L/C/TF Number(s) Closing Date (Original) Total Project Cost (USD) IDA-38870,IDA-46040,TF Dec ,493,000.00

Transcription

1 Public Disclosure Authorized Independent Evaluation Group (IEG) 1. Project Data Report Number : ICRR Public Disclosure Authorized Project ID P Project Name VN-URBAN UPGRADING Country Practice Area(Lead) Additional Financing Social, Urban, Rural and Resilience Global P Vietnam Practice L/C/TF Number(s) Closing Date (Original) Total Project Cost (USD) IDA-38870,IDA-46040,TF Dec ,493, Bank Approval Date 29-Apr-2004 Closing Date (Actual) 31-Dec-2014 IBRD/IDA (USD) Grants (USD) Public Disclosure Authorized Original Commitment 382,470, Revised Commitment 350,711, Actual 353,680, Sector(s) General water, sanitation and flood protection sector(80%):housing finance(10%):sub-national government administration(5%):housing construction(5%) Theme(s) Urban services and housing for the poor(25%):participation and civic engagement(25%):pollution management and environmental health(24%):municipal governance and institution building(13%):personal and property rights(13%) Public Disclosure Authorized Prepared by Reviewed by ICR Review Coordinator Group Hanne Charlotte Moen George T. K. Pitman Christopher David Nelson IEGSD (Unit 4) 2. Project Objectives and Components a. Objectives The project development objective according to the Development Credit Agreement (page 22) is: "to alleviate poverty in urban areas by improving the environmental and living conditions of the urban poor in the Project Cities through the use of participatory planning methods and the promotion of more inclusive and pro-poor urban planning processes". The development objective is phrased slightly differently in the Project Appraisal Document (page 3): "to alleviate poverty in urban areas by improving the living and environmental conditions of the urban poor using participatory planning

2 methods, and influencing planning processes to become more inclusive and pro-poor". The development objectives stated in the Development Credit Agreement is the basis for this Review. b. Were the project objectives/key associated outcome targets revised during implementation? No c. Components The project was implemented in four cities in Vietnam (Ho Chi Minh City, Can Tho, Hai Phong and Nam Dinh) and was expected to reach 865,000 low-income beneficiaries directly and 1,070,000 indirectly (PAD page 10). In each of the cities the project had six components. A seventh sector-level component financed the development of a National Urban Upgrading Program. 1: Tertiary Infrastructure Upgrading and Service Improvements (Appraisal: US$ 86.97million. Actual: US$88.60 million). This component financed a package of tertiary infrastructure and services in low-income areas, with improvements to water supply, drainage, paved access, electricity, sanitation and solid waste management. In some low-income areas it also included the construction or rehabilitation of market infrastructure, public restrooms, and educational or health facilities if identified as priorities by communities. Community committees were set up and were involved in preparing the Community Upgrading Plan (CUP), which set out the details of the infrastructure investments, participation process, community management structures etc. 2: Complementary Primary and Secondary Infrastructure (Appraisal: US$ million. Actual: US$ million) This component financed the improvement of primary and secondary infrastructure (drainage networks, pumping stations, road and bridge improvements, and water supply) where necessary to complement the tertiary improvements, building on experience showing that the effectiveness of community level infrastructure improvements depends on whether they are correctly linked to the larger network and whether the network is capable of absorbing the new additions. This work involved city departments such as the Department of Transport and Public Works. 3: Resettlement Housing (Appraisal: US$34.32 million. Actual: US$66.60 million) This component financed resettlement of migrants who had settled along canals and lakes, creating health hazards. It was estimated that 5,363 households would require resettlement. At the time of appraisal resettlement sites had been secured, and the project provided infrastructure for these sites. Bank funds also financed housing for the poorest whom were unable to afford to build houses on the serviced resettlement plots. The participating cities preferred to build mixed income developments with a combination of low-cost and higher cost buildings/plots. Bank funds only financed the proportion of site development costs related to the resettlement housing. 4: Land and Housing Management (Appraisal: US$ 1.70 million. Actual: US$1.40 million) Depending on the existing capacity of the relevant departments, this component financed technical assistance and training to strengthen the administration of land use certificates, which had just been decentralized to local governments. Assistance was provided in each city for the (i) establishment of a networked land administration and housing management information system; (ii) completion of cadastral maps to accelerate the issuance of certificates; (iii) support to administrative and organizational restructuring; and (iv) simplifying procedures for providing land use permits. In order to give women more equal access to assets womens names would be included on all certificates issued. 5: Housing Improvement Loan Program (Appraisal: US$15.52 million. Actual: US$13.10 million) This component provided access to credit to poor households to improve their houses and strengthened the housing microfinance system to facilitate the provision of housing improvement loans to the urban poor. Five partner organizations provided micro-credit to eligible households for housing improvements (These were the Women's Unions in Haiphong, NamDinh and Ho Chi Minh City; the Capital Aid Fund for Employmentof the Poor (CEP) in Ho Chi Minh City; and the Mekong Housing Bank (MHB) in Can Tho (in association with the Women's Union of Can Tho). This component financed capital to be on-lent as housing improvement micro-loans to low-income households according to arrangements outlined in an Operations Manual and individual Implementation Agreements with each partner. The amount of capital to be provided to each partner was determined according to the potential customer base and the partner's lending capacity. Technical assistance, training and equipment was also provided to strengthen the implementing partners. 6: Capacity Building (Appraisal: US$10.11 million. Actual: US$7.30 million) This component financed consulting services and goods/equipment for capacity building for the following activities: 1. Technical assistance and training at the city/district/ward levels for: a) participatory planning, design and implementation; and b)

3 strengthening the capacity of utility companies and District Authorities to improve operations and maintenance of infrastructure financed; 2. Technical assistance and training for communities to organize more effectively to participate in project planning and design; and Information Education and Communication (IEC) campaigns particularly to promote better sanitation practices; and 3. Technical assistance and training for the Porject management Unit (PMU) in each city in: a) project management; b) procurement and contract management; c) financial management; d) audits; e) RAP and Environmental Monitoring; and f) monitoring and evaluation of the project. Equipment to support the PMUs would also be financed. 7: Development of a National Urban Upgrading Program (Appraisal: US$1.00 million. Actual: US$ 0.60 million) This component provided TA and training to the Ministry of Construction (MOC) to help: (i) develop a National Urban Upgrading Program that would apply the principles of urban upgrading used for VUUP to cities across Vietnam, and (ii) evaluate the progress and impact of VUUP on a continuous basis. d. Comments on Project Cost, Financing, Borrower Contribution, and Dates Project costs: Total project costs at appraisal consisted of the costs for each of the seven components as well as physical and price contingencies (US$26.74 million) and taxes (US$14.51 million), amounting to US$ million. The total actual costs include contingencies and taxes and amount to US$ million. The components remained unchanged during project implementation, but their scope was expanded and costs and financing increased, especially for Component 2 in 2009 when Additional Financing (AF) of US$ 160 million was approved. The additional financing was used to co- finance (i) increases in the cost of Components 1, 2 and 3 arising from unforeseen increases in prices of labor and construction materials (US$31.20 million), and (ii) an expansion needed for improvements of the THLG canal in Ho Chi Minh City (about US$129 million, including the financing gap of US37.5 million created by the withdrawal of AFD (French funding) which had increased by US$50 million due to inflation, US$16 million for additional waste water interceptor, and US$24 million for price and physical contingencies). With approval of the AF, the estimated project cost rose to US$580 million. However, the unexpected increased price competition among contractors during the extension period resulted in substantial cost savings mainly in Ho Chi Minh City and together with all other unused contingencies amounts, the project had to cancel US$32.50 million. Financing: An IDA Credit of Special Drawing Rights (SDR) million, then equivalent to US$ million, was approved April 24, At the time of project closing US$ million had been disbursed and US$20.23 million was cancelled. An additional IDA Credit of SDR million, then equivalent to US$ million, was approved July 2, 2009 and at the time of project closing US$ has been disbursed and US$10.81 million was cancelled. The reduction in the value of the Credits over the life of the project was the result of SDR/US$ exchange rate variations. The Government of Japan provided US$ 3.80 million of the US$4.84 million anticipated at approval. Part of the initial funding plan was a loan of US$37.50 million from the Agence Franaise de Dveloppement (AFD, French bilateral aid agency) intended for infrastructure improvements in the Tan Hoa-Lo Gom (THLG) basin in Ho Chi Minh City. This loan did not materialize and this component was therefore later financed by the additional IDA financing. Borrower contribution: Borrower contribution was originally planned to be US$ million. Actual borrower contribution was US$151.1 million. In addition communities contributed US$0.70 million of the US$4.20 million estimated at appraisal. Dates: The project was approved on April 29, 2004 and was restructured July 02, 2009 when its closing date was extended by 24 months from December 31, 2012 to December 31, Two new outcome indicators were added in the 2009 Additional Financing Project Paper (AFPP) in order to measure the outcomes of the construction of the Tan Hoa - Lo Gom (THLG) Canal, to be financed from the Additional Financing. The AFPP also set targets for some of the outcome indicators and intermediate results indicators as none were set in the PAD. In line with the two-year extension of the Credit closing date, the date for achievement of the targets were changed from December 31, 2012 to December 31, Relevance of Objectives & Design a. Relevance of Objectives The objectives of the project are highly relevant and in line with both Vietnamese and Bank strategies and policies at appraisal as well as at

4 closing of the project. They are in line with the Government of Vietnams orientation Master Plan for Urban Development to Year 2020, where the lack of infrastructure, the lack of sanitary facilities and limited institutional capacity are highlighted. Moreover the objectives are highly relevant to the 2003 Vietnam Comprehensive Poverty Reduction and Growth Strategy, in which one of the social and poverty reduction objectives are provision of basic infrastructure services to extremely disadvantaged groups and urban poor people. The objectives are relevant to the needs of project beneficiaries. Project objectives are in line with the Bank's Country Assistance Strategy for , in which addressing the needs of the urban poor is highlighted under the thematic are of enhancing equitable, socially inclusive and sustainable development. The objectives are in line with the current Country Partnership Strategy , where urban challenges including environmental pollution, clean water and housing for the urban poor are addressed in several sections. Rating: High Rating High b. Relevance of Design The Development Objective was to alleviate poverty in urban areas by improving the environmental and living conditions of the urban poor in the Project Cities through the use of participatory planning methods and the promotion of more inclusive and pro-poor urban planning processes. The underlying project logic linking the project outputs to the intended outcomes is relatively clear. The outcome indicators are logical and link well to the development objectives. The causal chain between the main investments and intended outcomes is clear. The largest proportion of the project finances were used to provide infrastructure to improve water and sanitation. Coupled with improved housing conditions and credits for home improvements and income generation, this was intended to lead to better living conditions in terms of reduced incidence of diseases, reduced flooding and increased property values, which again would lead to poverty alleviation. This causal chain built on the assumption that improved infrastructure would lead to poverty alleviation. The project components and related activities are relevant measures to improve the environmental and living conditions for the urban poor. Adequate sanitary facilities would avoid discharge from toilets directly into drains, canals and lakes and thus improving public health. Drainage, flood control and infrastructure upgrading would mitigate disruption to increased service coverage and improve the quality of life and commercial activities of those living in hazardous locations. The effectiveness of these improvements is ensured through the provision of complementary primary and secondary infrastructure. Simplified procedures for issuing ownership and land use certificates, and improved access to loans for improved housing, would support home improvements and resettlement housing. Capacity- building of authorities and citybased project management units (PMUs) was expected to build the institutional capacity necessary to ensure adequate implementation and sustainability. Although not necessary, the component of developing a national urban upgrading program is contributing to maximize the impact of the program in terms of ensuring learning and replication in other locations. All the components are consistent with, and contribute to, the overall goal of alleviating poverty in urban areas. Moreover, the emphasis on participatory methods and inclusive planning processes were essential elements expected to contribute to more efficient implementation and increased sustainability. Rating. Substantial Rating Substantial 4. Achievement of Objectives (Efficacy) PHREVISEDTBL Objective 1 Objective To alleviate poverty in urban areas by improving the environmental and living conditions of the urban poor in the Project Cities.

5 Rationale Outputs: 29,687 new household water connections (Indicator also contains sewer and electricity connections, but achieved value refers to water connections only). No target, but an original estimate was 18, kms of drainage constructed or rehabilitated (AF target 120 km). 243 kms of new and/or rehabilitated primary and secondary access roads constructed and 340 kms of tertiary roads (no targets but the Bank's supervision reports indicate 150 km as a target). Construction of 2,641 land plots and 2,383 apartments. However, only 2,547 plots and apartments were utilized for resettlement within the project (original indicator was number of serviced plots and/or low-cost housing unit, provided to low-income households. AFPP amended this indicator as number of serviced plots or housing provided for resettlement) % of households in upgraded areas received Building Ownership and Land Use Certificates/Land Use Certificates (no target or baseline). Technical assistance and equipment were provided to relevant agencies as planned. 51,109 loans were extended for home improvement (AF target 29,000). Repayment rate of 98% of housing improvement loans to low-income households (AF target 99%). Repayment risks were less than 5% of portfolio for housing improvement loans to low-income households (AF target less than 5%). National Urban Upgrading Program (NUUP) developed and legalized June Intermediate outcomes: The project has achieved significantly improved service coverage for water from 40-50% at appraisal to over 98% (water supply) and over 94% (sanitation) at closing (combined achievements for all cities). It is likely that improved water supply and sanitation has led to a reduction of selected diseases, but there are no epidemiological studies confirming this. Appraisal had no baseline or targets, and the Mid-Term Review attempted to measure achievements, but results were inconclusive (ICR factsheet (a)). Increased property values in the project areas rose by % (no baselines or targets). However, this cannot be attributed to project interventions alone (ICR p. 17). There was substantial reduction in time and cost of acquiring Building Ownership and Land Use Certificates/Land Use Certificates in all project areas, from over 8 months to days. However, the target of 15 days was only achieved in Can Tho. Operational sustainability of participating microcredit lending institutions was achieved for four institutions but not for the three Womens' Unions Flooding in THLG Canal area was completely eliminated by closing Significant improvement of water quality in the THLG Canal from a COD level of 811 mg/l in 2009 to only mg/l in 2015 (although short of the overly ambitious target of 50 mg/l) and improvement in water quality in the other cities. Outcomes: Through the upgrading of service provision in low-income areas in the four cities the project was expected to improve the environmental and living conditions of the urban poor and thereby contribute to alleviating poverty. Many of the supporting project activities were substantially achieved, project investments generally met or exceeded expectations where there were measurable results, and more direct beneficiaries than anticipated were reached. Still, based on the information in the PAD and ICR the project faced problems with adequately capturing information to measure and record changes in the incidence of poverty in project areas. This was due to an inadequate M&E system and shortcomings related to the lack of baselines and targets as described in the ICR. Baseline data and targets were not included in the PAD and the AFP paper set targets for some of the outcome indicators and intermediate results indicators, but not for all. However, the Region has provided additional information and clarifications related to the alternative methodologies that were implemented to capture qualitative and quantitative information in order to measure progress against the PDO. The data, including the impact study which was conducted at the end of the project, shows substantial poverty reduction in the project areas during the project implementation. Based on this additional information IEG finds it plausible that the project resulted in substantial poverty reduction in the project areas. Attribution: As discussed several places in the ICR it is generally difficult to attribute the positive changes solely to the project due to the weak monitoring framework. It is clear that other external factors also contributed to the increased property values. With the additional information provided by the Region IEG still finds that there is sufficient evidence to suggest that the project has contributed substantially to the positive changes in the project areas. Rating: Substantial

6 Rating Substantial 5. Efficiency Economic and Financial Efficiency: The economic internal rates of return (EIRR) for infrastructure investments, which accounted for over 80% of the total costs, were estimated at appraisal, in 2009 at the time of AF, and after project closing. The EIRRs were higher at project closing than at appraisal, ranging between 58-67% for tertiary infrastructure and 33-60% for primary and secondary infrastructure. Because of the use of constant prices at appraisal and current prices at closing, the EIRRs are not directly comparable. Still, the ICR argues plausibly that even when correcting for this error the returns would not be reduced below 12% (ICR p 20 and ICR Annex 3 p 35). Operational and Administrative Efficiency: Project management was not fully efficient during the initial stages of the project due to lacking experience with the Banks requirements and with participatory planning as well as poor management and coordination between PMUs and other stakeholders. This led to substantial delays with procurement, in terms of securing community participation and changes in engineering designs. Delays in counterpart funds for land acquisition and compensation payments led to resettlement taking more time than anticipated. However, with increased capacity-building support the efficiency of the PMUs and other implementing partners improved significantly. All housing improvement loans were fully repaid at project closing indicating efficient fund use. Efficiency Rating Substantial a. If available, enter the Economic Rate of Return (ERR) and/or Financial Rate of Return (FRR) at appraisal and the re-estimated value at evaluation: Rate Available? Point value (%) *Coverage/Scope (%) Appraisal 0 ICR Estimate 0 0 Not Applicable 0 Not Applicable * Refers to percent of total project cost for which ERR/FRR was calculated. 6. Outcome The relevance of objectives is High and the relevance of design is Substantial. Despite the fact that the project has faced problems with the results framework, including missing targets, the ICR, the TTL interview and additional information provided by the Region has given plausible evidence and arguments for the achievements of the project. IEG therefore concludes that the project has achieved its objectives, and efficacy is therefore rated Substantial. The EIRRs for the main components of the project is found to be satisfactory and the efficiency is rated Substantial. a. Outcome Rating

7 7. Rationale for Risk to Development Outcome Rating The risk of inadequate operation and maintenance is low as the infrastructure is not technically complex and responsibility is delegated to experienced city government departments and other utilities. The housing improvement loan programs managed by the Womens Unions were, however, not financially sustainable, and therefore there is a risk related to finding other ways of sustaining the provision of access to credit for the poor households. There is a risk of natural disasters such as flooding, but all project cities have the financial capacity to pay for repairs. Despite these moderate risks the enhanced capacity of the city administration, implementing partners and the communities is likely to be sustained in the long run and the strong government ownership and the participatory approach reduce the risk to development outcome. a. Risk to Development Outcome Rating Modest 8. Assessment of Bank Performance a. Quality-at-Entry Appraisal benefited from an expert team adequately resourced with a number of technical specialists for two and a half years prior to the project approval. Comprehensive background analyses were based on a wealth of Bank experiences from similar upgrading projects in Asia and elsewhere. This constituted a comprehensive basis for a project design that was complex, but comprehensive and relevant to the beneficiaries. The strong government ownership, the decentralized decision-making and the participatory approach reduce the risk to development outcome. Safeguards were adequately covered and appropriate. A weakness was the Bank's inability to ensure a comprehensive results framework with clear targets able to measure the effect of the project on poverty alleviation from the onset of the project. Quality-at-Entry Rating b. Quality of supervision The Bank provided a multi-diciplinary supervision team with highly adequate expertise in engineering, project management, financial management, procurement, monitoring, and safeguards. Most of these experts were based in the country office and therefore available for short field visits and regular communiction with the implementing partners. Additionally the Bank had two full supervision visits and many shorter site visits a year. Through this close supervision the multi-disciplinary Bank supervision team were able to identify and proactively address challenges related to procurement, resettlement and community participation as they occurred. The Bank invested substantial efforts in working with city authorities to ensure efficient management of the project resources, it worked closely with the PMUs and played an instrumental role in setting up microfinance arrangements. Moreover, the Bank managed to ensure high compliance with safeguard policies. Through close financial management the Bank was able to identify shortcomings related to procurement and provided adequate training and technical assistance to PMUs and other implementing partners in order to rectify the problems. However, even though the additional financing provided the Bank with the opportunity to revise the results framework, establish adequate time-bound targets and the resources to measure poverty and strengthen the M&E component of the project, this was not sufficiently ensured across the project. Therefore the weakness of the results framework with missing targets remained partly unresolved throughout the project period. In light of this important aspect the quality of supervision was lower than expected. Quality of Supervision Rating Moderately Unsatisfactory Overall Bank Performance Rating

8 9. Assessment of Borrower Performance a. Government Performance The Government of Vietnam was highly committed to the project and contributed approximately 30% of the funding. The GoV supported the Bank in developing innovative solutions, maintained good relationships with other development partners and provided a timely request for the additional funding. In addition the government supported strong beneficiary/stakeholder consultations and involvement with active participation of community members, although this approach was seen to be new to Vietnam. The strong community participation was one of the success factors of the project and therefore important although it led to some initial delays. The government must, however, share the responsibility for the poor M&E framework and lack of targets that led to problems assessing outcomes and attribution. The baseline and final household survey do, however, mitigate the potential problems arising from the M&E shortcomings. Government Performance Rating b. Implementing Agency Performance The project was largely decentralized and implemented by the four city governments through their Project Management Units (PMUs). The city governments were committed to the project and willing to use their own financial resources for project preparations not covered by the project grants. They supported the participatory approach emphasizing the importance of active involvement of the community members. The implementing agencies executed their financial management adequately by employing qualified staff, and establishing and maintaining appropriate accounting and internal control systems. There were also some weaknesses in the performance of the city governments. Due to the PMU staffs limited experience with the Banks procurement requirements the project suffered severe delays in its initial phase. After receiving TA and training, however, the PMUs improved significantly. Even so, according to the ICR (page 10) the PMUs were understaffed during most of the implementation period which led to a lack of emphasis on M&E. The Region provides conflicting information, but information from the ICR and TTL interview has been given more weight on this issue. Moreover, the city governments centered their monitoring and reporting on inputs/outputs and not on the outcome/impact levels. This is a clear weakness, and has contributed to the difficulty in measuring results attributable to the project. The implementing agencies of the Housing Improvement Loan Program performed well with more loans than expected and a repayment rate near 100%. Implementing Agency Performance Rating Overall Borrower Performance Rating 10. M&E Design, Implementation, & Utilization a. M&E Design The M&E design at the time of appraisal was based on a logframe style results framework called Project Design Summary with 6 outcome/impact indicators and 10 output indicators. The objectives were clearly specified but the indicators did not clearly reflect the linkages between the project outputs and the overall intention of poverty alleviation. There were no indicators in the results framework to measure poverty reduction. Additionally, there were no targets set to measure achievements and some of the supporting indicators for intermediate outcomes were not measurable or sufficiently specific, such as reduced flooding or reduced incidence of selected diseases. Other indicators suffered attribution problems such as change in property values. The Project Design Summary refers to baseline surveys for measuring the 6 outcome/impact indicators and according to the PAD (page 23) a baseline socio-economic survey was carried out in the project areas during preparation. However, according to the ICR baselines were generally missing and no baseline values or targets were actually set in the PAD (PAD Annex 1 and ICR Fact Sheet and pp 10, 15 and 19). The fact that the project had no baselines and targets was confirmed in the TTL interview but this is disputed by additional information provided by the Region holding that baseline data was collected although not included in the PAD. The Region

9 provides an annex outlining the baseline and target values and based on this additional information the M&E design is assessed to be more satisfactory. b. M&E Implementation M&E had low priority due to the constant understaffing in the PMUs. The PMUs focused on reporting on output level indicators and to a lesser extent on outcomes. They needed Bank assistance to track non-physical indicators such as reduced incidence of selected diseases (ICR p 10). Two evaluations were financed under the project: (i) an evaluation of the housing improvement loan program in 2012; and (ii) an impact assessment covering the entire project in The impact assessment does to some degree compensate for the inadequate monitoring. At the time of the additional financing (2009) targets were set for several of the indicators, and some baselines were developed retrospectively. However, no indicators to measure the objective of alleviating poverty were included. Therefore the weaknesses in design, including indicator specification were only partly fixed during implementation. c. M&E Utilization The M&E data was collected by the PMUs and was, for the most part, used to measure inputs and outputs rather than focusing on outcomes and the fine-tuning of strategic choices. It was used by the Bank to follow the implementation of the physical work as stipulated in project plans. After the additional funding was released and some targets were set, the Bank monitored data to keep track of the indicators against the set targets, enabling the Bank to remedy problems and direct funding to the most appropriate areas. The PMUs, however, used the M&E system to monitor physical progress and compliance with plans only. Generally, the M&E activities paid too little attention to the outcomes to be achieved through the project, and they are therefore not seen to have been instrumental in influencing strategic decisions during the cource of the project. The Impact Assessment served the important purpose of measuring the achievements of the project. The evaluation of the home improvement loan program was conducted late into the project life and were therefore mainly used for the preparation of the Government's project completion report (see ICR Annex 7). M&E Quality Rating Substantial 11. Other Issues a. Safeguards The project was classified as Environmental Category A under OP/BP 4.01 Environmental Assessment. Two other safeguards were triggered: Involuntary Resettlement (OP/BP 4.12), and Indigenous Peoples (OP 4.20). The construction of the THLG Canal financed with the additional financing also triggered safeguards on Physical Cultural Resources (OP 4.11). The Borrower developed an Environmental Management Framework, a Resettlement Policy Framework and an Ethnic Minority Development Strategy (EMDS) in order to mitigate potential risks. Two types of environmental assessments were prepared for each project city, and a participatory approach was used to develop the Community Environmental Management Plans for tertiary infrastructure. Resettlement Action Plans were developed in each project city. As the project was seen to potentially adversely affect about 25 Khmer minority group households a Minority Development Strategy was agreed but since these households were scattered and did not form a cohesive group it was decided that a separate ethnic minority development plan was not needed (ICR p 12). In order to mitigate negative impacts on shrines and temples near the Tan Hoa-Lo Gom canal mitigation measures were developed in dialogue with the relevant property owners. Action plans for resettlement were developed based on the Resettlement Policy Framework, and initially 4,800 households were anticipated to be resettled permanently and 16,700 households were to be temporarily adversely affected. Efforts were made to reduce the need for resettlements and as a result the final relocations were 40% lower than anticipated. However, the number of partially-affected households were 14% higher than at appraisal. During project implementation, about 2,640 serviced plots and 2,400 apartments were developed for

10 resettled households, on a total area of about 63 ha (ICR annex 9). As described in the ICR there was satisfactory compliance with all relevant safeguards throughout the project, although resettlement and compensation to households suffered severe delays. Compensation and resettlement costs were higher than planned for because of unforeseen increases in land and asset prices (ICR pp 12, 13). b. Fiduciary Compliance Financial management: Financial management performance was generally satisfactory with appropriate systems, qualified staff and timely deliveries of financial reports of acceptable quality. Unqualified externally audited financial statements were submitted each year for all four cities with two exceptions. In both cases the auditors recommendations were subsequently implemented and the reports later received unqualified opinions (ICR p 13). The ICR does not comment on whether all Bank resources were fully accounted for by project closure or issues of corruption or misuse of funds. However, the TTL interview clarified that resources were fully accounted for and there had not been any corruption or misuse of funds. Procurement: During the initial years of implementation procurement was unsatisfactory with severe delays. This was due to lack of knowledge of Bank procedures and unnecessary bureaucratic procedures. Subsequently the Bank and the government intervened with technical assistance and training, which led to substantial improvements in contract management. According to the ICR (p 13) procurement was rated Satisfactory at closing and all the 450 contracts financed by the project adhered to the Banks procurement guidelines. c. Unintended impacts (Positive or Negative) --- d. Other Ratings Ratings ICR IEG Outcome --- Risk to Development Outcome Modest Modest --- Bank Performance --- Borrower Performance --- Quality of ICR Modest --- Reason for Disagreements/Comment Note 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, The "Reason for Disagreement/Comments" column could cross-reference other sections of the ICR Review, as appropriate. 13. Lessons

11 The ICR provides 10 lessons. The five most relevant, with editing, are summarized below (ICR p 26 and 44): Ensuring meaningful community participation among the urban poor is efficient and can deliver small-scale infrastructure that better meets the needs of communities than infrastructure projects planned in a traditional top-down fashion, but it is time- and labor-intensive. Investments in complimentary higher-level infrastructure are necessary to secure the benefits of community-based infrastructure. Transformational urban upgrading projects create wider benefits that cannot be captured by traditional measures. Urban infrastructure projects are inherently complex and require strong project (and especially contract) management skills. Improved access to credit is essential for an urban upgrading project, but does not need to be provided by the project itself. 14. Assessment Recommended? No 15. Comments on Quality of ICR The ICR is generally well written, consistent and concise and contains detailed and sufficient information to evaluate project implementation challenges and infrastructure and housing achievements. Because of the lacking evidence of project achievements due to the absence of baselines and targets, the ICR relies heavily on external studies and research to argue for the projects efficacy in terms of poverty reduction. However, although these studies may provide some support for the likelihood of asserted project results, they are not a substitute for precise and measurable outcome data directly attributable to project inputs. The additional information from the Region conflicts with the information in the ICR on several issues, which contributes to lowering the credibility of the ICR. a. Quality of ICR Rating Modest