Document of The World Bank Report No: ICR0000695 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-34600 IDA-H1980) ON A CREDIT IN THE AMOUNT OF SDR 31.3 MILLION (US$ 40.0 MILLION EQUIVALENT) AND AN ADDITIONAL FINANCING GRANT IN THE AMOUNT OF SDR 21.1 MILLION (US$ 30.6 MILLION EQUIVALENT) TO THE REPUBLIC OF BURUNDI FOR A PUBLIC WORKS AND EMPLOYMENT CREATION PROJECT June 26, 2008 AFTU2 (Water and Urban 2) AFCC2 (Central Africa 2) Africa Region CURRENCY EQUIVALENTS Exchange Rate Effective December 31, 2007 Currency Unit = Burundian Franc (FBU) FBU 1,000.00 = US$0.893 US$1.00 = FBU 1,119.54 FISCAL YEAR January 1 ­ December 31 ABBREVIATIONS AND ACRONYMS ABUTIP Agence burundaise pour la réalisation des travaux d'intérêt public (Burundi Agency for the Execution of Public Works) ADB African Development Bank AGETIP Agence d'exécution des travaux d'intérêt public (Agency for the Execution of Public Works) DCA Development Credit Agreement EA Environmental Assessment ESMF Environmental and Social Management Framework EMP Environmental Management Plan EIRR Economic Internal Rate of Return EU European Union FBU Franc burundais (Burundian Franc) ICR Implementation Completion and Results Report INT Department of Institutional Integrity ISC Inter-Ministerial Steering Committee ISN Interim Strategy Note ISR Information of Status and Results Reports LMC Local Maintenance Committee M&E Monitoring and Evaluation MAP Multi-Sectoral AIDS Project MPWE Ministry of Public Works and Equipment O&M Operations and Maintenance PWECP Public Works and Employment Creation Project PDO Project Development Objectives PPF Project Preparation Facility QEA Quality at Entry Assessment SDR Special Drawing Rights SME Small and Medium-sized Enterprise TS Technical Secretariat TTL Task Team Leader Vice President: Obiageli K. Ezekwesili Country Director: John Murray McIntire Sector Manager: Eustache Ouayoro Project Team Leader: Déo-Marcel Niyungeko ICR Team Leader: Déo-Marcel Niyungeko REPUBLIC OF BURUNDI Public Works and Employment Creation Project CONTENTS Data Sheet A. Basic Information i B. Key Dates i C. Ratings Summary i D. Sector and Theme Codes ii E. Bank Staff ii F. Results Framework Analysis iii G. Ratings of Project Performance in ISRs iv H. Restructuring v I. Disbursement Graph v 1. Project Context, Development Objectives and Design...............................................1 2. Key Factors Affecting Implementation and Outcomes...............................................5 3. Assessment of Outcomes ............................................................................................9 4. Assessment of Risk to Development Outcome.........................................................12 5. Assessment of Bank and Borrower Performance......................................................13 6. Lessons Learned........................................................................................................15 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners...........15 Annex 1. Project Costs and Financing ..........................................................................17 Annex 2. Outputs by Component..................................................................................18 Annex 3. Economic and Financial Analysis .................................................................20 Annex 4. Bank Lending and Implementation Support/Supervision Processes.............23 Annex 5. Beneficiary Survey Results ...........................................................................25 Annex 6. Stakeholder Workshop Report and Results...................................................27 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR .....................28 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders .......................37 Annex 9. List of Supporting Documents.......................................................................38 Annex 10. Evaluation Grid for Sub-Project Evaluation (Supplemental Annex) ..........39 Annex 11. Situation of basic infrastructure at Appraisal (Supplemental Annex).........40 MAPS A. Basic Information PUBLIC WORKS AND Country: Burundi Project Name: EMPLOYMENT CREATION PROJECT Project ID: P064961, P095028 L/C/TF Number(s): IDA-34600, IDA-H1980 ICR Date: 06/19/2008 ICR Type: Core ICR GOVERNMENT OF Lending Instrument: SIL Borrower: BURUNDI Original Total Disbursed Amount: XDR 51.1M XDR 31.3M Commitment: Additional Financing: XDR 21.1M Environmental Category: F Implementing Agencies: Ministry of Public Works and Equipment Cofinanciers and Other External Partners: Not applicable B. Key Dates Process Date Process Original Date Revised / Actual Date(s) Concept Review: 05/13/1999 Effectiveness: 06/29/2001 06/29/2001 Appraisal: 10/02/2000 Restructuring(s): Not applicable Not applicable Approval: 01/23/2001 Mid-term Review: 12/01/2003 10/08/2003 Closing: 12/31/2006 12/31/2007 Appraisal, AF 10/06/2005 Effectiveness, AF: 02/28/2006 05/12/2006 Approval, AF 01/05/2006 Closing, AF: 12/31/2007 12/31/2007 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Satisfactory Risk to Development Outcome: Moderate Bank Performance: Satisfactory Borrower Performance: Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Satisfactory Government: Satisfactory Quality of Supervision: Satisfactory Implementing Agency/Agencies: Satisfactory Overall Bank Overall Borrower Performance: Satisfactory Performance: Satisfactory i C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Performance Indicators (if any) Rating Potential Problem Project at Quality at Entry (QEA), No Not applicable any time (Yes/No): Original Project: Problem Project at any time QEA, Additional Moderately No (Yes/No): Financing: Satisfactory Quality of Supervision DO rating before Satisfactory (Original Project and Not applicable Closing/Inactive status: Additional Financing): D. Sector and Theme Codes Sector Code (as % of Total Bank Financing) Original Actual General education sector 21 0 Health 25 0 Roads and highways 21 22 Sub-national government administration 12 12 Water supply 21 0 General industry and trade 0 28 General water, sanitation and flood protection 0 25 Other social services 0 13 Theme Code (Primary/Secondary) Original Actual Conflict prevention and post-conflict reconstruction Secondary Primary HIV/AIDS Primary Secondary Municipal governance and institution building Primary Primary Other urban development Primary Primary Poverty strategy, analysis and monitoring Primary Not applicable Small and medium enterprise support Not applicable Primary E. Bank Staff Positions At ICR At Approval Vice President: Obiageli K. Ezekwesili Callisto E. Madavo Country Director: John McIntire Emmanuel Mbi Sector Manager: Eustache Ouayoro Letitia A. Obeng Project Team Leader: Déo-Marcel Niyungeko Eustache Ouayoro ICR Team Leader: Déo-Marcel Niyungeko Christian Eghoff and Déo-Marcel ICR Primary Author: Niyungeko ii F. Results Framework Analysis Project Development Objectives (from Development Credit Agreement) The objectives of the project were to: (i) generate productive employment to help reconstruction in the Borrower's territory through labor-intensive public works; (ii) promote the execution of cost-effective and economically and socially useful public works projects, including through small and medium enterprises; and (iii) build institutional capacity at the municipal level to identify, prioritize, operate, and maintain sub-projects. Revised Project Development Objectives (as approved by original approving authority) The PDO was not revised. However, indicators related to employment generation and training of small and medium-sized enterprises (SMEs) were adjusted to reflect the scaling up of Component 1 (see Indicators 1 and 2 below). (a) PDO Indicator(s) Original Target Formally Actual Value Indicator Baseline Value Values (from Revised Achieved at approval Target Completion or documents) Values Target Years Indicator 1 (Objective (i), Person-days of employment created. Comp. 1): There is no baseline. Value Performance is measured 750,000 2,750,000 6,082,722 against project progress. (2005: 3,421,206) Date achieved 10/02/2001 12/31/2006 12/31/2007 12/31/2007 221 percent achievement overall. 456 percent achievement for the original project and 133 percent achievement for the additional. The original project Comments generated 3,421,206 person-days of employment, largely surpassing the target of (incl. % 750,000 person-days. The more ambitious target for the additional financing was achievement) 2,000,000 person-days, exceeded by 661,516 person-days for a total of 2,661,516 person-days for the additional financing and 6,082,722 person-days total for the entire project. Indicator 2 (Objective (ii), Capacity building of local SMEs (person-days of training). Comp. 1): There is no baseline. Value Performance is measured 2,000 5,000 5,582 against project progress. (2005: 4,300) Date achieved 10/02/2001 12/31/2006 12/31/2007 12/31/2007 111 percent achievement overall. 215 percent achievement for the original Comments project and 183 percent achievement for the additional. The original project (incl. % provided 4,300 person-days of training, largely surpassing the target of 2,000 achievement) person-days. The target for the additional financing was 700 person-days, exceeded by 582 person-days for a total of 1,282 person-days for the additional financing and 5,582 person-days total for the entire project. iii Indicator 3 (Objectives (ii) and (iii), Percentage of rehabilitated or constructed infrastructure found to be operational Comp. 1 and and properly maintained after completion. 2): There is no baseline. Value Performance is measured 75% NA 95% against project progress. Date achieved 10/02/2001 12/31/2006 NA 12/31/2007 Comments (incl. % All infrastructures are operational and maintained except for five sub-projects achievement) that were not finalized at project closure and one market damaged by a landslide. Indicator 4 (Objective Percentage increase in municipal budgets (all municipalities) for infrastructure (iii), Comp. 2):maintenance. Value 0% 20% 80% (FBU 473,261,231) (FBU 567,913,477) NA (FBU 51,450,141) Date achieved 12/31/2004 12/31/2006 NA 12/31/2007 Comments The local governments made substantial efforts to increase revenues and largely (incl. % surpassed the target. Figures are only reliable from 2004 (baseline year), but the achievement) percentage increase in revenues would likely have been larger compared to 2001 if figures had been available for this year. Indicator 5 (Objective Staff and officials of municipalities trained by the Technical Secretariat on sub- (iii), Comp. 2):project identification, execution, operation, and maintenance There is no baseline. Staff and officials Value Performance is measured Staff and officials in NA in 129 against project progress. 80 municipalities. municipalities. Date achieved 10/02/2001 12/31/2006 NA 12/31/2007 Comments (incl. % 161 percent achievement. Staff of all 129 municipalities of Burundi received achievement) training. Indicator 6 (Comp. 3): HIV/AIDS awareness campaigns implemented. At least 65 percent of There is no baseline. sub-project sites by At least 66 percent Value Performance is measured 2003. At least 85 NA of project sites by against project progress. percent of sub- 2007. project sites by 2006. Date achieved 10/02/2001 12/31/2006 NA 12/31/2007 66 percent of project sites by 2007, based on the sites targeted by activities Comments implemented by the PWECP. The remaining sites benefited from the Multi- (incl. % sectoral HIV/AIDS project (MAP), which took over implementation of achievement) HIV/AIDS activities in 2004 on behalf of the PWECP, but without systematic reporting. iv (b) Intermediate Outcome Indicator(s) for Component 1 implementing agency (AGETIP Burundi/ABUTIP) Original Target Formally Actual Value Indicator Baseline Value Values (from Revised Achieved at approval Target Completion or documents) Values Target Years Indicator 1: Operating costs of implementing agency not exceeding 8 percent of the amount of executed works. There is no baseline as Value the agency was established with the 8% NA 5.9% project. Date achieved 10/02/2001 12/31/2006 NA 12/31/2007 Comments (incl. % The result for the entire project is 26 percent below the maximum. achievement) Indicator 2: The time period between receipt and payment of an invoice by the agency. Value NA Maximum 15 days NA 4 days Date achieved 10/02/2001 12/31/2006 NA 12/31/2007 Comments The result for the entire project is 73 percent below the maximum. The 4 day (incl. % time period for the implementing agency compares to an average turnaround achievement) time of around 90 days for the central Government in 2005. G. Ratings of Project Performance in ISRs Actual No. Date ISR Archived DO IP Disbursements (USD millions) 1 06/08/2001 Satisfactory Satisfactory 0.00 2 10/02/2001 Satisfactory Satisfactory 1.30 3 03/25/2002 Satisfactory Satisfactory 2.41 4 04/26/2002 Satisfactory Satisfactory 2.54 5 04/30/2002 Satisfactory Satisfactory 2.54 6 09/30/2002 Satisfactory Satisfactory 4.78 7 10/01/2002 Satisfactory Satisfactory 4.78 8 12/11/2002 Satisfactory Satisfactory 6.73 9 01/22/2003 Satisfactory Satisfactory 7.53 10 08/12/2003 Satisfactory Satisfactory 13.08 11 11/26/2003 Satisfactory Satisfactory 15.47 12 05/25/2004 Satisfactory Satisfactory 21.29 13 11/24/2004 Satisfactory Satisfactory 26.53 14 04/15/2005 Satisfactory Satisfactory 32.98 15 11/23/2005 Satisfactory Satisfactory 40.45 16 05/30/2006 Satisfactory Satisfactory 43.35 17 11/09/2006 Satisfactory Satisfactory 51.08 18 02/28/2007 Satisfactory Satisfactory 55.99 19 10/13/2007 Satisfactory Satisfactory 68.03 v H. Restructuring (if any) Not Applicable I. Disbursement Profile vi 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal At appraisal, the Republic of Burundi was slowly emerging from a civil war that had erupted in 1993 and claimed the lives of an estimated 300,000 people while displacing about 15 percent of the population. The population was nearly 7 million people, with an urbanization rate of about 10.6 percent. A landlocked country, Burundi is one of the smallest in Sub- Saharan Africa, with a total land area of only 27,834 square kilometers. The country had the second highest population density in the region at almost 250 persons per square kilometer, creating strong competition for natural resources and land. The combined effects of the civil war and a regional economic embargo from 1996 to 1999 resulted in economic and social decline. GDP dropped from US$180 per capita in 1993 to US$110 in 1999. The relative poverty rate nearly doubled from 35 percent in 1992 to 66 percent in 1998. Infant mortality rates were as high as 114 per 1,000 (2000) and gross primary school enrollment had fallen from 70 percent in 1993 to 44 percent in 2000. Overall social indicators were among the worst in Africa at the time. The war and the economic decline took a heavy toll on the country's infrastructure, which was either destroyed or seriously deteriorated due to poor maintenance and lack of investment. Water and electricity facilities were among the most affected as they were targeted by acts of sabotage. Many rural dwellers had sought refuge in semi-urban areas and particularly the Bujumbura periphery, contributing to a drop in the rate of urban water supply coverage from over 70 percent in 1993 to an estimated 57 percent at appraisal. Less than two percent of the country's households had electricity in their homes. The Bujumbura sewerage and wastewater treatment network covered only 38 percent of the population. Population health surveys showed that 84 percent of mortality and morbidity in children under five resulted from poor conditions of water supply, hygiene and sanitation, leading to water-borne diseases such as bacillary dysentery, amoebiasis, diarrhea and cholera. The capacity of the Government to manage public resources was largely eroded. Institutions had weakened and the legislation governing public finances and budget management in Burundi, which had been in effect since 1969, was obsolete. Much of the public sector's human resource base had been eliminated. In addition, the poor security situation and the lack of oversight and the absence of democratic institutions had eroded government accountability and transparency. In 2000, the international community sponsored a framework for national reconciliation and lasting peace among Burundians. In August 2000, a comprehensive peace agreement was concluded between seventeen political organizations in Arusha, Tanzania. Several rebel groups did not participate in these talks. At a donor conference held in Paris in December 2000, the Government of Burundi presented a transitional strategy for the period 2001-2003 to the international community to address the immense challenges faced by the county. This included economic stabilization and prevention of further-deepening poverty as the peace process evolved. The international community responded by pledging US$440 million in emergency humanitarian, reconstruction, and long-term development assistance to Burundi. 1 World Bank Response As the conflict and regional embargo rendered the 1995 Country Assistance Strategy (CAS) obsolete, the Bank prepared an Interim Strategy in 1999 to guide actions over a two-year period and provide an integrated framework for the Bank and other donors to address poverty and its underlying factors, including the triggers of repeated conflicts in Burundi. The Interim Strategy rested upon three pillars: (i) promoting governance and ownership through participation in rehabilitation and reconciliation; (ii) creating productive employment and rural purchasing power; and (iii) restoring key imports to support private sector activity and minimum levels of essential social services. The donor community aligned its assistance with this strategy. The 1999 Interim Strategy, specifically its second pillar, provided the basis for the Public Works and Employment Creation Project (PWECP), which was one of three immediate operations under this Strategy. The two others were the Social Action Project II, supporting a participatory approach to the rehabilitation and provision of social and economic infrastructure in rural areas, and an Emergency Economic Recovery Credit. The Development Credit Agreement (DCA) in the amount of SDR 31.3 million (US$40.0 million equivalent) was approved on January 23, 2001 and became effective on June 29, 2001. The original closing date was December 31, 2006. Context at appraisal of Additional Financing By January 2005, disbursement projections indicated that the credit would be fully disbursed by December 31, 2005, one year ahead of the original closing date. Therefore, the Government requested additional financing to scale up the project, as strong local government demand for infrastructure sub-projects largely exceeded the original funding envelope. Local governments and communities had already mobilized about FBU 602 million (US$0.6 million equivalent) in local contributions to support their requests for additional sub-projects. Emerging from twelve years of civil conflict, Burundi's security conditions were increasingly conducive to resuming economic activity at the local level and throughout the country. Fair and transparent municipal and legislative elections were held during 2005. It was evident that not implementing the additional PWECP sub-projects would have undermined the prevailing momentum and possibly promoted distrust between the population and the new Government, an unwelcome development at a time when the country was attempting to consolidate its national unity towards a lasting peace. A window of opportunity was open and experience in other countries suggests that rapid provision of exceptional levels of support was needed to take advantage of such opportunities. At the same time, a new Bank Interim Strategy Note (ISN) for Burundi covering FY06-07, approved by the Board on May 3, 2005 included an Additional Financing for the on-going PWECP. One of the objectives of the ISN was to ensure that communities and populations have access to basic social services and income-generating activities with the expectation that they would be, in turn, supportive of the peace process and the economic reform agenda. Therefore, the Additional Financing of the above Public Works and Employment Creation Project was prepared and approved by the Bank on January 5, 2006 for a total amount of SDR 21.1 (US$30.6 million equivalent) and became effective on May 12, 2006 bringing the total project amount to SDR 52.4 million (US$70.6 million equivalent). 2 1.2 Original Project Development Objectives (PDO) and Key Indicators As stated in the Development Credit Agreement (DCA)1, the objectives of the PWECP were to: (i) generate productive employment to help reconstruction in the Borrower's territory through labor-intensive public works; (ii) promote the execution of cost-effective and economically and socially useful public works projects, including through small and medium enterprises (SMEs); and (iii) build institutional capacity at the municipal level to identify, prioritize, operate, and maintain sub-projects. The key indicators focused on the generation of employment, capacity building, the performance of the public works implementation agency, operation and maintenance of infrastructure, and the fight against HIV/AIDS (see Data Sheet for details). 1.3 Revised PDO (as approved by original approving authority) and Key Indicators The PDO did not change during implementation, as the project remained consistent with the country conditions. However, indicators related to employment generation and training of SMEs were adjusted to reflect the scaling up of Component 1 (see Data Sheet for details). 1.4 Main Beneficiaries Local populations. The rehabilitation and construction of economic and social infrastructure were intended to improve the living conditions of the populations residing in the beneficiary urban centers. Unskilled workers, including youth, women, returning refugees, displaced and landless people and ex-combatants, were to benefit from short-term employment through the use of labor-intensive construction methods. Creating employment and thereby boosting local purchasing power was to provide a broad segment of the population with an economic stake in the reconciliation process and reduce the likelihood of triggering further conflicts. Local governments. The project was to improve the capacity of local governments to mobilize local resources, to identify, prioritize, and undertake sustainable investments relevant to the needs of their populations, and to maintain these assets through the development of a systematic approach to financing, operating, and maintenance arrangements. This was to further strengthen the nascent decentralization process. Small- and medium-sized enterprises. The project was to help strengthen the capacity of local consultants and contractors active in the public works and construction sector through training and the execution of sub-projects using labor-intensive construction methods. Sector ministries. The ministries in charge of public works and equipment, energy and mines, environment, and land management were to benefit from capacity building programs aimed 1The statement of the Project Development Objectives (PDO) on page 3 of the Project Appraisal Document (PAD) contains the same three basic elements as the Development Credit Agreement (DCA): employment generation, implementation of economically and socially useful public works projects, and institutional capacity building. This differs somewhat from Annex 1 of the PAD, which specifies demobilized combatants and returning refugees as the main beneficiaries of employment generation. The PDO version in the DCA is a better representation of the actual objectives, since the project sought to generate equal employment opportunities for all, irrespective of their role in the civil war, and thus not introduce potential sources of conflict. The PDO from the DCA (the one used for the purposes of this ICR) was reproduced in the Project Paper and Financing Agreement for the Additional Financing Grant. 3 at improving their ability to carry out social and environmental impact assessments as well as monitoring and evaluating mitigation action plans and resettlement plans. 1.5 Original Components Component 1: Sub-project Implementation and Capacity Building for Employment Creation (Appraisal: US$33.26 million. Approved Additional Financing: US$30.60 million. Actual: US$67.19 million) Based on requests from local governments (in consultation with their constituents), this component was to finance sub-projects to rehabilitate or construct basic local level infrastructure. In addition, the component was to strengthen the financial and operational capacity of a newly created delegated contract management agency, the Agency for the Execution of Public Works (Agence burundaise d'exécution des travaux d'intéret public ­ AGETIP Burundi, which later became ABUTIP), which would assume responsibility for procurement and implementation of Component 1. This component also included training programs for SMEs in business management and relevant technical skills. Component 2: Institution Building (Appraisal: US$4.54 million. Actual: US$6.06 million) Under this component, a Technical Secretariat was to provide advisory services, technical assistance and training to local governments in programming, implementing, financing, operating, and maintaining local level infrastructure, including sub-projects financed under Component 1. It was also to support capacity building activities for Government agencies and local governments in the area of resource management, budgeting and accounting, and social and environmental assessment. Component 3: Support for the Prevention of HIV/AIDS (Appraisal: US$1.00 million. Actual: US$0.43 million) This component aimed at preventing the spread of HIV/AIDS through a community information and education program in areas where sub-projects were implemented. Component 4: Project Monitoring (Appraisal: US$1.75 million. Actual: US$37,665) This component provided funds for limited studies and surveys. 1.6 Revised Components Component 1 was scaled up to finance additional sub-projects in relation to an Additional Financing Grant of US$30.6 million and the performance indicators were adjusted to reflect the expected increased impact (see Data Sheet for details). 1.7 Other significant changes The project implementation period was extended by one year, from December 31, 2006 to December 31, 2007 to provide adequate time to implement the Additional Financing Grant. 4 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry Quality at Entry Rating: Satisfactory The quality at entry is rated satisfactory based on: (i) the relevance of the project development objectives at appraisal of the original and additional financing; (ii) the soundness of the background analysis contained in the 1999 Interim Strategy and 2005 Interim Strategy Note, (iii) the incorporation of lessons learned from public works and employment creation agencies active in Sub-Saharan Africa into the project design; (iv) the Government's commitment to the original project and its expansion; and (iv) the proper identification of project risks and the comprehensiveness of the risk mitigating measures. The project was designed to overcome the weak institutional capacity in Burundi, the perceived high risk business environment associated with the post-conflict situation of the country and the limited local private sector capacity. Key features of the project design consisted of: (i) separating responsibilities for sub-project selection from sub-project implementation to ensure transparency and avoid conflict of interest; (ii) using demand, including beneficiary contribution, for selecting sub-projects and sustainability; (iii) advocating the use of local materials in construction and labor-intensive methods to support the development of small and medium-sized enterprises and employment generation, and (iv) inclusion of an environmental impact analysis into the sub-project approval process and close monitoring of implementation of its mitigation plan by the Technical Secretariat (TS). The Quality Assurance Group (QAG) assessed the quality at entry (QEA) of the additional financing in August 27, 2008 as moderately satisfactory (See paragraph 2.3, 2.4 and 5.1a for details). The strong client ownership of the project was demonstrated by the creation of a Technical Secretariat (TS) in the Ministry of Public Works and Equipment (MPWE) to assure global project oversight and assist local governments in submitting proposals while ensuring that sub-projects were demand-driven, identified, and partly financed by the beneficiary communities. Selection of sub-projects was based on a screening process using a series of filters including the level of community contribution, economic, social, and environmental criteria (Annex 10). A final arbitration of eligible sub-projects was undertaken by an Inter- Ministerial Steering Committee (ISC) to assure equitable distribution of works and avoid concentration of sub-projects in the capital city and a few major urban centers. AGETIP Burundi (later ABUTIP) was created to implement the sub-projects submitted by the TS. Rapid and transparent processing of procurement activities and quick payment of SME invoices were considered critical to show results and support the strengthening of the local construction and civil engineering industries. Furthermore, the Government also showed its commitment to the project by identifying, prior to appraisal, sub-projects for the first year of implementation. This helped to get the project rolling and build confidence in the local governments that the project would deliver and that they could participate and pay counterpart funds for subsequent sub-projects. In a fully transparent process, the Government recruited AGETIP Burundi's Director General to participate in the appraisal of the project. 5 2.2 Implementation Project implementation has consistently been rated satisfactory in ISRs and this is corroborated by the following factors: · The use of an experienced team for project supervision with most of the supervision done from the field; · Government commitment to showing results and a good communication strategy with beneficiaries; · Transparency of sub-project implementation, following on a participatory and transparent sub-project selection process; · Importance of competitive selection of all project staff with incentives and clear accountability; · Tailored capacity building program and rapid payment of invoice arrangements to SMEs, and the systematic application of penalties for default on contractual clauses · Regular technical and financial audits. Slight delays in sub-project implementation were experienced early on due to late payments of beneficiary contributions. This issue was addressed during the Mid-Term Review when, based on the findings of the periodic beneficiary assessment reports, the counterpart contribution for social infrastructure was lowered from 5 percent to 2 percent and in-kind payment of contributions was allowed. The first solution proved to be very effective; the second option delayed the completion of some of the sub-projects, and the quality of materials provided by the communities did not always meet the required standards. Despite a challenging institutional and governance environment for the project, the Government and the Bank were able to ensure that the project did not suffer from governance issues, particularly related to the management of AGETIP Burundi. The regular supervision missions identified issues with poor management and internal organization of the original AGETIP Burundi as well as with its governing structure, due to the fact that the Board lacked involvement of key stakeholders. The Government promptly acted upon the highlighted issues by dismissing the managing director while the Board of AGETIP was dissolved and the organization dismantled, 13 months after project effectiveness. A new agency, ABUTIP, was established to take over from the previous agency. During the transition, the TS and the Government ensured a smooth continuation of the project by maintaining key personnel in place in a temporary implementation arrangement. ABUTIP implemented the project in a satisfactory manner despite the dismissal of the new managing director two and one- half years later for conflict of interest related to two contracts, as substantiated by the Department of Institutional Integrity (INT) in collaboration with the client (Refer to Fiduciary Compliance, Section 2.4, for details). Due to a strong commitment and pro-active intervention by the client and close follow-up by the Bank, the Project implementation performance did not suffer from these issues. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization The M&E framework design was simple, taking into account the weak institutional capacity of the client, and flexible enough to accommodate the project's demand-driven approach. The key indicators focused on the generation of employment, capacity building, the 6 performance of the public works implementing agency, operation and maintenance of infrastructure, and the fight against HIV/AIDS (see Data Sheet for details). Monitoring of employment generation was based on direct reporting by works supervisors and has provided useful information on which types of works and technology are most efficient at generating employment for unskilled labor, including for women. The indicator on the infrastructure maintenance program was monitored on an annual basis by external consultants hired to assist municipalities with budget programming and assessed the actual execution of the budget and the quality of works. For the HIV/AIDS activities, awareness- raising activities were monitored through reporting by international and local NGOs hired by the TS. The operating cost of the implementing agency (ABUTIP) against the total investment for infrastructure was monitored and reported quarterly. In addition, the timeframe for payment of invoices by the public works implementing agency helped to monitor ABUTIP support to the emerging SME sector by assuring regular and timely flow of funds. 2.4 Safeguard and Fiduciary Compliance Safeguards compliance The project was rated Category B as no adverse long term impacts were anticipated. According to the original Project Environmental and Social Data Sheet, OP 4.01 Environmental Assessment was triggered by the project and the OP 4.12 on Involuntary Resettlement was applied. An Environmental Assessment (EA) was conducted prior to appraisal to develop an Environmental and Social Management Framework (ESMF) comprised of the following tools: (i) mitigation plans for environmental and social impacts, including a resettlement and compensation framework to be used in screening all sub- projects; (ii) monitoring plans and inventory of existing quarries and environmental guidelines for construction activities; and (iii) capacity building plans to addresses the needs of the Ministry of Environment, SMEs and civil society. These tools were systematically used to screen sub-projects for the original credit and the additional financing. Supervision of the environmental and social compliance of sub-projects was carried out by the TS environmental specialist in collaboration with the environment focal point in the line ministry. Environmental and social screening checklists have been adequately applied in sub-project processing, and in all cases each sub-project has a site specific Environmental Management Plan (EMP) that is reviewed by the Bank safeguard specialists. The Technical Secretariat has delivered 1,320 person-days of training on social and environmental issues to the Ministry of Environment staff and to the staff of the consulting and construction industries. The Quality at Entry Assessment (QAE) rated the additional financing moderately satisfactory, noting that the project documentation lacked detailed description and analysis of outstanding issues relating to resettlement and therefore the Involuntary Resettlement Safeguard should have been triggered. Furthermore, the QAE also noted that the project impact on specific groups, such women heads of households, ex-combatants, returnees and refugees, could have been better monitored. Although the project proved to be an opportunity for such groups, the approach of the project was to provide equal opportunities to ex-combatants, returnees and local poor communities and not to single out specific groups. In addition, while OP4.12 related to resettlement should have been triggered, a 7 Resettlement Plan Framework (RPF) and an ESMF were available as part of the preparation of the original project and both instruments were still valid and applicable to the Additional Financing. . In addition, sub-projects were selected on the basis of eleven eligibility criteria including safeguards, and all the sub-projects were implemented in full compliance with the safeguards guidelines. Fiduciary compliance Compliance with fiduciary requirements received close supervision in light of the country context. A financial and procurement capacity assessment was carried out at appraisal and an action plan developed to build capacity in the two implementing bodies. Regular supervision missions with participation of Bank financial management and procurement specialists were key to ensuring compliance and strengthening institutional capacity. Three technical audits and six financial audits were conducted and their recommendations were duly implemented. In addition, when the Bank received a complaint from a witness who alleged that the first General Director of ABUTIP was involved in a conflict of interest in awarding contracts, it referred the allegations to the Department of Institutional Integrity. The INT promptly investigated and substantiated the allegations of conflict of interest on two consultancy contracts totaling US$20,875. The Bank informed the Government of Burundi of deviations in project procurement procedures. The Government of Burundi immediately undertook its own investigation and confirmed the conflict of interest. The Director General of ABUTIP was removed from his position and misprocurement for the two contracts was declared. 2.5 Post-completion Operation/Next Phase The project has taken steps to ensure sustainability of the project's achievements in the areas of both infrastructure operation and maintenance (O&M). Throughout the country, 3,164 Local Maintenance Committees (LMCs) comprised of neighborhood residents have been established and trained to ensure proper maintenance of local infrastructure. The post- project operation of the commercial facilities constructed under the project is adequately addressed. Most of completed neighborhood markets are managed by local private operators under contracts signed with the municipalities. For the social infrastructure, local user associations have been trained to assure proper O&M of infrastructure. At the institutional level, the project has provided training and technical assistance to local governments on budgeting and resource mobilization, thus increasing the funds earmarked to maintain infrastructure, but this will need to be reinforced over time. The capacities created under the project will continue to be available to the municipalities. The same implementing bodies (ABUTIP and TS) are used to implement a US$14 million multi-sector reinsertion project financed by the African development Bank (ADB) and an IDA-funded education project. A follow-up operation has been strongly requested by the Government to cope with high demand from communities and to buttress the emergent decentralization process. Municipalities and communities have already mobilized about FBU 822 million (US$0.8 million equivalent) in local contributions to support their additional requests for sub-projects. In addition, the two growing cities of Bujumbura and Gitega urgently need adequate management tools to: (a) improve the programming, financing, and implementation 8 mechanisms of service delivery; (b) improve local resource mobilization; and (c) rationalize municipal management and infrastructure maintenance. The new Country Assistance Strategy for Burundi for 2009-2012 is scheduled for Board presentation around July 2008 and has earmarked resources for a follow-up operation to the PWECP and to strengthen the decentralization process as a result of the impacts achieved by the project. Stakeholders consulted during the preparation of the CAS have indicated that an appropriate communication strategy will be needed to address the strong demand from communities and to clarify the eligibility criteria in order to ensure a more equitable geographic distribution of the sub-projects. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation The project objectives accurately reflected the development priorities of Burundi at appraisal of the original credit and at the time of project closure and remain consistent with the Bank Interim Strategy Note (ISN) for Burundi (FY06-07), approved by the Board on May 3, 2005. One of the objectives of the 2005 ISN was to ensure that communities and populations have access to basic social services and income-generating activities. The objectives of the project are also aligned with the second pillar of the 2006 Poverty Reduction Strategy Paper which addresses increasing access to basic services, promoting the private sector, and generating jobs for the poorest segment of the population. 3.2 Achievement of Project Development Objectives All the PDO indicators were achieved and in some cases surpassed. The primary indicator of creating employment was surpassed. Sub-projects succeeded in improving social services and reviving economic activities, and reached more than a quarter of the total population of the country. Objective 1: Generate productive employment to help reconstruction in the Borrower's territory through labor-intensive public works. The project generated 6,082,722 person-days of employment compared to the combined target of 2,750,000 days for the original project and additional financing. Of this employment, 17 percent was generated for women. In addition, more than US$9.3 million was paid in salaries to unskilled workers, of a total project cost of US$70.6 million Objective 2: Promote the execution of cost-effective and economically and socially useful public works projects, including through small and medium enterprises. A total of 191 civil works contracts were signed with 52 local contractors for US$62 million, and 169 contracts were signed with 48 local consulting firms for US$4.3 million. These included the rehabilitation or the construction of: (i) 55 km of cobblestone roads, compared to 350 m with cobblestone before the project; (ii) 28 markets; (iii) 6 health centers; (iv) 230 classrooms; (v) 335 km of potable water system; (vi) 80 km of sewerage system providing wastewater evacuation system to 18,000 population equivalent; (vii) one stadium with a multipurpose hall for 300 visitors; (viii) two butcheries; (ix) 12.9 km of riverbank consolidation works carried out to protect residential plots, schools, hospitals, roads and two national bridges. Access to infrastructure has increased and the following results have been achieved: 9 · There has been a 2 percent increase in the level of access to water services with the construction of a 335 km water supply network providing water to 120,000 people; · Now 25 percent of the urban roads (55 km) are paved with cobblestone · With the construction of 80 km of sewerage lines, the network has increased in length to 260 km, bringing access to sewerage system services in Bujumbura city from 38 to 42 percent of the population; · Classroom occupancy has been reduced from 65 pupils to 50 with the 230 classrooms built and rehabilitated under the project, as compared to 100 pupils per class in the majority of the schools in the country; · Distances to health centers have been reduced from 12 km to 7 km with the construction of 6 health centers; · Three percent of the urban population has been protected against erosion as a result of the construction of 12.9 km of gullies and riverbanks; and · In terms of economic revitalization of the country, the 28 neighborhood markets constructed or rehabilitated by the project have created 30,679 points of sale and induced 5,000 new jobs from the increased economic activity generated in the vicinity of the markets. Objective 3: Build institutional capacity at the municipal level to identify, prioritize, operate, and maintain sub-projects: · Local administrations have increased their budgets for project maintenance by 80 percent against a target of 20 percent; 27 of 28 neighborhood markets constructed or rehabilitated are privately managed. 95% of the constructed/rehablitated infrastructures are operational (one market was damaged by landslide). · The project has provided more than 14,000 person-days of training in project identification, management, operation and maintenance across all 129 municipalities against a target of 80 municipalities. · Local project maintenance committees are in place over the entire country and are actively engaged in maintenance of infrastructure. · Positive experiences with privatization of management of nine markets constructed under the project are being scaled up across the country and the remaining municipalities are being supported by the Technical Secretariat to implement public- private partnerships under standardized management contracts. · In the construction sector, 5,582 person-days of training were provided to SMEs in business administration, financial management, and technical issues, surpassing the target of 5,000 person-days of training. The average invoice payment time was 4 days for a target of 15 days, a remarkable achievement, even by international standards. For ABUTIP, the contract management agency managed to keep its operating ratio at 5.9 percent of the cost of works for a target of 8 percent. · The average revenues of construction companies increased by more than 100 percent between 2002 and 2006 (from US$312,628 to US$790,503) while their number of staff increased from 16 to 29, with the average number of female employees increasing from 1.8 to 4.1. This was accompanied by an important annual investment in new equipment, corresponding to for more than 10 percent of their revenues for some of the firms, showing increase in implementation capacity of SMEs. 10 · Consulting firms saw their revenues increase by more than 100 percent between 2002 and 2004 (from US$25,010 to US$60,739). In 2000, Burundi had less than 10 registered construction consulting firms; in 2007 there were 48 firms and the six consultancy firms surveyed had invested 13 percent of their turnover in equipment (mainly computers and software) to assure continued business development. On HIV/AIDS awareness and prevention campaigns, 66 percent of project work sites benefited from the program, against a target of 85 percent. The remaining sites benefited from the Multisectoral HIV/AIDS project (MAP) which has been effective since FY03. 3.3 Efficiency Selection of sub-projects was based on a screening process using a series of filters including the level of community contribution, economic, social, and environmental criteria (Annex 10). At project completion, a detailed cost-efficiency study was carried out for the Implementation Completion and Results Report (ICR) for the various types of infrastructure. For 85 percent of physical investments, where cost-comparison across standards was possible. An analysis of the efficiency of investments in terms of unit rates of the various types of social and economic facilities was prepared and was also provided in Annex 3. This analysis shows that the unit rates compare favorably with the values observed in the country and elsewhere in Africa. For schools, and health centers, where the line ministries have provided standard drawings and guidelines, price comparisons were done by putting aside the earthworks item which are specific to each site. In addition, an economic analysis has been carried out for 18 of the 28 markets supported by the project. The analysis presented an NPV of around US$470,000 for an investment of US$11,849,723, with a weighted average IRR of 12.6 percent. Nine of the 18 surveyed markets demonstrated an IRR above 12 percent. For the remaining ones, revenues currently generated are less than in other markets as a result of lower economic activities. It is worth mentioning that the analysis did not assess additional economic benefits related to increased revenue to farmers and shop tenants and the improved services rendered to communities and consumers. The Economic Rates of Return calculated for the sub-projects are likely to understate some of the social and environmental benefits gained. 3.4 Justification of Overall Outcome Rating The project has achieved its objectives and the overall rating is satisfactory. The PDO continues to remain relevant considering the country context and the limited access to infrastructure services by the majority of the population and the pervasive poverty in Burundi. The PDO indicators were achieved and in some cases surpassed, including the primary indicator of employment creation. Sub-projects succeeded in improving social services and reviving economic activities, reaching more than a quarter of the total population of the country. Increasing funds for infrastructure maintenance was a major project achievement, laying the foundation for a sustainable management regime for local level infrastructure. SME capacity for sub-project implementation was strengthened through targeted training activities and learning by doing. Moreover, project efficiency has been evaluated as satisfactory. 11 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development Historically, Burundi has been characterized by a very centralized administrative system, with little public involvement in setting priorities and implementing developmental initiatives. The result has been an absence of community ownership of local infrastructure. The project has introduced changes in the previous approaches using fully participatory approaches at the municipal level to identify community needs, set priorities and implement, operate and maintain infrastructure. Furthermore, the project has relied on the labor force provided by women as they represented 28 percent of the unskilled workforce for the labor- intensive road paving activities. Many of these women are heads of households and the impact evaluation revealed that revenues generated through these activities have had a significant impact on empowering them and reducing poverty in their households. (b) Institutional Change and Strengthening Poor governance in the public works sector was a major problem in Burundi following years of conflict. The situation was characterized by lack of transparency, delays in invoice payments, inflated unit cost rates and cost overruns, poor quality of works and delays in execution. The project emphasized open competition and respect for contractual clauses and procedural manuals. Application of penalties for delays, unknown in the past, was generalized and contributed to introducing more professionalism in the construction industry. While 75 percent of contractors were penalized for delays in the execution of works for the first batch of sub-projects, this number had dropped to 14 percent in the final batch of sub- projects. The introduction of private operators in the management of commercial infrastructures has significantly increased the revenue of the municipalities and the quality of the service rendered to the communities. Increasing funds for infrastructure maintenance was also a major achievement of the project, laying the foundation for a sustainable management regime for local level infrastructure. (c) Other Unintended Outcomes and Impacts 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops Two beneficiary assessments were conducted during project implementation and an impact evaluation at project closure. The information collected has been used to adapt and improve the implementation arrangements over time. Contribution from the communities was lowered from 5 to 2 percent for social infrastructure sub-projects. In addition, the findings of the assessment were helpful in creating tailored training modules aimed at improving the capacity of SMEs and consulting firms. The details are presented in Annex 5. 4. Assessment of Risk to Development Outcome The risks to development outcomes remain moderate despite the mitigation measures implemented by the project. Although there has been a significant increase in budget allocation for O&M, the creation of LMCs and specific arrangements for maintenance of social and commercial infrastructure, there is still a moderate risk of poor municipal infrastructure maintenance due to inadequate funding. The 2010 election presents a potential 12 political risk of the country falling into another round of instability given the lack of strong political and economic institutions. 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Satisfactory The project design benefited from the lessons learned by the Bank through the Burundi Social Action Project I and the experience gained in the design and implementation of public works by the 12 implementing agencies (AGETIPS) active in Sub-Saharan Africa. These include: (i) separating responsibilities for sub-project selection from implementation to ensure transparency and avoid conflict of interest; (ii) using demand, including beneficiary contribution, for selecting sub-projects and increasing sustainability; and (iii) advocating the use of local materials in construction and labor intensive methods to support SME development and employment generation which have proved to be very effective on the ground. Bank performance in ensuring Quality at Entry is rated satisfactory. One drawback was the lack of assessment of local government's and communities' capacity to pay contributions in support of their requests and to set the appropriate level of contribution. The Quality at Entry of the project additional financing was assessed on August 28, 2007 (see para 2.4). (b) Quality of Supervision The Bank performance in supervision is rated satisfactory, as the team addressed prevailing issues in relation with the Borrower. Since project effectiveness, regular supervision missions visited the country and project sites and met with all key stakeholders, including contractors. Supervision was conducted in the field with a civil engineer and fiduciary team based in the country office. Issues that arose from the poor management of AGETIP Burundi at the beginning of the project were swiftly handled by the client and the Bank team, ensuring that the project continued to deliver. Later on, the Bank received a complaint from a witness for alleged cases conflict of interest in contract awards and for funds mismanagement. The case was referred to INT and the Bank proactively suspended disbursements. INT substantiated the allegations of conflict of interest on two consultancy contracts totaling US$ 20,785 and the Government of Burundi confirmed the conflict of interest. Therefore, the Director General of ABUTIP was removed from his position, after which the disbursement suspension was lifted. Based on the findings of the periodic beneficiary assessment reports, the counterpart contribution for social infrastructure was lowered from 5 percent to 2 percent and allowed in-kind payment of contributions. This change in the level of contributions accelerated the implementation of the project. 13 (c) Justification of Rating for Overall Bank Performance The overall rating of Bank performance is considered to be satisfactory on the basis of Quality at Entry and the effective supervision arrangements put in place, with strict supervision conducted by a Burundi-based task team leader since FY05. 5.2 Borrower Performance (a) Government Performance The Government performance is rated satisfactory. The client participated actively in the preparation of the project. It prepared the first group of sub-projects prior to appraisal to get the project rolling and developed local governmental trust by showing that the project would deliver and that their counterpart funds would be put to good use. The Government further assured a timely and transparent recruitment of the manager of AGETIP Burundi to participate in project preparation. The Government also supported the delegated contract management arrangement and separation of sub-project selection from their implementation. During implementation, the Government reacted appropriately to management problems at the implementing agency and worked to find solutions to the counterpart contribution issue. The Inter-Ministerial Steering Committee met regularly and assured equitable geographic distribution of sub-projects. Despite a tight financial situation, the Government managed to pay 100 percent of its counterpart funds. However, one issue remains regarding an outstanding payment of US$533,000 corresponding to about 20 percent of the required beneficiaries' contributions, but this has had limited impact on the completion of sub- projects. This balance is partially due to price increase of sub-projects between the appraisal and the implementation period (regional cement and fuel price hike). (b) Implementing Institutions Performance The project implementation performance is rated satisfactory, based on the contribution of the two implementing institutions and the client to the project outcome. Technical Secretariat The TS assured smooth coordination of the project and ensured capacity building at local and central levels, following up on local governmental performance of infrastructure operations and maintenance and assured timely transfer of the list of selected sub-projects to ABUTIP. In addition, the project included environmental impact analysis into the sub- project approval process and closely monitored implementation of its mitigation plans. The satisfactory performance of the TS has been recognized by other partners and this body is now being used to implement the AfDB-financed Multi-Sectoral Reinsertion Project. ABUTIP ABUTIP improved its operating efficiency by keeping its operating cost at 5.9 percent of total project investment in infrastructure, compared to the target of 8 percent. Average time for payment of invoices was reduced to 4 days instead of the targeted 15 days. This performance has been impressive and was greatly appreciated by the local SMEs. The strong enforcement of contractual clauses and systematic application of penalties for delays in 14 contracts execution has helped strengthen the capacity of the local construction industry and brought discipline in the construction sector. (c) Justification of Rating for Overall Borrower Performance Based on the analysis above, the overall borrower performance is rated satisfactory. 6. Lessons Learned The delegated contract management arrangement is well-suited to mitigating capacity constraints in a post-conflict environment. Showing peace dividends in a post-conflict environment requires swift implementation of programs while ensuring that fiduciary requirements are met. Delegated contract management arrangements using streamlined and transparent procurement procedures and faster payment methods of contractors have proved to be very efficient in delivering quick results on the ground. This arrangement has allowed the disbursement of more than US$75 million in 6½ years and access to infrastructure services has been increased tremendously in a very short period of time. Such performance could easily be attributed to project implementation arrangement. Demand-driven selection through participatory arrangements of communities' infrastructure needs is most suited to address social cohesion and sustainability of services in post-conflict environments. With a social fabric very fragmented as a result of years of conflict and the persistent perception of Government unfairness in allocating public resources, a demand-based approach to the selection and financing of community infrastructure needs is appropriate to ensure transparency in resource allocation and strengthened social cohesion, while at the same time improving the prospects for sustainable operation of financed facilities. However, a proper mechanism to ensure equitable distribution of sub-projects is needed to avoid project concentration in the capital city and a few major urban centers. In addition, strong ownership at all levels and the demand-based approach used by the project to select activities to be financed have precluded the project from failure and have mitigated the fiduciary risks faced by the project considering the multiple changes that took place at central, local and implementing agencies' levels. Short term employment creation is critical in post-conflict environments. Public works projects using labor-intensive technologies can address high unemployment levels and also the revenue-generating needs of conflict-affected groups such as single women, ex- combatants, returnees and refugees. Post conflict countries are faced especially with high levels of unemployment and poverty that can only be structurally addressed in the medium- to long-term timeframe. Burundi is blessed with abundant local materials and has developed labor-intensive technologies that have been successfully used for paving roads, and for building schools, markets and health centers. This approach has significantly reduced the need for imported materials and has reinforced the capacity of the local construction industry while supporting the distribution of revenues to an unskilled labor force. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies The draft ICR was translated and shared with the Client and the implementing agencies. The Client has provided its contribution (Annex 12). 15 (b) Cofinanciers There were no direct cofinanciers at the time of appraisal. The African Development Bank is using the same project implementation bodies and approaches for its ongoing operation. (c) Other partners and stakeholders (Annex 5 &6) 16 Annex 1. Project Costs and Financing (a) Project Cost by Component (in US$ million equivalent) Appraisal Additional Actual/Latest Components Estimate Financing Estimate Percentage of (US$ million) (US$ million) (US$ million) Appraisal A. SUB-PROJECT IMPLEMENTATION & CAPACITY BUILDING FOR 33.26 30.60 67.2 105.2 EMPLOYMENT CREATION B. INSTITUTIONAL BUILDING 4.54 0.00 6.06 133.6 C. SUPPORT FOR PREVENTION OF HIV/AIDS 1.00 0.00 0.43 43.2 D. PROJECT MONITORING 1.75 0.00 0.04 2.2 PPF REFUND 1.00 0.00 0.00 0.0 Total Project Costs 41.55 30.60 73.73 103.6 Physical Contingencies 0.00 0.00 0.00 0.00 Price Contingencies 0.00 0.00 0.00 0.00 Total Project Costs 0.00 0.00 0.00 Front-end fee PPF 0.00 0.00 0.00 .00 Front-end fee IBRD 0.00 0.00 0.00 .00 Total Financing Required 0.00 0.00 0.00 (b) Financing Appraisal Additional Actual/Latest Source of Funds Type of Percentage of Cofinancing Estimate Financing Estimate (US$ million) (US$ million) (US$ million) Appraisal Borrower 0.24 0.00 .00 IDA 40.00 30.60 73.73 103.6 Local Governments of Borrowing Country 1.31 0.00 .00 17 Annex 2. Outputs by Component Component 1: Sub-project Implementation and Capacity Building for Employment Creation Cost: Appraisal: US$33.26 million. Additional Financing: US$30.6 million. Actual: US$67.2 million. 122 sub-projects were constructed in 60 communes and the City of Bujumbura. · 27 new markets and one rehabilitated market · 2 butcheries · 24 road-paving sub-projects (covering 55.50 kilometers2- 60 lots) · 1 culvert · 9 drainage and dirt road resurfacing sub-projects (13.75 kilometers) · 3 sub-projects of land development (64.29 ha including 12.96 kilometers of dirt roads) · 30 school infrastructure projects totaling 230 classrooms and also including 5 teachers' residences · 6 health centers · 1 urban sewerage sub-project (80 kilometers of sewerage network) · 1 sub-project to create the basic elements for an Olympic stadium · 7 urban rainwater drainage sub-projects (8.150 kilometers3) · 9 water supply networks (324 kilometers) and rehabilitation of one water supply network (9 kilometers) · 1 communal office · 3 sub-projects to protect 5 ravines against erosion (7.4 kilometers total) · 2 sub-projects to protect river banks (5.5 kilometers) Component 2: Institution Building Cost: Appraisal: US$4.54 million. Actual: US$6.1 million. Training programs for consulting firms and construction companies have involved 814 managers and employees of more than 203 entities for a total of about 5,582 person-days of training in business, financial and technical management of companies, as well as carrying out project environmental and social impact evaluations. More than 14,963 person-days of training at both the central and decentralized levels were carried out in the following areas: (i) 4,852 person-days of training of the local maintenance committees in communal infrastructure maintenance; (ii) 882 person-days of training in management of privatized management contracts for markets and butcheries; (iii) 1,320 person-days of training in environmental concerns (iv) 2,349 person-days of training in local 2Including 250 meters of paved road in connection with works in Rumonge (construction of a culvert and paving of a road). 3 Including 1.32 kilometers of collectors in Rumonoge. To this should be added 0.4 kilometers of gutter in Kayanza and Kirundo. 18 development planning; and (v) 5,560 person-days of training in accounting. Further, the PWECP has undertaken 133,947 person-days of training for the cobblestone cutters. A total of 70 studies were undertaken by the TS/PWECP for an amount of FBU 4.5 billion. Component 3: Support for the Prevention of HIV/AIDS Cost: Appraisal: US$1.00 million. Actual: US$0.4 million. Awareness campaigns had been carried out in 100 percent of project sites by the end of 2004, when the World Bank financed Multi-Sectoral AIDS Project (MAP) took over the activities. Implementation of activities was returned to the PWEC after the MAP could no longer finance them beyond the end of FY 2006. Not counting activities financed by the MAP, 66 percent of work sites were affected over the project duration, against a target of 85 percent. Site-specific data are not available for the MAP-financed activities. Under this component, 43,946 persons had participated in awareness-raising activities by the end of 2004. For the second phase of activities, started in the second half of 2007, 3,540 persons were affected, and of these more than 17 percent were women. Component 4: Project Monitoring Cost: Appraisal: US$1.75 million. Actual: US$297,959. Three beneficiary surveys have been carried out, including the final impact evaluation. 19 Annex 3. Economic and Financial Analysis No economic evaluation was carried out at appraisal since the sub-projects were demand driven. For the ICR, a study on cost-efficiency has been carried out for the various types of infrastructure. In addition, the Internal Rate of Return (IRR) and the Net Present Value (NPV) have been calculated for a representative selection of market infrastructure. Cost-efficiency The cost-efficiency study shows that for each works contract tendered, more than seven bidders responded to the competition. The high turn-up of competitors has had an impact on the pricing of works. For 85 percent of investments under Component 1, where cost- comparison in terms of standard and service was possible, unit cost rates compared favorably with investments in Burundi and in the Region as shown below. Average cost per person served (US$/person) Type of facilities Project Regional Average Simple gravity systems 36 50-80 Major systems (>70 km) 66 >100 Average cost (US$/standard unit by type of infrastructure) Type of facilities Project National Average Average cost per classroom (US$/classroom, primary school) 27,420 27,680 Average cost per Health center type B (US$/health center) 304,130 330,000 Average cost per linear meter (lm) of paved road by cobble 215 302 stone (US$/lm) 328 (Rwanda) Average cost per square meter of covered neighborhood 127 138 market (US$/SM) N.B.: a. A basic standard primary school includes (i) a block of six classrooms covering 472 square meters, (ii) an administrative block; and (iii) a block of latrines (12 stances). b. For paved roads with cobblestone, the cost for a standard linear meter is comprised of an average width of 7 meters including drainage on both sides. Economic analysis of markets An economic analysis has been carried out for 18 of the 28 markets supported by the project. The markets account for 30.8 percent of sub-project investments of the project. The analysis uses a discount rate of 12 percent and assumes a 20-year4 investment horizon corresponding 4Inflation is estimated at 9% per year. 20 to the average life of the assets, a 12.5 percent annual increase in revenues generated by the operation of markets and a 5 percent annual increase in expenses. Assumptions. In terms of revenues, most of the constructed markets included on the average 100 shop windows (covered stalls), 200 kiosks, 300 open stands, 1 block of toilets, 1 butchery with a cold room and a parking lot for more than 100 vehicles, generating revenues ranging from US$25,000 to US$80,000 in the form of rent for the different services and patent fees. Out of the above revenues, the operation of markets incurs expenses that include mainly (i) employment cost for 32% of the total expenses, (ii) transport and maintenance for 3%, (iii) water/electricity for 5%, and (iv) depreciation/rent of the facility which accounts for 50% paid to the municipality. Since the majority of markets are run by local private operators, there is a corporate tax based on the benefit margin which accounts for 35% of the balance of the above total revenues minus expenses. The analysis presented an NPV of around US$470,000 for an investment of US$11,849,723, with a weighted average IRR of 12.6 percent. Nine of the 18 surveyed markets demonstrated an IRR above 12 percent. For the remaining ones, revenues currently generated are less than in other markets as a result of lower economic activities. It is worth mentioning that the analysis did not assess additional economic benefits related to increased revenue to farmers and shop tenants and the improved services rendered to communities and consumers. The Economic Rates of Return calculated for the sub-projects are likely to understate some of the social and environmental benefits gained, in particular: (i) Employment generation: by appropriately dimensioning public works using labor- intensive methods, the project has created employment opportunities for unskilled and semi- skilled labor and promoted SMEs in construction and urban services. The project efficiently used the US$67.2 million of Component 1 for infrastructure investments to generate 6,082,722 person-days of employment, with US$9.3 million paid in salaries to unskilled laborers, corresponding to 14 percent of investments going to unskilled labor. (ii) Improved physical living conditions and better environmental management has had a positive impact on the quality of life. 21 Annex 3: b) NPV and EIRR of the neighborhood markets NPV (NET PRESENT MARKET VALUE), INVESTMENT, N° BENEFICIARY COMMUNE DESIGNATION EIRR (%) FBU FBU 1 Matongo Bandaga 13.0% 18,141,769 473,432,094 2 Bubanza Bubanza 14.3% 95,217,546 544,622,133 3 Kirundo Kirundo 11.1% -78,812,365 596,728,336 4 Muyinga Muyinga 13.9% 78,037,146 594,664,367 5 Giteranyi Giteranyi 9.2% -261,405,926 907,611,000 6 Mpanda Musenyi 13.3% 68,224,817 950,585,475 7 Musigati Musigati 13.5% 38,915,990 429,474,521 8 Rugombo Rugombo 10.7% -108,734,285 631,955,604 9 Mugina Nyeshenza 9.5% -222,295,821 839,081,293 10 Bwambarangwe Mukenke 8.3% -284,891,073 795,989,742 11 Makamba Makamba 15.9% 186,438,264 550731198 12 Mugamba Tora 15.5% 92,746,212 305794073 13 Rutana Rutana 8.9% -225,748,119 686494430 14 Buyengero Muyama 3.6% -363,778,199 563629612 15 Kayokwe Mwaro 13.3% 29,985,772 452908547 16 Kabezi Kabezi 7.9% -156,581,100 400467823 17 Buyenzi Buyenzi 17.6% 1,797,289,455 3317152691 18 Cankuzo Cankuzo 8.8% -228,170,761 625,384,169 TOTAL 474,579,323 13,666,707,108 WEIGHTED AVERAGE EIRR 12.6% NB: The analysis uses a discount rate of 12 percent and assumes a 20-year investment horizon, a 12.5 percent annual increase in revenues generated by the operation of markets and a 5 percent annual increase in expenses. 22 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Responsibility/ Names Title Unit Specialty Lending Eustache Ouayoro Senior Water and Sanitation Specialist AFTU2 TTL Jainaba Kah Urban Specialist AFTU2 Economic analysis Pamphile Kantabaze Senior operation officer AFT HD Country coordination Prosper Nindorera Procurement specialist AFTPC Procurement Bernard Abeille Principal adviser Procurement AFTPC Procurement Isabella Micali-Drossos Counsel AFTPC Legal Agnes-Albert-Loth Finance Officer LOAFC Disbursement Abdul Haji Sr. Financial Specialist AFTQK FM aspect Ernestina Attafuah Sr. Language Program Assistant AFTU2 Cost table Sophie Hans-Moevi Language Program Assistant AFTU2 Concept/PAD consolidation Leoncie Niyonahabonye Program Assistant AFMBI Logistic suppport Supervision/ICR Otieno Ayany Financial Management Specialist AFTFM FM aspect Mary C.K. Bitekerezo Sr. Social Development Specialist AFTCS Social safeguard Bella Lelouma Diallo Sr. Financial Management Specialist AFTFM FM aspect Yvette Laure Djachechi Sr. Social Development Specialist AFTCS Social safeguard Sophie Hans-Moevi Language Program Assistant AFTU2 ISR Ludovic Joseph Kabran Sr. Operations Officer CSRRM FM aspect Jean Charles Amon Kra Sr. Financial Management Specialist AFTFM FM aspects Anta Loum Lo Language Program Assistant AFTEG Report writing Prosper Nindorera Sr. Procurement Specialist AFTPC Procurement Seraphine Nsabimana Program Assistant, AFMBI AFMBI Africa Eshogba Olojoba Sr. Environmental Specialist AFTEN Environmental 23 safeguard Eustache Ouayoro Sector Manager AFTU2 Quality assurance Deo-Marcel Niyungeko Municipal Engineer AFTU2 TTL, since FY05 Emmanuel Tchoukou Financial Management Specialist AFTFM FM aspect Richard Verspyck Consultant AFTU2 Monitoring & Evaluation (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle US$ thousand (including No. of staff weeks travel and consultant costs) Lending FY99 36.42 FY00 25 78.67 FY01 47 162.29 FY02 9 21.00 Total: 81 298.38 Supervision/ICR FY01 11.91 FY02 17 96.80 FY03 23 57.78 FY04 42 83.47 FY05 32 43.17 FY06 40 76.40 FY07 30 41.14 FY08 6 13.43 Total: 190 424.10 24 Annex 5. Beneficiary Survey Results (if any) Three beneficiary assessments were carried out during project implementation. The first one was carried during the first year of implementation (2002), a second one at mid-term review (2004), and the last one around the project closing date. The issues raised were immediately addressed, in close partnership with the Borrower These included (i) lowering the counterpart contribution for social infrastructure from 5 percent to 2 percent, (ii) accepting in-kind contribution, (iii) mainstreaming the annual stakeholders workshop, (iv) customizing SME and local government authority training programs based on recommendations of the training needs assessment report, and (v) expanding the Interministerial Steering Committee (ISC) to accommodate the participation of the ministries of health, water, energy and mines. The expansion of the ISC helped the project ensure the availability of health workers just after the completion of health centers financed by the project. Lowering the counterpart contribution for social infrastructure from 5 percent to 2 percent proved to be very effective but the second option related to the contribution in-kind delayed the completion of some sub-projects because the quality of materials provided by the communities did not always meet the required standards. Furthermore, the question of equitable distribution of sub- project countrywide was regularly addressed. ICR Survey The final beneficiary survey (the ICR impact study) built on focus group interviews of beneficiaries and in-depth interviews of key stakeholders from a selection of 28 sub-projects, with all types of infrastructures represented and covering the entire country. It has not been possible to verify the methodology for all aspects of the study. According to the interview guide presented in the survey, the interviews focused on changes in the behaviors of the beneficiaries of sub-projects. Furthermore, the survey made use of a method to synthesize findings of focus group interviews on qualitative aspects into simple scores on a scale of satisfaction/dissatisfaction of various project outputs. The survey presents findings and general trends on a number of quantitative aspects, such as increase in revenues of market vendors, increase in land prices and rent, increased urban mobility. The findings are presented below: The general perception of project outputs is positive. Markets have been positively received. More and more women are starting up businesses, goods are better protected in the new buildings, and the formalization of the management of markets has contributed to weeding out unwanted dishonest informal vendors. The permanent nature of the buildings associated to the fact that the structures are environmentally friendly was also highlighted as a positive aspect. Road paving and drainage have impacted positively on urban mobility in general. The road paving and drainage works benefited more than 65,000 people living directly next to roads, improving urban mobility in general and specifically reducing incidences of flooding from two to three per year in the affected neighborhoods before the works to none or one maximum after the completion of works. 25 Construction of school classrooms has improved teaching conditions in general. All survey respondents highlighted the improved schooling conditions as a positive aspect. 26 Annex 6. Stakeholder Workshop Report and Results (if any) Supervision missions systematically held annual implementation workshops with stakeholders (municipalities, community leaders (CDCs), contractors, consulting firms, NGOs, implementing bodies (ABUTIP, TS, ISC), chamber of commerce and representative of the ministries in charge of public works, environment, mines, education and health) and the team took such opportunity to improve implementation performances. The main issues raised during this annual forum included (i) the annual update and classification of contractors and consulting firms based on their specialization, capabilities and experiences taking into account the new registered SMEs, (ii) the need to update the sub-project eligibility criteria to promote equitable distribution of sub-project countrywide and the need to avoid complex sub-projects which are expensive and require very skilled labor to operate and maintain, (iii) the need to provide water supply and electricity to the constructed /rehabilitated infrastructure by the project for an effective operation and maintenance, (iv) the tailoring of training programs of the local maintenance committees and the provision of first maintenance kits, and (v) the need to conceive an agreement between the GoB and the User Association (UA) or NGOs on the use of some government infrastructure such as schools operated by NGOs and churches. Furthermore, the stakeholders recommended that the GoB takes advantage of the capacities created under the project and the accumulated experiences of the two implementing bodies (ABUTIP and TS) to continue implementation of the government infrastructure programs and reinforce the capacity of the municipalities. . 27 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR The summary from the Borrower's ICR: Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR The summary from the Borrower's ICR: EXECUTIVE SUMMARY This completion report relates, respectively, to a) the evaluation of the Public Works and Employment Creation Project (PWECP) at its design and implementation stage, b) the performance appraisal of the borrowing client which in this case is the Government and c) the performance appraisal of the other stakeholders, i.e. IDA, the research departments and other partners. The completion report is carried out, as its name suggests, at the time of project closure and aims to highlight the strengths and weaknesses identified during the design and implementation stages and to identify the lessons learned in order to make use of them in comparable future projects. · Appraisal of the Project objectives The Project objectives aimed to respond to Burundi's particular situation, i.e. a post conflict situation characterized by the breakdown of basic infrastructure, the deterioration in the level of social services, widespread impoverishment, etc. The Project also responded to the recommendations that came out of the Burundi Peace Negotiations in Arusha, signed in August 2000. One of these recommendations cited reconstruction efforts and productive employment creation as being key to consolidating peace. A labor-intensive job creation approach was used to rehabilitate basic infrastructure and large scale reconstruction projects throughout the country helped the recovery of SMEs and consulting firms. This approach aimed to provide a foundation for a swift return to peace, which was the overall objective of the Project. · The Government's role in the Project design While Government's role may have been less proactive during the project identification and appraisal stages, it had a positive influence that contributed to the success of the Project. It was able, for example, to adjust the beneficiary contribution to the cost of sub-projects from 10% to 5% and later to 2% for social sub-projects. It also strengthened institutional arrangements for implementing the project with a Technical Secretariat responsible for liaising with grassroots communities for sub-project identification, programming and maintenance, as well as for coordinating Project and other donor operations, building synergies and ensuring consistency with the overall development guidelines. Also in support of strengthening implementation institutional arrangements, the Government entrusted the implementation of the Project's infrastructure component to a Maître d'Ouvrage Délégué (MOD), and to that effect, established an autonomous executing agency. 28 Separating the sub-project planning, programming and implementation functions also helped to ensure continued beneficiary involvement and develop ownership, which is contributing towards the sustainability of project-built infrastructure. · Role of the Government in implementing the Project The Government fulfilled its obligations outlined in the credit Agreement by providing the resources and assistance necessary to implement the Project and by introducing conservatory measures when necessary to protect the Client's interests and also enforce compliance with the financing agreement clauses particularly when the MOD contract arrangement with the Agence d'exécution des travaux d'intérêt public - AGETIP (Agency for the Execution of Public Interest Works) was terminated in 2002. Apart from the termination of the agreement between AGETIP and the Government, which will be discussed below, and some of the delays associated with the signing of the MOD contract agreements between the communes and the Agency, the various agreements between the parties that regulate the Project proved to be successful. The rules governing project operations have been documented in the procedure manuals and have proved to be effective and flexible enough to sustain project implementation under changing circumstances, including various agreements which had been continued under the COG arrangement and the arrangements linked to the additional grant awarded at the beginning of 2006. These rules contributed to the building of a level playing field and, particularly with regard to procurements, to the satisfaction of the stakeholders in as far as there were no fundamental conflicts related to misinterpreting or incorrectly applying the rules. The Government and the donor proved to be flexible with regard to tailoring the basic texts of the Agreement to the reality in the field. Certain conditions relating to the National Bidding threshold were revised upward, certain steps in the no objection request were reviewed and, lastly, studies and monitoring activities were combined for each office in order to speed up the implementation process. The Project implementation arrangements were based on two essential assumptions: (i) the comparative advantage of private management over public management with the introduction of an Agency that operates on a private basis with limited government involvement; (ii) the centralization and strengthening of the Agency Director's authority in order to make him individually accountable and to put pressure on him to achieve success. With regard to this assumption, the October 2000 assessment mission noted "that in view of the private management nature of the Agence burundaise pour la réalisation des travaux d'intérêt public - ABUTIP (Burundi Agency for the Execution of Public Works) it is essential to encourage individual responsibility by giving the Director General full scope to select his immediate staff and to negotiate their contracts" and of course terminate their contracts if their performance is unsatisfactory. 29 The Project was affected by two crises (the termination of the MOD contract agreement between AGETIP and the Government and the termination of the 2nd Director General's contract) which exposed the limitations of these two assumptions, i.e. the absence of an early warning mechanism in the case of internal management failures. The key reasons of the latter are a lack of joint management and shared responsibility at the management level of the Agency, the absence of mechanisms allowing prompt decisions to be made or sanctions to be carried out as a result of the limited level of government involvement in the management of the Agency, and the lack of clarity with regard to the respective responsibilities of the Government and of the Agency's management (Executive Committee) with regard to decision making. The implementation of the Project also revealed the inappropriate nature of the Agency's legal status. The Agency is governed by the 1992 law relating to the organization and operation of non-profit organizations. The various delegated contract management arrangements link the Agency to different partners with the result that it has relationships of a private contractual nature with inherent responsibilities. The payment system that has been kept means that the Agency is responsible for operating costs. This does not put it in a favorable position with regard to meeting the other obligations associated with fulfilling its contracts. The pending action in the courts against the Agency illustrates the ambiguity and the limitations of the current legal status. The obligations and accountability imposed by its legal status is inconsistent with the level of its resources and autonomy. · Project performance Despite these failings the Project has been able to deliver impressive results. For example, the credit was fully disbursed one year ahead of the original closing date given in the Agreement and almost the entire grant was absorbed within just 16 months of operation. While its initial closing date was December 31, 2006, the credit was in fact fully disbursed by December 31 2005, although the grant was only released at the end of 2006. In terms of results, the Project supported 122 sub-projects which are listed in the annex. These included, in particular, 55 sub-projects associated with basic infrastructure including 23 road paving sub-projects, 7 sanitation and unpaved road resurfacing sub-projects, 6 rain water drainage sub-projects, 2 urban waste water disposal sub-projects, 5 river and gully management sub-projects, 9 drinking water supply sub-projects and 3 building site development sub-projects. The Project also supported 37 socio-economic facility sub-projects including 29 school facility sub-projects, 6 health centers, 1 Olympic center and 1 communal office facility. Lastly, the Project supported 30 market sub-projects including 28 markets (including a rehabilitated market) and 2 slaughterhouses. These 28 markets supported the creation of 22,238 points of sale, 2,555 enclosed stalls, the development of 5,381 indoor market sites, 505 butchery spaces and 13,637 sites directly close to these markets. They also supported the creation of 5,000 additional jobs inside and around the market as well as the development of new distribution chains. 30 These achievements have improved the population's access to basic services, created spin- off services, improved quality of life and hygiene, created jobs and injected revenue into the economy. The road paving, urban improvement and sanitation sub-projects have prompted residents to become more aware of environmental issues, an appreciable improvement in the conditions related to the movement of goods and people as well as hygiene conditions. The river and gully management sub-projects have served to protect 200 public infrastructures and 22,000 households and has at the same time generated revenue-earning activities in the redeveloped areas. The sewage disposal sub-projects in the commune of Ngagara allowed 2,275 households and 20 public infrastructures to be reconnected to the system and the area to be cleaned up. Pools of water were also removed which were breeding grounds for mosquitoes. The drinking water supply sub-projects have allowed the construction of 570 water supply points providing access to 120,000 beneficiaries including 19,200 pupils in more than 51 establishments. These sub-projects contributed to a 2% increase in the level of drinking water supply in rural areas as well as a reduction of 50% in the time spent by women and children collecting water, etc. The school infrastructure and equipment sub-projects have helped to reduce overcrowding in schools with the number of pupils in each classroom falling from 65 to 50 pupils, and they have also led to a 70% reduction in the distance walked by each student to school as well as an improvement in the hygiene and working conditions for the teachers. Lastly, the health center construction sub-projects have resulted in (i) a reduction in the distance covered in order to access to health care, which fell from 12 km to 7 km in the communes concerned; (ii) improvement in health care coverage and reduced pressure on the previously existing health care infrastructure. The project had generated 6,082,722 person-days of temporary employment by December 2007, 17% of which was generated for women. This is equal to 3,000 full time jobs over a period of six years, or 10% of full time jobs in the informal sector in Burundi and 5% of the jobs in the public and parapublic sector in accordance with the 2004 report on human development. This employment resulted in revenue for the unskilled labor force amounting to USD 9.3 million. The Project has prompted the development of new management tools and reform initiatives. This includes, in particular, the drawing up and introduction by private operators of a standard management contract for the drinking water supply networks and markets and an up-to-date inventory of public infrastructure and equipment for which, together with the maintenance requirements, the local administrations are responsible. Other examples include the creation of a national map showing the geographic distribution of the natural paving stone quarries and the development of standard prospection permits for the quarries, issued by the local administration. 31 · Project performance against the conditions stipulated by the credit Agreement For the most part, the Project results were on target, i.e. operating costs were reduced to less than 8% of the cost of works, i.e. 5.87% for the whole Project. The Project has led to a drastic reduction in the average period of time between invoice receipt and payment to 4 days compared to the 15 days outlined in the Agreement. Similarly, 7 companies submitted bids for works contracts instead of 3 as originally expected. The rehabilitated infrastructure was utilized and has been effectively maintained. Target groups were trained according to plans and campaigns to increase HIV/AIDS awareness were introduced and reached more members of the public than forecast, i.e. 100% of the work sites in 2007 compared with the 85% targeted. However, a total of 66% of the work sites benefited from the program during the whole period of the Project. The only delays related to the maximum period between sub-project approval and the initiation of studies (6.1 months instead of 2). This was to due to various factors such as the minimum period required to recruit consulting firms, the lack of project officers at the start of the Project, insufficient requests for proposal and the absence of or insufficient beneficiary contributions in certain cases. These delays had an impact on the other efficiency parameter since they led to delays in starting up the sub-projects. 50% of sub- project works were supposed to start 9 months after transmittal of the TS but this was delayed to 15 months due to the reasons given above. · Project visibility The Project's visibility and the population's confidence in it is reflected by the improved collection rate of beneficiary contributions as well as the level of anticipated payments amounting to BIF 822 million at 12/31/2007 for 1,162 approved sub-projects. The Project has had more of an impact than any other previous stakeholder operations. Construction sites were opened up in 60 communes throughout the country where billboards provided information about the Project. · World Bank performance For its part, the Bank fulfilled its obligations outlined in the Agreement by ensuring timely disbursements, following up on the no objection requests within the required time frames and providing guidance and support for the Project through its supervision missions. As mentioned before, the Bank proved to be flexible with regard to changes in procedures when constraints in the field so demanded. This flexible approach was also evident when it agreed to provide the country with a grant in response to the pressing demands of the beneficiaries when the expiration date of credit was reached. This avoided the long process of negotiating further credit. · Performance of other partners The Project allowed 47 consulting firms to be set up (compared with the 4 that were regularly in operation in 2000). These firms were awards BIF 5.331 billion worth of contracts including 171 studies and works control contracts. 32 Revenues from these contracts help these firms to become more organized and to purchase modern equipment. Their involvement in the various aspects of the construction business allowed them to increase their competencies and widen their experience. The Project also supported 44 different companies and 12 groups with contracts of approximately BIF 62 billion which contributed to the country's wealth and helped companies to become self-reliant with regard to equipment and personnel recruitment. These companies and research offices benefited from training in business administration, financial management and technical skills. 5,582 person-days in training were provided. The communes also benefited from capacity building with regard to planning, management and infrastructure maintenance with 8,635 person-days of training provided in 2003 and 14,963 person-days of training in 2007. Thanks to the privatized management of contracts introduced within the framework of the Project, the communes now earn more money and no longer have to worry about management concerns. In fact, expenditure such as security and maintenance which can sometimes put a strain on the commune's budget is now assumed by the management which is able to collect more revenue due to an improved collection system. As such, Mabanda market now earns BIF 2 million per month, Makamba market earns BIF 3 million and Matana market earns BIF 2.5 million. The beneficiary population took control of mobilizing contributions and infrastructure maintenance through local maintenance committees. · Lessons learned · Lessons learned with regard to Project design With regard to client involvement at the identification stage of the Project. The Client's involvement made it possible to lay the foundations for the Project's operating mechanisms which, since they were tailored to the situation in Burundi, ensured the Project's success. The client effectively changed two key components of the donor's proposals. These related to the creation of a national structure to link the Agency and the beneficiaries and to provide support to the beneficiaries with regard to preparing and implementing the sub-projects and also to a review of the beneficiary contribution rate for infrastructure projects which was initially decreased from 10% to 5%. A further revision of the contribution rate allowed the start-up of the sub-projects. It became clear later on that if the rate had remained at 10% and if the contribution had been due before start-up then very few communes would have been able to readily participate. With regard to defining objectives The Project's objectives were both relevant and practical. They created a framework aimed at dealing with the challenges that existed at the start-up phase of the Project when the country faced uncertainty, damaged infrastructure, a lack of access to basic services and widespread impoverishment, etc. 33 Beneficiary commitment is dependent on how well the Project objectives are perceived and how well they are tailored to the environment. · Lessons learned with regard to Project implementation With regard to Project implementation arrangements. On a more technical note, institutional responsibilities were divided between a works implementation structure (Project component A), and an institutional capacity building and sub-project monitoring structure (Project component B and D). Given the importance of both functions and the distinct specialized skills they require, the dual institutional structure proved to be more cost effective than the combination of both functions in one structure as suggested during the negotiation stages of the Project. Combining both functions in a single agency would have resulted in time delays or could have led to one function having priority over the other. With regard to the adjustment mechanisms that guide the partners and Project implementation structures, The relationships between the various partners including the Bank and the Client (credit Agreement), the Client and the agency (MOD agreement), the TS and the Agency (portfolio agreement), the Bank and the agency (Project Agreement), as well as local governments and the agency (specific MOD contract) were governed by agreements stipulating the rights and obligations of each of the contracting parties. Arrangements are based on the principle that specialization and separating responsibilities leads to greater efficiency as mentioned above, but also on a realistic understanding of the stakeholders' competencies. Obviously, at the start of the Project none of the communes had the necessary skills to be able to prepare, implement or monitor the projects themselves. Striving for efficiency also called for specific private sector-based procedures which helped to avoid the lengthy and complicated public procurement procedures, particularly useful in a crisis situation undermined by all sorts of demands. The works contractors preferred to bid for Project contracts rather than those offered by the State despite strong competition and tighter pricing, proof that the efficiency of this system was considered to be more equitable. Half of all contractors surveyed indicated that if the working conditions and the resources available to them meant that they could only choose between bidding for a Project contract or a government contract then they would automatically bid for the Project contract as long as the government contract value was not more than double that of the estimated Project contract value. The contractors stated that their decisions were governed by three factors: risks associated with procurement (project procurement viewed as less risky), shorter payment terms and a more reassuring contract management approach, particularly with regard to the frequency of project implementation monitoring and the management of potential conflicts. These same contractors also indicated that even if the estimated value of the government contract was double that of the Project contract, the expected benefits from the government contract would always be undermined by the impractical payment terms, which are particularly strict if a bank guarantee is involved. 34 With regard to the Client contribution The Client and the beneficiaries met their obligations satisfactorily once they had been convinced of the merits of the Project. They paid their contributions and provided the necessary support to ensure that the Project ran smoothly. Lessons learned showed that clearer Project objectives meant beneficiaries paid their contributions more readily. With regard to Project operation The Project was governed by well-thought rules, guaranteeing its smooth operation whatever the type management structure or activity. Key features of the management rules included: - neutrality (no conflict or very little) - adaptability (the rules applied under the COG, AGETIP and ABUTIP systems, both with regard to credit and grants) - flexibility (the Project has the advantage of being able to adapt the rules to the context without surrendering its key elements of efficiency, competition and fairness. With regard to the sustainability of the structure: Two factors threaten the Agency's sustainability: - the constraints associated with the Agency's status and the payment methods that have been put in place - limited government support The Agency is governed by the 1992 law on the management of not-for-profit organizations. The Agency is, however, not a not-for-profit organization in the classic and legal sense of the term since its operating methods are contractual in nature with associated obligations and duties and it uses a delegated management contract approach. Its status means that payment can only be negotiated on the basis of the known operating costs without taking development or risk compensation costs into account. However, the identification mission suggested a payment of 5% of the works contract to the Agency with no other precisions regarding the amount. Settling the potential liabilities associated with losing the lawsuit against the Agency is obviously problematic in view of the current state of affairs. With regard to government support, the government has not demonstrated any sustained commitment to approving works in favor of the Agency. The government's contract procurement and implementation institutions should, however, take a guarded approach to a competitive institution. There is also the issue of mobilizing local resources both for maintaining the existing infrastructure and also to expand it. A tax system tailored to the local economy has been introduced to respond to this requirement. This was put forward in the recommendations. 35 · Conclusions and recommendations The Project has produced substantial results thanks to solid management tools, stakeholder compliance with the conventions, stakeholder commitment and the experience of some of its managers. The following recommendations advocate capitalizing on the advances that have been made and adjusting certain internal management mechanisms in order to guarantee the Project's long-term success. The recommendations relate to 7 key areas: i) The project should be extended based on its performance, visibility and the continued demand that justified its implementation and the scope of its operations ii) Contractual commitments should be pursued (collection of overdue payments) and the litigation settled in order to meet the project extension requirements iii) Progress made by the Project should be sustained iv) The institutional management framework should be fine-tuned in order to correct past deficiencies by: a) Specific actions aimed at improving local governance with regard to: · Mobilizing local resources · Financial and accounts management · Urban planning and maintenance of infrastructure · Environmental management b) "Cross-sectional" activities to support central and provincial government with regard to training programs, sectoral studies and cross-sectional activities for all stakeholders including the BTP (Construction and Public Works sectors). These activities aim to ensure the sustainability of major investments made within the framework of the Project, to continue with the agreed efforts to improve financial relationships and encourage the communes to achieve a level of resources more in line with their level of socio-economic development and demand for basic services from the beneficiary population. v) Agency support through continued government commitment to assign works to ABUTIP within the framework of the EIB or other multilateral or bilateral funding. vi) The communes should be given maximum opportunity to access future Project funding vii) The successful management practices suggested by the donor should be consolidated. All of these recommendations are discussed further in the following sections. 36 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders 37 Annex 9. List of Supporting Documents · Project Appraisal Document (dated December 28, 2000). · Minutes of Negotiations (December 20, 2000). · Development Credit Agreement (February 14, 2001). · Amendment to Development Credit Agreement (October 12, 2004). · Project Paper, Additional Financing (December 6, 2005). · Minutes of Negotiations for Additional Financing (November 21, 2005). · Financing Agreement, Additional Financing (February 13, 2006). · Project Aide-Mémoires. · Project Closure Impact Study. · Cost-Effectiveness Study. · Technical Audits. · Project Status Reports (PSRs). · Implementation Status and Results Reports (ISRs). · Government Implementation Completion Report. · Official correspondence concerning additional financing, restructuring, and extension of project closing date. · Interim Strategy (1999). · Burundi Poverty Note (1999). · Transitional Support Strategy (2002). · Interim Strategy Note (2005) · Poverty Reduction Strategy Paper (2006). 38 Annex 10. Evaluation Grid for Sub-Project Evaluation (Supplemental Annex) Criteria Score Weight Criteria Score Weight 1 Cost of sub-project 2 6 Size of enterprise required 3 < 100 million 6 Small 3 100-200 million 5 Medium 2 200-350 million 4 Large 1 350-450 million 3 450-550 million 2 > 550 million 1 Sub-total 1 (max) 6 12 Sub-total 6 (max) 3 9 2 Labor content in the cost of 4 7 Use of imported construction 3 sub-projects materials 10-20 percent 1 > 30 percent 0 20-30 percent 2 20-30 percent 1 > 30 percent 3 10-20 percent < 10 percent 3 Sub-total 2 (max) 3 12 Sub-total 7 (max) 3 9 3 Duration of employment 3 8 Type of equipment used 1 < 2 months 1 No mechanical equipment 3 2-3 months 2 No significant use of 2 3-4 months 3 mechanical equipment 4-5 months 4 Significant use of 1 > 5 months 5 mechanical equipment Sub-total 3 (max) 5 15 Sub-total 8 (max) 3 3 4 Participation of the population 2 9 Technical and financial in the initiation of sub-project capacity to maintain Participation 2 infrastructure No participation 1 Sub-total 4 (max) 2 4 5 Social and economic viability 9a Technical capacity 2 5a Social aspect 4 Exists 2 Market 3 Does not exist 1 Roads and drainage 3 Sub-total 9a (max) 2 4 Health center 2.5 9b Financial capacity 2 Primary school 2.5 Exists 2 Communal high-school 2.5 Does not exist 1 Water supply 3 Sub-total 9b (max) 2 4 Communal office 1 10 Complementarity with other 1 Electrification 1.5 projects Teachers' homes 1 Yes 2 Sub-total 5a (max) 3 12 No 1 5b Economic aspect - economic 4 rate of return Sub-total 10 (max) 2 2 < 12 percent 0 11 Impact on the environment 1 12 percent 1 Negative impact 0 12-20 percent 2 Slight improvement 1 > 20 percent 3 Significant improvement 2 Sub-total 5b (max) 3 12 Sub-total 11 (max) 2 2 Total 11 criteria: 100 39 Annex 11. Situation of basic infrastructure at Appraisal (Supplemental Annex) TABLE1: NUMBER OF CLASSROOMS/ 1000 INHABITANTS 5 4 3 2 1 0 Provinces TABLE2: HEALTH CENTERS PER 10,000 HABITANTS ACROSS PROVINCES 1.50 1.00 0.50 - RUSI A O I I YA EGA RO A ZA O ND OKE GA ANA URI KA AMV GIT AN NGOZ YIN AMBA MWA RUYIG KIRU CIBIT BUR MU RUT CANKUZ MUR BUBANZ KAY BUJUMBUR MAK TABLE3: RATIO OF MARKETS /20,000 HABITANTS ACROSS PROVINCES 2.50 2.00 1.50 1.00 0.50 - O I RUSI EGA ARO NZA A ND OKE GA TANA YIGI URI KA AMVYA GIT UMBURA MW KUZO YANZ NGOZ AMBA RU BUR BUBA KIRU CIBIT MUYIN RU CAN MUR KA BUJ MAK TABLE 4: METERS OF ROAD PER SQUARE KM2 1400 1200 1000 800 600 400 200 0 A A O A RUSI YA ZI UR ARO ANZA ANZ KE TANA AMB UZO URI KA RAMV GITEGA UMB YINGA MW NGO BUB KAY KIRUND CIBITO NK RUYIGI MU RU BUR CA MU MAK BUJ 40