Mozambique - Country assistance strategy completion report (CASCR) review for the period FY2004-07 : IEG review

88813 EG CASCR Review independent Evaluation Group For Official Use Only May 15,2007 Country: Mozambique CAS Year: FY2004 I CAS Period: FY2004-07 2. Executive Summary This CASCR Review examines the implementation of the FY2004-07 Mozambique CAS and evaluates the 2007 CASCR. The FY04 CAS was a results-based strategy prepared jointly with IFC and MIGA. A CAS Progress Report, an assessment of the Bank's program during the first two years of implementation of the CAS, was presented to the Board in March 2006. The CASPR proposed a harmonization of the outcome indicators in the CAS and the retrofitted Performance Assessment Framework of the Poverty Reduction Strategy (PARPA) undertaken in 2006, but failed to explain the rationale for changes in individual indicators. The FY04 CAS supported the development objectives of the Government through the three pillars of the Mozambique Poverty Reduction Strategy defined as: improving the investment climate, expanding service delivery, and building capacity and accountability. The Bank's program aimed to help achieve these objectives by providing US$659.3 million in 14 operations to Mozambique and producing a number of analytic and advisory services. This Review rates the overall outcome of the implementation of the FY2004-07 CAS as moderately satisfactory, the same as the CASCR. Regarding the first pillar, the review found that Bank Group support for reforms to improve the business environment was well-focused and included the right actions to reduce the cost of doing business, although the outcomes were mixed. A notable achievement was the reduction of the average time required for registering a business (from 174 days to 113 days, just short of the 40 percent declined projected by the CAS), but this did not necessarily translate into a significant in the cost of doing business as there are still 13 registration steps required to launch a business, compared to 15 at the start of the CAS. Labor market reforms also fell short of changes required to improve competitiveness as the draft new labor law was watered down, limiting its effect on private-sectorjob creation. Bank support also focused on improved access to telecommunications, transport connections (roads, ports and railways), and access to power. Except for the major improvement in teledensity (doubling to 16 per 1,000 inhabitants), most reforms targeting privatization in transport and power made slow progress. Reforms to sustain the management of natural resources and to support the adoption of new farm technologies also had mixed results. On the other hand, the economy recorded strong growth (in excess of 7 percent per annum during 2003-05), which could be attributed to the expansion in construction and communications, the rebound of agriculture and the increases in exports aided by the megaprojects in mining and petroleum. On account of these outcomes, the achievement of the objectives of the first pillar is rated moderately satisfactory. The achievement of the objectives of the second pillar is also rated moderately satisfactory likewise on account of mixed outcomes. Capacity constraints on the part of Government, design deficiencies and poor coordination among donors limited the effectiveness of Bank support to reduce the incidence of HIVIAIDS. Policy support for the health sector, on the other hand, led to an increase in the budget allocation to the sector from 3.4 percent of GDP in 2003 to 3.5 percent by 2005, exceeding the CAS target of maintaining the health budget at the 2003 level with respect to GDP. In the area of education, the outcome of Bank support has been significant in building capacity and improving access, especially at the primary and tertiary levels. But quality issues persisted: the primary completion rate CASCR Reviewed by: Peer Reviewed by: Group Manager James Sackey Chad Leechor Ali Khadr Lead Economist, IEGCR Sr. Evaluation Officer, IEGCR Sr. Manager, IEGCR  For Official Use Only EG CASCR Review Independent EvaluationGroup 2 remained low at less than 50 percent, raising the risk that the MDG targets in this area may not be attained. The third major area of Bank support (pillar 3), also had mixed outcomes and is likewise rated moderately satisfactory. Support for reforms of the fiscal structure had satisfactory outcomes (the budget deficit declined from -4.5 percent of GDP in 2004 to -2.0 percent in 2006, consistent with targets agreed with the IMF; priority expenditures improved; and structural reforms supported by the IMF and the Bank are on track). Nevertheless, interventions in support of governance reforms (including the reduction of corruption) yielded disappointing outcomes. Mozambique's ranking for anti- corruption control remains low; the 2007 Doing Business Report ranked Mozambique 93 out of 146 countries for corruption (the inherent deficiencies of such measures notwithstanding). IEG concurs with the lessons of the CASCR which touch on the need for better collaboration among IDA, IFC and MlGA for strengthened support to private sector development; improved appreciation of the country's political economy during project design, timing and implementation, especially with respect to reforms dealing with governance; and better design of monitoring and evaluation systems to aid project supervision. In addition, the findings of the CASCR Review suggest the need to: (i) ensure that governance reforms are based on a comprehensive program of sensitization and capacity-building at all levels of government; and (ii) substantially scale up capacity- building activities in core competencies if service delivery and governance objectives are to be realized. Both require a high level of government ownership and commitment. 3. CASCR Summary Overview of CAS Relevance: The CAS was relevant because: It constituted a continuation of the previous CAS by supporting the development objectives of the Government as defined in the PARPA (the Mozambique PRSP). The alignment with the PARPA enabled the Bank to be selective and to capitalize on its comparative advantage under each of the three pillars. Under pillar 1, the Bank supported the Government's long-term goal of sustaining GDP growth per capita by improving the investment climate and facilitating public-private partnership in infrastructure development. Under pillar 2, the Bank supported the Government's effort in the area of human development and in improving quality of life through interventions to: control HIVIAIDS prevalence; reduce child mortality; improve access to safe water; raise the primary education rate; and increase lblozambican participation in business management and the professions. Under pillar 3, the Bank helped the Government to improve public expenditure management and contain corruption. The Bank also sought to help strengthen the rule of law, including systems supporting contract enforcement, and to enhance the Government's M&E capacity. The proposed lending and non-lending assistance was generally consistent with the areas of focus of the Bank (taking into consideration ongoing operations from the previous CAS). The flagship of the lending program was programmatic support through four rolling Poverty Reduction Support Credits (PRSCs), in line with the effort to implement the Paris Declaration on donor alignment and harmonization (A & H) and development effectiveness. The shift to programmatic lending through the PRSCs was underpinned by core diagnostic economic and sector work (ESW), including annual Public Expenditure Reviews (PERs) but experienced scheduling problems. Selected investment projects targeted institutional development and capacity-building (Decentralized Planning and Finance Project, Financial Sector Capacity Project); infrastructure (Beira Railway SIL, Roads and Bridges APL); and human development (in addition to the ongoing Education Sector Strategy Project, the Technical and Vocational Education and Training Project). The Bank also participated in a SWAP for the agricultural sector (PROAGRI), which spanned 1999 through 2006, but the experience was not repeated  For Official Use Only EG CASCR Review Independent Evaluation Group 3 because of challenges encountered in reconciling IDA'S fiduciary policies and procedures with those of other donors and the government. As the Bank moves towards adopting country systems in line with improvements in the country's procurement and financial management systems, SWAPS could increasingly become a useful instrument for donor A & H in IMozambique. The CAS Progress Report (CASPR), submitted to the Board in March 2006, highlighted areas of discrepancy between the CAS Results Matrix and the retrofitted Performance Assessment Framework (PAF) of the PARPA undertaken in 2006, as well as the indicators for the country monitoring tool of the Africa Action Plan (AAP). It proposed a harmonization of the outcome indicators, which resulted in a revised CAS Results Framework (Appendix II of the CASPR). The revised CAS Results Framework retained the three pillars of the CAS, but modified some of the intermediate outcome indicators in order to take account of progress made in implementing the CAS (Annex I, Attachment 1, of the CASCR). The reformulation of some of the outcome indicators aimed at providing realistic benchmarks for assessment, but in effect watered down some of the quantitative benchmarks (such as those for education and access to safe water) and adopted process benchmarks for reforms (such as those for the financial and agricultural sectors). While this is not explicitly stated in the CASPR or the CASCR, it appears that most of the quantitative benchmarks in the original CAS Results Matrix were either not attainable (over-ambitious) or were not verifiable (because of data limitations). The discussions of achievements by objective below, will nevertheless, take into account (to the extent possible) both the original targets and the reformulated targets. I The implementation of the program was broadly consistent with the approach proposed in the CAS. The Bank delivered 14 operations as proposed in the CAS, including one GEF and three additional investment operations that were not originally programmed. Total commitments were US$659.3 million, compared to the CAS base case of US$560 million. Three of the operations were Poverty Reduction Support Credits (PRSCs), totaling US$250 million, close to the proposed volume in the CAS. While the commitment followed the base case, most operations were approved during the latter part of the CAS period, reflecting slippages in preparation. The CASCR provided no commentary concerning these delays. The flagship of the lending program was three PRSCs (as opposed to the four planned), which supported policy reforms underpinning all the pillars of the CAS. The scheduling of the PRSCs encountered problems with respect to the preparedness of Government, which limited the timeliness and effectiveness of the sequence. Selected investment projects were also implemented and they targeted institutional strengthening, capacity-building, transport infrastructure, water, agriculture and communications (see Annex Table I), but also encountered scheduling problems and long delays in initiating implementation. The performance of the portfolio was mixed during the CAS period; it was rated low risk in FY04, but deteriorated substantially in FY05 and improved marginally during FY06-07 (see Annex Table 4). There were two reasons for the deterioration in the portfolio: slow project start-up, which could be attributed to weaknesses in project preparation, and slow disbursement, which could be traced to problems with procurement, financial management, and counterpart funding (including the refund of VAT payments). The percentage of the portfolio at risk was however broadly comparable to that for the Africa Region and Bank-wide. While the portfolio was overaged (in December 2006, the CASCR noted that 9 of the 19 projects in the portfolio were at least 5 years old), IEG ratings of projects that exited during FY04-06 were generally satisfactory. Of the 7 projects that exited the portfolio during FY04-06, 3 had moderately satisfactory Bank program outcome ratings and the remainder had satisfactory ratings (see Annex Table 3).  EC CASCR Review Independent Evaluation Group For Official Use Only 4 The mid-term CAS Progress Report noted additional projects added to the portfolio, and changes made in mid-stream to enhance the implementation of the CAS (including retrofitting of the Results Matrix). The new operations included the Maputo Municipal Development Program to support the Government's decentralization agenda; legal and judicial capacity-building undertaken through the restructured Public Sector Reform project; and aligning the PRSC with the Budget cycle, thereby advancing the PRSC3 by six months and merging it with PRSC2. Emphasis was also placed on deepening coordination (which was long overdue) with IFC and MlGA on tourism, SMEs and HIVIAIDS Issues; and technical assistance in trade facilitation, water resource management and telecommunications infrastructure. Non-Lending Services. Ten of the 18 ESWs planned in the CAS were completed. An additional 13 un-programmed ESWs vvere delivered to make up for the shortfall. The programmed ESWs included core analytic works such as the CEM, PER, Private Sector Competitiveness, Rural Strategy, Institutional and Governance Review and Country Status Report on Health. However, some important ESWs pertinent to the objectives of the CAS were dropped. These included the PSlA on agriculture, the CPAR, and the HIVIAIDS Retrospective review. The CASCR does not explain the factors behind the choice of the un- programmed ESW and why a number of key studies were dropped, but highlighted problems with planning, coordination, funding and dissemination. The CAS Progress Report noted that the non- lending program was strengthened to include analytic work on: constraints to agricultural productivity; rural access to telecommunications; growth poles and regional development; updated assessment of poverty and gender; and environmental and social framework. The quality of analytic work (ESW) was generally high, although it could have benefited from better programming, as well as been more demand-driven and better coordinated with other donors. Overview of Achievement by Objective: The Mozambique CASCR provides a detailed assessment of the achievement of CAS objectives by tracking performance benchmarks through the revised CAS Results Matrix. The outcomes associated with the CAS objectives (based on the original and revised CAS Results Matrix) are briefly reviewed below. Pillar I: Improving the Investment Climate. The first pillar of the CAS dealt with the growth agenda. Bank support emphasized growth through private sector investment, improved regulation and supervision of financial systems, improved delivery of infrastructure services, sustainable management of national resources, and increased use of new farm technology. A sixth expected outcome, an efficient flexible labor market, was subsumed under improved business environment in the revised Results Matrix. Improved business environment for the private sector. Bank Group support for reforms on the business environment was well-focused but faced limitations associated with Government commitment. The PRSC series and the Enterprise Reform Project, complemented by analytic work through the CEM, the PERs and the Doing Business surveys by the IFC, were instruments used for dialogue and reform. But the outcomes were mixed and overall below expectations. A notable achievement was the reduction of the time required in registering a business (from 174 days to 113 days, just short of the 40 percent decline projected in the CAS), although this did not necessarily translate into a significant reduction in the cost of doing business as there are still 13 registration steps required to launch a business, compared to 15 at the start of the CAS. The strong growth that occurred during the period can be attributed to the expansion in construction and communications, the rebound of agriculture and the increases in exports aided by a series of megaprojects. Megaprojects in mining and petroleum also account for the high growth in foreign investment. The labor market reforms (required under PRSC I and II) also fell short of changes required to improve competitiveness of the  For Official Use Only EG CASCR Review Independent Evaluation Group 5 market. The draft of a new labor law that was intended to increase flexibility in the labor market was watered down, limiting its effect on private-sectorjob creation, although the government had promised to re-examine the issue following pressure from donors. Reinforcement of the regulation and supervision of financial systems. The expected outcomes in this area were expanded from a narrow focus on efficient, profitable banks in the original Results Matrix to a broader overview of the financial sector. Bank support was provided through the multi-donor Financial Sector Technical Assistance Project and the PRSC series, working along with other development partners such as the IMF and DflD. The achievement of the outcomes could be rated moderately satisfactory on account of the slow rate of implementation, both with respect to the commercial banks (the target for compliance with IFRS standards was shifted from 2006 to 2007) and the financial system as a whole. The Central Bank introduced IFRS accounting, which is being extended to the rest of the financial sector; while the strategy for the Government to divest from BIM is still under discussion. Through the support provided by the Enterprise Development Project and assistance from the IFC, some progress was made in increasing access to credit by SMEs, but no hard data on coverage are available. Slow implementation of this component can be attributed to poor collaboration between IDA and IFC on which business model to adopt. Coordination with other donors (notably the AfDB) also exhibited room for improvement. Improved delivery of infrastructure services. The expected outcomes in the original Results Matrix were to reduce infrastructure costs and to increase access. The former benchmark was dropped in the revised Results Matrix. Bank support was targeted at improved access to telecommunications, transport connections (roads, ports and railways), and household access to power. But even under the scaled-down benchmarks, outcomes were below target as only one out of the three revised broad benchmarks was fully met. On the positive side, reform of the telecommunications sector (supported by the well-designed Communications Sector Reform Project under the previous CAS) led to major improvement in teledensity (doubling to 16 per 1,000 inhabitants) and a dramatic drop in the cost of mobile telecommunication. But the privatization of the state-owned operator remained unresolved. With respect to the transport sector (supported through the Roads and Bridges APL), the objective of Bank assistance was to ensure the long-term financial sustainability of investments. But while substantial policy reforms had been initiated with respect to private sector participation, road safety and the establishment of a Road Fund, deficiencies in provincial works departments and procurement delays undermined performance and road rehabilitation and maintenance targets were not met. Finally, in the power sector, policy reversals with respect to privatization and private sector participation on the part of the new Government made it impossible to meet the target for improved access. The Energy Reform and Access Project, which was the main instrument for the reforms, was approved just before elections and was marred by weak commitment on the part of the new Government and was put under consideration for restructuring. Sustainable management of natural resources. The focus in the revised Results Matrix was shifted from the narrow concern with the rational use of land in the original matrix to the broader objective of facilitating more efficient and sustainable land management, while helping to improve the environment for tourism led by the private sector. Outcomes on both original and retrofitted indicators were mixed; IDA and GEF programs of assistance experienced substantial implementation bottlenecks (including weak Government capacity and commitment) and most of the intermediate indicators established were not attained. Increased use of new farm technologies. In both the original and retrofitted results frameworks, the objective of the Bank's assistance was to assist in the adoption of new technologies so as to improve agricultural productivity. Bank assistance was largely through analytic work and SWAPS in advancing research, extension, animal health support systems and promoting the security of land tenure for smallholders. There was progress in the adoption  For Official Use Only EC CASCR Review Independent EvaluationGroup 6 of short-cycle, protein-rich varieties of maize; disease tolerant cassava clones; and nutrient enhanced sweet potatoes. But although the narrow benchmarks identified in the CAS matrix were attained, poor policy execution continued to be an issue. The common approach through the SWAP failed as major donors (DflD, Netherlands, Italy, USAlD and IDA) pulled away because of poor performance of PROAGRI and concerns with procurement and fiduciary issues. Pillar II: Expanding Service Delivery. The second pillar of the Bank's strategy aimed at addressing service delivery issues that could be a drag on efficiency and efficacy, with special focus on reducing the incidence of HIVIAIDS; improving coverage of health services; increasing access to safe water and sanitation; increasing primary enrollment and completion rates; and eliminating gender and regional disparity in education. The outcome benchmarks in the revised Results Matrix were radically different from those of the original matrix, in part because Government had made slow progress in service delivery reforms, despite Bank efforts to help through the series of PRSCs, the public sector reform project, sectoral intervention and analytic work. The review below is based mainly on the revised Results Matrix since most of the benchmarks in the original matrix were not met or could not be evaluated because of data limitations, and for a lack of specificity in definition. Reduced incidence rate of HIVIAIDS. Bank support through the HIVIAIDS Response Project, the regional HIVIAIDS Treatment Acceleration Project, and policy dialogue through the multi-donor Partnership Forum was broad-based and the Bank also provided about 60 percent of total financing for HIVIAIDS. But the outcome was mixed: geographical coverage did not improve in line with expectations and key benchmarks were not attained. Capacity constraints on the part of Government, design deficiencies associated with program intervention and coordination problems among donors accounted for poor performance. Improved coverage of health services. Although the Bank did not provide direct financing to support the objectives of improving the coverage of basic health services, its analytic work was instrumental in influencing policy reform and service delivery options. Under the original Results Matrix, the key directions of Bank policy dialogue were the need to increase health expenditures; use an integrated service delivery approach; and focus on less-se~ed rural areas. Through the triggers of the PRSC series, the overall budget for the health sector increased from 3.4 percent of GDP in 2003 to 3.5 percent in 2005, which exceeded the target of maintaining health expenditures at 3.4 percent of GDP. However, it is not possible to evaluate the impact of reform on coverage in view of data limitations. The deliveries, coverage and utilization benchmarks specified in the revised Results Matrix were all met. Increased access to safe-water and sanitation. The goal of Bank assistance to Government was to reduce by half the number of people without access to clean water and sanitation. Bank support was implemented through two back-to-back National Water Development projects. But although Bank intervention assisted in substantial institutional reforms, outcome targets identified in both the original and revised Results Matrices were not met in part because the National Water Development projects were aimed at institutional development reforms targeted at five cities. There was clearly a disconnect between the goals of the CAS and the Bank instrument, and access increased only marginally for water while there was little evidence of significant improvements in sanitation. lncreased Access and Quality of Primary Education System. There was no substantial modification in the benchmarks in the original Results Matrix with respect to this objective. Bank support to the education sector ranged from access issues at the primary, secondary and tertiary levels to quality concerns. Three projects supported this effort: the Education Sector Strategy Program (1999-2006), the Higher Education Project (2002-2007) and the Technical and Vocational Education and Training project (2006-2010). The outcome was significant in  For Official Use Only EG CASCR Review Independent Evaluation Group 7 terms of building capacity and improving access, especially at the primary level through the ESSP, approved prior to the CAS under review in this CASCR. But quality issues ~ersisted: the primary completion rate remained low at less than 50 percent, raisingVthe risk that the MDGs in this area will not be attained. The targets for passing examinations would also not be met. The lack of donor coordination, in the absence of a common fund or SWAP, contributed to implementation problems, which was also exacerbated by design constraints. Efforts at eliminating gender disparity at all levels of the education system seemed to succeed as the ratio of girls to boys at the primary and secondary levels had witnessed a systematic increase. The Higher Education Project assisted in the fulfillment of targets for the key performance indicators related to internal efficiency rates, annual number of graduates and the introduction of new degrees. The target for the diversity indicator was only partially met. Improved public service delivery planning, budgeting, and financial management. The outcome benchmarks in the original Results Matrix were process-oriented (involving the enactment of new legislation which was complied with). They were replaced by similar process-oriented benchmarks in the revised Results Matrix but with more emphasis on the implementation of the decentralization program. Bank support through the Public Sector Reform (PSR) project (approved under the previous CAS), the Decentralized Planning and Finance (DPF) project, and the Maputo Municipal Development project helped initiate the process of reforms. But progress has been slow because of design problems associated with Bank interventions. For example, the requirement of a project implementation unit (PIU), which competed with sectoral ministries for control of reforms in a complex political environment acted as a drag on the reform process. Overall, most benchmarks specified in the revised Results Matrix were not met. They included: government approval of salary policy for the medium term; functional analyses for national and municipal agencies (directly supported by the PSR project); restructuring of seven designated ministries; and implementation of the new Procurement Code which had been approved in 2005. Pillar Ill: Building Capacity and Accountability. The revised Results Matrix reformulated the outcome indicators away from the HlPC expenditure tracking indicators towards more general improvements in budget allocation and execution. Bank assistance was to help bring about improvements in public expenditure management, thereby reducing the perception of corruption, and to strengthen the rule of law. The expected outcomes identified included improved budget allocation and execution; improved M&E capacity in Government; reduced corruption; and improved justice system. Improved budget allocation, execution and monitoring. Using the PRSCs and the PERs, the Bank worked closely with the IMF in supporting Government reforms on macroeconomic issues. Collaboration with Program Aid Partners in providing budget support, through the use of a common Performance Assessment Framework (PAF), was also instrumental in helping the Government achieve most of the objectives of public finance reform. Total revenues as a share of GDP rose from 12.6 percent in 2004 to an estimated 14.4 in 2006. Supported by an inflow of official grants that have risen from 7.5 percent of GDP in 2004 to 10.7 percent in 2006, the budget deficit shrank from -4.5 percent of GDP in 2004 to an estimated -2.0 percent in 2006. Within this favorable fiscal environment, the IMF in its January 2007 Review of the PRGF noted that priority expenditures rose steadily and should be on target. Implementation of the fiscal structural reform agenda was also on track, albeit slower than anticipated. In particular, good progress had been made in integrating donor-financed projects into the 2007 budget ceilings. Finally, M&E capacity in Central Government had improved, especially for the ~ monitoring of the PARPA. However, at the sectoral agencies, provincial and district levels, capacity remained weak as no results frameworks were in place.  For Official Use Only EG CASCR Review Independent Evaluation Group 8 Governance and reduced corruption. Progress on governance related reforms has been slow. IDA support through the various PRSCs, complemented by TA from WBI, did not yield the expected result. The perception of corruption in Mozambique remained high, while Government's response remained weak. The Government's governance surveys completed in 2004 indicated that Mozambique posted worse-than-average performance with respect to the control of corruption as compared with the countries in Southern Africa. An Anti-Corruption Strategy had been in place but an action plan was yet to be finalized. Similarly, with respect to the justice sector, priorities remain poorly defined by Government. The Bank has only recently (in 2006) initiated a program to assist capacity-building and strengthen institutions at the provincial/appellate jurisdictional levels with the objective of raising awareness of citizens' rights through outreach programs; and improve access to courts, conflict resolution mechanisms and legal services. Progress to date is minimal. Achievement of CAS Objectives Objectives CASCR Rating IEG Rating Explanation 1 Comments Analytic support was provided for improved business environment but was not sufficiently linked to other governance concerns or to government commitment and Pillar I: Improving the capacity. Delivery of infrastructure Investment Climate. services failed to deal adequately with weak capacity at the provincial level. Support for sustainable management was piecemeal and did not deal with capacity limitations and weak ~overnment commitment. Design limitations and weak donor coordination affected outcomes for Bank support for HIVIAIDS. Support to increase Pillar II: Expanding access to water and sanitation Service Delivery. was largely institutional and had narrow coverage. Quality concerns affected interventions in education, although access improved. While support for fiscal reforms Pillar Ill: Building was satisfactory, Bank support for Capacity and governance reforms (including the Accountability. reduction of corruption) was not sufficiently focused. There were weaknesses in capacity, as well 1 as in Government commitment. MS: Moderately Satisfactory  EG CASCR Review Independent EvaluationGroup For Official Use Only 9 Comments on Bank Performance: The strategic instrument for Bank assistance to Mozambique was the programmatic support through the PRSC. The shift to programmatic lending was underpinned by Bank's effort to improve coordination with other donors (developed in tandem with the Joint ReviewIPerformance Assessment Framework) and to facilitate A & H as required by the Paris declaration. The approach was rightly complemented by an investment portfolio that was aligned with the three pillars of the CAS. Wh~le the lending program was consistent with the base scenario outlined in the CAS, slippages in project preparation meant that most of the projects were implemented beginning the latter half of the CAS period. Slow start-up and implementation delays also led to frequent extension of closing date. Implementation was also constrained by design limitations (which affected projects such as the Public Sector Reform, the Coastal and marine Bio-Diversity Management), and problems with counterpart funding resulting from limitations on the Government's budgeting process. The quality of ESW was generally good, although it could have benefited from tighter scheduling, better and wider dissemination and better coordination with other donors. Efforts also needed to be made to ensure that analytic work was demand-driven. Technical assistance provided for strengthening capacity of the public sector in the preparation of the PARPA was in the right direction, although such support needed to be extended to the sectoral agencies and the provinces to aid in implementation. Cluster and Task-Team leaders based in Maputo provided just-in-time support to the Bank's program and interacted effectively with government counterparts. Deficiencies in the M&E system were widespread. The CASCR noted the less-than-satisfactory design of the results framework and failure on the part of government agencies to implement agreed M&E arrangements, which often made it difficult to assess progress in implementation during supervision. Towards the end of the CAS period, the Country Team had begun to pay more attention to this issue (as manifested in the revised Results Matrix) and improvements in supervision and increased attention to M&E took place. But the timiqg of the retrofitting of the results framework posed both conceptual and practical problems for evaluating the outcome of Bank assistance to Mozambique. The change in some indicators for key policy areas made it conceptually difficult to evaluate the first two years of program implementation. Furthermore, the evaluation of performance is constrained by new indicators whose benchmark periods were not clear. The retrofitting did not provide any clear advantage and instead weakened the M&E system. In general, the lack of realism in establishing outcome indicators accounts for the relatively poor evaluative performance of Bank Group assistance to the Mozambique PARPA. 4. Overall IEG Assessment The support of the Bank, along with the IMF and other donors, of prudent macroeconomic policies and the first wave of structural reforms in public finance management, privatization, decentralization, etc. have yielded good results in terms of strong growth and moderate inflation. But the support for structural reforms had mixed outcomes: Bank support for governance lacked precision and was not properly integrated with the justice, public finance, and procurement reforms. Bank support for M&E was also not comprehensive until towards the end of the CAS period. While capacity-building and TA were provided for the development of M&E systems under the PRSP, support for statistical capacity building at the agency was uneven. Coverage of vital registration system remained inadequate. As regards survey data on poverty and social indicators, these are in need of improvements. Finally, although good progress has been made in FPM, the challenges remain in executing donor-financed projects through the integrated financial management information system (eSISTAFE), which was expected to strengthen the use of country systems for the donor funded project. The participation of the Bank in this process was not clear, although Bank's policy framework provides for the use of  For Official Use Only EC CASCR Review Independent Evaluation Group 10 country systems and institutions where these are assessed to be adequate and able to provide fiduciary assurance on the use of Bank proceeds. 5. Assessment of CAS Completion Report The CASCR provides a comprehensive and clear assessment of the implementationand outcomes of the Mozambique country program during FY2004-07, although a more detailed treatment of the 2006 retrofitting of CAS goals, including spelling out the change in targets and indicators and the rationale for each modification, would have improved the report. IEG acknowledges the concern that including a detailed discussion on the retrofitting (which was partially done in the CASPR) would have increased the length of the CASCR. A better reference to the role played by other donors could have improved the discussion on the role played by the Bank Group and the outcome of Bank Group intervention. The CASCR could also have benefited from a better explanation of the deviation between the proposed lending and non-lending programs as contained in the CAS and the actual implementation. IFC had no input in the preparation of the CASCR even though the CAS was prepared jointly. Furthermore, the CASCR has no information on IFC's activities in the country pertaining to either investment operations or advisory services, except acknowledging IFC's support under Pillar I (improving the investment climate, and especially in the area of promoting SMEs, privatization of Petromoc, equity participation in a gas project, and in the South East African Tourism Investment) and reference in the concluding section for better collaboration between IDA, IFC and MIGA. 6. Findings and Lessons The CASCR identifies a number of lessons around three key themes: organization, project design, and project supervision and portfolio monitoring. Organization: It notes that better collaboration among IDA, IFC and MIGA, as well as improved internal coordination across sectors are important to support efforts to strengthen private sector development. Weaknesses in Bank Group coordination affected Bank support for the SME and PSD development, as well as the effort to mobilize other donors on issues such as HIVIAIDS. It is also important to build government capacity in dealing with the Bank Group's fiduciary requirements (financial management, p.rocurement and audits). Project Design: Improved appreciation of the country's political economy could help enhance project design, timing and implementation. Project design should take into consideration the limited implementation capacity within Government. This could also help ease the problems associated with politically risky projects by dealing with realistic policy reforms, performance targets, and implementation time-frames within the CAS results framework. Project Supervision and Portfolio Monitoring: It notes that recurring portfolio issues such as counterpart funding and VAT refunds should be dealt with upfront at project launch. Project teams need to assist sector agencies to develop M&E systems, which will facilitate project data collection and reporting. IEG agrees with the lessons drawn in the CASCR. Additionally, IEG notes that it is essential to pay close attention to issues of capacity-building and governance. The findings of the CASCR Review suggest the need to: Ensure that reforms of governance supported by the Bank Group apply to all agencies of government and are based on comprehensive training program of sensitization and capacity- building. It should bring together both judiciary (rule of law, appeals mechanism, etc.) and administrative (public finance management, procurement, etc.) aspects of governance in order to have a meaningful chance of success.  For Official Use Only EC CASCR Review Independent EvaluationGroup Substantially scale up capacity-building activities at all tiers of government in core competencies such as development planning, financial management, procurement and M&E, if service delivery and governance objectives of Government are to be realized. If capacity building efforts are restricted to only one or two tiers of government, the benefits of reform cannot pass through to all levels of government. Finally, strong Government commitment is important in politically sensitive reforms, such as reducing corruption and improving transparency. In these circumstances, Bank strategy and support should be built on programs developed by Government and promoted through a participatory process (that is, internally-owned). ESW can play an important role under these circumstances if topics are collaboratively selected, analytic work done jointly with Government and other development partners, and dissemination outlets/downstream engagement broadened.  Annexes EG CASCR Review Independent Evaluation Group 13 Annex Table 1: Actual vs. Planned Lending Annex Table 2: Analytical and Advisory Work: actual vs planned Annex Table 3: IEG Project Ratings Annex Table 4: Portfolio Status lndicators by Year Annex Table 5: IBRD IIDA Net Disbursements and Charges Annex Table 6: Net Aid Flows (net or gross, ODA or ODAtOOD, as relevant) Annex Table 7: Econorr~ic and Social lndicators Annex Table 8: Millennium Development Goals  EG CASCR Review Independent Evaluation Group Annexes 15 - Annex Table 1: Mozambique Planned and Actual Lending, FY04-07 (US$ million) Planned Approval Planned Approved Programmed Projects FY FY $ $ Decentralized Planning and Finance 2004 2004 42 42.0 S. Africa Power APL2 2004 slipped 13 NWDP Il -Supplemental Credit 2004 2004 15 15.0 PRSC I 2004 2005 60 60.0 Beira Railway SIL 2005 2005 70 110.0 Sustainable Rural Development (Tranfrontier Conservation Areas (TFCA) and Tourism Dev) 2005 2006 20 20.0 Financial Sector Capacity 2005 2006 10 10.5 Legal Sector Capacity 2005 dropped 5 PRSC 2 2005 2006 60 120.0 Roads and Bridges 2 APL 2006 2007 85 85.0 Technical and Vocational Education and Training 2006 2006 20 30.0 PRSC 3 DPL 2006 2007 70 70.0 Public Sector Reform 2 2007 dropped 20 PRSC 4 2007 slipped 70 Approval Planned Approved Non Programmed Projects FY $ $ Energy Reform and Access SIL 2004 40.3 Market Led Smallholder Development 2006 20.0 Maputo Municipal Development 2007 30.0 Market Led Smallholder Development (GEF) 2007 6.5 Total FY04-07 560 659.3 Regional projects HlVlAlDS TAP 2004 25.0 Sources: Mozambique CAS 2003 and WB Business Warehouse as of March 9th, 2007.  EG CASCR Review Independent Evaluation Group Annexes 16 Annex Table 2: Mozambique - Planned Non-Lending Program and Actual Deliveries, FY04-07 Programmed Planned FY Delivered FY Agriculture PSlA 2004 dropped Country Status Report on Health 2005 PER dropped CPAR dropped Legal and Judicial Assessment 2003 RurallAgriculture Strategy 2005 Private Sector Competitiveness 2005 Labor Markets and Tec. Voc. Ed 2004 Poverty Update dropped Institutional Governance Review 2005 PER 2005 CEM 2006 HIVIAIDS Retrospective Review dropped Water Management 2005 PER TA 2006 InfrastructureAssessment dropped PER dropped Pay reform PSlA dropped Actual (not included in the CAS) Planned FY Delivered FY Technical and Vocational Education 2004 Contract Farming and Supply Chain Financing 2005 Impacts of Extension Services 2005 PSlA Reducing Primary School Fees in Moz 2005 Natural Resources in Mozambique 2005 Marginal Budgeting through Bottlenecks TA 2006 Achieving the Health MDGs (Health Status Report) 2006 Public Financial Management Assessment (PEFA) 2006 Decentralization & Local Service Delivery Policy Note 2006 Mining Policy TA 2006 Country EnvironmentalAssessment/Country Social Assessment 2007 Value Chain Analyses 2007 Rural Strategy (Horticulture Development Sector Study) 2007 Sources: Mozambique CAS 2003, WB Business Warehouse, IRIS, and Integrated Controller's Systems, as of March loth, 2007.  EC CASCR Review Independent Evaluation Group Annexes 17 - Annex Table 3a: Mozambique Project Ratings, Exit FY04-07 Project Exit FY name Net Commitments IEG Outcome IEG Sustainability IEG ID Impact GEF Transborder MODERATELY 2004 Parks SIL 0.0 SAI'ISFACTORY UNLIKELY MODEST Health Sector Recovery SIL (FY96) 79.9 SATISFACTORY LIKELY SUBSTANTIAL MODERATELY EMPSO 134.2 SA'I'ISFACTORY LIKELY MODEST PRSC 60.0 SATISFACTORY LIKELY SUBSTANTIAL Natl Water 1 36.9 SATISFACTORY LIKELY SUBSTANTIAL PRSC 2 120.0 SKI'ISFACTORY LIKELY SUBSTANTIAL Edu Sec Strtgy Prgm MODERATELY ESSP TAL 69.8 SATISFACTORY NA NA - Annex Table 3b: Mozambique Portfolio Ratings, Exit FY04-07 lnst Dev Total Total Outcome lnst Dev Impact Evaluated Evaluated Outcome % Sat Impact % Subst Sustainability Sustainability ($MI (No) % Sat ($) (No) % Subst ($) (No) % Likely ($) % Likely (No) Mozambique 500.8 7 100.0 100.0 69 66.7 100.0 83.3 AFR 7,287.1 171 77.9 70.2 58 49.7 74.6 68.8 Bank wide 47,503.9 759 88.4 80.8 65 56.3 89.4 81.7 Source: WB Business Warehouse fables 4a.5 and 4a.6 as of March 9th, 2007.  EG CASCR Review Independent Evaluation Group Annexes 18 Annex Table 4: Mozambique- Portfolio Status Indicators, FY04-07 (US$ million) CountrylFiscal year 2004 2005 2006 2007 Mozambique # Proj Net Comm Amt # Proj At Risk % At Risk Comm At Risk % Commit at Risk Kenva # Proj Net Comm Amt # Proj At Risk %At Risk Comm At Risk % Commit at Risk Lesotho # Proj Net Comm Amt # Proj At Risk % At Risk Comm At Risk % Commit at Risk Malawi # Proj Net Comm Amt # Proj At Risk % At Risk Comm At Risk % Commit at Risk Tanzania # Proj Net Comm Amt # Proj At Risk % At Risk Comm At Risk % Commit at Risk Uganda # Proj Net Comm Amt # Proj At Risk % At Risk Comm At Risk % Commit at Risk  EG CASCR Review Independent Evaluation Group Annexes 19 Annex Table 4: Mozambique- Portfolio Status Indicators, FY04-07 (US$ million) (continued) CountrylFiscal year Zambia # Proj Net Comm Amt # Proj At Risk %At Risk Comm At Risk % Commit at Risk JF ' & # Proj Net Comm Amt # Proj At Risk %At Risk Comm At Risk % Commit at Risk Bank wide # Proj Net Comm Amt # Proj At Risk % At Risk Comm At Risk % Commit at Risk Source: WB Business Warehouse Table 3a.4 as of March 9th, 2007.  Annexes EG CASCR Review Independent Evaluation Group 20 - Annex Table 5: Mozambique IBRDllDA Net Disbursements and Charges, FY04-07 (million US$) FY Disb. Amt. Repay Amt. Net Amt. Charges Fees Net Transfer 2004 183.5 9.2 174.3 8.5 1.4 164.4 Total 954.7 38.1 916.7 33.6 8.3 874.7 Source: Client Connection as of April 23T 2007  Annexes EG CASCR Review Independent Evaluation Group 21 Annex Table 6: Total Net Disbursements of Official Development Assistance and Official Aid to Mozambique 2004-07, (US $ Million) Donor Calender years 2003 2004 2005 Total Australia 4.59 2.02 0.69 7.3 Austria 3.35 5.17 4.31 12.83 Belgium 8.71 10.57 12.21 31.49 Canada 26.7 27.34 56.19 110.23 Czech Republic 0.01 0.01 0.01 0.03 Denmark 66.43 67.42 64.87 198.72 Finland 21.99 25.66 24.75 72.4 France 16.59 14.61 13.71 44.91 Germany 37.91 38.65 42.64 119.2 Iceland 1.02 1.24 1.56 3.82 Ireland 39.9 48.69 ' 48.31 136.9 Italy 15.09 26.99 21.56 63.64 Japan 35.27 19.41 14.77 69.45 Korea 0.1 0.05 0.16 0.31 Luxembourg 0.73 0.35 1.35 2.43 Netherlands 47.27 54.7 64.46 166.43 New Zealand 0.61 0.43 0.48 1.52 Norway 54.11 61.06 67.94 183.11 Portugal 19.11 24.25 22.56 65.92 Slovak Republic 0.18 0.2 0.1 0.48 Spain 22.58 32.48 29.35 84.41 Sweden 56.53 67.92 79.25 203.7 Switzerland 20.82 27.65 24.61 73.08 United Kingdom 63.37 65.92 80.84 210.13 United States 135.4 109.96 95.97 341.33 All bilateral donors 2701.37 2736.75 2777.65 8215.77 AfDF (African Dev.Fund) 31.94 91.39 73.42 196.75 Arab Agencies 11.63 26.45 13.1 1 51.19 Arab Countries 1.58 2.34 3.92 EC 90.17 151.1 162.57 403.84 GEF 0.95 0.89 1.58 3.42 Global Fund (GFATM) 16.38 16.38 IDA 159.05 194.22 242.7 595.97 IFAD 6.02 7.01 6.24 19.27 SAF+ESAF+PRGF(IMF) -8.95 -20.31 -25.46 -54.72 UNDP 8.91 8.51 7.41 24.83 UNFPA 9.04 8.96 5.89 23.89 UNHCR 1.54 2.2 2.44 6.18 UNICEF 7.8 8.51 8.73 25.04 UNTA 2.88 2.44 2.82 8.14 Nordic Dev. Fund 7.04 7.77 5.42 20.23 WFP 8.77 5.21 6.38 20.36  EG CASCR Review Independent Evaluation Group Annexes 22 Annex Table 6: Total Net Disbursements of Official Development Assistance and Official Aid to Mozambique 2004-07, (US $ Million) (continued) Donor Calender years 2003 2004 2005 Total DAC Countries, Total 697.06 731.25 770.82 2199.13 DAC EU Members, Total 419.56 483.38 510.17 1413.11 Non-DAC Bilateral Donors,Total 2.89 3.84 1.83 8.56 G7, Total 330.33 302.88 325.68 958.89 Multilateral, Total 336.79 510.73 513.25 1360.77 All Donors,Total 1036.74 1245.82 1285.9 3568.46 Source: OECD DAC Online database, Table 2a. Destinationof OfficialDevelopment Assisfance and Official Aid Disbursements as of March 19, 2007  Annexes EG CASCR Review Independent Evaluation Group 23 - Annex Table 7: Mozambique Economic and Social Indicators Low Mozambique Malawi Uganda Zambia SSA income 2001 2002 2003 2004 2005 Average 2001-2005 Growth and Inflation GDP growth (annual %) 13.1 8.2 7.9 7.5 7.7 2.7 5.4 4.8 4.3 6.0 GDP per capita growth (annual %) 10.7 6.0 5.8 5.4 5.7 1 0.4 1.9 3.0 2.0 4.1 GNI per capita, Atlas method (current US$) GNI per capita, PPP (current international $) Inflation, consumer prices (annual %) Composition of GDP Agriculture, value added (% of GDP) Industry, value added (% of GDP) 27.8 26.3 28.1 29.2 30.0 1 17.1 21.O 26.2 30.9 27.1 Services, etc., value added (% of GDP) External Accounts Exports of goods and sewices (% of GDP) Imports of goods and services (% of GDP) Current account balance (% of GDP) Total debt service (% of exports of goods, services and income) External debt (% of GNI) Gross domestic savings (% of GDP) Fiscal Accounts, per FY Total Revenue, before grants (% of GDP) Total Expenditure (incl. adj. to cash, %of GDP) Overall Budget Balance, before grants (% of GDP) Overall Budget Balance, afler grants (% of GDP)  Annexes EG CASCR Review Independent Evaluation Group 24 - Annex Table 7: Mozambique Economic and Social Indicators (continued) Low Mozambique Malawi Uganda Zambia SSA income 2001 2002 2003 2004 2005 Average 2001-2005 Social Indicators 11 Health Immunization, DPT (% of children ages 12-23 months) 70.0 72.0 72.0 72.0 81.8 75.3 80.0 57.9 63.8 Improved sanitation facilities (% of population with access) 32.0 61.O 43.0 55.0 37.0 38.3 Improved water source (% of population with access) 43.0 73.0 60.0 58.0 56.2 75.1 Life expectancy at birth, total (years) 41.9 41.9 41.8 39.9 47.8 37.7 46.0 58.6 Mortality rate, infant (per 1,000 live births) .. 104.4 .. 109.8 80.2 102.0 100.5 79.5 Education School enrollment, preprimary (% gross) 3.4 15.0 25.0 School enrollment, primary (% gross) 80.5 83.9 94.9 .. 133.5 128.1 84.0 90.4 97.2 School enrollment, secondary (% gross) 6.8 8.4 10.8 31.I 17.6 25.2 32.4 44.1 Population Population growth (annual %) 2.1 2.1 2.0 1.9 1.9 2.3 3.4 1.7 2.3 1.9 Population, total (million) 18.3 18.7 19.1 19.4 19.8 12.3 26.9 11.3 710.2 2,270.1 Urban population (% of total) 31.5 32.2 33.0 33.7 34.5 1 16.4 12.4 34.9 34.3 29.3 Note: Some data for recent years are stille stimates Source: WB World Development Indicators (March 2007 update) for all indicators I 1Some of these indicators are not available on an annual basis, so some averages are based on fewer observations.  EG CASCR ReviewIndependent Evaluation Group Annexes 25 Annex Table 8: Mozambique Millennium Development Goals 1990 1995 1998 2001 2004 2005 Goal 1: Eradicate extreme poverty and hunger Income share held by lowest 20% 6 .. Malnutrition prevalence, weight for age (% of children under 5) .. 27 26 26 24 Poverty gap at $1 a day (PPP) (%) .. 12 12 .. Poverty headcount ratio at $1 a day (PPP) (% of population) .. 38 38 .. Poverty headcount ratio at national poverty line (% of population) 69 .. Prevalence of undernourishment (% of population) 58 .. 45 Goal 2: Achieve universal primary education Literacy rate, youth total (% of people ages 15-24) 49 .. Persistenceto grade 5, total (% of cohort) 34 .. 46 49 Primary completion rate, total (% of relevant age group) 27.1 .. 13.4 18.9 29 29 School enrollment, primary (% net) 43 .. 49 60 71 71 Goal 3: Promote gender equality and empower women Proportion of seats held by women in national parliament (%) 16 .. 25 30 30 34.8 Ratio of girls to boys in primary and secondary education (%) 71.6 .. 74 76.8 82.3 82.3 Ratio of young literate females to males (% ages 15-24) 47.9 .. Share of women employed in the nonagriculturalsector (% of total nonagricultural employment) 11 .. Goal 4: Reduce child mortality Immunization, measles (% of children ages 12-23 months) 59 71 64 74 77 77 Mortality rate, infant (per 1,000 live births) 158 145 .. 122 104 104 Mortality rate, under-5 (per 1,000) 235 212 .. 178 152 152 Goal 5: Improve maternal health Births attended by skilled health staff (% of total) .. 44.2 .. 48 Maternal mortality ratio (modeled estimate, per 100,000 live births) .. 1000 Goal 6: Combat HIVIAIDS, malaria, and other diseases Children orphaned by HIVIAIDS .. 510000 510000 Contraceptive prevalence (% of women ages 15-49) 6 .. 17 Incidenceof tuberculosis (per 100,000 people) 167.1 .. 460.2 460.2 Prevalence of HIV, female (% ages 15-24) I1 II Prevalence of HIV, total (% of population ages 15-49) 16 16 Tuberculosis cases detected under DOTS (%) .. 53.8 47 43.9 45.9 45.9 Goal 7: Ensure environmental sustainability C02 emissions (metric tons per capita) 0.1 0.1 0.1 0.1 Forest area (% of land area) 26 .. .. 25 25 25 GDP per unit of energy use (constant 2000 PPP $ per kg of oil equivalent) 1 2 2 2 2 Improved sanitation facilities (% of population with access) 20 .. 32 32 Improved water source (% of population with access) 36 .. 43 43 Nationally protected areas (% of total land area) 8.4 8.4 Goal 8: Develop a global partnership for development Aid per capita (current US$) 74.7 67.1 60.7 51 63.2 63.2 Debt service (PPG and IMF only, % of exports of G&S, excl. workers' remittances) 17 33 41 3 3 3 Fixed line and mobile phone subscribers (per 1,000 people) 3.5 3.8 4.8 13.2 26.9 Internet users (per 1,000 people) 0 0 0.2 1.6 7.1 7.1 Personal computers (per 1,000 people) .. 0.9 2.3 3.8 5.8 5.8 Total debt service (% of exports of goods, services and income) 26 35 18 8 5 5  EG CASCR Review I n d e p e n d e n t Evaluation G r o u p Annexes 26 Unemployment, youth female (% of female labor force ages 15-24) Annex Table 8: Mozambique Millennium Development Goals (continued) 1990 1995 1998 2001 2004 2005 Unemployment, youth male (% of male labor force ages 15-24) Unemployment, youth total (% of total labor force ages 15-24) Other Fertility rate, total (births per woman) 6.2 6 5.9 5.5 5.4 5.4 GNI per capita, Atlas method (current US$) 170 140 200 210 270 310.0 GNI, Atlas method (current US$) (billions) 2.3 2.2 3.5 3.8 5.2 6.1 Gross capital formation (% of GDP) 22.1 30.5 24.2 25.9 20.7 22.2 Life expectancy at birth, total (years) 43.2 43.7 43.8 41.9 41.8 41.8 Literacy rate, adult total (% of people ages 15 and above) 33.5 .. Population, total (millions) 13.4 15.9 17.1 18.3 19.4 19.8 Trade (% of GDP) 44.2 56.6 40.3 72.2 70.2 71.7 Source: World Development Indicatorsdatabase, September 2006 Note: Figures in italics refer to periods other than those specified.