Provision of the services of an experienced c) An engineer/manager and a Engineer/Manager and a Financial Controller for LWB and financial controller were employed. management support for other activities in the sector; d) A pilot project to demonstrate low-cost, appropriate d) Completed. Construction of sanitation technology in Lilongwe; sanitation low-cost waste disposal system in progress. e) Procurement of equipment and supplies essential for e) Equipment was provided to LWB, preparation of the Lilongwe (Phase III) Water Supply Project including trucks, meters, and a and support of the pilot, low-cost sanitation component; computer and software support, which enabled LWB to update its billing and inventory control procedures. f) Assessment of the sector's manpower and training needs f) Not carried out. LWB's needs were including recommendations for a training program; identified with Bank assistance and funded through ODA. g) A study of the sector's organization and, if necessary, g) Not carried out. assistance to Government to develop and review proposals for reorganization or improved coordination within the sector. 34 Annex A Results The Engineering Credit achieved its specific objectives with respect to Water Supply: (a) A Master Plan was produced. (b) A Feasibility Study was carried out. (c) An engineering design and contract documents were produced leading to the Phase III Lilongwe Water Supply Project. (d) An engineer/manager and a financial comptroller were employed. The Engineering Project also supported the employment of a low-cost sanitation adviser and the development of a demonstration project. Equipment was provided to LWB, including a computer and software support, which enabled LWB to update its billing and inventory control procedures. The provision for sector-wide manpower training was not utilized, as it was agreed that this area was already addressed by other studies conducted for the GOM. Specific training needs of LWB were reviewed with the assistance of the Bank, and support obtained from Overseas Development Administration (ODA) to implement the training program. All of the above was accomplished essentially within budget, the overall underspending being principally the result of not proceeding with the sector organization and training needs assessment study. 35 Annex A Second Lilongwe Water Supply Project (Credit 1742) 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 seven percent additional capacity was to 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. Project Components Planned (Cost: Kwacha 55.0 million) Actual (Cost: Kwacha 77.7 million) as per PCR a) Construction of a new earthfill dam on the Lilongwe River a) Completed seven months after the near Masula, which would be 25 meters high and 750 meters scheduled completion date. 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 b) Completed on time. one meter, on an existing dam serving Lilongwe; c) Extension of water treatment plant capacity with additional c) Completed five months after the 27 ml/d; revised completion date. d) Extension of the water transmission and distribution system, d) Completed. 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; e) A new office building was also built. f) Consultancy services for engineering and construction f) Good consultancy services. supervision; g) Strengthening of LWB through management staff support g) Management staff support and and training; and training program, LWB's management capacity, necessary to operate and maintain the system, was developed. h) Implementation of the on-site sanitation scheme and preparation of the feasibility study of Lilongwe's long-term h) Completed. sanitation and sewerage options. 36 Annex A Results 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. The physical targets of the project were achieved with less than anticipated cost (in US dollar terms) and much earlier than the time anticipated at appraisal. This was achieved because LWB was able to come up with the counterpart funding earlier than anticipated during appraisal. The project also provided financing for a feasibility study of Lilongwe's sanitation and sewerage and implementation of an on-site sanitation system. 37 Annex B Figure 1 Household Income Distribution for the City of Lilongwe 100 - 90 .8 b 0 - 80 60 C 4J 60 S40 20 0 100 200 300 400 500 600 700 800 900 1000 1100 Household Monthly Income, Kwacha Total number of households: 51,880. Source: Center for Social Research; collected by Lilongwe Water Board, 1995. Notes: a. Household of six; consumption 20 lcd; cost Kl0 per month; 10% of household income. b. Household of five; consumption 100 lcd; cost K40 per month; 5% of household income. HWatkins M:\SKYTTA\MALAWI\MALAW1.DOC March 17, 1997 4:18 PM  39 ATTACHMENT COMMENTS FROM THE BORROWER Lilongwe Water Board TELEPHONE: 744 366 Office Your Ref P0 Box 96 744 071 Compiaints LILONGWVF TELEX: 44517 Madzi MI Our Ref .. . 3.43 MALAV% FAX. 741259 20 January 1997 Mr Yves Albouy Chief Infrastructure & Ener-gy Division Operations Evaluation Depar-tment The World Bank 1818 H Street NW WASHINGTON DC 20433 United States of America Dear Mr Elbouy, RE - MALAWI: LILONGWE WATER SUPPLY ENGINEERING PROJECT CCREDIT 1272 - MAI3 AND SECOND LILONGWE WATER SUPPLY PROJECT ECREDIT 1742 - MAI3 DRAFT PERFORMANCE AUDIT REPORT Thank you very much for your letter dated 4 December 1996, enclosing a draft Performance Audit Report for our, comments. Please find attached our comments on the draft Performance Audit Report. Yours sincerely, MACLEAN J MZUMARA GENERAL MANAGER Encl. MJM/mas 40 LI L_QN43WE= WF-TER E3=RD LILONGWE WATER SUPPLY ENGINEERING PROJECT ECREDIT 1272 - MAI) AND SECOND LILONGWE WATER SUPPLY PRQJECT ECREDIT 1742 - MAI3 COMMENTS ON THE DRAFT PERFORMANCE AUDIT REPORT 1. GENERAL The Lilongwe Water Board is in agreement with the Performance Audit Report. The report really reflects the true position of the Lilongwe Water Board after the implementations of the projects. In fact, a lot of reference has been made to the Project Completion Report EPCR3 and the Staff Appraisal Report ESAR3 on which the Lilorgwe Water Board had commented in detail. 2. MINOR AMENDMENTS The only minor amendments to be incorporated in the report are as follows: . In paragraph 3.12 an page 21, substitute the words "and with the return of the Financial Controller from overseas training in October 1993" with the words "and with the appointment of a local Financial Controller in October 1993." . On page 33, in the table of Project Components, item (c) Under planned, substitute "Chief Accountant" with "Financial Controller." On page 35, in table of Project Components, item (c), substitute 32 Mi/d with 27 Mi/d. M J MZUMARA GENERAL MANAGER MJM/mas    IMAGING Report No.: 16430 Type: PPAR