IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA IDA IDA-47610) ON A CREDIT

Public Disclosure Authorized Document of The World Bank Report No: ICR2482 Public Disclosure Authorized IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-0 IDA-1 IDA-47610) ON A CREDIT Public Disclosure Authorized IN THE AMOUNT OF SDR 143.6 MILLION (US$ 205.6 MILLION EQUIVALENT) TO THE PEOPLES REPUBLIC OF BANGLADESH FOR THE MUNICIPAL SERVICES PROJECT Public Disclosure Authorized Urban Sector Unit Bangladesh Country Unit South Asia Region December 27, 2012

CURRENCY EQUIVALENTS (Exchange Rate Effective November 30, 2012) Currency Unit = Bangladeshi Taka 1.00 = US$ 0.01 US$ 1.00 = Tk. 81.3 FISCAL YEAR July 1 June 30 ABBREVIATIONS AND ACRONYMS ADB BMDF CAS ERR FOAP GOB IDA LGED MDF MPRC MSU NPV PMU PPA Asian Development Bank Bangladesh Municipal Development Fund Country Assistance Strategy Economic Rate of Return Financial and Operational Action Plan Government of Bangladesh International Development Association Local Government Engineering Department Municipal Development Fund Municipal Performance Review Committee Municipal Support Unit Net Present Value Project Monitoring Unit Participatory and Partnership Approach Vice President: Isabel M. Guerrero Country Director: Ellen A. Goldstein Sector Manager: Ming Zhang Project Team Leader: Zahed H. Khan ICR Team Leader: Janis D. Bernstein

PEOPLES REPUBLIC OF BANGLADESH Municipal Services Project CONTENTS Data Sheet Page A. Basic Information B. Key Dates C. Ratings Summary D. Sector and Theme Codes E. Bank Staff F. Results Framework Analysis G. Ratings of Project Performance in ISRs H. Restructuring I. Disbursement Graph Main Document 1. Project Context, Development Objectives and Design... 1 2. Key Factors Affecting Implementation and Outcomes... 7 3. Assessment of Outcomes... 15 4. Assessment of Risk to Development Outcome... 24 5. Assessment of Bank and Borrower Performance... 24 6. Lessons Learned... 27 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners... 29 Annex 1. Project Costs and Financing... 30 Annex 2. Outputs by Component... 31 Annex 3. Economic and Financial Analysis... 64 Annex 4. Bank Lending and Implementation Support/Supervision Processes... 72 Annex 5. Beneficiary Survey Results... 74 Annex 6. Summary of Borrower's ICR and/or Comments on Draft ICR... 80 Annex 7. List of Supporting Documents... 99 Map IBRD 33368

A. Basic Information Country: Bangladesh Project Name: Municipal Services Project ID: P041887 L/C/TF Number(s): IDA-0,IDA- 1,IDA-47610 ICR Date: 07/19/2012 ICR Type: Core ICR Lending Instrument: SIL Borrower: GOB Original Total Commitment: XDR 100.00 M Disbursed Amount: XDR 141.86 M Revised Amount: XDR 143.60 M Environmental Category: B Implementing Agencies: Local Government Engineering Department, Bangladesh Municipal Development Fund Cofinanciers and Other External Partners: B. Key Dates Process Date Process Original Date Revised / Actual Date(s) Concept Review: 08/31/1997 Effectiveness: 10/25/1999 10/25/1999 Appraisal: 06/13/1998 Restructuring(s): 2/10/2005 12/21/2007 08/23/2010 Approval: 03/16/1999 Mid-term Review: 03/02/2002 Closing: 06/30/2005 06/30/2012 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Risk to Development Outcome: Bank Performance: Borrower Performance: Satisfactory Moderate Moderately Satisfactory Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Satisfactory Government: Moderately Satisfactory Quality of Supervision: Moderately Satisfactory Implementing Agency/Agencies: Satisfactory Overall Bank Performance: Moderately Satisfactory Overall Borrower Performance: Satisfactory i

C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments (if Indicators Performance any) Potential Problem Project at any time (Yes/No): Problem Project at any time (Yes/No): DO rating before Closing/Inactive status: Yes Yes Moderately Satisfactory Rating Quality at Entry (QEA): Highly Satisfactory Quality of Supervision (QSA): Moderately Unsatisfactory D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) General water, sanitation and flood protection sector 20 20 Other social services 17 17 Sub-national government administration 58 58 Urban Transport 5 5 Theme Code (as % of total Bank financing) Municipal finance 22 22 Municipal governance and institution building 23 23 Participation and civic engagement 11 11 Pollution management and environmental health 22 22 Urban services and housing for the poor 22 22 E. Bank Staff Positions At ICR At Approval Vice President: Isabel M. Guerrero Mieko Nishimizu Country Director: Ellen A. Goldstein Frederick Thomas Temple Sector Manager: Ming Zhang Frannie F. Humplick Project Team Leader: Zahed H. Khan Jonathan S. Kamkwalala ICR Team Leader: Janis D. Bernstein ICR Primary Author: Janis D. Bernstein F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The Project Development Objective (PDO) is to improve environmental and infrastructure service delivery in urban areas. See section 1.2 for the entire PDO. ii

Revised Project Development Objectives (as approved by original approving authority) The Project Development Objective did not change during the implementation of the original project or the Additional Financing. (a) PDO Indicator(s) Indicator Baseline Value Original Target Values (from approval documents) Formally Revised Target Values Actual Value Achieved at Completion or Target Years Indicator 1.1 Municipalities able to plan, prioritize, and implement service improvement by end of project. Value Phase 1: 16 AF 2010 (Quantitative or 0 183 Qualitative) Phase 2: 40 183 Date achieved 03/16/1999 06/30/2001 06/30/2012 06/30/2012 By 2010, 92 cities had received MSU capacity building and support for community Comments mobilization. During the AF 2010, the MSU extended the capacity building to another 91 cities. Indicator 1.2 Management reform in municipalities completed. Value (Quantitative or Qualitative) 0 Phase 1: 16 Phase 2: 40 AF 2010 183 183 Date achieved 03/16/1999 06/30/2001 06/30/2012 06/30/2012 Comments By 2010, 133 cities achieved management reform. By June 30, 2012, another 50 completed reforms in project planning, tax and water tariff collection, and managing municipal accounts. Indicator 1.3 Demand driven and cost recovery measures introduced for revenue generating sub-projects. Value (Quantitative or Qualitative) 0 Phase 1: 16 Phase 2: 40 AF 2010 168 168 Date achieved 03/16/1999 06/30/2001 06/30/2012 06/30/2012 By 2010, 112 cities had introduced these measures; under the AF2010, another 56 cities were Comments required to involve participation in project selection and to initiate measures to improve cost recovery. Indicator 2.1 MDF created and operational for on-lending to municipalities. BMDF operations BMDF operations streamlined and Value streamlined and its BMDF created and coordination (Quantitative or MDF not established coordination with operational with LGED and Qualitative) LGED and other other relevant agencies strengthened agencies strengthened Date achieved 03/16/1999 09/1/2004 06/30/2012 06/30/2012 iii

Comments Indicator 2.2 Value (Quantitative or Qualitative) Under the AF2010, a new indicator was introduced - to streamline its operations and to strengthen coordination with LGED (for details on all of the measures taken, see Sec. 3.2 (b) of the ICR main text. Municipal Performance Review Committee (MPRC) established and operational by end of Phase 1 MPRC does not exist MPRC established and operational Established on July 5, 1999 Operational on March 16, 2002 Date achieved 03/16/1999 03/16/2002 Comments Indicator 3.1 Value (Quantitative or Qualitative) The MPRB was established on July 5, 1999 and fully operational by March 16, 2002, when the first meeting was held. Physical investments targeting poverty and environmental improvements through community participation. No target established in PAD because of 595 works 0 the demand driven 595 works packages packages nature; by the AF of completed with BMDF completed with 2010, 454 works funding in 154 towns BMDF funding packages had been financed. Date achieved 03/16/1999 06/30/2005 06/30/2012 06/30/2012 Phase II investments involved participation in planning and implementing sub-projects aimed Comments at poverty and environmental improvement. Specific targets for Phase 1 are described below. (b) Intermediate Outcome Indicator(s) Indicator Baseline Value Original Target Values (from approval documents) Indicator 1.1 Municipal Development Fund established by July 1, 2000. 0 BMDF established by BMDF established July 1, 2000 Date achieved 03/16/1999 03/9/2002 03/9/2002 The delay was due mainly to the severe floods of 2000 which diverted project staff attention to Comments finance urgent repairs, and the transitional government putting final steps on hold until new government took over. Indicator 1.2 Municipal Performance Review Committee (MPRC) established by June 30, 1999. 0 Established by June MPRC established 30, 1999 Date achieved 03/16/1999 07/5/1999 07/5/1999 Comments Indicator 1.3 Municipal Support Unit (MSU) fully staffed and working, as agreed with the Bank, by July 1, 1999. 0 MSU fully staffed and working by July 1, 1999 MSU fully staffed and working iv Formally Revised Target Values Actual Value Achieved at Completion or Target Years

Date achieved 03/16/1999 10/25/1999 10/25/1999 Comments Indicator 1.4 Partnership and Participatory Approach (PPA) designed, in agreement with the Bank, by September 30, 1999. PPA designed in PPA designed in 0 agreement with the agreement with Bank Bank by September 30, 1999 Date achieved 03/16/1999 09/30/1999 09/30/1999 Comments Indicator 1.5 All Phase I participating city corporations and municipalities implement Financial and Operational Action Plan by June 30, 2000. 0 All Phase I CCs and municipalities implement FOAP by June 30, 2000 All Phase I CCs and municipalities implemented FOAPs Date achieved 03/16/1999 07/30/2000 07/30/2000 Comments Indicator 1.6 At least 40 other municipalities implement FOAP by June 30, 2004. 16 Phase I CCs and municipalities implemented FOAPs 40 40 Date achieved 07/30/2000 06/30/2004 06/30/2004 Comments Indicator 2.1 Roads constructed/rehabilitated. 0 50% Phase I road works completed by June 30, 2000 100% Phase I road works completed by June 30, 2001 Works completed Date achieved 03/16/1999 12/31/2006 12/31/2006 Comments By 2005, 95% of the works were completed. Due to the need to redirect LGED s efforts to urgent repair works related to the 2000 floods, the project did not meet original target dates. Indicator 2.2 Drainage facilities improved. 67% (25 km) completed by June 30, 2000 0 Works completed 100% (37 km) completed by June 30, 2001 Date achieved 03/16/1999 12/31/2006 12/31/2006 Comments By 2005, 95% of the works were completed. Due to the need to redirect LGED s efforts to urgent repair works related to the 2000 floods, the project did not meet original target dates. Indicator 2.3 Construction of public toilets completed. v

0 68% (32 toilets) completed by June 30, 2000 Works completed 100% (47 toilets) completed by June 30, 2001 Date achieved 03/16/1999 12/31/2006 12/31/2006 Comments By 2005, 95% of the works were completed. Due to the need to redirect LGED s efforts to urgent repair works related to the 2000 floods, the project did not meet original target dates. Indicator 2.4 Rehabilitation of water supply facilities completed. 75% (9 km) of water lines completed in Khulna by June 30, 2000 0 Works completed 100% (12 km) of water lines completed in Khulna by June 30, 2001 Date achieved 03/16/1999 12/31/2006 12/31/2006 Comments By 2005, 95% of the works were completed. Due to the need to redirect LGED s efforts to urgent repair works related to the 2000 floods, the project did not meet original target dates. Indicator 2.5 Municipal markets upgraded. 50% (8) completed by June 30, 2000 0 Works completed 100% completed by June 30, 2001 Date achieved 03/16/1999 12/31/2006 12/31/2006 Comments By 2005, 95% of the works were completed. Due to the need to redirect LGED s efforts to urgent repair works related to the 2000 floods, the project did not meet original target dates. Indicator 2.6 Bus/truck terminals constructed or rehabilitated in each city corporation and municipality. Two new bus terminals constructed by June 30, 2002 0 Seven new bus terminals constructed by June 30, 2001 Three bus terminals rehabilitated by June 30, 2001 Works completed Two new truck terminals constructed by June 30, 2001 Date achieved 03/16/1999 12/31/2006 12/31/2006 Comments By 2005, 95% of the works were completed. Due to the need to redirect LGED s efforts to urgent repair works related to the 2000 floods, the project did not meet original target dates. vi

Indicator 2. 7 Facilities for solid waste management put in place Disposal sites identified/improved in all 16 phase 1 participating cities and municipalities by June 30, 2001 0 All community bins constructed and operated by June 30, 2001 Works completed All waste transport equipment procured and operational by June 30, 2001 Date achieved 03/16/1999 12/31/2006 12/31/2006 By 2005, 95% of the works were completed, and the original targets were not met due to the Comments 2000 floods. Disposal sites were not improved for all 16 Phase I cities due to land acquisition issues. New Indicator MDF-funded sub-projects in 56 more municipalities providing grant and loan funding for for AF 2010 schemes selected locally and implemented and loan repayments underway BMDF working with 112 municipalities BMDF working with 168 municipalities BMDF worked with 154 municipalities 826 km roads 1,156 km roads 1,128 km roads Value (Quantitative or Qualitative) 8500 street lights 175 km drains 75 public toilets 11,500 street lights 265 km drains 87 public toilets 24,960 street lights 260 km drains 87 public toilets 75 kitchen markets 90 kitchen markets 211 kitchen markets Value (Quantitative or Qualitative) 137 km water supply pipelines 167 km water supply pipelines 173 km water supply pipelines Date achieved 06/1/2010 06/30/2012 06/30/2012 Comments By 2010, BMDF had funded 454 packages that were completed. By 6/30/2012, 595 projects had been completed in 154 ULBs. The loan recovery rate is 75%. New Indicator Capacity building inputs by MSU extended to 50 more municipalities with expected increase for AF 2010 in holding tax collection in the 48-70% range in 20 additional municipalities. Ongoing capacity building in Capacity building in Capacity building in 92 municipalities 142 municipalities 142 municipalities 75 municipalities with tax billing computerized 125 municipalities (total) tax billing computerized 142 municipalities tax billing computerized vii

7,405 municipal staff provided training 8,400 municipal staff (total) provided training 8,400 municipal staff provided training 20% increase in tax collection in 20 municipalities 25% increase in tax collection by municipalities (average/range) viii 17.5% increase in tax collection by municipalities (average/range) Date achieved 06/1/2010 06/30/2011 06/28/2012 The 17.5% increase in tax collection is based on a sample of 39 cities; 23 had tax collection Comments rates in the 48-95% range; another 7 had collection rates in the 40-47% range. G. Ratings of Project Performance in ISRs No. Date ISR Actual Disbursements DO IP Archived (USD millions) 1 04/22/1999 Satisfactory Satisfactory 0.00 2 07/02/1999 Satisfactory Satisfactory 0.00 3 01/06/2000 Satisfactory Satisfactory 6.29 4 03/22/2000 Satisfactory Satisfactory 6.29 5 08/28/2000 Satisfactory Satisfactory 6.29 6 03/01/2001 Satisfactory Satisfactory 10.55 7 04/24/2001 Satisfactory Satisfactory 12.82 8 10/29/2001 Satisfactory Satisfactory 20.65 9 12/12/2001 Satisfactory Satisfactory 27.70 10 06/06/2002 Satisfactory Satisfactory 30.76 11 12/24/2002 Satisfactory Satisfactory 44.37 12 06/24/2003 Satisfactory Satisfactory 47.51 13 12/22/2003 Satisfactory Satisfactory 47.51 14 06/08/2004 Unsatisfactory Unsatisfactory 59.24 15 12/23/2004 Unsatisfactory Unsatisfactory 59.50 16 04/25/2005 Satisfactory Satisfactory 63.22 17 12/07/2005 Satisfactory Satisfactory 74.37 18 06/29/2006 Satisfactory Moderately Satisfactory 92.26 19 12/01/2006 Moderately Satisfactory Satisfactory 99.41 20 06/08/2007 Satisfactory Satisfactory 113.23 21 12/30/2007 Satisfactory Satisfactory 124.53 22 06/27/2008 Satisfactory Satisfactory 129.20 23 12/30/2008 Satisfactory Satisfactory 135.53 24 05/26/2009 Satisfactory Satisfactory 147.57 25 11/29/2009 Satisfactory Satisfactory 159.65 26 05/25/2010 Satisfactory Satisfactory 161.91 27 12/12/2010 Satisfactory Satisfactory 161.91 28 11/10/2011 Satisfactory Moderately Satisfactory 170.27 29 06/25/2012 Moderately Satisfactory Moderately Satisfactory 200.19 30 07/10/2012 Moderately Satisfactory Moderately Satisfactory 200.19

H. Restructuring (if any) Restructuring Date(s) Board Approved PDO Change ISR Ratings at Restructuring DO 2/10/2005 S S IP Amount Disbursed at Restructuring in USD millions 12/21/2007 S S 124.53 23/06/2010 S S Reason for Restructuring & Key Changes Made Additional funding of US$11 million reallocated from another IDA project, and reallocation of $15 million from the BMDF component for infrastructure rehabilitation in response to a major flood in September 2004 Addition of US$25 million and extension of closing date in response to a major flood for urgent repairs to flood damaged infrastructure, and to further operationalize BMDF. Additional financing of US$42 million to scale up BMDF operations. I. Disbursement Profile ix

1. Project Context, Development Objectives and Design 1.1 Context at Appraisal Country and Sector Background. At the time of project appraisal, the path of economic growth in Bangladesh was uncertain. As discussed in the Bangladesh Country Assistance Strategy (CAS) for the period 1998-2001, it could either remain a desperately poor country or become a nascent tiger economy, but it never managed to be either. Though better than in other developing countries at that time, the country s growth was too slow to have a significant impact on poverty. Nearly 36 percent of the population remained below the poverty line for the very poor; 53 percent below the poverty line for the poor. The Government of Bangladesh (GOB) also was confronting severe challenges in overcoming the institutional constraints to pursuing sound development activities. Despite a slowdown of population growth since independence (1971), the population (70 million) was expected to rise over the next 30 years by 60 million. The urban population had been growing at a rate of about 4.2 percent per annum. Until the early 1990s, this rapid growth was occurring mainly in the country s four metropolitan areas (Dhaka, Chittagong, Khulna, and Rajshahi). By the latter 1990s, smaller municipalities were experiencing rapid growth rates as well. Regardless of their size, all of the municipalities in Bangladesh at the time of project appraisal were facing: (a) inadequate urban infrastructure services; (b) weak management and financial capacity; (c) fragmentation of urban sector responsibilities among numerous central government agencies; and (c) an increasing level of urban poverty and accompanying environmental degradation. To respond to these issues, as highlighted in the CAS, the Bank aimed to assist the country in improving the coverage and efficiency of basic urban services; support private sector growth and urban poverty reduction through the formulation and implementation of appropriate strategies for institutional and financial reforms in municipalities; and support fiscal decentralization. In this context, the GOB with World Bank assistance designed the Bangladesh Municipal Services Project (MSP) to address the weak institutional and financial capacities that were undermining sustainable service delivery, and to promote sound urban management. Rationale for Bank Assistance. The World Bank s involvement in the project was valuable for several reasons. First, the Bank had supported the development of an urban strategy which mapped the policy, institutional, and financial reforms necessary to underpin and sustain the investments to be made in this sector, and thus improve municipal service delivery. Guided by this strategy, the Bank also had experience supporting Bangladesh s urban sector through one urban development project, five water supply projects, and its continuing dialogue on key urban development issues. Involvement in the MSP would offer the Bank the opportunity to lead and shape the urban development agenda, particularly on policy reforms, and use its leverage to help the GOB prepare and implement institutional and financial reforms likely to have maximum impact. Lastly, the World Bank s participation in the project would afford the Bank the opportunity to apply its broad experience in the urban sector to help the GOB and donors coordinate appropriate responses to the urban challenges in Bangladesh. 1

1.2 Original Project Development Objectives (PDO) and Key Indicators The original PDO was to improve environmental and infrastructure service delivery in urban areas. Specifically, the project would: (a) strengthen the institutional capacity of selected municipal corporations and secondary towns to plan, finance, implement, and operate urban infrastructure services in an efficient and sustainable manner, (b) improve resource mobilization and allocation, and fiscal discipline through the creation of an improved financing mechanism for urban infrastructure investment; and (c) support the GOB and municipalities to reduce urban poverty and improve environmental conditions of urban communities through the financing of critical urban infrastructure and services. The original indicators that would be used to measure progress in meeting these outcomes included: (a) Number of municipalities able to plan, prioritize, and implement service improvements by the end of the project (Phase 1:16; Phase 2:40); (b) Number of municipalities that completed management reform (Phase 1: 16; Phase 2: 40); (c) Demand driven and cost recovery measures introduced for revenue generating sub-projects; (d) Municipal Development Fund (MDF) created and operational for on-lending to municipalities; (e) Municipal Performance Review Board established and operational by the end of Phase 1; (f) Physical investments targeting poverty and environmental improvements implemented through community participation. 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification The project was restructured three times to respond to two major floods in 2004 and 2007, and to scale up the demand-driven MDF operations. In 2005, US$11 million was added to the project, together with relocation of US$15 million from the BMDF component to the Flood Damage Rehabilitation (FDR) component, in response to a major flood in September 2004. After another severe flood in 2007, the Project was allocated an additional credit of SDR15.8 million (US$24.5 million) as part of the Bank s Post Flood Recovery Assistance Program. In 2010, an Additional Financing of US$42 million was approved to continue and scale up the project s successful program of infrastructure and capacity building support. Thus, overall, the project received reallocations and additional funding three times for a total of US$77.5 million. The PDO was not revised. The indicators were revised to reflect the additional financing. The revised indicators are: (a) Municipalities (183) able to plan, prioritize, and implement service improvements by end of project. (b) Management reform in municipalities completed in 183 municipalities. (c) Introduce demand driven and cost recovery measures; the target increased to 168 municipalities. (d) Streamline BMDF operations and strengthen coordination with LGED and other relevant agencies. This indicator was introduced under the AF of 2010 to replace the one requiring MDF operational for on-lending to municipalities and Municipal Performance Review 2

Board established and operational because these latter two targets had been achieved during the original investment. (e) Add 150 more works packages targeting poverty and environmental improvements through community participation funded by BMDF. This indicator was introduced in the AF or 2010. At that time, 454 works packages had been completed with BMDF funding; the new target became 595 works packages. 1.4 Main Beneficiaries As described in the PAD, the main direct beneficiaries of the project would be: (a) the residents of participating municipalities who would receive improved infrastructure services; and (b) staff at both national and local levels, who would benefit from capacity building. The PAD also notes that long-term project sustainability through community participation in the operation and maintenance (O&M) of urban infrastructure would be achieved through the adoption of the Participatory and Partnership Approach (PPA). This approach would encourage greater involvement of women in planning and managing activities that would directly contribute to improving their living conditions. The entire municipal population, including the poor and women, would benefit from the improved coverage of services as well as upgrades of the existing municipal infrastructure. The PAD also notes that any improvements in service coverage and quality would have a disproportionately positive impact on lower income groups because they were least likely to have these services at the time of project preparation. 1.5 Original Components (as approved) The original project had seven components divided into two phases. The first phase included civil works, equipment, and technical assistance (TA) to be implemented through the Local Government Engineering Department (LGED). Physical investments would be financed according to the traditional method of selecting investments for specific municipalities meeting IDA requirements. The second phase would support a MDF that would finance civil works, equipment, goods, and TA. First Phase Component 1 Civil Works (US$35.8 million). This would include the following: (a) Water Supply - rehabilitation of existing systems and equipment for O&M in 2 municipalities as well as urgent works (e.g., rehabilitation of hydrants, valve chambers, and overhead tanks). Khulna, which had the most severe water supply problems, would receive rehabilitation of 12 km of the distribution system (the largest amount). (b) Sanitation - construction of 47 public toilets (latrines and washing facilities for men and women) in 15 municipalities, and a pilot small-bore sewerage program in Khulna. Communities and/or the private sector would be involved in O&M of public toilets as well as the pilot sewerage in Khulna through the PPA. 3

(c) Roads and Drainage - upgrading and rehabilitation of existing road and drainage networks and systems. The road improvements would include renovation of deteriorated roads; conversion of earthen, gravel, or brick roads to bitumen roads; road widening; bridge renewal; and construction of footpaths and culverts. Drainage improvements would include drain reconstruction and conversion of earthen drains to brick-walled drains. This first phase would include about 220 km road improvements and about 37 km of drainage improvements. A PPA would be used in drain cleaning, and in carrying out a public awareness campaign about the need to keep drains clear from solid waste. (d) Bus and Truck Terminals- construction and rehabilitation of bus and truck terminals in all participating city corporations and municipalities. (e) Slum Improvement - pilots in Khulna and about five other municipalities. Physical investments would include improved water supply, on-site sanitation, footpaths, and street lighting. Slum communities would be organized with NGOs' assistance so they could identify their own needs and priority investments programs. Communities would be actively involved in planning, implementation, and O&M. (f) Solid Waste Management - equipment for primary collection. A PPA would be used in selecting sites for communal bins. Existing disposal sites would be improved in Khulna and Rajshahi from open to controlled dumping. Municipalities without formal disposal sites would receive technical assistance for identifying a suitable disposal site. (g) Markets - upgrading and/or rehabilitation of 16 existing municipal markets in the municipalities participating in this phase. (h) Land Acquisition to implement the bus/truck terminal and sanitation components. Component 2 Equipment (US$3.9 million). This component includes (a) solid waste equipment to improve primary collection, including waste transport vehicles (tipping trucks and trailers) in Khulnaand Rajshahi; and (b) motor vehicles and computers for personnel located in the field, and computer equipment and consumables for support personnel at each pourashava. Component 3 Technical Assistance (US$ 7.9 million). This would include: Institutional Development most of the TA would aim at (i) strengthening and institutionalizing the Pourashava Support Unit of LGED into a Municipal Support Unit (MSU) to support city corporations and pourashavas; and (ii) creating a municipality performance monitoring unit within the MSU to develop a municipality information data base and related applications systems. The MSU would consist of: (a) a Central Support Unit located at LGED; (b) a City Corporation Unit, located at the City Corporations; and (c) field advisors assigned to groups of 3 pourashavas. In addition, a central Monitoring Unit would be set up, and consultants would be recruited for the Central and City Corporation Support Units as well as for the Monitoring Unit to design and set up systems and procedures and train MSU personnel. 4

Municipal Data Base - TA would be provided to create and implement a municipal information data base and related applications systems. The project also would finance the development of the database/systems and related training and implementation. Community Development and Participation - TA would be provided for (i) hiring NGOs to develop a community participation program; (ii) conducting a workshop; and (iii) conducting a micro-credit study. The project also would finance consultants in community participation consultants to work with the MSU at various levels. Municipal Development Fund - TA for establishing the Municipal Development Fund (MDF) involving (i) preparation of the structure and incorporating the MDF as a legal entity; (ii) establishing the management structure of the MDF, including defining the initial organization structure, writing job descriptions, and assisting in recruiting suitable individuals to man key positions; and (iii) drafting policies and procedures both for operational areas and for financial management and accounting. Completion of all TA activities would be considered triggers for the implementation of Phase II. Studies Phase I studies included: (i) Environmental Management Action Plans in Khulna and Rajshahi for sanitation and solid waste management, and (ii) a hydrogeological investigation for Khulna. Project Implementation - TA for project management unit for design and construction supervision for two years initially, to be extended as needed during the second phase. Training - provided to MSU, LGED, and municipalities. Most training would be on-the-job in accounting, planning, and management. Project Staffing and Incremental O&M - additional staff for MSU and MDF as well as incremental O&M costs for project assets and MSU operating costs. Component 4 - Special Consideration (US$7.6 million). Also referred to as the Chittagong Hill Tracts, Component 4 would finance physical investments in three municipalities (Rangamati, Bandarban, and Khagrachari) which had lagged behind other municipalities in service provision and urban development. Component 5 - Flood Damage Rehabilitation (US$16.2 million). This component would finance rehabilitation of municipalities affected by the 1998 floods. The rehabilitation work would be determined upon completion of a needs assessment. Second Phase Component 6 - Municipal Development Fund (US$78.0 million). A Municipal Development Fund (MDF) would be established at the beginning of Phase II to start operating at the beginning of the third year of project execution. The Fund would finance municipal subprojects consisting of civil works, equipment, goods, and technical assistance. It would be established as an 5

autonomous agency, with resources provided through a US$70 million line of credit from IDA and GOB contributions. Component 7 Technical Assistance (US$4.6 million). This includes (a) support to the MDF specialists in sub-project appraisal, including technical review of proposals for compliance with environmental, social, economic, and financial feasibility; (b) technical support to municipalities to help them prepare sub-project proposals and FOAPs to make them eligible for MDF financing; and (c) incremental O&M costs for the MDF and MSU during the first three years. 1.6 Revised Components The components were not revised. 1.7 Other Significant Changes Although the components did not change, there were changes in funding and the project schedule (see table below). As a result of severe flooding in 2004, and in view of LGED s demonstrated success in implementing the first phase of investment, US$15 million of the original credit was reallocated from the BMDF component, along with US$11 million from another project, to the Flood Damage Rehabilitation component to be used to restore critical communication and transportation networks in the affected municipalities. At that time, the original closing date of June 30, 2005 was extended to June 30, 2008. After another severe flood in 2007 and in recognition of the good implementation performance under the original project, an AF of SDR 15.8 million (US$25 million equivalent) was approved in 2008 to help rehabilitate damaged urban infrastructure as part of the Emergency 2007 Flood Restoration and Recovery Assistance Program. The funds were allocated to the MSP to be used to rehabilitate urban roads, bridges, culverts, and drainage channels in flood-affected municipalities. Also at that time, the credit closing date was extended from June 30, 2008 to June 30, 2011 because activities were to be scaled up and targets enhanced. In 2010, there was another AF of SDR 27.8 million (US$42 million equivalent) to scale up BMDF-financed improvements in urban infrastructure as well as to continue the municipal financial and management capacity building under LGED. This new credit extended financing for urban infrastructure improvements to an additional 170 towns and extended the municipal capacity strengthening program to about 142 out of the 315 urban ULBs in the country. The AF also tightened fiduciary controls and supported the development of a program of improvement in urban infrastructure and institutional development over the medium term that would become the basis of projects to be funded by the Bank or other agencies. In addition, the BMDF would provide follow-up financing to 16 municipalities that made particularly good use of the first financing and could carry out further investments, and MSU would continue its trouble shooting for the 133 municipalities that had received support for capacity building. The AF also resulted in another extension of the closing date to June 30, 2012. 6

Component Appraisal Estimate Changes in World Bank Financing for MSP 1999-2012 Reallocation of $15 m from BMDF to FR and US$11 m added from other project Additional Financing (US$25 m) for FR (Cr - 1) Additional Financing (US$42m) (Cr 4761- BD) Total (1999-2012) Year 1999 2004 2008 2010 2012 Phase I- Physical Investments 32.5 32.5 Phase II- MDF 70 (15 m of orig. credit reallocated to FDR) 36.84 91.84 Chittagong Hills Tracts 7.6 7.6 TA and Project Staffing Flood Damage Rehabilitation 10.8 4.41 15.21 16.2 26 (US$15m +$11m ) 25 67.2 (FDR) Incremental O&M Costs 1.5.75 2.25 Total 138.6 26 25 42 216.6 Note: The US$11million allocated to the MSP from another project does not appear in overall project financing in the World Bank project data sheets. The total reported by the Bank system is US$205.6 million. Also note that the overall total takes into account the reallocation of US$15 million from the Phase II-MDF in 2004 to the FDR component. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry QAG rating Highly Satisfactory QAG provided the highly satisfactory rating on QAE on the basis of the following main points. First, the team made a serious effort to analyze the difficult institutional issues and came up with the means for addressing them. QAG also noted that the project had benefited from an earlier decision to delay project preparation in order to undertake urban sector work that provided the core of the analysis and the proposed solutions as well as the time to build up support within the country. QAG further acknowledged the serious effort made by the team to incorporate social development issues in a sensitive, focused way that would help navigate the institutional and 7

bureaucratic difficulties, clear team commitment and support in LGED, and an understanding of the risks taken and the appropriateness of the measures contained in the project to address them. The ICR agrees with the points raised by QAG, but not the rating. The following are additional factors that would support a satisfactory rating. The design was appropriate to meet the project s objectives. The project s development objectives were clear and the components were well designed to address them. The project would address the (i) urgent need for basic urban services, particularly in small ULBs with lowincome areas, (ii) weak local government institutions through the provision of technical assistance to improve the institutional and financial capacity of participating municipalities; (iii) poor local resource mobilization and cost recovery through the establishment of a municipal development fund and capacity building; (iv) prior neglect of the O&M aspects of infrastructure by focusing on rehabilitation, and provision of technical assistance towards maintenance activities; and (v) need to improve sustainability of services through a participatory and partnership approach (PPA) in the planning, implementation, and O&M of these services. The design took into account the Bank's experience with earlier urban projects worldwide and in Bangladesh. The team responsible for project design was aware that many of the earlier multisectoral urban projects that focused on poverty alleviation through investments in basic infrastructure and housing included poorly integrated and ill-prepared components were too complex to be properly implemented. The design took into account the need to simplify the design, be realistic in determining small municipalities financial capacities, and emphasize strengthening of their financial and institutional frameworks to help ensure sustainability. Prior to the MSP, the Bank s involvement with the urban sector in Bangladesh included the Urban Development Project (approved in FY88), which focused on slum upgrading in Dhaka and drainage improvements in Dhaka and Chittagong, and five IDA-financed urban water supply projects. Although physical progress on the completed projects was satisfactory, the institutional and financial performance was not adequate and the projects confronted problems related to land acquisition, counterpart financing, and procurement. Based on the lessons from these earlier projects, the design emphasized the need for: (a) generating broad stakeholder commitment and interest; (b) establishing the willingness of the municipal authorities to borrow for the improvements before funding would be committed, and (c) directing attention to such issues as land acquisition, sub-project design, and procurement arrangements during project preparation. Based on the lessons from the completed projects, the design also responded to the need to: (a) achieve sustainable institutional and policy improvements by emphasizing underlying institutional problems rather than on achieving specific institutional and financial targets; and (b) establish a clear link between underlying problems, project objectives, and the project components. These issues were addressed in the project mainly through the demand-driven approach inherent in the BMDF. Because the introduction of the BMDF represented the first time that such a fund would be established in Bangladesh, an important lesson learned from experience in other countries was that the sustainability of such a funding mechanism would depend on the creditworthiness of its borrowers, and the need for a strong commercial orientation of the fund, which would operate without political interference. Taking into account the weak capacity of the ULBs in Bangladesh, 8

however, the design incorporated a realistic approach to assessing eligibility by requiring a participating ULB to prepare a Financial and Operational Action Plan (FOAP), which included an assessment of capacity and a plan for improving the ULB s financial standing so it could sustain an investment in urban infrastructure. The design also took into account the need to establish the BMDF as a commercially oriented, autonomous agency that would introduce transparency and equity into resource allocation for urban infrastructure nationwide. Project design took into account alternative approaches, the trade-offs between them, and the rationale for the final decisions. The three alternative approaches included: traditional urban lending versus the MDF approach, traditional grant financing versus loan grant financing mix, and single project versus phased approach. Unlike the traditional approach with pre-selected municipalities and components, introducing the MDF was considered optimal because the ULBs applied for demand driven financing (and differed from the supply driven approach) for providing long term and sustainable infrastructure financing and capacity building. The loangrant financing mix was considered preferable to traditional grant financing because it had the potential for leveraging funds for greater cost recovery. Project design benefitted from public participation and consideration of social issues. The project took into account the results of a social assessment, and the possibility of causing resentment on the part of the municipalities that are lagging in reforms to receive funding that had traditionally come from the central government. To mitigate this risk, the BMDF would be set up as an autonomous agency under the Ministry of Finance, with financing decisions made by a Steering Committee/Board of Directors from different spectra of society. The design also incorporated a PPA to help ensure ownership and the social sustainability of the project. The design could have been improved if it included funds for carrying out beneficiary assessments during project implementation to determine the extent to which the ultimate beneficiaries were satisfied with the project-related processes and improvements. Notwithstanding the above, the ICR team noted QAE issues that were not adequately addressed during the design of the investment. The main project outcome indicators were not clearly defined. Although the main PDO objectives were well articulated, there could have been clearer definitions of the PDO indicators such as municipality is able to plan, prioritize, and implement service improvements by end of project and management reform was completed. With regard to the outcome relating to support for reducing poverty and improving environmental conditions in urban communities through community participation, it would be difficult to establish targets for these in the PAD when a large part of the project s physical investments would be demand driven and determined only after the project was under implementation. The design underestimated the time needed to establish and operationalize the BMDF. Based on the original design, the BMDF was expected to be established as a legal entity with a clear ownership and management structure, registered with all concerned agencies, and ready to administer $78.0 million of project funds after two years of project implementation. The team greatly underestimated the amount of time it would take to obtain all the necessary approvals and financial flow arrangements for the MDF to be registered and operational. 9

The design could have included mechanisms to better coordinate local capacity building with BMDF investments throughout the entire project period. Better planned and programmed technical assistance could have been available throughout the project to help ensure that the ULBs that received funding would have the necessary capacity to adequately maintain them. 2.2 Implementation The MSP had a 12-year implementation period. The project became effective on October 25, 1999 and closed on June 30, 2012, seven years after the original closing date. The total project financing increased by $77.5 million during the 12-year period with three reallocations and additional funding. Several factors influenced project performance and justified the need for multiple project extensions. The three most important factors were: (a) the implementation competence of LGED and successful implementation of Phase I investments; (b) the delay in establishing and operationalizing the BMDF, and its subsequent success leading to an AF in 2010; and (c) massive floods in Bangladesh during the years 2000, 2004, and 2007 which led to additional allocations of funds in 2004 and 2008 in order to respond to the urgent need to repair damaged infrastructure. These and other factors are discussed below. Phase I Performance. Despite initial start up delays, the implementation of Phase 1 investments was carried out successfully. More than 95% of the original MSP components (LGED portion) were completed by June 30, 2005, the original closing date. Capacity building through the Municipal Support Unit (MSU) was set up within LGED and continued throughout the life of the project to build institutional and financial capacities in the ULBs, and to assist municipalities in meeting the eligibility criteria for project support. The good performance under Phase 1 reflected LGED s strong leadership which continued throughout project implementation. The strong performance under Phase 1 and the need to address urgent repairs caused by the floods of 2000, 2004, and 2007 also led to additional project funding, which greatly increased the number of activities to be led by LGED. Thus some of the original target dates could not be met. BMDF Delay and Subsequent Performance. Although the Borrower was fully supportive of the concept of BMDF during project preparation and appraisal, the Fund was not formally registered until 2002, and not fully operational until 2004, three years behind the original schedule. The main reasons for the delay related primarily to: (a) the fact that the government was in transition in 2001 when the BMDF was expected to become operational, and the transitional government put the decision to formally register the BMDF on hold until the new government took over; and (b) the severe floods of 2000, which diverted the attention of the project staff to finance urgent works to repair critical transportation and communications networks. Despite the start up delays and its difficulties in securing good quality technical staff and consistent leadership, the BMDF ultimately became an effective instrument for allocating funds to ULBs for critically needed infrastructure improvements. This assistance would not have been available to many ULBs, especially the small municipalities, if they continued to depend solely on the block grants administered from the national governments. As a result of BMDF s successful operation, in 2010, US$42 million was added to the project to help finance the scaling up of improvements in urban infrastructure as well as municipal 10

financial and management capacity. At that time, the project was performing well. There were significant benefits achieved in the participating municipalities as well as a strong demand for additional support. The AF would help to extend the financing of urban infrastructure improvements to some 170 towns, and extend the municipal capacity strengthening program to about 183 municipalities (out of 315 urban local bodies in Bangladesh). The AF of 2010 also would support development of the next generation of urban improvement programs and require tightening of fiduciary controls. By May 2010, practices adopted under the project for municipal institutional and infrastructure development represented significant improvements, including: (a) greater authority and responsibility of the municipalities in selecting and implementing the investments with matching operational and financial plans, (b) substantial self and loan financing requirements, (c) improved project evaluation and management, (d) improved municipal management and revenue systems, and (e) improved procurement practices. Demand for BMDF Financing. Throughout project implementation and after closing, there has been a continuous demand for BMDF support for infrastructure projects, and sense of ownership on the part of the participating ULBs and project communities. By the time of project closing, there was a backlog of over 200 project proposals seeking funding. Floods and Related Changes in Project Financing, Activities, and Schedule. Following the serious flood and heavy rainfall during the 2004 monsoon, a joint decision was made by the GOB and World Bank that urgent rehabilitation works in 67 flood-affected municipalities would be undertaken with IDA support through a credit reallocation of US$15 million from BMDF to LGED. The project also received a reallocation of US$11 million from the Private Sector Infrastructure Development Project (Cr. 2995). Accordingly, the DCA was amended and the credit closing date was extended to June 30, 2008. In response to another flood crisis in 2007, through the Emergency 2007 Flood Restoration and Recovery Assistance Program, the MSP received additional financing of US$25 million to support rehabilitation of urban roads, bridges, culverts, and drainage channels in municipalities damaged by the floods. This AF was allocated to MSP on the basis of LGED s strong performance under Phase I. Through this, LGED again demonstrated its effectiveness in expediting the urgent works needed to address critical flood damage in numerous municipalities. The change also meant an extension of the closing date to June 30, 2011. Finally, the AF of 2010 (US$42 million) was allocated to the MSP to allow the BMDF and LGED to further strengthen their investment and capacity building programs. The AF also supported the preparation of a follow up investment. Mid-Term Review. The MTR (2002) provided a comprehensive assessment of progress toward development objectives and implementation performance. While the review recognized that the overall implementation of the project s first phase was satisfactory, it raised serious concerns that the project encountered what was then a two year delay in establishing the BMDF. The Bank also noted in its review that it may have been overly optimistic given the country context, the weak institutional capacity of municipalities, and inherent governance problems to expect completion of all first phase investments, involving over 1,000 contracts within two years. Thus, the main recommendations of the MTR were to revise some of the project outcomes and target 11