Mozambique - Urban Rehabilitation Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 16747 IMPLEMENTATION COMPLETION REPORT REPUBLIC OF MOZAMBIQUE URBAN REHABILITATION PROJECT (CREDIT 1949-MOZ) June 20, 1997 Water, Urban and Energy I Eastern and Southern Africa This document has a restricted distribution and may be used by recipients only in the perfornance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENT (Period average) Currency Unit = Metical 1987 Mt 291 1988 Mt 525 (appraisal) 1989 Mt 745 1990 Mt 929 1991 Mt 1,434 1992 Mt 2,517 1993 Mt 3,874 1994 Mt 6,039 1995 Mt 9,024 1996 Mt 11,294 1997 (March) Mt 11,004 FISCAL YEAR Government and Public Enterprises: Calender Year ABBREVIATIONS AND ACRONYMS AdM Agua de Maputo (Water Company of Maputo) AdB = Aguas da Beira (Water Company of Beira) BPD = Banco Popular de Desenvolvimento DNA = Direccao Nacional de Aguas (National Directorate of Water) DNEP = Direccao Nacional de Estradas e Pontes (National Directorate of Roads and Bridges) GOM = Government of Mozambique GPE Gabinete de Promocao de Emprego (Office for the Promotion of Employment) HABITAR Housing Management and Supervision Unit MCA/MOPH = Ministerio da Construcao e Aguas (Ministry of Construction and Water), renamed Ministerio das Obras Publicas e Habitacao (Ministry of Public Works and Housing) MAE = Ministerio da Administracao Estatal (Ministry of State Administration) MT = Ministerio do Trabalho (Ministry of Labor) PIU Project Implementation Unit PROL = Programa de Reforma dos Orgaos Locais (Local Government Reform Program) PRU = Projecto de Reabilitacao Urbana (Urban Rehabilitation Project) ROCS = Roads and Coastal Shipping Project(s) Vice President: Callisto Madavo Country Director: Phyllis Pomerantz Technical Manager: Jeffrey Racki Task Manager: Bernard Becq ICR prepared by: Catherine Seibert  MOZAMBIQUE FOR OFFICIAL USE ONLY URBAN REHABILITATION PROJECT (CREDIT 1949-MOZ) IMPLEMENTATION COMPLETION REPORT CONTENTS PREFACE EVALUATION SUMMARY Introduction ............................................................i Project Objectives ............................................................i Implementation Experience and Results ........................................................... ii Future Operations, Summary of Findings, and Key Lessons Learned ................................................... iii PART I: ASSESSMENT FOR THE BANK'S PERSPECTIVE A. Introduction .........................................................I B. Project Objectives and Description .........................................................1 C. Achievement of Project Objectives .........................................................3 D. Implementation Record and Major Factors Affecting the Project .......................................................6 E. Project Sustainability ........................................................8 F. Bank Performance .........................................................9 G. Borrower Performance ........................................................ 10 H. Assessment of Outcome ........................................................ 10 I. Future Operations .........................................................11 J. Key Lessons Learned .........................................................11 PART II: STATISTICAL TABLES Table 1: Summary of Assessment Table 2: Related Bank Loans/Credits Table 3: Project Timetable Table 4: Loan/Credit Disbursements: Cumulative Estimated and Actual Table 5: Key Indicators for Project Implementation Table 6: Key Indicators for Project Operation Table 7: Studies Included in Project Table 8A: Project Costs Table 8B: Project Financing Table 9: Economic Costs and Benefits Table 10: Status of Legal Covenants Table 11: Compliance with Operational Manual Statements Table 12: Bank Resources: Staff Inputs Table 13: Bank Resources: Missions Appendix 1: March 1996 Supervision/Completion Mission Aide-Memoire This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.   IMPLEMENTATION COMPLETION REPORT REPUBLIC OF MOZAMBIQUE URBAN REHABILITATION PROJECT (CREDIT 1949-MZ) PREFACE This is the implementation completion report (ICR) for the Urban Rehabilitation Project (PRU) in Mozambique, for which Credit 1949-MOZ in the amount of SDR 44 million was approved in August 1988 and became effective on February 14, 1989. The credit was closed October 31, 1996, ten months after the original closing date of December 31, 1995. The last disbursement took place on January 9, 1997, at which time a balance of SDR 5.3 million (about US$7.4 million) was canceled. In addition to the Government counterpart contribution, cofinancing for the project was provided by the governments of the Netherlands, Denmark, Spain, and Finland. The ICR was prepared by Catherine Seibert (financial analyst) of the Water, Urban and Energy 1 Unit of the Africa Region (AFTU 1). It was reviewed by Phyllis Pomerantz, country director, Jeffrey Racki, technical manager, AFTU1, Jacomina de Regt, principal operations officer, Operations Support 1, and former task managers including Bernard Becq, senior operations officer, Richard Beardmore, senior operations officer, Zambia, and Roberto Chavez, resident representative, Mozambique. Preparation of the ICR was begun during the supervision/completion mission in February 1996 and continued during subsequent missions. It is based on material in the project file and mission findings. Because of several months' delays in the Borrower's final project cost report, this ICR has been finalized based on estimates of project costs, using disbursement reports. Comments were requested from the Borrower and from cofinanciers, but by the time the ICR was finalized, only informal comments had been received from the Borrower.   IMPLEMENTATION COMPLETION REPORT REPUBLIC OF MOZAMBIQUE URBAN REHABILITATION PROJECT (CREDIT 1949-MZ) EVALUATION SUMMARY Introduction 1. The Urban Rehabilitation Project (PRU, Cr. 1949) was the first Bank project in the urban sector, and it was designed in a difficult environment. A bloody civil war, that had begun about 15 years before the project, was still raging, resulting in destruction to infrastructure, large scale movements of the population, and loss of employment opportunities. The urban sector faced major problems, stemming from inadequate financial resources, weak institutions, deteriorated infrastructure and housing stock, and a burgeoning population with a substantial refugee population in squatter camps. PRU was designed in the context of a Bank strategy to promote economic reform supported by a series of adjustment operations and investment projects that contained the elements of emergency projects. IDA lending to Mozambique began in 1985 with the Rehabilitation Program (Cr. 1610); the Second Rehabilitation Program (Cr. 1841) followed in 1987. PRU was one of the emergency-type projects, along with the Education and Manpower Development (Cr. 1907) and Health and Nutrition (Cr. 1989) projects. Project Objectives 2. The primary objective of the project was to assist the Government of Mozambique (GOM) to stem the deterioration in basic urban infrastructure and services, and to mitigate the social costs of structural adjustment through the implementation of a program of urban rehabilitation and employment generation. The project comprised five components: (1) rehabilitation of infrastructure in the major cities of Maputo and Beira; (2) housing rehabilitation in Maputo and Beira; (3) upgrading of solid waste management in Maputo and Beira; (4) employment generation, through two lines of credit; and (5) strengthening of institutions andproject administration, through technical assistance and training for implementing agencies. The objectives and design tried to establish a balance between physical components and capacity building, with some enabling policy reforms. The design of some components, notably the housing component, showed a creativity not normally found in Bank projects. Notable covenants in the Development Credit Agreement related to the financial sustainability of the Maputo and Beira water companies, and to the development of a housing policy. The strategy underlying the project objectives, of pursuing an integrated approach to the rehabilitation of the key subsectors of infrastructure in the two largest, and fastest  - ii - growing cities in Mozambique, was appropriate. In general, however, the scope of the project proved to be too ambitious, and the project design too complex, for the limited human and financial resources of the country. Implementation Experience and Results 3. The project was identified in October 1987 and went to the Board in August 1988, only 10 months after identification. The project became effective on February 14, 1989, and closed on October 31, 1996, 10 months after the original closing date of December 31, 1995. Two extensions of the closing date, one of four months to April 30, 1996 and a second of six months to October 31, 1996, were granted to allow completion of activities. Of the credit amount of SDR44 million, SDR 5.4 million or 12 percent was canceled. Total project costs are estimated at US$65.7 million, compared to US$83 million at appraisal. By category, the most significant deviation from appraisal estimates concerned the provision for consulting services, training and surveys; estimated at appraisal at about SDR 4.4 million, over SDR 11 million was eventually disbursed. 4. Overall, achievement of project objectives has been very mixed, with significant variations between components. The best performing subcomponent under the project was the roads subcomponent of the rehabilitation of infrastructure component, which fully met the objectives set under the project. Achievement of the objectives related to the other components, however, was partial at best. Implementation of all components, in particular the housing component, was adversely affected by limited availability of counterpart funds. The project also overestimated implementation capacity within the country, on the part of Government and the local private sector that was to be involved in project activities. The sustainability of the various components of the project is uncertain. Although much of the physical investment took place, sustainability of the investments rests on the willingness and ability of Government to introduce appropriate policy measures, in particular cost recovery. Inattention to the latter has resulted in lack of sustainability for several of the PRU components, in particular for the solid waste and water supply activities. The development of local governments, in particular, the devolution of fiscal responsibilities to the local level, was required to sustain the achievements of PRU; although the design of the project recognized the importance of viable local governments and tried to support their development through strengthening of local units and creation of a center for urban management, implementation of these aspects of the project was not satisfactory. The impact of the project on the labor market also needs to be highlighted as a sustainability issue. The paying of significant salary supplements to civil servants began a practice of salary top-ups by some donors that continues today. This, in combination with the paying of salaries to the Mozambican consultants in the Project Implementation Unit (PIU) that were well out of line with the local salary structure, contributed to significant distortions in the local labor market. Both practices have been stopped in all IDA-financed projects. 5. Bank performance during project design was satisfactory. The project was designed in record time for a non-emergency project in a high-risk environment. The  - iii - integrated subsector approach, emphasizing local structures, was conceptually appropriate. Ultimately, the design of the project proved too demanding for the limited human and financial capacities in Mozambique; however, it is not clear whether the extent of the limited resources would have been known without extensive sector work, which would have resulted in major delays to project investments. Overall, implementation of the project was fair, although from both the view of the Bank and the Government, the earlier years of the project were marked by more assiduous implementation. Time spent on supervision dropped after the first three years, due to a number of factors (change in task managers, limited funding for missions, overshadowing by preparation of subsequent projects). 6. While the Bank drove the design stage of the project, GOM was involved and in agreement with the objectives. However, the project clearly tested the capacity of the involved institutions. The Direccao Nacional de Estradas e Pontes (National Directorate of Roads and Bridges, DNEP), which oversaw the roads component, proved itself a adept implementor; however, other implementing agencies had limited success. Across the board, counterpart staff struggled to follow Bank procedures related to disbursements and procurement, at the same time they followed complex GOM procedures. The performance of the Project Implementation Unit (PIU) was mixed. In the early years of the project, the PIU appeared to operate effectively; however, like the Bank team, the borrower's implementation team suffered from changes in staff. Communication problems between the PIU and other implementing agencies on several occasions resulted in less than satisfactory implementation. As the project wound down, and staffing of PIU reduced, PIU became much less effective. In addition, as the bulk of PIU work was done by consultants, the impact of the project on the development of skills within the parent ministry. 7. Evaluation of the project was hampered by the lack of information; in fact, the evaluation process highlighted the weakness of the information systems for the project, which increased in the last half of the project. In particular, the PIU was unable to produce final project costs for each component and subcomponent. In view of the mixed results in implementation and the questionable sustainability of project achievements, the project outcome has been rated overall as unsatisfactory. Future Operations, Summary of Findings, and Key Lessons Learned 8. As noted above, the strategy underlying the design of PRU was well thought out, and appropriate. However, it became clear during implementation that the scope of the project was too optimistic, particularly in terms of borrower capacity. Learning from the experience of PRU, the design of the operations that followed PRU, and developed the initiatives begun under PRU, were more tightly focused in design. The first and second Roads and Coastal Shipping projects (Credits 2374 and 2599) have emphasized policy reform and institutional development along with a large investment program in the roads sector. The Local Government Reform and Engineering Project (Cr. 2530) has expanded and intensified PRU initiatives in local government and urban planning to three other  - iv - cities in Mozambique. A National Water Development Project is also under preparation; learning from the lessons of PRU, the major objective of this project, which is the first in the water sector, is to improve sustainability of the water supply function through private sector management and market-oriented tariff structures. 9. The key findings and lessons emerging from this operation, in relation to projects in Mozambique and in the infrastructure sector are: In a complex and uncertain environment, marked by weak capacity in the country, project design has to be well focused on the key issues and constraints. Design should be clear and preferably simple, and above all, focus on building the capacity of the involved agencies. While all the objectives of this project were laudable, Government resources - human and financial - were clearly overwhelmed by the carrying out the complex project activities (along with numerous other donor activities). Implementation capacity- financial but especially human capacity - must drive design, if the achievements of the project are to be sustainable after the project has ended. The design team needs to carefully debate and come to agreement on this issue, and the resulting decisions made known, in order to manage the expectations of all involved, including clients and beneficiaries. Implementors must be diligent in enforcing appropriate policy measures. Without such, the sustainability ofphysical improvements is undermined. While the design of PRU was fairly forward looking in terms of the emphasis on policy reform, especially cost recovery, implementation should have been more aggressive in ensuring that more of the agreed upon reforms were made. In some cases - for example, the measures related to development of local government structures, the decision to defer implementation to a follow on operation specifically focused on local governments, was appropriate in view of limited government capacity. On the other hand, the Bank should have pressed more strongly for cost recovery mechanisms: The sustainability of improvements in water supply, sanitation, housing and solid waste have all been put into question by uneconomic tariffs and charges. The experience of the project underscores Bank experience to date with project implementation units: while PIUs may successfully implement the specified project activities, they rarely make a lasting improvement in the implementation capacity of the involved government. The PRU PIU had little impact on capacity building within the Ministry. Because consultants to the unit did most, if not all, the work related to project supervision, procurement and disbursements, the PIU had little impact on developing capacity within MCA/MOPH to manage projects. Following from the above lesson concerning project design and capacity, if the project design requires the specialized services of an implementation unit, the design team would do well to revisit the complexity of the operation.  v - More emphasis needs to be put on creating strong information systems and insisting on regular and timely progress reporting, throughout the life of the project. In the case of a ministry or similar government implementing agency, such systems should be applicable to general operations as well as to the project. One of the most useful outputs that a project may leave behind is a reliable information system, and more generally, an appreciation on the part of the implementing agency of the importance of timely and accurate information. Clearly, without such a system, evaluation of outcomes will be difficult, if not impossible. While an attempt was made to put such systems into place for PRU, the systems and procedures were not maintained and getting accurate information on project outcomes was difficult.   IMPLEMENTATION COMPLETION REPORT REPUBLIC OF MOZAMBIQUE URBAN REHABILITATION PROJECT (CREDIT 1949-MZ) PART I: ASSESSMENT FROM THE BANK'S PERSPECTIVE A. INTRODUCTION 1. The Urban Rehabilitation Project (Projecto de Reabilitacao Urbana or PRU, Cr. 1949) was the first Bank project in the urban sector, and it was designed in a crisis environment. A bloody civil war that had begun about 15 years before the project was still raging, resulting in destruction to infrastructure, large scale movements of the population, and loss of employment opportunities. The availability of trained people, capable of planning and operating urban services, was virtually nil, as the Portuguese who had run the systems fled the country at Independence, having trained few Mozambicans. Population growth was (and remains) a serious problem: at the time of project design, about 10 percent of the population lived in the two main cities of Maputo and Beira, with population growth rate in the urban areas about seven percent per annum. The urban sector faced major problems, stemming from inadequate financial resources, weak institutions, deteriorated infrastructure and housing stock, and a burgeoning population with a substantial refugee population in squatter camps. 2. PRU was designed and put in place in the context of a Bank strategy to promote economic reform supported by a series of adjustment operations and investment projects that contained the elements of emergency projects. IDA lending began in 1985 with the Rehabilitation Program (Cr. 1610), which was followed in 1987 by the Second Rehabilitation Program (Cr. 1841). The adjustment process was supported by a series of investment projects that, although not classified technically as emergency projects, contained the elements of such projects, including a simplified and shortened implementation schedule under a defined action plan aimed at mitigating the social effects of adjustment and the effects of the civil war. The Urban Rehabilitation Project (Cr. 1949) was one of these emergency-type projects, along with the Education and Manpower Development (Cr. 1907) and Health and Nutrition (Cr. 1989) projects. Because of the need for quick action, no sector work was undertaken as a basis for PRU. B. PROJECT OBJECTIVES AND DESCRIPTION 3. The primary objective of the project was to assist the Government of Mozambique (GOM) to stem the deterioration in basic urban infrastructure and services, and to  -2 - mitigate the social costs of structural adjustment through the implementation of a program of urban rehabilitation and employment generation. The objective of the project was clearly spelled out in the project documentation, and there appeared to be good understanding and commitment to this objective on both the Government's and the Bank's sides. The project comprised five components: (1) rehabilitation of infrastructure in the major cities of Maputo and Beira including roads, water/sewerage and storm drainage, and coastal protection (about 45 percent of base project costs); (2) housing rehabilitation in Maputo and Beira (33 percent of base project costs); (3) upgrading of solid waste management in Maputo and Beira (six percent); (4) employment generation, through two lines of credit, one for loans to small businesses with employment generating capacity and the other to consumers purchasing building materials (seven percent); and (5) strengthening of institutions andproject administration, through technical assistance and training for implementing agencies (nine percent). Base project costs were increased by 20 percent to allow for physical and price contingencies. 4. The SAR lists seven institutional beneficiaries of the project, plus small enterprises. The beneficiary and implementor of the infrastructure rehabilitation component was the Ministerio da Construcao e Agua (MCA/MOPH)'. Within MCA/MOPH, the Direccao Nacional de Aguas (DNA) implemented the water and sewerage component, and the Direccao Nacional de Estradas e Pontes (DNEP) implemented the roads and coastal protection component. MCA/MOPH, through HABITAR, also implemented the housing component. The upgrading of solid waste removal capacity was implemented jointly by MCA/MOPH and the City Councils of Maputo and Beira. The lines of credit were implemented by the Ministerio do Trabalho, through the Gabinete de Promocao de Emprego (GPE). The above beneficiaries and implementors, plus the Ministerio da Administracao Estatal (MAE), which has responsibility for development of local governments, were the beneficiaries of the technical assistance components. To assist in implementation, two project implementation units were set up within MCA/MOPH. The Project Implementation Unit (PIU) monitored the implementation of all components except housing, and comprised about seven staff from MCA/MOPH, supported by a minimum of three non-Government staff. The second implementation unit, HABITAR, was established to monitor and implement the housing component; it was comprised of about three staff within MCA/MOPH and field implementation units in Maputo and Beira. 5. The project was supported by a set of legal documents, including the Development Credit Agreement, Maputo Implementation Agreement, Beira Implementation Agreement, Maputo Water Agreement, Beira Water Agreement, Financing Agreement between the Banco Popular de Desenvolvimento (BPD) and MCA/MOPH for the self-help construction line of credit, and Financing Agreement between BPD and GPE for the small business line of credit. Notable covenants in the 'Later during project implementation, MCA was renamed the Ministry of Public Works and Housing (MOPH). For the purposes of this ICR, the Ministry will be referred to as MCA/MOPH.  -3 - Development Credit Agreement related to: (i) the financial condition of the water companies, including water tariff studies for the water companies in Maputo and Beira and generation of internal funding equal to at least 25 percent of the annual average of capital expenditures; and (ii) development of a housing policy. 6. The strategy underlying the project objectives, of pursuing an integrated approach to the rehabilitation of the key subsectors of infrastructure in the two largest, and fastest growing cities in Mozambique, was appropriate. The objectives and design tried to establish a balance between physical components and capacity building, with some policy reforms intended to support both. The design of some components, notably the housing component, showed a creativity and propensity for risk taking not normally found in Bank projects. In general, however, the scope of the project proved far too ambitious, and the project design too complex, for the limited human and financial resources of the country. In retrospect, a tighter focus to the design, perhaps on a subset of activities (either sectors or cities), might have improved the chances for timely implementation and sustainability. 7. A significant problem with the design, but perhaps more so in the implementation, was that the project was never adequately involved the city councils, making them feel responsible for the success and sustainability of the works. The design of the project recognized the importance of the development of viable local governments; it included mechanisms to develop such local structures on a pilot basis (e.g., through strengthening a specific unit in solid waste collection or in road construction), and intended to support the establishment and development of a Center for Urban Management as a tool for the development of local government expertise. However, implementation of these initiatives was inadequate, with the result that PRU was not able to influence a sustainable change in the functioning of local governments. The works undertaken to improve drainage in Palmeira (Beira) illustrate the dangers of this approach: at the same time that works were being undertaken to rehabilitate drainage, the City Council engaged in intensive mining of sand close to the works, which threatened the completion and sustainability of the rehabilitation. Despite IDA warnings and discussions, this mining was still being done one year later. C. ACHIEVEMENT OF PROJECT OBJECTIVES 8. Overall, achievement of project objectives has been very mixed, with significant variations between components. While a good proportion of the physical works was completed, achievement of the enabling policies required for the sustainability of the physical improvements, in particular, that of cost recovery, has been uniformly disappointing. Achievement of objectives in relation to capacity-building initiatives also has been disappointing. In general, components that were located in Beira were not carried out as well as similar components in Maputo (e.g., SME line of credit, coastal protection); this probably underscores the weak capacity at the level of the municipality  -4- and the need for the support from the national level, and the ineffectiveness of programs aimed at developing capacity. 9. The best performing subcomponent under the project was the roads subcomponent of the rehabilitation of infrastructure component, which fully met the objectives set under the project. All the physical activities were completed, and in a timely fashion and close to budget. Because of donor support that materialized after appraisal, the scope of this subcomponent was increased to include the Machava Road in Maputo, which has proved to be a very heavily traveled road and an important addition to the Maputo road infrastructure. As a result of the work begun under this project and developed under the subsequent Roads and Coastal Shipping projects (Credits 1790 and 1804), DNEP has developed into a capable planning and implementing institution. 10. Achievement of the objectives related to water supply activities has been problematic. In Maputo, only one of the two contracts (the Maputo water works rehabilitation and extension) envisioned at appraisal was let; the Mahotas extension was to be picked up by another donor, but had not been done at the time of the ICR. Implementation of the Maputo water network and subsequent achievement of objectives was adversely impacted by several factors. The lack of capacity and/or inexperience of the local contractor undertaking the works caused major delays, resulting in increased costs related to supervision by the consulting engineers, and exposed weaknesses in DNA's implementation capacity. In Beira, the water distribution network and sewage system were rehabilitated. The physical works in water were to have been supported by tariff improvements for Agua de Maputo (AdM) and Aguas da Beira (AdB) that were needed to sustain the rehabilitation. However, the French-financed tariff study for AdM was completed only in April 1995, several years behind schedule, and the tariff study for AdB was never done. In view of the delays, and in recognition of the need to look beyond the two major cities, the Bank, with participation from Dutch and Danish aid, agreed to finance a tariff study to look at all 13 provincial capitals; this study was expanded to cover technical, legal and environmental aspects. The draft report was completed in September 1993 and the final in early 1996; the results of this work are being incorporated in the design of the National Water Development Project. 11. Achievement of the objectives related to coastal protection measures were mixed, with activities in Beira significantly delayed. The technical assistance contract for coastal and road maintenance was canceled during implementation due to poor coordination between MCA/MOPH and the two city councils. It was decided to put the onus for implementation squarely on the city councils. 12. Supported by other donor assistance, the solid waste activities got off to a fairly strong start, and achieved noticeable differences in Maputo in particular in the early part of the project. However, the lack of cost recovery and subsequent problems with counterpart funding eventually resulted in the collapse of project activities.  -5 - 13. Physical objectives related to housing rehabilitation were only partially met. The success of this component rested on the recycling of the revenues from the completion of three buildings and some apartment units and subsequent rental to expatriates, to other subcomponents - rehabilitation of apartment units, building of core houses and development of serviced sites - that would benefit Mozambican households. Because of difficulties in mobilizing counterpart funding, Government turned the three buildings over to three state-owned entities, who in turn got a 100 percent stake in the future stream of income and rented the apartments out to relatively well-heeled employees. The collapse of revenues from this subcomponent effectively limited funding for the other components benefiting the Mozambican households. Accordingly, only about 340 of the 774 low-cost apartment units in Maputo and about 80 of the 220 units in Beira were rehabilitated. The success of the rehabilitation of low-income apartments rested on achieving triangulation of residents from the dilapidated low-income apartments to core houses during the rehabilitation, and then out of the core houses or serviced sites and back to either the rehabilitated apartments or other housing, which ever was affordable. This component stalled primarily because virtually none of the apartment dwellers could afford the new rents on the rehabilitated houses. All corehouses were built, and despite uneconomic rents, maintenance and upkeep appears to be good as a result of renter efforts. Resettlement under the project was handled well, in accordance with socio- economic surveys and guidelines. One positive indicator of the satisfaction of resettled families may be in the high levels of maintenance and investment they are making to the core houses. 14. The two lines of credit had little positive impact. The line of credit for building materials was never disbursed, and only about half of the credit for loans to small businesses was disbursed. Follow up on the loans was poor, due at least in part to confusion between the implementing agency and the commercial bank as to which was responsible. Documentation on the results of the SME line of credit is poor; it appears that little if any reliable information exists on the status of the businesses, ex poste employment or repayment of loans. It has been estimated that recovery of the line of credit is lower than 30 percent. 15. In terms of institutional development andproject management, the achievement of project objectives again was mixed. The institution building of DNEP begun under PRU and expanded under the ROCS projects has clearly left a much strengthened institution that is capable of planning and implementing sectoral programs. DNA, the water directorate, continues to be a weak institution, and has struggled to manage the relatively small Maputo water works contract; strengthening DNA will be a focus of the upcoming National Water Development Project. HABITAR, the housing agency, was intended to emerge under this project as a center of strategic thought and implementation for the housing sector; however, HABITAR was disbanded in 1994 and housing activities were put under the national directorate for civil construction until April 1996 when two "subordinate institutions" were created in the Ministry, a fund for promotion of housing (Fundo de Fomento de Habitacao) and a national institute for housing and urbanism (Instituto Nacional de Habitacao e Urbanismo). The action plan that MAE was to  -6 - develop related to the development and strengthening of the municipalities was very delayed, and relatively unsatisfactory, and efforts to develop the Center for Urban Management as a tool for the development of local government expertise never took off. However, the follow on project to PRU, the Local Government Reform and Engineering Credit (Cr. 2530, PROL) supports measures to address these objectives. For the other institutional beneficiaries, the project had little positive impact on the institutional strength. 16. While an economic rate of return was estimated at appraisal, the available information in relation to costs and benefits is inadequate to re-estimate the ERR. D. IMPLEMENTATION RECORD AND MAJOR FACTORS AFFECTING THE PROJECT 17. Summary of Costs. Table 8 compares the expected project costs at appraisal to estimated project costs. At the time of finalizing this ICR, seven months after the closing date, the PIU had not yet presented to the Bank the finalized project costs, on a component by component basis. The costs in Table 8 have been estimated by the Bank, using IDA disbursements and contract information, available donor information, and audits of GOM contributions to the project. Total project costs are estimated at US$66 million2, compared to US$83 million at appraisal. Of the IDA credit amount of SDR 44 million, SDR 5.3 million (equivalent to about US$7.4 million) or 12 percent was canceled. Actual expenditures by category varied significantly from appraisal estimates for several categories. The most significant deviation from appraisal estimates concerned the provision for consulting services, training and surveys; estimated at appraisal at about SDR 4.4 million, over SDR 11 million (equivalent to about US$15.8 million) was eventually disbursed. Expenditures related to many consulting contracts came in higher than the original contract price because of changes to the contracts, in particular extensions; this underscores the very thin human resources in Mozambique, which was underestimated at the time of appraisal. GOM contributed the metical equivalent of US$10.9 million to the project; it is estimated that most, if not virtually all, of GOM's contribution went for customs and duties. 18. Implementation Schedule. The project was identified in October 1987 and went to the Board in August 1988, only 10 months after identification. The project became effective on February 14, 1989, and closed on October 31, 1996, ten months after the original closing date of December 31, 1995. Two extensions of the closing date (one for four months, a second for six months) were granted to allow completion of activities; a GOM request for a longer extension to cover preparatory activities related to the National Water Development project, was not granted. 2Not including cofmancing from Finland and Spain  19. Implementation Experience. Issues regarding implementation, by component, have been brought out in the previous section. However, there were several implementation issues that affected virtually all the components of the project. All components were adversely affected by the worsening budget situation which resulted in limited availability of counterpart funds. While in some cases it was possible to move ahead by drawing in other donor resources, the lack of counterpart funding has seriously undermined the sustainability of many of the works undertaken by the project. Efforts, both on the part of the Bank and the implementing agencies, to introduce the cost recovery mechanisms required for sustainability, were not successful; water tariffs, for example, dropped precipitously in real terms over the project period. The project also overestimated implementation capacity within the country, on the part of Government and the local private sector that was to be involved in project activities. Government implementation required reinforcement by consultants throughout the project; when consultant resources were cut, near the end of the project, there was an immediate deterioration in implementation. 20. Government procurement procedures, involving the Ministry of Finance and Bank of Mozambique, were complicated, and resulted in major delays in start up for many contracts. The Maputo water works contract illustrates procurement problems (and also problems in working with perhaps less than adequate consulting services and the weak local contracting industry). Expected at the time of appraisal to be completed by 1991, the contract was not advertised until January 1991, and the order to the contractor to begin work did not come until October 1992. Once under way, the work continued to experience delays, and completion date slid from June 30, 1994 to June 30, 1996, posing problems for the closing of the IDA credit and contributing to the need to extend the closing date. Virtually all contracts involving local contractors ran into problems, due to financial and management inadequacies; in one case, a bank guarantee had to be effected because of non-performance. However, because of resistance from the guaranteeing bank and lack of follow-up from PIU, four years elapsed between the first request to make good the guarantee, and the eventual paying of the guarantee. 21. An interesting issue that arose during implementation related to the impact of low Government salaries on project staff. Stemming from the scarcity of skilled human resources in Mozambique, particularly within GOM, and in order to retain and stimulate government staff working on the project, the Bank agreed to pay salary supplements in US dollars to 42 staff in the various implementation agencies. Supplements ranged from US$500 per month to National Director level staff (a little over three times the Government salary), to US$200 per month for stenographic secretaries (about four times Government salary), to US$100 for administrative assistants. As project implementation wound down, the number of Government employees receiving supplements also went down; as of late 1995, 13 were receiving supplements. PRU was one of the first projects in Mozambique to top up salaries, a practice which has caused much concern over the subsequent years. In addition, because of lack of information about the Mozambican labor market at the time, the Bank agreed to pay salaries to the Mozambican consultants who staffed the PIU that were several times their previous levels of pay within  - 8 - Government, and the pay of the counterparts with whom they worked; in one case, the increase was more than 40 times the previous pay. These practices have been stopped in recent IDA-financed projects. 22. Compliance with key credit covenants was less than satisfactory. Key covenants related to tariff studies and internal fund generation for Agua de Maputo and Aguas da Beira were never met; in fact, the real value of tariffs dropped 50 percent over the life of the project, which has seriously undermined the water sector component. It should be brought out that when these covenants were put in place, neither AdM nor AdB were preparing accounts that were audited by external auditors; the accounts in particular of AdB remain unauditable to date. In the light of such poor and unreliable information, it can be questioned how useful a covenant on internal fund generation was to start with. Implementation of an action plan to strengthen local government institutions and finances, which was to have been at least begun by June 1989, was not implemented under the PRU, although the follow-on PROL project, which was approved in June 1993 and became effective in July 1994, has squarely addressed this issue. E. PROJECT SUSTAINABILITY 23. The sustainability of the various components of the project is mixed, leading to a rating of "uncertain" for overall project sustainability. Whether the infrastructural improvements will be sustainable rests on the willingness and ability of Government to introduce the appropriate policy measures, in particular cost recovery. Inattention to the latter under the project resulted in lack of sustainability for several of the PRU components, in particular for the solid waste and water supply activities. The development of local governments, in particular, the greater involvement of local governments in providing services was an important element in achieving sustainable results from PRU; however, it was recognized early during implementation that PRU was an inadequate vehicle to address this objective, and agreement was reached with Government to begin design of a separate project, specifically targeted on the development of local governments. The Local Government Reform and Engineering Project (Cr. 2530) became effective in July 1994, and implementation proceeded in parallel with PRU. 24. The impact of the project on the labor market through the experience of the PIU and the paying of salary supplements needs to be highlighted as a sustainability issue. Such "topping up" is typically done in response to low civil service salaries and only for those people working on special projects, thereby introducing a significant element of inequity into Government pay structures without addressing the real underlying problem of civil service salaries. The paying of salaries to Mozambican consultants that were well  - 9 - out of line with the local salary structure (although well under what expatriate consultants might have been paid), contributed to significant distortions in the local labor market.3 F. BANK PERFORMANCE 25. Bank performance during project design and the early stages of implementation was good. The project was designed in a very difficult and high-risk environment, but was designed in record time for a non-emergency project. The integrated subsector approach, emphasizing local structures, was conceptually appropriate. Ultimately, the design of the project proved too demanding for the limited human and financial capacities in Mozambique; however, it is not clear whether the extent of the limited resources would have been known without extensive sector work, which would have resulted in major delays to project investments. 26. Overall, implementation of the project was fair, although from both the view of the Bank and the Government, the earlier years of the project were marked by more assiduous implementation. The project was supervised intensively during the first three years of implementation; Bank missions visited Mozambique four times in each of the first three years, averaging a total of about 80 person days in the field. Mission composition was diverse, and usually included both technical and finance people, and aide-memoires were frequently left in both English and Portuguese. Early Bank missions took a flexible approach to the project, showing a willingness to make changes to the design in response to changing circumstances, such as the availability of donor funding for a component (Mahotas water) that subsequently freed up IDA funds for a new activity (Machava Road). Time spent on supervision dropped after the first three years, due probably to a number of factors. Task managers changed fairly rapidly; due to unforeseeable events (e.g., transfers to another division and to field assignment) in the space of about 14 months, responsibility for the project passed among four task managers before being resolved. Limited funding for supervision missions also limited the size and skill mix of supervision teams. Finally, and as is frequently the case, the implementation of PRU was overshadowed as the attention of both the Bank and the Government shifted to the development of the next project, the Local Government and Engineering Project (PROL). Bank missions, particularly in early implementation, were effective in working and coordinating with donors involved in the sector. 27. Although technical support provided by Bank teams was good, supervision missions were less insistent on the policy reforms aimed at under the project. While some work was done in regard to the strengthening of local government structures, it was recognized during implementation that this objective could be better addressed under the follow project, PROL. The Bank did not press for timely completion of the tariff studies, and the financial condition of the water companies continued to deteriorate, largely as a 3 The topic of donor (especially Bank) involvement in the labor market in Mozambique is treated in depth in a 1995 discussion paper by Ame Disch, Technical Assistance and the Labor Market in Mozambique.  - 10- result of inadequate tariffs. In response to this situation, a National Water Development Project is under preparation, which emphasizes cost recovery for the sector. The Bank reacted more strongly in regard to on this issue in relation to solid waste, but to no avail; this issue will be addressed in the context of the Urban Environment Project, now under design. G. BORROWER PERFORMANCE 28. While the Bank drove the design stage of the project, GOM was involved and in agreement with the objectives. Implementation of this complex project clearly tested the capacity of the involved institutions. The borrower's performance varied widely by component. DNEP, which was in charge of the roads component, proved itself a very adept implementor of projects. In contrast, DNA, GPE, MAE and the other implementing agencies clearly struggled, with limited degrees of success, in implementing their components. Across the board, counterpart staff struggled to understand and follow Bank procedures related to disbursements and procurement, at the same time they followed the complex GOM procedures. There were major weaknesses in moving procurement and disbursements through the Bank of Mozambique, which adversely impacted implementation; these weaknesses, although improved in some respect, continue to cause problems today. 29. The performance of the PIU was mixed. In the early years of the project, the PIU appeared to be well organized and submitted detailed quarterly reports on time. However, like the Bank, the borrower's implementation team suffered from rapid changing of staff; the project coordinator changed four times during the project. Several supervision missions highlighted communication problems between the PIU and the other implementing agencies, particularly on financial matters. As noted previously, PIU was unable to finalize project costs at the time of the ICR. The bulk of the implementation work -- and corresponding development of procurement, contract monitoring, disbursement, etc. skills - was done by consultants to the unit. Accordingly, the impact of the project on the development of skills within MOPH has been minimal at best. The experience of PRU in this respect underscores the experience elsewhere in the world with project implementation units staffed by non-government officials. H. ASSESSMENT OF OUTCOME 30. PRU was the first Bank operation in the urban sector in Mozambique, and it was designed and largely implemented during a tumultuous period. The combination of these factors -- possibly the poorest country in the world, civil war, impoverished human resources, and poor information base -- combined to make PRU a very risky project, which the complexity of the design heightened. However, while in general the formal successes of this project are relatively few, PRU did achieve important physical and  institutional success in the roads sector, and some very limited, but notable, success in terms of the provision of core houses and serviced sites. In addition and more broadly, the project set the stage for several subsequent operations in the roads and urban sector, which have learned from the lessons of PRU and taken different approaches to achieving development in those sectors. These operations are discussed in the next section. I. FUTURE OPERATIONS 31. The strategy underlying the design of PRU was well thought out, and appropriate the needs of the sector. However, it became clear during implementation that the scope of the project was too optimistic, particularly in terms of borrower capacity. Learning from the experience of PRU, the design of the operations that followed PRU, and developed the initiatives begun under PRU, were more tightly focused in design. The first and second Roads and Coastal Shipping projects (Credits 2374 and 2599) have emphasized policy reform (sectoral planning and cost recovery) and institutional development at the same time a large investment program was launched. The Local Government Reform and Engineering Project (Cr. 2530) has picked up the dialogue on local government and urban development that was begun under PRU, and has expanded the PRU initiatives in urban planning to three other cities in Mozambique. Preparation is beginning on an Urban Environment Project that will support the development of urban infrastructure working within the context of the local governments. A National Water Development Project is also under preparation; learning from the lessons of PRU, the major objective of this project, which is the first in the water sector, is to improve sustainability of the water supply function through private sector management and market-oriented tariff structures. J. KEY LESSONS LEARNED 32. The key lessons emerging from this operation are: In a complex and uncertain environment, marked by weak capacity in the country, project design has to be wellfocused on the key issues and constraints. Design should be clear and preferably simple, and above all, focus on building the capacity of the involved agencies. While all the objectives of this project were laudable, Government resources - human and financial - were clearly overwhelmed by the carrying out the complex project activities (along with numerous other donor activities). Implementation capacity- financial but especially human capacity - must drive design, if the achievements of the project are to be sustainable after the project has ended. This may involve a much more limited scope of project, or it may involve a different approach to design, stressing the least expensive, basic levels of service. The design team needs to carefully debate and come to agreement on this issue, and the resulting decisions made  - 12 - known, in order to manage the expectations of all involved, including clients and beneficiaries. Implementors must be diligent in enforcing appropriate policy measures. Without such, the sustainability ofphysical improvements is undermined While the design of PRU was fairly forward looking in terms of the emphasis on policy reform, especially cost recovery, implementation should have been more aggressive in ensuring that more of the agreed upon reforms were made. In some cases - for example, the measures related to development of local government structures, the decision to defer implementation to a follow on operation specifically focused on local governments, was appropriate in view of limited government capacity. On the other hand, the Bank should have pressed more strongly for cost recovery mechanisms: The sustainability of improvements in water supply, sanitation, housing and solid waste have all been put into question by uneconomic tariffs and charges. The experience of the project underscores Bank experience to date with project implementation units: while PIUs may successfully implement the specified project activities, they rarely make a lasting improvement in the implementation capacity of the involved government. The PRU PIU appears to have had little impact on capacity building within the Ministry. Because consultants to the unit did most, if not all, the work related to project supervision, procurement and disbursements, the PIU had little impact on developing capacity within MCA/MOPH to manage projects. Following from the above lesson concerning project design and capacity, if the project design requires the specialized services of an implementation unit, the design team would do well to revisit the complexity of the operation. More emphasis needs to be put on creating strong information systems and insisting on regular and timely progress reporting, throughout the life of the project. In the case of a ministry or similar government implementing agency, such systems should be applicable to general operations as well as to the project. One of the most useful outputs that a project may leave behind is a reliable information system, and more generally, an appreciation on the part of the implementing agency of the importance of timely and accurate information. Clearly, without such a system, evaluation of outcomes will be difficult, if not impossible. While an attempt was made to put such systems into place for PRU, the systems and procedures were not maintained and getting accurate information on project outcomes was difficult.  PART II: STATISTICAL TABLES  Table 1: Summary of Assessment A. Achievement of Objectives Substantial Partial Negligible Not Applicable Macroeconomic Policies 0 0 O N Sector Policies 0 0 0 Financial Objectives 0 0 0 Institutional Development 0 0 0 Physical Objectives N0 0 0 Poverty Reduction 0 0 0 O Gender Concerns 0 0 0 N Other Social Objectives 0 0 0 O Environmental Objectives 0 0 0 N Public Sector Management 0 0 0 O Private Sector Development 0 0 0 O Other (specify) 0 0 0 N B. Project Sustainability Likely Unlikely Uncertain 0 0 j 1ighlv C. Bank Performance Satisracfory Satisfactory Deficient Identification 0 0 Preparation Assistance 0 0 Appraisal 0 0 Supervision 0 0 D. Borrower Performance Satisfacfory Satisfactory Deficient Preparation 0 0 Implementation 0 O Covenant Compliance 0 0 Operation (if applicable) 0 0 0 E. Assessment of Outcome Satisfact/ory Satisfactory Unsatisfactory Unsatisfactory 0 0 0 0  Table 2: Related Bank Loans/Credits Loan/Credit Title Purpose Year of Approval Status Preceding operations 1. None Following operations 1. Roads and Coastal Development of local FY 92 on-going Shipping (Cr. 2374) transport network 2. Local Government Development of local FY 93 on-going Reform and Engineering goverments and (Cr. 2530) related services 3. Roads and Coastal Development of local FY 94 on-going Shipping II (Cr. 2599) transport network Table 3: Project Timetable Steps in project Cycle Date Planned Date Actual/Latest Estimate Identification (Executive Project Summary) October 1987 Preparation/Preapprisal January - February 1988 Appraisal April 1988 Negotiations June 20-24, 1988 Board Presentation August 2, 1988 Signing October 19, 1988 Effectiveness February 14, 1989 Midterm Review (if applicable) March 1992 Project Completion December 1994 September 1996 Credit Closing December 31, 1995 October 31, 1996 l As provided, for example, in the Staff Appraisal Report (SAR).  Table 4: Loan/Credit Disbursements: Cumulative Estimated and Actual (US$ thousands) FY 89 FY 90 FY 91 FY 92 FY 93 FY 94 FY 95 FY 96 ]FY 97 Appraisal estimate 1.00 8.00 22.00 38.00 52.00 58.00 60.00 60.00 60.00 Actual 7.81 9.05 18.30 30.05 43.09 48.08 50.95 52.08 53.6 Actual as %ofestimate 781% 113% 83% 79% 83% 88% 86.8% 89.1% 89.3% Date of fnal disbursement 1/9/97 Table 5: Key Indicators for Project Implementation Not Applicable Table 6: Key Indicators for Project Operations Not Applicable Table 7: Studies Included in Project Study Purpose as Defined at Status Impact of Study Appraisal/Redefined Action plan based on To develop local Work delayed by almost Negligible study to strengthen local government capacity, one year; findings not government institutions especially fnance applied and finances mobilization National housing To develop via the Completed by using Negligible strategy housing strategy UNDP funds Water and sanitation To promote fnancial Maputo study completed Could be significant, tariffs study viability of water just before closing date; Maputo study opened up companies in Maputo Beira not done discussion of regional and Beira tariff levels  Table 8A: Estimated Project Costs Item Appraisal Estimate (USSM) Actual (US$M) Local Foreign Total Local Foreign Total Costs Costs Costs Costs Infrastructure Rehabilitation 7 24.3 31.3 23.211 Housing Rehabilitation 4.2 19 23.2 15.433 Solid Waste Management 0.8 3.3 4.1 0.95 Equipment, vehicles, etc.'] 3.328 Lines of Credit: 0 4.9 4.9 1.5 SME 0 Building Materials 0 Institutional Development 0.1 3.3 3.4 2.149 Project Administration 0.4 2.7 3.1 1.924 CON, TRG, Survey'] 6.23 Base Cost 12.5 57.5 70 10.943 54.725 65.668 Physical Contingencies 1.2 5 6.2 Price Contingencies 1.4 6.2 7.6 Total Cost 15.1 68.7 83.8 10.943 54.725 65.668 1. Not able to allocate by components Table 8B: Project Financing Source Appraisal Estimate (USSM) Actual (USSM) Local Foreign Total Local Foreign Total Costs Costs Costs Costs IDA 6 54 60 53.678 Cofmancing: 3.1 14.7 17.8 Finnida 2.32 11.12 13.44 Spain 0.76 3.59 4.35 Denmark 0.5 Netherlands 0.547 Government 6 0 6 10.943 Total Financing 15.1 68.7 83.8 10.943 54.725 65.668 Table 9: Economic Costs and Benefits Information inadequate  Table 10: Status of Legal Covenants Agreement Section Covenant Present Original Revised Description of Covenant Comments Type Status Fulfillment Date Fulfillment Date DCA 2.03 13 12/31/95 8/31/96 Closing date Extended 10 Months DCA 3.03 11,12 CP 3/31/89 GOM will (a) prepare and furnish to IDA: (i) surveys done; (i) the results of socio-economic sample surveys to (ii) action plan assess impact of price increases on affordability of unsatisfactory project components; and (ii) the action plan based on the 1988 study to strengthen local government institutions and finances. DCA 3.03 (b) CP 6/30/89 (b) exchange views with IDA on the recommendations from the surveys and action plan; and DCA 3.03 (c) NC (c) implement such recommendations, including No cost recovery cost-recovery mechanisms. DCA 3.04 12 CP June 30, 1989 (a) Done, date GOM will, review with IDA, the progress in: (a) Housing plan unclear designing and adopting a national housing strategy; designed, but not (b) Not done and (b) implementing an action plan to strengthen implemented local government agencies DCA 3.05 11 CP 12/31/89 See comments GOM will: Maputo done in (i) carry out a water and sanitation tariffs' study. late 1995; Beira not done. Study on 13 cities added, and finalized in April 1996. DCA 3.05 2,11 NC June 30, 1990 (i) review with IDA the recommendations of the study, and (ii) agree on an action plan to implement the recommendations, including targets for the water and sanitation agencies in Maputo and Beira, to achieve their financial viability by FY 1993, and (iii) promptly thereafter, carry out the action plan according to a timetable agreed with IDA. DCA 3.06 5 CP GOM will conduct annual Project implementation reviews. DCA 3.07 7 GOM will implement the resettlement plan for persons to be relocated under the housing rehabilitation component. DCA 3.08 5 CP GOM will maintain PIU and HABITAR with such Habitar disbanded responsibilities, powers, services, facilities, and staff. in 1994. PIU staff to include director, project coordinators,  Agreement Section Covenant Present Original Revised Description of Covenant Comments Type Status Fulfillment Date Fulfillment Date financial controller, procurement advisor; Habitar staff to include director, project coordinator, financial controller, and field implementation teams in Maputo and Beira. DCA 4.01 1 CP GOM will maintain or cause to be maintained records A de M accounts and accounts adequate to reflect the operations, inadequate for resources, and expenditures, in respect of the Project, much of project; of the departments or agencies of GOM responsible A da B accounts for carrying out the Project or any part thereof, unavailable. including those related to all operations carried out by A de M and A da B in the water and sanitation sector. DCA 4.02 (i) 2 NC Starting with FY (i) A de M and A da B will produce, for each fiscal Not accomplished 93 (December years, funds from intemal sources equivalent to not by either A de M 31) less than 25% of the annual average of capital or A da B. expenditures incurred or expected to be incurred, as calculated from capital expenditures for that year, the previous fiscal year and expected for and the three following fiscal years. DCA 4.02 (ii) 2 NC By September 30, (ii) GOM will, on the basis of forecasts prepared by Not done of each year A de M and A da B and satisfactory to IDA, review whether they meet the above requirement. DCA 4.02 (iii) 2 NC (iii) GOM will promptly take all necessary measures Not done (including, adjustments of the structure or levels of its water and sanitation tariffs). Covenant types: Present status: I = Accounts/audits 8 = Indigenous people C = covenant complied with 2 = Financial performance/revenue 9 = Monitoring, review, and reporting CD = complied with after delay 3 = Flow and utilization of project funds 10 = Project implementation not covered by categories 1-9 CP = complied with partially 4 = Counterpart funding 11 = Sectoral or cross-sectoral budgetary or other resource NC = not complied with allocation 5 = Management aspects of the project or 12 = Sectoral or cross-sectoral policy/regulatory/institutional executing agency action 6 = Environmental covenants 13 = Other 7 = Involuntary resettlement  Table 11: Compliance with Operational Manual Statements Statement Number and Title Describe and Comment on Lack of Compliance |I1 None 3. Table 12: Bank Resources: Staff Inputs Planned f Revised Actual Stage of Project cycle Weeks US$ Weeks US$ Weeks US$ Through appraisal 71.4 165.9 Appraisal-Board 9.0 21.4 Board-effectiveness 14.0 25.5 Supervision 109.3 270.3 Completion 13 30.3 TOTAL Sources: COSR20  Sheet1 Table 13: Bank Resources: Missions Stage of the |Number of Specialized staff project cycle Monthlyear |persons Days in field skills represented Performance Ratings Types of problems I Overall Status or Development Impact Implementation or Development Project Avail. Through____ app______sal____ Status Objectives Legal Mgmt Funds Through *ppraIsal: _______ iiiZZ IiZZ Identification Oct47 5 12 a. d. g not applicable Preappralsal Jan48 6 t5 a, d, e, f. g not applicable Appraisal to Apr-88 16 a, d, e, 1. g, h, I not applicable Board Board to Effectveness _ I Sep-88 5 5 d, h, m, 1, g not applicable Jan-89 d not applicabe instutional Supervision = May-89 2 1 la, e 11 Ju-9 1 2t no 590 Sep89 6 2 a, j no 590 Oct-89 5 12 d, e, g, i 17 1 1 1| 2 instiutional Jan-90 2 14 d, z no 590 Mar-90 1 11 9 no 590 May-90 4 18 d, e, g, k ' | ' 1 1 1 procurement Sep-9_ 2 2 a. e no 590 Oct-90 4 12 d, h. 1i e 12 1 1 1 procurement Feb-91 2 2 a, e no 590 Feb-91 3 30 d, e 1 21 1| 21 1 coastal component Sep 91 2 4 a, z no 590 _ ________ Nov-91 3 11 d,e.n 21 2 1 2 1 Mar-92 3 13 d,e.n 21 1 1 1 2 weak ocal capacty Oct-92 2 6 d6 e no 590 weak local capacity. sustainabilky Apr-93 1 2 o no 590 Jun-93 2 6 d, e no 590 capacity, finance Sep-93 1 3 o no 590 Jan-94 1 7 o no 590 Jan-94 2 11 d. a 2 1 2 1 1 1 2 solid waste, sustainabilit Jun-9,4 2 3 d,z no 590 Oct-94 2 2 d, e no 590 Feb-95 3 d S IS 2 2 1 21 2 9_ _ 95 1 3d no 590 Jul-95 2 ia, d 21 2T 2 Nov-95 2 _ _ d.a I I Feb-96 2 3 d, a U U 31 3 2 management Legend for skills: c - transport pi f - housing I - sanitation I r procurement o = operations a = rmancial anaiyst d - urban plan g institutonal d j - water m = Bank manager z = not specified b transport economist * - engheer h - legal k geographer n = GIS Page I   NIOZAMBIQUE URBA,N REHABILITATIO, PROJECT (CR. 1949) AIDE-ME MOIRE I Catherine Seibert, financial analyst, visited Mozambique during the period February 18 through Mlarch 8. 1996, during which time she supervised the Urban Rehabilitation Project (PRU). Bernard Becq. task manager for the project, joined Ms Seibert for fLnal discussions. The mission met with members of the Project Implementation Unit, DNA, DNEP, City Council of Mlaputo, and GPE, and wishes to thank all involved for their assistance. As the project is scheduled to close April 30, 1996, this mission reviewed pending marters and preparations for the Implementation Completion Report (ICR). Extension of the Closing Date 2. Since the last mission, Government and the Bank agreed to extend the closing date from December 31, 1995 to April 30, 1996, in view of the number of activities that had not been completed. The following tables summarizes those activities; the second and third columns indicate the pending disbursements and expected dates of completion, as listed in Minister White's communication to Ms Katherine Marshall, dated in December 1995, and the last two columns indicate revised dates: Activity Remaining Expected Date Expected Date Revised Date of Revised Date Disbursements of Completion of Last Completion of Last Disbursement Disbursement Maputo Water USSO.9 M Jan. 31, 1996 April 30, 1996 March 31. 1996 April 30. 1997 Supply USS0.9 M lan. 316 r316 r3 96 i0_ Beira USSO. I M completed Feb. 28, 1996 Sew erage__ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Maputo Roads USS 0.2 M Maintenance Brigade Beira Roads USS 0.4 M Maintenance Brigade 3. Progress in implementation and achieving disbursements since the last mission and subsequent extension of the closing date has been negligible. It was agreed during the mission that PIU would update the above table, including revised datafor completion and disbursements, andfax to the World Bank by March 15. While the above activities are discussed in more detail in the following paragaphs, the mission emphasi to the Ministry and the Project Implementation Unit (PIU) that with the exception of the fincial controller and the project coordinator, no disbursements wll be made on activies that are competed after tke fevfsed closing date of April 30, 1996. In view of difficulties in completing the essential activities of the project, the mission agreed that any additional activities to those listed above (including the thdee tasks added by the Bank - beneficiary assessment, familiarization visits in regard to AGETIP and UN resettlement report) would not be carried out. Rehabilitation and Extenslon of Water Works - Maputo 4. The most serious problems in implementation relate to the rehabilitation and extensioa of the Maputo water network being carried out by CETA and supervised by Mott McDonald: timely completion, payment of final payment, resolution of payments to Mott McDonald. DNA furnished the mission with a stas report on the component.  5. Timely Completion.. At the time of the extension of the closing date, substantial completion was expected by January 31, 1996 with completion of disbursements by April 30, 1996. Because of further delavs on the part of the contractor, DNA has revised estimates for completion and "hopes it will be the end of Mtarch 1996". In addition to delays in completion, disbursements appear to have slowed as CETA has been very slow in submitting invoices, and as of March 1. 1996. had not yet submitted invoices for October 1995. The mission emphasized to DN/A and directly to CETA that all activities, must be completed by .4pril 30, 1996; any activities not completed by that date would not befinanced under the CrediL 6. Disbursements of Final Payments. Because of unexpected delays in completion of the water network activities, the retention payment to CETA. and any related payments to Mort McDonald for issuance of the completion certificate, are scheduled to take place in March/April 1997, one year after completion of works. After consultation with the Bank's disbursement and legal officers, the mission confirmed to PIU that Cr. 1949 would not be able tofinance these expenditures, as they take place after the closing date. Accordingly, these expenditures, estimated to cost about USS 100,000 to USS 110,000, will have to be borne by Government. 7. Pending Issues with Supervising Engineer. Several other matters were discussed in relation to this component. The mission received a revised Addendum 48 of Mott McDonald contract. While some of Mr. Becq's earlier comments have been taken into account, others have not; in particular, the Bank cannot finance any expenditures incurred after April 1996, so Section B of the expenditures should be dropped. Attached to this Aide-Memoire is a marked up copy of the revised Addendum, for the use of DNA and PIU. DNIA will revise the Addendum and resubmit to IDA for the no objection. Refund of Advance Payment to Indkco 8. PIU appears to have made little progress since the last mission in resolving the issue of the repayment of the advance payment to Indico. In brief, Indico Construcoes, SARL, undertook to the completion of the building 33 Andares, for which it received advance and direct payments totaling USS 1,303,450.93 and Mt 1,275,913,956.53. However, Indico only accounted for part of that amount, leaving USS528,843.89 and Mt348,680,084.88 to be refunded to the project. Because of Indico's financial problems (it was subsequently privatized), the then-Ministry of Construction and Water approached the Bank of Mozarnbique in January 1992 to request that BdM pay the guarantees provided on Indico's perfornance. Follow up correspondence from the Ministry to BdM was sent in May 1992, March 1993, and April 1994 apparently without response or action from BdM. 9. Clearly, the Bank of Mozambique has been remiss in not acting on the guarantees, and the mission urged the Ministry to pursue this issue vigilantly with BdM. However, as IDA funds were involved in what has turned out to be an inappropriate disbursement, and as the disbursement period is drawing to a close, the World Bank is obligated to get a refund on these amounts. To effect this, the mission instructed the Loan Department at the Bank to deduct the amount involved in the guarantee from future disbursement on the project (for example, legitimate invoices submitted for Maputo water supply, would not be paid or reimbursed, but "applied" to the USSS28,843.89 outstanding on 33 Andares). The Bank's actions notwithstanding, bowev, it would be in the bests interemsts of the Ministry if PIU more dllgendy fo/Jew op wvtI WE on &e guarantee. Beira Sewerage 10. One of the activities on which disbursements are still expected is the rehabilitation and extension of the Beira sewerage network. The mission noted that in regard to consultancy fees, the amount paid up to the maintenance certificate was about USS 100,000 more than the original contract value of USS 174,000. The mission would appreciate receiving the date of the Bank's no objection for this increae. Disbursement Issues it. The mission discussed several disbursement issues with the PHU. First, PIU noted that the special account was being recovered at a time when a significant volume of disbursements is still pending, and requested a replenishment or partial replenishment of dhe special account The mission pointed out that there  were clear procedures related to special accounts and their recoverv. and that any future disbursements should be effected by direct payment. Second. the mission pointed out that the PIU needs to account for about USS370.000 which has already been disbursed from the special account. but for which no supporting documentation has been submitted to the Bank. Finally, the mission agreed to discuss with management the possible granting of a four-month grace period, in which to complete disbursements. Audit of Project Accounts 12. The mission agreed to PIU's suggestion that activities over the four months from January to April 1996 be covered in the 1995 audit, rather than as a separate audit. PIU should send a revised contract or addendum to the Bank on the extension. There appears to be some difficulty in establishing a register of the assets acquired under the project; the mission emphasized the importance of such a register, as there appear to have been many smaller items purchased under the project, which must be accountedfor. Preparation of the Implementation Completion Report (ICR) 13. PIU and the mission discussed the schedule for the preparation of the ICR, as suggested by the Bank mission: Closing date: April 30, 1996 Draft of Bank's report to Government June 15. 1996 Government's report to Bank July 15, 1996 Emerging Themes of the Bank's ICR 14. Attached to this Aide-Memoire is the draft table of contents for the Bank's evaluation (note that the Government's evaluation in included in the document which goes to the Board). Although the report is still in the process of being formnulated, some themes are emerging conceming project design and implementation. Conceming project design, the project appears to have been well-designed, in line with objectives that were carefully crafted for the country. The design of some components showed creativity. However, like many other projects of its age, insufficient attention was paid to policy measures during implementation, which places in doubt sustainability. 15. Project implementation overall was marginally satisfactory, with broad variation between components. TIhe road maintenance component was well executed from the borrower's point of view. Less successful was the housing rehabilitation component, which had a riskier design, and least successful were the lines of credit for businesses and for construction materials. The DNA supervised components, particularly in regard to Maputo water, are not yet completed; however, it is clear that implementation has been very problematic. From both the view of the Bank and the Government, the earlier years of the project were marked by more assiduous and enthusiastic implementation. Frequent changes of key staff, on both the part of the implementing agencies and of the Bank, probably adversely imnpacted aspects of implementation. In addition, as is frequendy the case, the implementation of PRU become of lesser interest and perhaps received fewer resoures as both the attention of the Bank and the Government shifted to the development of the next project, PROL. which was to take up and develop the policy issues. 16. As a result of both design and implementation factors, the sustainbtility of many of the activities supported by the project is questionable. This possible lack of sustainability is put into question by a number of factors: inadequate attention to policy measures in project design and implementation, especially the needfor cost recovery: As with many projects of its age, the design of the PRU was somewhat loose on policy measures, although it did discuss the need for changes related to water tariffs, housing policy, and local government development. However. implementation could have been more aggressive in ensuring that the agreed reforms were mad. rho sustauiability of improvements in water supply, sanitation, housing and solid waste are aJI pw am question by uneconomic tariffs and charges.  the implementation capacity of involved entities: The project put large demands on the implementing agencies, all Government, in terms of the breadth of project scope. the numbers of agencies involved, and the deliverables (tariff studies. civil works, etc.). The GQvernment had and continues to have limited human resources, which do appear to have been overtaxed by carrying out the project activities (along with numerous other donor activities). The efforts made through the project to address inadequate human resources at Government, including the establishment of a project implementation unit and the topping up of some Government salaries, responded to the problem only temporarily, and may have in fact worsened the problem. In addition to the lack of human resources, the lack of financial resources in terms of counterpart financing adversely affected several components, including the housing component. March 1996   IMAGING Report No.: 16747 Type: ICR