ICRR 13156 Report Number : ICRR13156 IEG ICR Review Independent Evaluation Group 1. Project Data: Date Posted : 04/05/2010 PROJ ID : P009121 Appraisal Actual Project Name : Rural Water Supply & US$M ): Project Costs (US$M): 117.0 78.8 Sanitation Project Country : Uzbekistan Loan /Credit (US$M): Loan/ US$M ): 75.0 58.4 Sector Board : WAT Cofinancing (US$M ): US$M): 29.2 0 Sector (s): Water supply (87%) Sanitation (7%) Health (4%) Central government administration (2%) Theme (s): Pollution management and environmental health (40% - P) Rural services and infrastructure (40% - P) Other communicable diseases (20% - S) L/C Number : L4216 Board Approval Date : 08/21/1997 Partners involved : KfW and the Kuwait Closing Date : 12/31/2005 06/30/2008 Fund Evaluator : Panel Reviewer : Group Manager : Group : George T. K. Pitman John R. Heath IEGSE ICR Reviews IEGSE 2. Project Objectives and Components: a. Objectives: The Staff Appraisal Report (SAR) states that the project has two objectives . "The first is the provision of safe drinking water along with improved hygiene education and sanitation facilities for the Republic of Karakalpakstan and Khorezm Oblast. Specifically, this objective would decrease the incidence of water -borne diseases among the population, particularly diarrheal diseases among children . The second objective is to strengthen institutional capacity for management, operation and financial performance of the regional water supply and sanitation utilities as well as the regional Centers of Health and the Sanitary Epidemiology Stations ." The Loan Agreement (LA) provides a less complex and more straightforward description of objectives that was adopted by the ICR and is used in this ICRR : "To improve water supply, sanitation, and health in the Project area through : (a) the provision of safe drinking water and sanitation facilities; and (b) the strengthening of the financial, operational, and managerial capacities of water supply and sanitation ." The main beneficiaries of the project were expected to be 1.5 million people living in rural areas and 0.4 million urban residents who would gain increased access to improved, adequate, safe, and reliable water supply services . About 50,000 rural and peri-urban inhabitants were to be provided with improved sanitation facilities . Key performance indicators at approval were : 1. Increased access to water through house and yard connections, public standpipes, and hand pumps; 2. Quantity of water produced and distributed; 3. Quality of water produced and distributed (percentage of water quality samples meeting bacteriological quality standards); 4. Percentage of the rural population in the project area covered by hand pumps and low -cost sanitation; 5. Improvement in knowledge, attitudes and practices related to water, sanitation, and health; 6. Establishment of a set of key monitoring indicators on managerial, operational, and financial performance and training; 7. Preparation by January 1998 of a plan for restructuring the water agencies (described as a "draft Corporate Development Plans for the water agencies "). This was later revised at Mid-term Review (MTR) to January 2000 to reflect a more realistic timetable for achievement . 8. Implementation of the restructuring plan according to a schedule agreed with the Bank . b.Were the project objectives/key associated outcome targets revised during implementation? Yes If yes, did the Board approve the revised objectives /key associated outcome targets? No c. Components (or Key Conditions in the case of DPLs, as appropriate): 1. Water Supply and Distribution (Estimated cost, US$ 68. 58 .4 million; of which the 68 .2 million; Actual cost, US$ 58. Bank financed US$ 38. 38 .4 million ). (a) Main Pipeline Supply System . This included replacement or rehabilitation of sections of the water distribution systems and water rural distribution centers in the project area, rehabilitation of two and expansion of one water treatment plants and of five ground water sources; and installation of one new trunk pipeline. (b) Development of Local Water Supply and Distribution Systems . This included provision of spare parts and training of operators for about 300 desalinization units; rehabilitation and development of ground water sources and expansion of demand -based rural water supplies. (c) Water Demand Management and Loss Reduction . This included metering trials to test different approaches to metering water usage, a consumer awareness program to assist reduction of losses and optimization of water use and design and implementation of leakage reduction program . (d) Equipment for Operation and Maintenance . 2. Sanitation, Health and Hygiene Component (Estimated cost, US$ 11. 11 .2 million; Actual cost, US$ 2.3 million; of which the Bank financed US$ 2.3 million ). (a) Provide Rural Sanitation to provide around 7,500 improved latrines and hand washing facilities, covering about 5 percent of the rural population in both regions . 20% of the construction costs was to be contributed by beneficiaries. It was to provide also equipment, transportation, consultant services and training for rural, community-based committee members and other staff in management and low -cost sanitation technologies . (b) Promote Health Promotion and Hygiene Education to the project area to complement the community-based sanitation and hand pumps sub -components of the project. This subcomponent also covered the costs of an inter -sectoral committee aimed at reviewing and assisting project activities . (c) Improve Water Quality Monitoring and Strengthen Sanitary Epidemiological Services (SES) provided to households and communal facilities . It also aimed to enhance the institutional and organizational efficiency of the SES through better M&E and integrating regional data reporting into the project ’s Geographic Information System (GIS) on a demonstration basis. 3. Technical Assistance Component (Estimated cost, US$ 8.2 million; Actual cost, US$ 5.6 million; of which the Bank financed US$ 5.6 million ). (a) Institutional Strengthening of Water Utilities and Bulk Providers to enable a program for regular preventive maintenance, monitoring and reduction of operating costs and reduction and management of accounts receivable. It included design and operation of a utility accounting system, including cost accounting, and local and foreign training programs . (b) A Tariff Study to lay the foundation for the introduction of new water and sewerage tariff rates for the various consumers groups . It included consultancy services for the formulation and implementation of the new tariff structure and office equipment . (c) Refinancing of the Feasibility Study carried out by the Kuwait Fund. 4. Project Management, Design and Supervision Component (Estimated cost US$ 8.4 million; Actual cost US$ 9.4 million; of which the Bank financed US$ 8.9 million ). This supported project management, project design, and construction supervision by the Project Implementation Unit (PIU), consultant services, and support staff . 5. Refinancing of Pilot Water Supply Engineering Project Component (Estimated cost US$ 5.0 million; Actual cost, US$3 US$ 3.2 million; of which the Bank financed US$ 3.2 million ). The Pilot project had funded the construction of a self -managed water supply scheme in the peri -urban areas of Nukus and the preparation of detailed designs and bid documents by foreign consultants . d. Comments on Project Cost, Financing, Borrower Contribution, and Dates: It was planned that Bank lending for this project was complemented by cofinancing from the Kuwait Fund for Arab and Economic Development of US$19.9 million, and US$ 19.4 million was disbursed as parallel funding . The KfW project was initially identified as a co -financed activity for US$9.4 million, but it was executed independently through the Ministry of Foreign Economic Relations, Investment, and Trade . In early 2000 following protracted difficulties in establishing an effective PIU, the Government entered into a technical and administrative assistance agreement with UNDP for a grant of US$ 0.45 million to finance the PIU. In 2003 a total of $1.67 million of the loan was cancelled due to misprocurement of equipment by the Borrower . The project was restructured on July 12, 2005 at the request of the Borrower; primarily to cancel most of the portion of the loan allocated for sanitation, health, and hygiene . Overall, US$12 million was cancelled and the LA was amended to reflect the reallocation between loan categories . The project was extended twice . The first extension of 2 years to December 2007 allowed for completion of ongoing contracts for civil works and institution strengthening . The second extension of 6 months to June 2008 was to permit completion of a water treatment plant and desalination facilities . 3. Relevance of Objectives & Design: High ) Relevance of Objectives (Rating : High) The multi-donor Aral Sea Program addressed the adverse impacts on health and productivity, particularly on water supplies, of poor environmental management caused by the Aral Sea crisis . This is a continuing international priority. The project targeted the poorest people and the regions most adversely affected by the Aral Sea crisis. The project was included in the Bank ’s Country Assistance Strategy (CAS) for the period 1995 - 1998 supporting two of its four pillars (macroeconomic stabilization along with structural and sector reforms, and protection of vulnerable groups through establishment of an effective safety net and more efficient service delivery). The 2002-2004 CAS focused on supporting long -term actions that would enhance the efficiency of resource allocation and use in social infrastructure and services; the Interim Strategy Note of 2006 maintained this focus. The 2008-2011 CAS emphasizes the need to improve the quality of basic service delivery, and to improve environment management. The objectives have remained relevant with respect to the Government of Uzbekistan’s overall development priorities. In 2008 the Government completed a national inventory of water supply and sanitation services that has made investment to improve services a priority over the next ten years . Relevance of Design (Rating : Modest ). At appraisal Uzbekistan was in its first years of separation from the former Soviet Union . Institutions were in flux, and many were severely weakened during the transition . Despite being the first Bank investment operation in the country, the design of the project did not sufficiently cater to building local institutional capacity for financial management and procurement, and insufficient attention was given to good governance . There were delays in completing pilot projects that were intended to lay the foundations for the community-based water supply and rural sanitation sub -component as well as the urban and rural health and hygiene components. Consequently, by appraisal, design of these components had not been completed, even though there were the results of a series of five social assessments to draw on . Appraisal overestimated Borrower commitment to an IBRD Loan for to the low -cost, community-based approach to hygiene and sanitation. As a result this component stalled and was later cancelled . Overall, relevance is rated Substantial . 4. Achievement of Objectives (Efficacy): While it is clear that the project contributed to the desired outcomes, attribution is made difficult by the large number of complementary activities and investments by other donors at the same time . USAID invested in improvements to the production capacity and physical efficiency of the Karakalpastan water companies during the life of the project. The ADB, the European Union, UNICEF and bilateral donors were also active in the water supply and sanitation subsectors in the two regions covered by the project . The original overall objective to improve water supply, sanitation, and health was partially achieved because the government cancelled the health and sanitation components . However the ICR (para 66), the Region and the TTL note that sanitation and health education were supplied by others (UNICEF) through grants (in comparison to an expensive IBRD Loan). Although the achievement of that part of the objectives that remained was substantial, overall outcome must be rated against the original objectives if they were not formally restructured . Thus overall efficacy is rated modest . ( a) Provide safe drinking water facilities . This was substantially achieved with some shortcomings . (Rating (1) (a ( Rating : Substantial ). The project provided improved water supplies and security to 1.2 million people. This is 22% less than the target of 1.5 million as a result of the reduction in expenditure due to loan cancellation . This includes 0.4 million people who gained new access to water supplies (surpassing the appraisal target of 0.3 million) and 0.8 million people that had their water supply rehabilitated (target was 1.2 million). The share of the beneficiary population receiving piped water supplies (via house and yard connections, standpipes and neighbors ) increased from 29% in 1995 to 46% in 2008. In comparison only 32% of the population in 'without project' randomly selected rayons had access to piped water supplies . We do not know the net incremental benefit because of the deficiencies in project M&E (see section 10). 80% of all beneficiaries were more securely supplied with water either through pipeline distribution systems or from electrically pumped wells. The balance in the rural areas, 239,000 (or 20% of total beneficiaries, ICR Table 3.5), were provided with new hand pumps only that drew on groundwater sources and are less reliable : the ICR notes (para 55) that about half of these hand pumps may be non -operational due to their poor quality construction. If this is correct then secure water supplies reach only about two -thirds of the target population . At appraisal (PAD, pages 91-94) it was expected under component 1 that 220,000 people would be supplied from "alternative water resources" (i.e. hand pumps to access groundwater ), and another 592,000 people would similarly benefit under component 2. As most of component 2 was cancelled it is inferred that only 3% (19,000 people) of its access target was achieved . Most beneficiaries (618,000 or 53%) lived in rural areas (the PAD target was 73%, SAR Table 3.7). Even so, poverty targeting was successfully achieved . Random sampling of the project and control areas found that project beneficiary households (HH) were generally poorer than in the non -project areas. In Khorezm the project's median HH income was Uz150,000-200,000 compared with control HHs that were Uz 250,000-300,000. A similar difference in median HH income was found in Karakalpakstan . Water consumption was reduced . For the utilities, the aim was to maintain ex -ante residential supply volumes of 280 liters/capita/day (l/c/d) plus an annual growth of 1% when conservation and pricing measures became effective. However, water conservation measures and pricing impacts, allied with the loss of industrial demand, were so effective that total water consumption fell by 54%. Average residential consumption fell to in the Ugench utility from 551 to 121 l/c/d for 87,340 people; in the Nukus utility from123 to 74 l/c/d for 87,633 people; and in the Trest utility from 98 to 30 l/c/d for 393,300 people. Even so, these latter are above the WHO basic consumption levels (basic access is defined as equal or better than 20 l/c/d; 50 l/c/d id defined as intermediate access). Water quality is substantially fit for human consumption . According to the TTL, the ICR team could not access water quality information and relied on verbal reports from the SES . On this basis the ICR rates 98% of water supplied 'safe' for domestic consumption . This value is based on piped water supplies (that are treated) and pumped groundwater (that is desalinated), and hand-pumped groundwater (that is not treated). If hand-pumped wells are excluded the proportion of 'safe' water is 80%. (b) Provide sanitation facilities . This was partially achieved . (Rating 1) (b (Rating : modest ). Improved access to low-cost sanitation was partially achieved - 38,000 compared with the target of 50,000. Almost nothing was achieved by the project on improving knowledge, attitudes and practices related to water, sanitation and health because the health and sanitation component was cancelled by the government . (2), (a) Strengthen the financial, operational, and managerial capacities of water supply was substantially achieved with shortcomings (Rating : Substantial ). Reorganization was successful . The Corporate Development Plan for the six original water utilities resulted in their rationalization and reorganization into three utilities . Successful implementation of the Corporate Development Plan for the six original water utilities resulted in their rationalization and creation of three utilities . In Khorezm the bulk water supplier (Tuyamuyun Urgench) and the urban and rural utilities were merged into one. In Karakalpakstan the bulk supplier and the urban utility were merged into one (Tuyamuyun Nukus) but the rural utility (Trest) remained separate. The customer base increased for both urban utilities . In Urgench the number of water connections rose by 77% and in Nukus by 24%. Sewerage outcomes were mixed. In Urgench connections fell by 1% while in Trest they increased by 22%. Operational efficiency generally improved . Water conservation measures (e.g. metering, leak detection and repair) and better utility management, accounting and billing has significantly reduced unaccounted -for-water. This declined from an average for the project area of 42% ex ante (SAR page 102) to a weighted average of 28% for the three utilities; the target was 34%. As a result total annual water losses for all utilities shrank from 37.8 million cubic meters (Mcm) to 22.6 Mcm - an important saving in a water-short country and a significant saving in the energy used for pumping . Staffing efficiency improved: in Ugench it fell from 3.5 staff per 100 connections to 1.3 (target 2.7) and in Nukus it fell from 2.1 to 1.6 (target 1.7). Similar data for Trest are not available. Billing remains a problem. Accounts receivable fell from 146 days to 127 days in Nukus, from 88 to 74 days in Trest, and from 153 days to 127 days in Ugench. While they were improvements they all were far short of the 60 day target. Even so, performance was better in the project area than in non -project areas. The 'with-without' survey found that 46% of consumers in the project area said they paid for water; in the without -project controls only 21% paid for water. Water tariffs were significantly increased but do not cover total supply costs . In 2009 prices the tariff charged by the Nakus utility rose 35-fold to US$0.13/m3, in Trest 13-fold to US$0.10/m3 and 50-fold in Urgench to US$0.07/m3. Most of these large tariff increases reflect the high inflation rate . The ICR states (Annex 2, page 10) that costs were as estimated at appraisal . Considering only variable costs the cost per m 3 was estimated at US$0.19 for piped supplies; US$0.56 for desalinated water and US$0.03 for water from hand pumps. If depreciation and interest were included these costs would be very much higher . The ratio of operating costs to operating revenues improved for all three utilities . The operating ratio in Nukus fell from 2.19 in 1995 to 0.99 in 2007 and over the same time period it went from 2.05 to 0.46 in Trest. Comparative data for Urgench are not available but the ratio in 2007 was 0.82. When all costs are included subsidies to the three utilities were UzSum 0.578 billion in 2007. The appraisal target was UzSum 2.148 billion corrected to 2007 prices. The outcome is thus far better than expected . (2), Strengthen the financial, operational, and managerial capacities of sanitation . Non-Non -evaluable . There is insufficient information on sanitation to judge outcomes attributable to the project . 5. Efficiency (not applicable to DPLs): The economic analysis at appraisal assumed a set of benefits that essentially did not materialize due to major changes in the available supply of water and changes in consumption patterns . One of the major benefits envisaged was increased consumption from increasing access and improving service quality, but the projections underestimated the impact that improved service efficiency and higher water tariffs would have on household water consumption. Although many more households gained access to water, the actual consumption per household decreased. This is not an adverse outcome . As in this project, in the majority FSU countries increased attention to the efficiency of water delivery generally reduced per capita consumption as wastage and leaks were eliminated. At completion the project has a positive net present value primarily due to cost savings . Revenues from residential consumers have been relatively small due to the low willingness to pay for water services . Even though water rates have increased rapidly over time – also in real terms – the actual water rates are still very low . The ICR presents a very though economic analysis that yielded a financial rate of return of the project is 13 percent. And on a per-capita cost basis actual costs were US$ 45/capita compared with the appraisal estimate of US$58/capita. By both measures, FRR and per capita costs, efficiency was substantial , despite the design and implementation delays. ERR )/Financial Rate of Return (FRR) a. If available, enter the Economic Rate of Return (ERR) FRR ) at appraisal and the re- re -estimated value at evaluation : Rate Available? Point Value Coverage/Scope* Appraisal Yes 17.9% 58% ICR estimate Yes 13.1% 74% * Refers to percent of total project cost for which ERR/FRR was calculated. 6. Outcome: The combination of substantial relevance, modest efficacy and substantial efficiency translates into an outcome rating of moderately satisfactory . a. Outcome Rating : Moderately Satisfactory 7. Rationale for Risk to Development Outcome Rating: The primary risks are financial. Continued subsidies will be required to keep the utilities financially viable and ensure that adequate replacement investment is made to keep the water supply and treatment systems in operation . Energy costs are increasing. And in 2013 the utilities will be required by national policy to begin servicing the debt incurred by this Loan and loans from other agencies . Increasing revenues from water sales will be difficult as the TTL says that the tariff issue is highly politicized because of the prevailing and high poverty levels . a. Risk to Development Outcome Rating : Significant 8. Assessment of Bank Performance: Quality at Entry . The Bank underestimated the magnitude of institutional weaknesses in the implementing and executing agencies and the governance challenges . It misjudged government's support for community -driven development and for sanitation and health improvements . Procurement risks were seriously underestimated . Compounding this it designed a complex project . Given that this was the first operation in this subsector a more extended period of dialogue, piloting and capacity -building may have secured a higher level of government buy-in. Alternatively, a series of more focused and smaller investments focusing on infrastructure and the utility reorganization only would have been more pragmatic . Supervision . Despite a high turnover of task team leaders (7 in all) supervision was generally effective .The variable level of Borrower ownership and frequent changes among key national management and coordination staff within the Ministry of Economy was particularly challenging . Recognizing the problems the Bank's TTL requested a Supervision Enhancement Review by QAG in early 2000 and also initiated two studies funded through the Bank Netherlands Water Partnership Fund to improve project performance and lessons learning from the experience. This operation was classified as a problem project in the periods 1998-2000 and 2003-2004. As a result supervision missions were increased to 3-4 a year and became very expensive, totalling US$ 1.28 million. Despite this increased attention, supervision missed the opportunity to revise the development objectives to reflect changes resulting from project restructuring and loan cancellation in 2005. It should be noted that the last TTL initiated significant improvements to M&E to inform the ICR . at -Entry :Moderately Unsatisfactory a. Ensuring Quality -at- b. Quality of Supervision :Moderately Satisfactory c. Overall Bank Performance :Moderately Satisfactory 9. Assessment of Borrower Performance: Government . The government was fully committed to rehabilitating and building new infrastructure . Its subsequent decision to utilize UNICEF and other grant financing for the sanitation and health aspects of the project was pragmatic given the cost of the IBRD Loan . Lack of readiness or capacity for implementation led to extended delays in establishing an effective project implementation unit (PIU). (The PIU was operated by government staff and was effectively a part of government: therefore, the Government performance rating embraces the PIU ). According to the ICR, procurement was fraught with corruption and delays and this adversely affected implementation. Financial management, reporting and attention to M&E was also weak . The PIU was disbanded following corruption allegations . Subsequently, with help from UNDP technical assistance, the performance of the PIU improved substantially . Implementing Agencies . The regional water companies and other regional agencies that implemented the project demonstrated a high degree of commitment and technical competence . However, financial management of the beneficiary water utilities – “Tuyamuyun Urgench� (formerly OJSC Horezm “Suvokava�) and “Tuyamuyun ukus� (formerly Production Board “Suuakaba�), in particular the accounting and internal control systems, remained unsatisfactory throughout the implementation period (ICR page 12). This is in spite of efforts made to strengthen accounting and internal controls, based on the financial management action plan developed as part of institutional strengthening . a. Government Performance :Moderately Unsatisfactory b. Implementing Agency Performance :Moderately Unsatisfactory c. Overall Borrower Performance :Moderately Unsatisfactory 10. M&E Design, Implementation, & Utilization: Design . Many of the indicators were poorly defined, focussing on inputs and outputs . The SAR’s Results Framework included target values for only some of the outcome indicators . There were no baseline indicators of health status and desired outcomes . Similarly, baseline date for agency performance were very patchy making incremental improvements difficult to evaluate . Setting up the M&E arrangements and their implementation was only scheduled to take place in 2000--three years after the project was expected to become effective . The new TTL in the final year of the project designed a beneficiary survey to capture some of the missing outcome data . This survey adopted a single -difference approach given there were negligible longitudinal data . It used a two-stage sampling process to select 400 sequentially selected random project households and 400 similarly sampled control rayon households for comparative analysis . There is no discussion in the ICR Annex of what steps were taken to match the treatment and control groups and it is not clear if there was selection bias . Households may have been randomly selected for the survey but it is unclear if their assignment to the project was random. Despite these reservations the beneficiary survey adds significant value to evaluating and reporting outcome - but it was too late to upgrade the overall design rating which is modest . Implementation . The focus was primarily on inputs and outputs and these were regularly monitored . Data to assess project impacts on health were supposed to be collected from government health statistics along with periodic knowledge, attitudes, and practices surveys; but this did not happen . The survey at towards the end of the project was an important achievement . Rating Modest. Utilization . Although data were collected, these were not systematically used (apart from the Bank's ICR survey). As a result, the M&E was used almost exclusively to respond to Bank and Borrower supervision demands and needs for project monitoring data rather than for more in -depth evaluation that could have generated lessons and guided refinements to project implementation . Rating Negligible. a. M&E Quality Rating : Modest 11. Other Issues (Safeguards, Fiduciary, Unintended Positive and Negative Impacts): Safeguards . This was an Environmental category B project that complied with Bank environmental and social safeguards during implementation. At negotiations is was agreed that the dam for the Tuyamuyun Hydro System would be inspected annually by independent experts . A satisfactory inspection report was provided to the Bank in 2000 following a detailed engineering and geodetic evaluation . Subsequent reports were not provided to the Bank as they are considered to be of strategic national security importance . Fiduciary . Project financial management arrangements were rated marginally satisfactory in the last financial management supervision (March 2008). Financial management of the beneficiary water utilities remained unsatisfactory throughout the implementation period in spite of efforts made to strengthen accounting and internal controls . Audit reports were frequently delayed and often qualified reflecting serious weaknesses in accounting and internal control systems. Audited financial statements of the water utilities and tariff structures also show noncompliance with the financial covenants . Procurement was a major problem. A corruption investigation by the Borrower resulted in replacement of all of the PIU national staff in May 2002. Misprocurement of earth-moving equipment was declared in 2003 and the related portion of the loan was cancelled . A case of fraud was proved on another contract of over US$ 11 million and this delayed a major civil works contract by 4 years. Although the Bank undertook investigation of these allegations of fraud and misprocurement the ICR (page 13) does not say if this was done by the Region as part of supervision or was referred to INT . Given the amounts involved and systemic nature of fraud and corruption it should be referred to INT . 12. 12. Ratings : ICR IEG Review Reason for Disagreement /Comments Outcome : Moderately Moderately Satisfactory Satisfactory Risk to Development Significant Significant Outcome : Bank Performance : Moderately Moderately Satisfactory Satisfactory Borrower Performance : Moderately Moderately Poor governance, poor accounting and Satisfactory Unsatisfactory corruption plagued the project . Quality of ICR : Satisfactory NOTES: NOTES - When insufficient information is provided by the Bank for IEG to arrive at a clear rating, IEG will downgrade the relevant ratings as warranted beginning July 1, 2006. - The "Reason for Disagreement/Comments" column could cross-reference other sections of the ICR Review, as appropriate . 13. Lessons: On first entry to a subsector it is better to have a simple project design than a complex one that may have low Borrower ownership and run into problems of implementation . In this case the Bank's desire for a comprehensive approach to WSS and health was too ambitious given the dire state of the country's basic infrastructure, its poor operation and maintenance and the political imperative to get things working . A phased approach would have worked more effectively as Borrower knowledge and confidence was built . Bank procurement rules can result in major cost savings providing local capacity and governance constraints are addressed . Unfamiliarity with the Bank's procurement rules initially resulted in significant delays in implementation that could have been avoided if governance challenges had been better understood and adequate resources had been allocated for procurement training and monitoring . Once this was addressed, however, application of the Bank's procurement rules led to cost savings that enabled the project to be more cost efficient than originally anticipated . Monitoring and Evaluation needs to be designed at appraisal, not during implementation . This project invested extensively in social assessments and pilot projects during preparation, but these efforts were not reflected in a reliable, evaluable, and comprehensive baseline indicators because detailed design of M&E was delayed for 3 years. 14. Assessment Recommended? Yes No Why? The project was one of several donor -financed initiatives on water and sanitation . It would be useful to examine how well the Bank worked in partnership with the other donors and how well the new water supply utilities are performing particularly in serving the rural populations . 15. Comments on Quality of ICR: This is a very thorough and candid ICR with excellent annexes . a.Quality of ICR Rating : Satisfactory