Document of The World Bank Report No.: 84667-MZ PROJECT PERFORMANCE ASSESSMENT REPORT THE REPUBLIC OF MOZAMBIQUE DECENTRALIZED PLANNING AND FINANCE PROJECT (IDA-HO670-MOZ) February 18, 2014 IEG Public Sector Evaluation Independent Evaluation Group ÿþii Currency Equivalents (annual averages) Currency Unit = Mozambican metical (MZM) 2004 US$1.00 MZN 22,144.71 2005 US$1.00 MZN 22,850.81 2006 US$1.00 MZN 25,758.32 (January to July) 2006 US$1.00 MZN 25.89 (July to December) 2007 US$1.00 MZN 25.79 2008 US$1.00 MZN 24.19 2009 US$1.00 MZN 27.58 2010 US$1.00 MZN 34.24 Abbreviations and Acronyms CCAGG Concerned Citizens of Abra for Good Government CPIA Country Program and Institutional Assessments DGA Development Grant Agreement DPFP Decentralized Planning and Finance Project FCA Fundo de Compensação Autárquica FIL Fundo de Iniciativa Local FRELIMO Frente de Libertação de Moçambique GTZ German Technical Cooperation ICR Implementation Completion Report IDA International Development Association IEG Independent Evaluation Group IEGPS IEG Public Sector Evaluation M&E Monitoring and Evaluation MOPH Ministry of Public Works and Housing MOZ Mozambique NDPFP National Decentralized Planning and Finance Project PAD Project Appraisal document PARPA Action Plan for Reduction of Absolute Poverty PEDD Plano Estratégico de Desenvolvimento Distrital PEFA Public Expenditure and Financial Accountability PESOD Plano Económico Social e Orçamento Distrital PPAR Project Performance Assessment Report PRSCs Poverty Reduction Strategy Credits PSRP Public Sector Reform Project UNCDF United Nations Capital Development Fund UNDP United Nations Development Program UNICEF United Nations Children's Fund Fiscal Year Government: January I - December 31 Acting Director-General, Independent Evaluation Mr. Richard Scobey Director, IEG Public Sector Evaluation Mr. Emmanuel Jimenez Manager, IEG Public Sector Evaluation Mr. Mark W. Sundberg Task Manager Ms. Lourdes Pagaran 111 Contents Principal Ratings ..........................v....... ................v The Republic of Mozambique Decentralized Planning and finance Project (P00 1807) .... v Key Staff Responsible..............................................v Preface........................................................ vii Summary ....................................................... ix 1. Background and Context.............................. ..................1 2. Objectives, Design, and their Relevance .................................2 Objectives.....................................................2 Relevance of objectives ............................... ..................2 Design.......................................................3 Components..................................................3 Design of Monitoring and Evaluation................. ...............4 Implementation Arrangements.................... .................5 Relevance of Design..............................................5 3. Implementation.................................................6 Planned vs. Actual Costs by Component..............................6 Implementation Experience.......................................7 Implementation of Monitoring and Evaluation.............. ............8 Safeguards Compliance ..........................................8 4. Achievement of the Objectives.......................................8 Outputs..................................................... 9 Outcome................................................... 16 5. Efficiency.................................................... 17 6. Ratings...................................................... 18 Outcome..................................................... 18 Risk to Development Outcome...................................... 18 Bank Performance............................................... 19 Borrower Performance........................................... 21 Monitoring and Evaluation........................................ 22 7. Lessons...................................................... 23 References...................................................... 25 Annex A. Basic Data Sheet.......................................... 27 Annex B. Mozambique Decentralization timeline .................. ........ 31 iv Annex C. Poverty in Mozambique............ ................... ..... 39 Annex D: Elements of Intergovernmental design supported by DPFP ... ............. 41 Annex E. List of Persons Interviewed .......................... ......... 44 Boxes Box 1. Decentralization Model ........................................... 6 Tables Table 1. Recurrent Expenses by Jurisdiction, 2011-2012 (in Millions of Meticais) ........ 11 Table 2. Execution of Small-Scale Infrastructure Projects .......... ........... 12 This report was prepared by Clay Wescott, who assessed the project in July 2012. The report was peer reviewed by Brett Liebresco, and panel reviewed by Michael Lay. Yezena Yimer provided administrative support. V Principal Ratings The Republic of Mozambique Decentralized Planning and finance Project (P001807) ICR* ICR Review* PPAR Outcome Satisfactory Satisfactory Satisfactory Risk to Development Low or Negligible Negligible to Low Modest Outcome Bank Performance Satisfactory Satisfactory Satisfactory Borrower Peforme Satisfactory Moderately Satisfactory Moderately Satisfactory Performance * The Implementation Completion Report Results (ICRR) is a self-evaluation by the responsible Bank department. The ICR Review is an intermediate IEG product that seeks to independently verify the findings of the ICR. Key Staff Responsible Division Chief/ Project Task Manager/Leader Sector Director Country Director Appraisal Lance Morrell Jaime M. Biderman Darius Mans Completion Ali Alwahti Jaime M. Biderman Peter Nicolas Vi IEG Mission: Improving World Bank Group development results through excellence in independent evaluation. About this Report The Independent Evaluation Group assesses the programs and activities of the World Bank for two purposes: first, to ensure the integrity of the Bank's self-evaluation process and to verify that the Bank's work is producing the expected results, and second, to help develop improved directions, policies, and procedures through the dissemination of lessons drawn from experience. As part of this work, IEG annually assesses 20-25 percent of the Bank's lending operations through field work. In selecting operations for assessment, preference is given to those that are innovative, large, or complex; those that are relevant to upcoming studies or country evaluations; those for which Executive Directors or Bank management have requested assessments; and those that are likely to generate important lessons. To prepare a Project Performance Assessment Report (PPAR), IEG staff examine project files and other documents, visit the borrowing country to discuss the operation with the government, and other in-country stakeholders, and interview Bank staff and other donor agency staff both at headquarters and in local offices as appropriate. Each PPAR is subject to internal IEG peer review, Panel review, and management approval. Once cleared internally, the PPAR is commented on by the responsible Bank department. The PPAR is also sent to the borrower for review. IEG incorporates both Bank and borrower comments as appropriate, and the borrowers' comments are attached to the document that is sent to the Bank's Board of Executive Directors. After an assessment report has been sent to the Board, it is disclosed to the public. About the IEG Rating System for Public Sector Evaluations IEG's use of multiple evaluation methods offers both rigor and a necessary level of flexibility to adapt to lending instrument, project design, or sectoral approach. IEG evaluators all apply the same basic method to arrive at their project ratings. Following is the definition and rating scale used for each evaluation criterion (additional information is available on the IEG website: http://worldbank.org/ieg). Outcome: The extent to which the operation's major relevant objectives were achieved, or are expected to be achieved, efficiently. The rating has three dimensions: relevance, efficacy, and efficiency. Relevance includes relevance of objectives and relevance of design. Relevance of objectives is the extent to which the project's objectives are consistent with the country's current development priorities and with current Bank country and sectoral assistance strategies and corporate goals (expressed in Poverty Reduction Strategy Papers, Country Assistance Strategies, Sector Strategy Papers, Operational Policies). Relevance of design is the extent to which the project's design is consistent with the stated objectives. Efficacy is the extent to which the project's objectives were achieved, or are expected to be achieved, taking into account their relative importance. Efficiency is the extent to which the project achieved, or is expected to achieve, a return higher than the opportunity cost of capital and benefits at least cost compared to alternatives. The efficiency dimension generally is not applied to adjustment operations. Possible ratings for Outcome: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. Risk to Development Outcome: The risk, at the time of evaluation, that development outcomes (or expected outcomes) will not be maintained (or realized). Possible ratings for Risk to Development Outcome: High, Significant, Moderate, Negligible to Low, Not Evaluable. Bank Performance: The extent to which services provided by the Bank ensured quality at entry of the operation and supported effective implementation through appropriate supervision (including ensuring adequate transition arrangements for regular operation of supported activities after loan/credit closing, toward the achievement of development outcomes. The rating has two dimensions: quality at entry and quality of supervision. Possible ratings for Bank Performance: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. Borrower Performance: The extent to which the borrower (including the government and implementing agency or agencies) ensured quality of preparation and implementation, and complied with covenants and agreements, toward the achievement of development outcomes. The rating has two dimensions: government performance and implementing agency(ies) performance. Possible ratings for Borrower Performance: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. vii Preface This Project Performance Assessment Report (PPAR) covers the Decentralized Planning and Financing Project. The objective was to improve the institutional performance of Mozambique's District Administrations to plan and manage small infrastructure investments in response to community demands. Total actual cost was $48.68 million, of which $45.17 million was financed by IDA Grant No. H067 MOZ, and $3.51 million by the Government. The grant was approved on November 10, 2003, became effective on February 16, 2004, and was closed on March 31, 2009, 9 months after the original closing date. This report was prepared by Clay Wescott, IEG consultant. The report presents findings based on review of the Project Appraisal Document, the Implementation Completion and Results Reports, aides-memoire and supervision reports, and other relevant materials. An IEG mission visited Mozambique July 8 - 28, 2012 to interview government officials, the staff of non-governmental organizations, project staff, donor representatives, and other stakeholders (see Annex E for list of persons interviewed). The mission visited and interviewed officials in Tete and Sofala Provinces; some of these officials had worked previously in Manica and Zambezia Provinces, which were also supported by the project. Bank staff members, donor representatives, and other informants were also interviewed at headquarters and by telephone and videoconference. This work was carried out jointly with a PPAR of another operation, the Public Sector Reform Project. The assessment aims, first, to serve an accountability purpose by verifying whether the operations achieved their intended outcomes. Second, the report draws lessons that are intended to inform future operations of this nature in Mozambique and other low-income, post-conflict states. Following standard IEG procedures, a copy of the draft Project Performance Assessment Report (PPAR) was sent to the Borrower for comments. No comments received from the Borrower.  ix Summary The objective of the Decentralized Planning and Finance Project (DPFP) was "to improve the institutional performance of its District Administrations to plan and manage small infrastructure investments in response to community demands." At the time of appraisal, an estimated 54 percent of the population lived in absolute poverty. A project objective that promised to improve local government capacity to provide essential infrastructure was substantially relevant to addressing the economic challenges in some of Mozambique's poorest districts. The design was also substantially relevant. It built on the lessons of a pilot project supported by the United Nation's Capital Development Fund (UNCDF), with funding from Netherlands and Norway, in Nampula Province, which was in turn built on the lessons of comparable work in Uganda. The project also built on the experience of Gesellschaft fUr Technische Zusammenarbeit (German Technical Cooperation-GTZ) funded activities in Sofala (since 1998), Manica (since 1997) and Inhambane (since 2002, in coordination with Development Cooperation Ireland) in a total of 18 districts. The Bank and other partners financed a parallel program of support for financial and fiscal management improvement, which was delivering gradually improved results as measured by Public Expenditure and Financial Accountability (PEFA) assessments and Country Program and Institutional Assessments (CPIA). There were also sector-wide assistance programs inter alia in health, education and agriculture. The DPFP benefited from both streams of work. The approach was to support infrastructure in areas where sectors were already working; that is, provide a school house where teachers were already conducting classes without a proper building. The achievement of objective was substantial. There are two aspects of the objective: improving local government resource management capacity, and doing so in a participatory and transparent manner. On the first aspect, provincial budget transfers to districts were established, and some financial statements were audited. Annual performance evaluations of implementation of local investment grants found that the target of good or excellent performance was exceeded in many cases, thanks to the training of local financial staff and opportunities for learning by doing. The project helped provide prior conditions for, though did not directly support, a new intergovernmental transfer system for unconditional block grants based on defined criteria, which provided added opportunities for the learning-by-doing approach to capacity development. On the second aspect, the project did well to help set up an institutional framework for transparent district planning and management. This process of community participation within a deconcentrated framework was a politically feasible step towards improved accountability. All districts were covered, with broad representation. Key administrative and fiscal instruments combined with planning and monitoring by District Consultative Councils helped to support transparency and accountability. Survey data and the IEG field mission found evidence that citizens were satisfied with the participatory process, and the accountability it provided. One overall issue is that Government did not prepare a decentralization policy framework as it had committed to do under DPFP, the Public Sector Reform Project (PSRP), and the Poverty Reduction Strategy Credits (PRSCs), which supported a related decentralization strategy. While many policy papers were x prepared, and donors agreed on an overall aid framework for decentralization, the decentralization policy framework and strategy have not yet been adopted. Another overall issue is that over the course of project implementation, poverty was reduced only in Tete, one of the four provinces targeted, and in that case, the reason was mainly due to overall economic trends driven by rapid expansion of coal production. However, the project was not designed to reduce poverty by itself. Reducing poverty will require a much better resourced government effort in future years. The risk to the institutional outcomes achieved is rated modest. Key elements of sub- national governance are laid out in the Constitution, enabling legislation and decrees. Government spending is increasing both on sub-national operating and investment costs. The financing of private investments outside the district planning process, while not supported by the project, is a risk for patronage and other forms of corruption. The whole point of a participatory district planning process was to generate accountability through community input, and so this exception, while declining in share over time, could undermine achievement of the objective. Bank performance is rated as satisfactory. The design was aligned with the Government's Action Plan for the Reduction of Absolute Poverty (PARPA) and the Bank's Country Assistance Strategy, and remains relevant to the latest Country Partnership Strategy. The design draws on lessons from the Bank and other donors in Mozambique and neighboring countries. The design also draws appropriately on analytical work. The selection of the targeted districts reflected a reasonable division of labor among donors. There was extensive supervision, and no major fiduciary or safeguards issues. Borrower performance is rated moderately satisfactory. The Government adopted enabling legislation, decrees, guidelines and strategies laying out its commitment to the project objective. The Government instituted a fiscal transfer system mid-way through project implementation, modeled on the system piloted by the project, validated by annual performance evaluations, and building on the improved subnational capacity that the project contributed to. However, the shift to supporting "productive investments", while not supported by the project, is at odds with the objective of community-based accountability, and raises concerns over the Government's commitment to the decentralization program. Other moderate shortcomings included: inability to produce a decentralization policy as had been agreed, delays in providing counterpart funds, delays in full compliance with environmental and social safeguard procedures, and lack of clarity on criteria for selection of private investment projects for funding. The Ministry of Planning and Development constituted a DPFP National Steering Committee to oversee the project. The Ministry motivated and guided provincial directors of planning in achieving project objectives. The monitoring framework was written up during preparation in a monitoring and evaluation (M&E) manual that provided clear instructions for use by provincial and district administrations. M&E implementation was supported by project financed M&E specialists to compensate for lack of qualified staff in local administrations. They were hired by the district administrations following project closure to ensure sustainability of the M&E function. M&E information was used to address shortcomings in the quality of district plans and underlying participatory processes. xi The lessons from this operation are: * For countries at an early stage in the decentralization process, a learning-by-doing approach can be effective where the focus is on solving locally nominated and defined problems, encouraging experimentation and experiential learning, and engaging a broad set of agents so that development efforts are viable, legitimate, relevant and supportable. * In political settings without democratically elected subnational authorities, a process of controlled/limited participation within a deconcentrated framework can be politically feasible, and help to ensure accountability and efficient use of resources. * While managing small scale, public infrastructure investments is useful for building subnational administrative capacity, managing lines of credit that are poorly regulated and vulnerable to manipulation can weaken legitimacy and the good will built up through participatory processes. Richard Scobey Acting Director-General Evaluation  1 1. Background and Context 1.1 Mozambique's first Constitution defines a unitary state: "which respects the principles of autonomy of local authorities (autarquias locais) in its organization. The Provincial Government is the body charged with ensuring the implementation, at provincial level, of centrally defined Government policies, and it shall exercise administrative supervision over local authorities, in accordance with the law."(Republic of Mozambique, 2004). 1.2 Following the 1992 Peace Accord and the first multiparty elections in 1994, the government adopted enabling legislation that heightened the importance of subnational entities, while selectively providing more citizen participation and autonomy. These initiatives recognized that the country couldn't be governed solely from Maputo. However, unlike Mozambique's neighbors that mainly had British colonial legacies, there was not a tradition of local government in Mozambique; thus, the country has been behind its neighbors in adopting elements of decentralized governance. 1.3 The elections of 1999 allowed for the first time elected municipal governments as part of the reform to provide representative government and to promote the decentralization of political authority. These covered 23 cities and 10 of 128 district towns judged to have sufficient capacity to execute their new responsibilities. In 2008, this was expanded to include 43 cities and district towns. For other district-level entities, the central government would retain control of the planning process through appointed district administrators until sufficient capacity was built. Provincial assemblies were elected starting in 2009. Provincial assemblies approve the provincial government's program, and supervise and monitor compliance with it (Republic of Mozambique, 2004)1. Community Authority decrees were adopted in 2000, and Local Administration Reform legislation in 2003. 1.4 Citizen participation and local autonomy received a boost in the Government's National Development Strategy or Poverty Reduction Strategy Paper: Action Plan for the Reduction of Absolute Poverty (PARPA) 2001-05, which supported decentralized decision making authority for planning and finance to provide basic infrastructure and services to accelerate poverty reduction, while reducing costs; to increase transparency, accountability and quality of planning and execution of investment projects; and to raise the efficiency and effectiveness of public service (Republic of Mozambique, 2001). In addition, the Government's closely linked 2001 Global Strategy for Public Sector Reform included measures to shift responsibilities of central authorities from managing to facilitating service delivery by provincial and district authorities. The following review will show that the efforts of Government and other donors continued through 2012 to deepen and expand support to strengthening district administration and district-level citizen participation. 1.5 These policy initiatives were likely spurred on initially by two key factors. First, Frelimo won the 1999 election by a very close margin over the second place party: Renamo won 48 percent of the presidential vote, and 39 percent of the parliamentary vote (African 1 While provincial governments were provided for in the 1990 Constitution, elected Provincial Assemblies were only introduced in the subsequent 2004 Constitition. 2 Elections Database, undated). This gave the Frelimo Party's reform wing credibility in calling for improving public services through deepening decentralization, since it would be popular with voters and improve Frelimo's chances in future elections. Second, decentralization was popular with Mozambique's donors. Between 2002-2003 there was a sharp decline in aid disbursements, from $2.2 billion to $1.0 billion. While 2002 saw unusually large aid flows, aid in 2003 was a bit less than in 1995, in part because Mozambique was losing its "post conflict" aid premium. As a result, the government wanted to seem to be supporting what donors wanted in order to get more aid (De Renzio and Hanlon, 2007). In summary, there was support for deepening decentralization led by the Frelimo reform wing and their supporters among civil servants and civil society, and reinforced by a desire by the Government to align with a policy advocated by donors, to help facilitate increased aid flows. 1.6 In the early 2000s, there were two views on decentralization and participatory development, as spelled out in World Development Report 2000/2001 devoted to poverty alleviation. One view supported community driven development, one of the foundations of the Comprehensive Development Framework. The other advocated the devolution of authority to legally constituted and accountable local government authorities (World Bank 2000, Mansuri and Rao 2004). The Bank's response to the Government's heightened interest in decentralization was to launch a new operation in 2003: the Decentralized Planning and Finance Project (DPFP), combining the notions of devolution of authority to legally constituted local government authorities, with accountability and priority setting supported by community-based participatory processes. This PPAR will analyze the extent to which decentralization and community participation have gained traction in Mozambique since 2003, and the extent to which this project contributed to this. 2. Objectives, Design, and their Relevance Objectives 2.1 The objective as stated in the Project Appraisal Document (PAD) was "to improve the capacity of local government to manage public financial resources for district development in a participatory and transparent manner." (World Bank, 2003a: 7). It was differently formulated in the Development Grant Agreement (DGA): "to improve the institutional performance of its District Administrations to plan and manage small infrastructure investments in response to community demands" (World Bank, 2003b: 14). This Review will use the latter formulation, because it is more specific and monitorable. Relevance of objectives 2.2 High. A project objective that promised to improve institutional performance to provide essential infrastructure was highly relevant to building up the capability of district administrations as the legitimate face of government in some of Mozambique's poorest districts. Although there was a legal framework for elected local municipal government, Mozambique's 128 rural districts were run by deconcentrated entities linked to 10 centrally appointed provincial governments. The objective is aligned with the Government's 3 commitment to greater citizen participation in providing poverty-reducing infrastructure and public services at the local level through the Poverty Reduction Strategy Paper (PARPA), and the 2001 Global Strategy for Public Sector Reform (World Bank 2003a: 5. It was aligned with the Bank's Country Assistance Strategy (2003 - 6) at appraisal, in particular its pillars on building capacity and accountability, and on expanding service delivery. It also continues to be aligned with the Bank's Country Partnership Strategy's (2012-15) pillar on governance and public sector capacity, supported by a follow-on operation: National Decentralized Planning and Finance Project, approved in 2010. 2.3 There was a clear political drive to push more resources to the sub-national level, but no policy document that formally stated what goals the Government intended to achieve; thus the political economy drivers need to be inferred from available documents and interviews with key players. The political impetus from the 1999 election results, and the need to address concerns of donors, have been discussed above. In addition, other political factors affected decentralization: the concern for national unity (having recovered from a civil war), and the heritage of Frelimo with a strong emphasis on popular participation. Ethnic/regional identity was also an issue, that is, coming from the south, center or north, is managed carefully through process of appointing governors and district administrators (Thomsen and Saide 2011: 15). Matrilineal structures in some areas are another feature calling for a contextualized approach. By supporting controlled and limited participation within a deconcentrated framework, the project was aligned with these political economy concerns. It should be acknowledged that the amount of resources being pushed to the sub-national level is far less than needed to achieve the ambitious targets of the Governments Development Strategy (PARPA), as evidenced by the modest success in reducing poverty since 2003 (see Annex C). However, given the political context, it was reasonable to focus at this stage on building an institutional framework that could accommodate larger resource transfers and democratically elected authorities when and if they became economically and politically feasible. Design COMPONENTS 2.4 The project had the following five components: Participatory District Planning (Appraisal cost: US$8.76 million, Actual: US$9.94 million). This component was to provide support to district administrations to establish and operationalize District Planning Teams and to prepare strategic and annual plans. To achieve this, the operation supported development of district planning manuals, training of Provincial Planning Support Teams, and of District Technical Planning Teams, seminars for local administration personnel including District Administrators, support to District Administrators in informing District Consultative Councils on the progress of district plan implementation, and annual workshops in each province for sharing of experience. The component would also develop capacity of civil society and rural communities to effectively participate in planning and ensure local administration accountability. To achieve this, the operation supported dissemination of guidelines for the functioning of District 4 Consultative Councils, contracting and training of district-based civil society facilitators to work with the Councils to help them understand their role, and development of a provincial participation networks to provide capacity building and exchange experience. Lastly, the component would support the development and implementation of a communications strategy. * Local Investment Grants (Appraisal cost: US$19.24 million, Actual: US$19.96 million). This component was to provide grants for implementation of small rural infrastructure investments. Three grant types were envisioned. Type I was to finance investments in district administrative infrastructure, managed by the province. Type 2 grants would finance investments in small-scale public social infrastructure prioritized by communities, co-managed at the province and district level. Type 3 grants would provide small discretionary funds to District Administrations for direct contracting for micro-projects. * Capacity Building for Local Administration (Appraisal cost: US$8.52 million, Actual: US$7.84 million). This component aimed to strengthen the training system to increase the capacity of local administration officials to undertake key local government functions more effectively (on the provincial and district level). Activities were likely to include training in basic public administration, planning and budgeting, and selected technical subjects. * Support for Decentralization Policy Reform (Appraisal cost: US$1.51 million, Actual: US$2.98 million). This component was to finance (demand-driven) technical assistance to the Government of Mozambique to improve decision-making through policy analysis and for the further development of the institutional and policy framework for decentralization. * Project Coordination (Appraisal cost: US$8.26 million, Actual: US$7.96 million). This component would consist of a small core team of project coordination staff at central and provincial level who would fulfill some World Bank-specific fiduciary functions and provide technical assistance to the government officials responsible for project implementation and the Government's fiduciary functions. This component would also contribute to the development of a common decentralization Monitoring and Evaluation (M&E) framework, and project resources under this component could also be used for the preparation of a future national decentralization program. Two safeguard policies were triggered by the Category B project: environmental assessment and involuntary resettlement. To ensure compliance, an environmental assessment and resettlement policy framework were prepared. DESIGN OF MONITORING AND EVALUATION 2.5 There was a clear monitoring framework documented in a comprehensive M&E manual. The project design is consistent with a reasonable theory of change that actions supported by the operation, and aligned with recommendations from political economy diagnosis (Serrano, 2002) commissioned to inform the participation and downward accountability processes, along with complementary support from the Bank and other partners, would lead to desired results. 5 IMPLEMENTATION ARRANGEMENTS 2.6 The implementing agency at the national level was the Ministry of Planning and Development. Key implementing partners in the four participating provinces included the Provincial Directorate of Planning and Finance, the District Administrations, and District Consultative Councils and local fora. Provincial Steering Committees in each province provided coordination across Provincial Directorates, and ensured mobilization of required technical staff. 2.7 The project initially focused on four provinces in the central zone of the country: Zambezia, Sofala, Tete, and Manica--with a total of 49 districts. They were reportedly selected because other donors were providing support to decentralization capacity building in other provinces and districts because, inter alia, of government requests and perceived needs, while the selected districts in the four project provinces hadn't yet received much support. In 2007, training and capacity building was provided for the province of Maputo (7 districts), but no investment grants were made in this province. Relevance of Design 2.8 Substantial. There is a clear statement of objectives in the DGA which is specific and monitorable. There is a broader formulation of the objectives in the PAD which is consistent and mutually supportive. The design is based on a clear and convincing logical chain between Bank funding, outputs and intended outcomes according to a model of decentralization that had been previously tested and refined in Mozambique, and other countries in the region. The Bank's funding to support consultative processes, local investment grants, capacity building for local administration, policy reform and project coordination is clearly linked to the decentralization model being supported (see Box 2.1), and would logically contribute to achieving the objectives. The networks, manuals, guidelines, and strategy all helped ensure that the sub-objectives of participation and transparency were achieved. The budget transfers and financial rules helped to plan and manage small-scale infrastructure investments. The overall approach of participatory and transparent planning and implementation of financial resources is a reasonable building block towards possible deeper institutional reforms such as democratically elected district councils in the future. The design was conditioned on, and helped to influence, the timeline of decentralization in Mozambique, as outlined in Annex B. The key elements of design for the operation, and for related intergovernmental operations supported by the Bank, are outlined in Annex D. 6 Box 1. Decentralization Model The model of decentralization being supported comprises appointed district administrations that mobilize resources from different sources to implement small scale infrastructure projects prioritized by district development plans. Plans are formulated with participation from District Consultative Councils. The process of soliciting ideas from the Councils, recommending priorities, and holding district administrations accountable was facilitated by an appropriate civil society organization based in each district. For districts meeting eligibility criteria, a maximum $200,000 per district per annum grant was provided to support small infrastructure investments prioritized in district plans. The approach was to strengthen capacity for financial and works management through learning-by-doing using government systems. There was close coordination with sector directorates at the provincial level to ensure compliance with works standards, and available funds for operations and maintenance. Source: World Bank 2003a 2.9 The project team considered and rejected four alternative designs: (1) a social fund would be efficient due to a vertically integrated structure, but would be duplicative in a context of scarcity of technical personnel, and not strengthen ongoing efforts to build local implementing arrangements; (2) direct financial support to local authorities would have been appropriate if there had already been devolution of authority and fiscal autonomy to provinces and districts, but that was not the case; (3) supporting sectoral programs would have allowed easy integration of investments and institutional improvements, but would have lacked a necessary multi-sectoral dimension, enhanced by local choice and prioritization; and (4) an area development approach was considered, but rejected because of the high-level coordination requirements going beyond available capacity at the local level, and because such approaches did not adequately reflect local demand (World Bank, 2003a: 12-13). Challenge grants were considered and rejected because of concerns to maintain equitable support across districts. 2.10 This was a safeguard category B project. An environment assessment and resettlement policy framework were prepared and disclosed according to Bank policies. 3. Implementation 3.1 The Decentralized Planning and Finance Project (DPFP) was approved November 20, 2003, and became effective February 16, 2004. The initial closing date of June 30, 2008, was extended twice. The first extension of six months was to accommodate delays in the local investment grants. The second extension to March 31, 2009, was to allow time to undertake a study of community participation in district planning, to finalize several studies, and to prepare inputs for the follow-on National Decentralization Planning and Finance Program. PLANNED VS. ACTUAL COSTS BY COMPONENT 3.2 The total costs at appraisal were US$46.20 million (US$42 million from the International Development Association (IDA) and US$4.20 million from the government), while the actual costs were US$48.68 million (US$45.17 million from IDA and US$3.51 million from the government). The increased dollar amount from appraisal to actual is due to exchange rate changes. Planned vs. actual costs by component are given above in paragraphs 7 2.4-2.8. There was an increase in the decentralization policy component to finance greater than expected costs in policy development. Other actual component costs were broadly in line with appraised costs. IMPLEMENTATION EXPERIENCE 3.3 The project disbursed using advances from government counterpart funds. The government was reimbursed based on quarterly financial management reports. The August 8, 2006 Mid-Term Review pointed out that full financial management reports had mainly not been produced, so that many ad hoc activities were financed not linked to strategic plans. However, based on a review of Implementation Supervision Reports and field consultations during the mission, this problem was addressed starting in late 2006, when there was clear improvement in project performance. 3.4 A concern was the changes over time of the block grants provided to districts through the Government's budget (not financed by the project). Starting in 2006, the Government provided an initial US$300,000 equivalent for unconditional block grants for each district, changed in 2007 to an average Metacais 7 million2 (US$273,000). The original intent was to supplement DPFP resources for small scale infrastructure and related public works. Provincial budget allocations to districts were established for district infrastructure maintenance through budget instructions to the line ministries (education, health, water, and the like). Based on agreed criteria relating to commitment, capacity and past performance, provincial governments indicated which districts had satisfied the eligibility criteria and from these which will gain access to local investment grants in each year. The development of the eligibility criteria was supported by PRSCs, and is a good example of complementarity between the PRSCs and the DPFP. 3.5 Then, at the direction of President Guebuza in 2007, the intent of the block grants was changed to support "productive investments" ( Orqamento de Investimento de Iniciativa Local). This policy initiative was not supported by the Bank, either through DPFP or the PRSCs, but it raises concerns over the Government's commitment to the decentralization program. The policy caught district officials by surprise: work on ongoing projects need to stop as funds were reallocated. It wasn't initially clear whether the funds comprised loans or grants, and if the former, how they were to be repaid. Each district initially received roughly the same amount for these investments. The expressed logic was that sectoral ministries already were providing resources for public investments, and that the critical gap was for private, job-creating investments. However, these "productive investments" were typically not presented in District plans, so were not well integrated into the participatory processes supported by the project. While the grants were intended as loans to private businesses, reportedly only 7-17 per cent have been repaid. There was no known monitoring framework to determine if the projects were successful. However, starting in 2007, contracts were signed by grantees specifying the repayment schedule and interest rate. Some district consultative councils have reportedly worked to ensure that these contracts are abided by. While there are anecdotal estimates of the number of jobs created by these businesses, there is no formal analysis available on this. The field mission found that during President's open government 2 Known as the "seven million" program. 8 consultations with citizens, criticism of the private grants was repeatedly brought up. The Bank and other donors suggested to Government in policy dialog that it reconsider its approach, since it was not well linked to the district planning and capacity building effort. In addition, productive investments are normally allocated by specialized institutions with the expertise to evaluate private business proposals and not by district level administrators as was the case here. 3.6 The Government's response in 2008 was to split the district transfers into two parts, one for productive investments, and the other for infrastructure and related operations and maintenance support, called FIL (Fundode Iniciativa Local). Each district received an equal allocation for FIL. While initially only a small part of the block grant, the Government committed through the medium-term expenditure framework to increase the proportion of delta in the total block grants from 24 percent in 2009 to 55 percent in 2012. While these changes are providing increasing resources to fund district plans, the continuing allocations for productive investments raise questions about the coherence of the Government's decentralization strategy. On the other hand, a positive sign is that the overall size of investment budget transfers to districts has increased in recent years, from 3.6 percent in 2008 to an estimated 4.9 percent in 20123 IMPLEMENTATION OF MONITORING AND EVALUATION 3.7 There was full reporting on the status of most indicators, except for the indicator on citizen perceptions, which wasn't reported on until the last year of implementation. M&E information was used to address shortcomings in the quality of district plans and underlying participatory processes. SAFEGUARDS COMPLIANCE 3.8 The 2008 review of safeguard compliance found that insufficient attention had been paid to environmental and social issues in two projects out of 25 reviewed. The reason identified was that there was a high turnover of local works technicians, and insufficient training provided to new staff. The Bank advised the government to take action to mitigate the environmental and social impacts of the two projects found to have issues, and to prepare a training plan to prevent future problems of this nature; the government took action as advised. 4. Achievement of the Objectives 4.1 Substantial. The achievement of efficacy of objective will be measured first by considering the evidence achievement of outputs, and then to examine whether the outputs achieved contributed to achievement of the overall objective of improved performance of district administrations in planning and managing small infrastructure projects, as formulated in the DGA. ' Republic of Mozambique, 2009: 26; 2012a: 21. 2012 estimate based on expenditures in the first three months. 9 OUTPUTS 4.2 Achievement of outputs is examined first by looking at project development objective (PDO) targets. The Bank's funding to support consultative processes, local investment grants, capacity building for local administration, policy reform and project coordination all were part of a results chain intended to lead to achievement of the project objective. Thus, it is appropriate to evaluate if these intended outputs were in fact achieved. PDO indicator-1: Percentage of districts with Development Plans and Annual Investment Plans approved 4.3 All 56 districts supported by the project were covered, which exceeded the target of 80 percent. PDO indicator -2: Percentage of districts with the District Consultative Councils organized and approving district plans and implementation reports 4.4 District consultative councils were set up and approved district plans in all 56 districts supported by the project. In addition, 713 local consultative councils, 168 local councils in the administrative posts, and 489 local councils in the localities were also set up. DPFP supported training for 11,300 consultative council members, 555 district technical specialists, and 28 provincial technical specialists (World Bank, 2009: 29, and confirmed by field mission). While there was no detailed study carried out of the effectiveness of training, the evidence of achievement of outputs produced by these trainees suggests that the training contributed to their performance. 4.5 The field mission carried out consultations in Tsangano, Chiuta and Changara Districts in Tete Province, and Nhamatanda and Gorongosa Districts in Sofala Province. Local officials and community consultative council members interviewed confirmed that plans and project ideas at the village level were prioritized within budget ceilings, vetted and prioritized again moving up the other three levels. Some projects were technically vetted by district technicians, for example, asking if a proposed dam was strong enough to hold. Other projects such as health clinics, hospitals, schools and bridges served multiple districts, so were prioritized at the district or provincial level as appropriate. Once approved, implementation of plans was supported by the Councils to the extent of their competence, sometimes with support from non-governmental organizations such as the Lutheran World Federation. Such monitoring was more likely in some areas (for example, schools), and less likely in others (for example, road construction, where technical knowledge may be lacking). In some years, only 60-70 percent of the priorities could be funded, but those not funded in one year were typically funded in the next. Unexpected visits from senior officials including the Head of State could also eat into budgets, and lead to further implementation delays. Data on actual costs and delays were not available. PDO indicator-3: Percentage of districts in which participants in planning processes express satisfaction with responsiveness and accountability of district administration 4.6 The target was that participants in 60 percent of districts participating in the Local Investment Grants would be satisfied with the responsiveness and accountability of their 10 district administration. This could not be directly monitored during the operation. However, a March 2009 consultant study funded by the project included a survey of people's perception of the performance of district administrations. It revealed that 81 percent of respondents were aware of district investments, and 88 percent of these expressed satisfaction with the results. A majority believed there was now much more interaction between government and citizens, and that plans and budgets reflect community priorities. It is not known what the sample size was, how the sample was selected, and what specific questions were asked. PDO indicator-4: Percentage of districts rated good or excellent on annual local investment grant performance. 4.7 With a target of 50 percent, DPFP achieved good or excellent performance in 67 percent of districts. The training of provincial and district financial staff contributed to this positive result, including 49 accountants and 51 works technicians, of whom 95 were still in their posts at the end of the project, with 35 in permanent positions. These staff were initially financed by DPFP, and then shifted to the district administration. Staff at the Departments of Public Works, Housing and Education were also trained. It would have been useful for the Government to have carried out an impact assessment of the training provided to demonstrate the link between training provided and improved performance on the job. In addition, the field mission found that a complaint of the training provided was that no certificates were provided that could enhance an officer's credentials. This has since been addressed by giving responsibility for training to the local Center for Training in Public Administration, Local and Municipal Governance, which will provide a certificate upon successful course completion. In addition, the Government has shown a clear commitment to increase financing for subnational personnel. For example, in 2007, 25 - 35 percent of recurrent spending was at the subnational level (Thomsen and Saide, 2011: 20; Boex and Nguenha, 2007: 106), a proportion mainly unchanged since 2003. In the most recent period for which spending data are available, 51 percent of all recurrent spending was at the subnational level, up from 43 percent the previous year , see Table. 1 4 Comparing spending in the first quarters of 2011 and 2012. Republic of Mozambique, 2012: 18. 11 Table 1. Recurrent Expenses by Jurisdiction, 2011-2012 (in Millions of Meticais) ClageS AM 211 AM212 Yxb* 4 ovgua must so org.wnIo ar wa . s o. 3y112 Terkriwl Anal Jan-Ma Rtaliz. nIrA AmAl .imla ialL (%) ar AsteCe*r 39308. 10352.4 2&3 46 U*. 46814.0 9 110.3 22.7 -&.% iamPoiaal 16999.1 45754 20t 22 323J 22 35.8 5601.7 251 18.7% 53 1 3541 02 29 15 5.1 607A 309.4 2A.2 4 6% CDo Dep> 2 676.4 65. 244 257 ,1 2097.9 837.1 20.9 2I 1% 52756.3 65.5 25.0 172,2 315 W.0 7673 24.1 136% bna 1490.5 3422 21.0 1 S3.9 1 6W2.5 4465 2619 26,5% Te 14.3 320.8 22.8 161L 1637 5 432,1 26A 06% wtc 1 551 LU.3 318 181MS2 1 867.1 40A 24.1 -1I 6% 50W4 2259.1 Mt.8 230 2577J 2611.5 6733 26.0 267 Iftrawt 969. 213-0 22.0 1 150.2 1 171,0 295A 252 Ut 5% 2za 1 021.5 227.5 223 1242.3 1 250.6 291.5 232 24.7% Ap. 1441.3 281.3 19. 1 90.1 1 92.0 438. 2.8 512% C"l4 e mapu 206.9 L63.9 22A 2 453.7 25242 S724 22.7 19.7% a*ft Dimaitl 14 339.2 a040.9 212 14 O6.1 14 2944 4 99 21.6 30 4% Dsis de nsa 813.I 131.3 162 83.1 W.1 2702 32.0 99 5% D5is C* Cab DeA 1236.0 M. 23.1 1 259 0 1259. NA 26.5 13 4% DilS Oe mNrp.1a 3048.5 60 19.9 242.4 247.4 673.5 27.5 7.75 Vtfns bet 2 258.7 SS2.0 24.4 2266 8 2507.0 736 29-3 20% Dstos 0@ Tv* 1215.2 289.6 2 1253.3 1244.0 365.9 29A '2 5% Diwsdewrica 9834 7.A 79 10S 1019.9 2A 31.7 X1ALv 10rs M 5000 216.7 263.5 21,7 130K.1 1310.3 3S9 25. 26 9% DStit*s Oe lwware 1 326.3 327.5 241 1407-a 14172 419,0 296 24 05 UVAts de az 1 18.5 275.3 232 1 11.0 1 185.0 33.1 23.3 18 3% Ds5t d, tp=l I0526 211.9 22.0 105.t 1 058.9 2882 27.2 20.5% iMbt AutiQuic 1001A 217.2 211 1 206A 1206.6 319. 26.5 42.7% TOW 73648.7 18185.9 iL 84 62.0 6440.8 19721.7 213 S 7% Source: Republic of Mozambique, 2012: 18. 4.8 The operation's Local Investment Grants were a critical part of the capacity building strategy, giving accountants and technicians experience using government systems at the provincial and district level to manage small-scale socio-economic infrastructure projects. A total of about 826 small projects were initiated, of which 774 are considered successfully completed, see Table 2. Success was measured as a project completed according to design, with proper financial accounting. This is an appropriate measure, given the operation's focus on strengthening managerial and participatory skills. The bulk of the financing ($13.27 million) went to type 2 projects, financing social infrastructure prioritized by communities. $4.7 million went for type 1 projects, financing district administrative infrastructure managed by the province, and $1.72 million went for type 3 projects, financing direct contracting by district administrations. Type 2 and 3 grants were allocated based on a participatory planning process with four levels of consultation and priority setting: at village, locality, administrative post and district level. Provincial governments carried out their roles of managing and reporting on local investment grant use by the districts under their authority. Financial management procedures were those used by the Government, strengthened as deemed necessary to meet World Bank standards and to facilitate improvements in quality 12 and timeliness. The project manual provided guidelines on the size of grants, and on what could and could not be financed. The stakeholders consulted in the sample of 5 districts told the field mission that community requests were taken into account where possible, and that the criteria for selection were followed. Prior to this, the government had no system for rules based fiscal transfers to districts. The field mission found some evidence presented in Section 5 that these projects represented better value for money than would have been available from alternative channels. Table 2. Execution of Small-Scale Infrastructure Projects Category Expenditure (million Number properly Number initiated US$) executed but not executed successfully Type 1 4.7 73 13 Type 2 13.27 298 13 Type 3 1.72 403 26 Totals 19.69 774 52 Source: World Bank, 2003: 29. 4.9 Once projects were completed, maintenance became the responsibility of the relevant sectoral authorities, with simplified maintenance manuals and extra training provided at the provincial and district levels. For example, roads were handled differently depending on their grade. Trunk roads were the responsibility of the National Roads Administration, though district and provincial authorities could complain when there are issues. Tete Province was reportedly a pioneer in mapping priorities for repair on all lower tier roads. District governments could carry out emergency maintenance while awaiting a more thorough job from national and provincial authorities. If roadwork was delayed, citizens might complain, and the government needed to explain why work wasn't being done. There was no evidence of communities trained to do basic technical monitoring of issues such as pavement composition and thickness, as has happened in other parts of the world (e.g. CCAGG, undated), although their involvement in basic functions such as fetching sand may over time help build understanding of proper standards for road work. PDO indicator-5: Percentage of districts at national level adopting DPFP approach and methodologies for participatory planning and governance. 4.10 The operation also aimed at extending its methodology regarding participatory district strategic and budgeting planning to other districts that fell outside the direct scope of the project. By the end of the project, all 128 districts in Mozambique had adopted the operation's approach, which exceeded the fifth PDO indicator's target of 60 percent. The field mission found that all the districts visited are using the project-supported strategic planning and annual planning and budgeting approaches. Some districts are on their second or third generation plans. In a few cases, these plans, together with sector level and provincial plans, feed into an operational plan and budget: Plano Econ6mico Social e Orqamento Distrital (PESOD). 4.11 Progress in this area was linked to UNCDF's innovative piloting in Nampula province that was the basis for the project's approach, and in turn the Government's guidelines for institutionalization of participation and community consultation. These were 13 seen as alternatives to a process of electing district level officials. They were approved and gazetted by Government as a condition for project effectiveness. Also important was UNCDF's subsequent piloting in Nampula (and Cabo Delgado beginning in 2003) of the structured operational plan and budget. Much more ambitious than in the original DPFP program design, this was the basis for the planning guidelines developed by the Government in 2005-6, and rolled out in 2007. On the other hand, DPFP went much further than the original Nampula design in giving district officials the authority to manage funds and account for their use. The field mission found that the project took a graduated approach to capacity challenges. Some districts in 2003 had not recovered from the damage done by the civil war, with no public water supply, marketplaces, or housing for public officials; the project helped officials to understand the income generating potential of their districts, and to start putting basic infrastructure in place to catalyze the local economy. A key motivating factor for attracting staff to positions in districts is provision of housing; to address this, the project supported construction of housing for technical staff under type 1 grants. There are possible risks with such a scheme; for example, when employers control the housing situation of employees, patronage could lead to improper decisions. However, other alternatives such as a pay enhancement system for employees in remote areas would not be sufficient in many administrative centers lacking suitable housing for rent. One district official explained that construction of housing was prioritized as follows: housing for the two technicians first (one an accountant, the other a construction engineer), for the district administrator second, and for school principals third. The reason for prioritizing technician housing first was that an administrator without technicians couldn't do his job. The project also hired essential technical staff, and when their contracts were finished they were in many cases hired by government under the same terms. 4.12 In addition to insufficient housing, attracting qualified staff to serve in remote districts faced additional challenges because of limitations of essential infrastructure, power, water, fueling stations, banks, phone and Internet service, schools, and health facilities. Staff may have to travel long distances on very rough roads to cash their paychecks, obtain fuel, and pursue further education, at great inconvenience and expense. 4.13 The parallel Public Service Reform Project, being assessed in another PPAR, supported deployment of senior level staff to provinces, one of which (Sofala) was supported by the operation. As staff were deployed and capacity expanded, provinces increased responsibilities to district administrations. The PSRP had some modest success in improving service quality in areas including land titling, visa services, commercial licensing, school registration system, exam registration system, vehicle imports, and driving licenses. There were many more improvements documented at the provincial level over the period 2006-8 in health care, education, justice, transport, and many other fields, as measured generally by reduction in time to receive a particular service (Republic of Mozambique: 2012a). PDO indicator-6: Key administrative and fiscal instruments adopted supporting transparent district planning and management. 4.14 This PDO indicator supported the adoption of five key administrative and fiscal instruments. The target of four out of five was achieved, including: 14 * Publication in the official gazette of guidelines for community participation and consultation in district planning, * The adoption of a law on the local organs of state and its implementing regulations that formalized the district as a budget unit, * The publication of operational rules and procedures for district finance. * The establishment of provincial budget allocations to districts for infrastructure maintenance through budget instructions to line ministries. 4.15 The fifth instrument called for a 3 percent per year increase in the state investment budget allocated for provincial planning and management with district input. This was not as well specified as the other four (for example, did it mean 3 percent increase in the nominal budget, or 3 percent increase in the budget share; was it a formal undertaking, or an indicative target) so its achievement can't be confirmed. However, as discussed earlier, there has been an increase in the share of the district investment budget in recent years. Although there are challenges in comparing figures across years from different sources, there is evidence that overall sub-national investment as a proportion of total public investment may have increased from 12 percent in 2007 to 16 percent in 2012, and the share of investment at the district level has increased from 3.6 percent in 2007 to 4.9 percent in 2012 (Boex and Nguenha, 2007: 107 and Government of Mozambique, 2012: 2). During the same period, overall public investment has more than doubled. Thus, although the proportionate amount managed at the district level is small (but growing), the amount of funds being managed is considerably more than it was. 4.16 The operation's design called for technical assistance to several key ministries to help implement these instruments. For example, the project along with other development partners provided critical assistance to the Tribunal Administrativo to conduct audits of the districts' financial accounts--which contributed to upward accountability in all provinces, including those supported by the operation. 4.17 According to three PEFA assessments carried out in the period 2006-10, at least 90 percent of allocations are rules based and transparent, the highest rating possible under indicator PI-8(i): Transparent and rules based systems in the horizontal allocation among subnational governments of unconditional and conditional transfers from central government (both budgeted and actual allocations). However, the first two assessments look only at transfers to municipalities, so aren't relevant to this operation. The latest one, based on 2010 field work, gives the same rating, but the basis for this isn't clear, as it says that it is unclear whether there is a formula or specific methodology to determine the size of sectoral and geographic allocations to Provincial Governments, while there is a formula for releases to the district level. There is presently a rule-based process to allocating investment budget transfers to provinces and districts. For provinces, transfers are weighted by population (70 percent) and poverty (30 percent), and for districts by population (35 percent), surface (20 percent), own revenues (15 percent) and poverty (30 percent)5. However, the field mission was told that the actual district allocations are also influenced by a political process that may not always follow these rules. 5 Republic of Mozambique, 2012b: 22-3. 15 4.18 In addition, by the end of the project, there was an improvement in PEFA indicator PI-8(iii): Extent of consolidation of fiscal data for general government according to sectoral categories. This improvement can reasonably be attributed to the capacity building supported by DPFP and other comparable projects in other provinces. Other achievements and issues 4.19 In addition to achievement as measured by these PDO indicators, there were many other achievements supported by the operation: * Three networks for community participation were established. * 13 standardized manuals for district participatory planning were adopted, covering issues such as district diagnostics, preparation of development strategies and action plans, production of district economic and social plan and budget, monitoring and evaluation, and organization and functioning of local consultative councils. * Guidelines for the preparation of the district land-use plans were prepared, and some 18 land-use plans developed. The work included diagnosis of the present situation, production of thematic maps of current land use, administrative divisions, population, infrastructure and facilities, physical conditions, and trends in settlement expansion. * A communication strategy was formulated to help ensure that citizens became better aware of the decentralization reforms. Based on this, radio and television production staff were trained, a regional seminar on the new law on local organs and participatory planning conducted, and sessions of open radio debates carried out. * Audits of 48 districts' financial statements were conducted in 2006, the first audits of district finances in the country's history. 107 district financial statements were audited in 2007, and more in subsequent years. However, no details were provided to the mission on what share were audited, and on trends in compliance. In addition, extensive support was provided to the Administrative Tribunal, including preparation of an Audit manual, institutional analysis and preparation of a project for addressing priority needs, improvement of legislation, and provision of training. The result is that district administrators were expected to comply more strictly with procedures to more carefully manage funds under their control. 4.20 One disappointment is that Government did not prepare a Decentralization policy framework as it had committed to do in its PARPA, with support from DPFP. As discussed above, the theory of the project was that that district administrations would mobilize resources to implement small scale infrastructure projects prioritized by district plans, that had been produced using participatory processes, and supported by an appropriate policy framework. The Government did produce numerous papers on decentralization, but they didn't present a coherent and comprehensive approach. There was also the decentralization joint review workshop in 2006, where participants identified key principles for a national decentralization program and four pillars focused on participation, execution, audit, and human resources development (see section 6 below); however, this exercise was never formalized by the Government. 4.21 The result has been uneven clarity on many key aspects. The 2003 Law of the Local Organs of the State sets out the principles for organization, competencies and functioning of 16 provinces, districts, administrative posts and localities. It clearly states that the district is the most important sub-national territorial unit, and that 'finance should follow function. However, the details of expenditure assignments are less clear. For example, the Law and its regulations are clear that local water supply systems are assigned to district governments to construct and manage. By contrast, in the education sector, it isn't clearly stated whether districts have the responsibility to construct, operate or maintain schools; instead they are given the job of guaranteeing the good functioning of schools and to promote education. The field mission was told that this lack of clarity leads to problems such as expansion of schools in areas already relatively well-served, leading to problems in access for more needy school aged children. This contributes to challenges in the quality of education, including 44 percent of primary school children being over-aged, only 20 percent of secondary school age children attending secondary school, and gender gaps in education and literacy. 4.22 In addition, the desired timing of decentralization isn't clear, with few timetables or benchmarks (Boex and Nguena, 2007: 36-7). Another sign of a lack of a clear policy framework is the initiative to support "productive investments". The selection and implementation of these investments with neither community inputs nor transparent monitoring and follow-up seemed to work against the DPFP objective. In recent years, the Government has intensified work supported by the PRSCs to prepare a decentralization strategy to clarify responsibilities across different levels of government and to serve as the unifying framework for future reforms. However, this hasn't yet been adopted. 4.23 The Bank and other partners financed a parallel program of support for financial and fiscal management improvement, which was delivering gradually improved results as measured by Public Expenditure and Financial Accountability (PEFA) assessments and Country Policy and Institutional Assessments (CPIA). There were also sector wide assistance programs inter alia in health, education and agriculture, and support from PRSCs. The DPFP benefited from all these streams of work. The approach was to support infrastructure in areas where sectors were already working; that is, provide a school house where teachers were already conducting classes without a proper building. OUTCOME 4.24 This section reviews the evidence that the achievement of the outputs examined above contributed to the achievement of the overall objective of improving the institutional performance of district administrations to plan and manage small infrastructure investments in response to community demands. 4.25 774 small-scale infrastructure projects were successfully delivered according to agreed selection criteria, with appropriate links to sectoral units to ensure quality standards were met, and maintenance costs would be covered. Sixty seven percent of districts achieved good or excellent performance on annual local investment grant performance. These small- scale projects were prioritized in district development plans and annual investment plans, that were prepared in consultation with district consultative councils. According to survey evidence, participants expressed satisfaction with the responsiveness and accountability of their respective district administrations. The Bank project provided technical assistance and training that supported the outputs discussed in the previous section that contributed to 17 achieving these outcomes, including the preparation of district development plans and annual investment plans in 56 districts, and also supported setting up and training district consultative councils in these 56 districts. The Bank project also financed the small-scale infrastructure projects. A disappointment is that Government did not prepare a decentralization policy framework as it had committed to do under its PSRP and the PRSCs, which supported a related decentralization strategy. However, many policy papers were prepared with support from the project, and donors agreed on an overall aid framework for decentralization. 5. Efficiency 5.1 Substantial. The ICR rates efficiency as: not applicable. This was incorrect: according to Bank guidelines, all investment projects should be rated on efficiency. The Project Appraisal Document didn't calculate an economic rate of return. However, Annex 4 of the Document gives extensive data from other African countries on favorable economic rates of returns for small scale infrastructure investments. This is consistent with extensive research in developed countries finding that decentralization generally increases the production efficiency of service delivery (Ahmad and Brosio 2009: 141-2). This implies that ensuring local autonomy is consistent with meeting national sector goals. In the case of this project, there were some delays in payments caused by lack of timeliness of provision of government counterpart funds as reported in the Mid-Term Review; however, these issues were subsequently addressed, as discussed below under Government performance. 5.2 Considering project management, an indicator of good efficiency is that expenses for project coordination were below the budgeted cost, despite a 9 month extension of the closing date. Aside from the payment delays mentioned above, procurement and fiduciary management was carried out to ensure that funds were used as intended. One informant believes that small scale infrastructure in provinces not supported by the Bank such as Nampula, Cabo Delgado and Gaza was more cost effective, because procurement rules were more flexible. This allowed greater use of local materials and contractors, less theft, and more economic stimulus to the communities. By 2006, these processes led to the creation of 33 small contractors with nearly US$1 million earnings equivalent in Nampula. However, this anecdotal evidence could not be confirmed. 5.3 The field mission was informed by officials in Tete that there were preferences in some procurements supported by the Bank project for local artisans, but that they often needed to be trained in basic business practices, and teamed up with more experienced contractors from outside the province. The mission could not find any formal studies to confirm this, and no evidence to indicate any serious issues concerning efficiency. In many localities it was reportedly difficult to attract the interest of contractors because the projects were so small and widely dispersed. To address this, districts tried to bundle together construction projects funded by the Bank, the United Nations Children's Fund (UNICEF) and other partners to increase the contract size, and make bidding on them more attractive. There has also been a helpful change in Government regulations. Contractors can now bid on contracts when they need to hire necessary equipment; previously, they could only bid if they owned the equipment. 18 6. Ratings Outcome 6.1 Satisfactory. The objectives were highly relevant, the design was substantially relevant, and efficacy and efficiency were both substantial. The operation supported participatory, district planning, which inter alia supported the design and implementation of small investment projects in all target districts. Staff in these districts were trained. All other districts in the country have also adopted the same approach. Provincial budget transfers to districts were established, internal financial reporting was carried out in a proper manner, and some financial statements were subject to external audit. The Government has increased the budget allocation to districts through 2012. There was a generally favorable perception of this effort as measured by a citizen survey and two pilot citizens' charters. These and other outputs supported by the project contributed to achievement of the objective of improving the institutional performance of district administrations to plan and manage small infrastructure investments in response to community demands. Risk to Development Outcome 6.2 Modest. Key elements of sub-national governance are laid out in the Constitution, enabling legislation and decrees. While there were delays in the early years in accessing counterpart funds, by the time of project closing 82 percent of the agreed funds had been provided. The initial design was broadly "adopt a district" with other partners piloting similar approaches in different places. The outcome was a single system, rather than parallel systems being built up and persisting. Starting in 2006, the government launched its own intergovernmental transfer system, drawing on the most attractive features from DPFP and other pilots when there were minor differences. The size of investment and recurrent budget transfers to sub-national authorities has increased since 2007. There was a joint decentralization review by donors in 2006, including the Mid-term Review of the Bank project, and reviews of similar support including PPFD-Norte (UNCDF, UNDP, Norway, Netherlands, Ireland, and Switzerland) and PRODER (GTZ). Following the recommendation of the joint review, the government agreed there would be two sources of partner support for decentralization: a common or basket fund overseen by donors and budget support where funds are provided to support agreed priorities, and monitored through the Government's budgeting system. The common fund for donor support to decentralization was launched in 2008, and there is an active decentralization working group, with sub-working groups for each pillar (see below). DPFP-Centro is the precursor of the current NDPFP-National in which all donors active in the decentralization sector joined forces (in a SWAP program, led by the Bank) to support the government decentralization program. The design of the NDPFP- National was built on the many of the lessons from DPFP-Centro in various areas like community participation and district planning. The follow-on Bank project became part of the common fund managed by the Ministry of Planning and Development. Donors and government agreed that future support would be aligned around a national program comprising four pillars: 19 * Pillar A Participatory Planning and Budgeting Coordinated by Ministry of Planning and Development: * Includes participation IPCCs strategic planning PEDDs annual planning and budgeting PESOD, * Pillar B Execution Coordinated by Ministry of Planning and Development: * Focuses on financial management and procurement, * Pillar C Control Coordinated by Tribunal Administrativo: * Focuses on Internal and External Audits and Procurement, * Pillar D Human Resource Development and Knowledge Management Coordinated by Ministry of State Administration: * Focuses on Human Resource Management Capacity Building and Knowledge Management. 6.3 In addition to support through the preferred modalities or common fund or budget, project support would be allowed as a third choice if required to meet particular donor rules; in all cases, donors were encouraged to align with the four pillars (World Bank, 2006). This harmonized approach was intended to help to ensure that future assistance from the Bank and other partners was productive. The advances in regulatory framework, experience base, and well-coordinated donor support all worked to ensure sustainability of the operation's development outcome. On the other hand, the difficulties in preparing a decentralization strategy to clarify responsibilities across different levels of government showed that there was not yet a political consensus on how to move forward. This risk is mitigated by the fact that the Government has adopted a series of laws, regulations and decrees over the last two decades that have had the effect of broadening and deepening key aspects of decentralization. The sustainability of the small infrastructure investments supported is not certain, since it has not been monitored. This risk is being mitigated by the roll-out of an improved conceptual model for M&E for PARPA implementation to target sectors, provinces and districts, supported by the World Bank (2011). The experience in other countries is that when there is community involvement in priority setting and implementation by local administrations, the chances of sustainability increases for local investments (see for example, Anokbonggo and others 2004). Finally, the allocation process of "productive investments" goes against the priority setting and participatory system of planning supported by the project, and raises concerns over the Government's commitment to the decentralization program. This risk is mitigated by the fact that the proportion of such investments in district allocations has decreased in recent years. Bank Performance 6.4 Quality at entry. Satisfactory. The design built mainly on the lessons of a pilot project supported by the United Nation's Capital Development Fund (UNCDF), with funding from Netherlands and Norway, in Nampula Province, which was in turn built on the lessons of comparable work in Uganda (cf. Pyndt and Steffensen, 2005). Drawing on Uganda's experience was ambitious in Mozambique. Analysts have classified Uganda as a high- functioning intergovernmental system well on the way to consolidating decentralization, 20 where Mozambique is considered a low-functioning intergovernmental system mainly relying on deconcentrated structures. The fact that Mozambique drew on cutting edge ideas from Uganda can be characterized as pushing the boundaries of the institutional environment. The country took on additional risk in doing this, but it was justified by confidence that local government and civil society leaders would support the changes (cf. Serrano-Berthet et al, 2010: 125). 6.5 Building on this experience, the design improved on the structures for civil- society/district dialog, increased transparency about public procurement, and created conditions for downward accountability about plan implementation. It also built on the experience of GTZ funded activities in Sofala (since 1998), Manica (since 1997) and Inhambane (since 2002, in coordination with Development Cooperation Ireland) in a total of 18 districts. Each of these parallel programs demonstrated that participatory decentralized planning, local investment grants and financial management could be adopted in Mozambique as building blocks for future decentralization. The design was aligned with the Poverty Reduction Strategy Paper (PARPA) and the Bank's Country Assistance Strategy, and remains relevant to the latest Country Partnership Strategy. The design draws on lessons from the Bank and other donors in Mozambique and neighboring countries. The design also draws appropriately on analytical work, e.g. Serrano, 2002. The selection of the initial four provinces reflected a division of labor among donors. With hindsight, the choice seems appropriate for the additional reason that these provinces had demonstrated among the best performance in reducing poverty up to the time of appraisal, and were thus good prospects to raise their performance to an even higher level. Within the target provinces, the Bank engaged at provincial, district and community level on the project. Another Bank project worked with municipalities. The M&E framework was clearly designed, supported by a comprehensive manual and largely complete, with minor shortcomings such as the lack of an impact assessment demonstrating the link between training provided and improved performance on the job. 6.6 The learning-by-doing approach is consistent with the latest thinking on iterative adaptation, where the focus is on solving locally nominated and defined problems, encouraging experimentation and experiential learning, and engaging a broad set of agents so that development efforts are viable, legitimate, relevant and supportable (Andrews, Pritchett and Woolcock 2012). 6.7 Quality of Supervision. Highly Satisfactory. There were 13 supervision missions over a 5 year period. Although there were 3 Task Team Leaders, each had been part of the project team before taking over, so handovers were smooth. Implementation Supervision Reports rated project performance realistically, and the team focused on the goal of mainstreaming the approach across Government, even at the expense of quicker disbursement. There was good donor coordination, with the Bank contributing to three joint reviews of decentralization projects, one of which constituted the DPFP's mid-term review (World Bank 2006), which was comprehensive in raising issues and making practical recommendations. Following the midterm review, the project showed a clear improvement in performance and the implementation performance rating was upgraded from moderately unsatisfactory to satisfactory. The Bank and other development partners frankly raised sensitive issues with the authorities such as the use of district funds to support productive 21 investments, and the need for a decentralization policy framework. In the case of the former, the government agreed to renew its support for block grants for infrastructure. In the case of the latter, Government deliberation is still continuing. The addition of Maputo province at a later stage due to demand from the Government is acceptable, given the challenges of low- income communities there of recent migrants from rural areas. The 2008 review of safeguard compliance found issues in two projects, for which it advised the government to take action, and to prepare a training plan to prevent future problems of this nature; the government took action as advised. Taking the above into account, overall Bank performance is rated Satisfactory. Borrower Performance 6.8 Government Performance. Moderately Satisfactory. Prior to project approval, the government had expressed its commitment to strengthening local administrative capacity through the piloting of 5 year district strategic and development plans, and adoption of legislation creating municipalities with full financial and political autonomy, a Decree legally recognizing community authorities as interlocutors between rural communities and district administration, the PARPA, and the Global Strategy for Public Sector Reform. Since then, the Government adopted further enabling legislation, decrees, guidelines and strategies laying out its commitment to the project objective. For example, the Government instituted a fiscal transfer system mid-way through project implementation, modeled on the system piloted by the project, and building on the improved sub-national capacity that the project contributed to. The shift to supporting "productive investments", while not supported by the project, suggests that the political leadership did not fully understand or accept the approach of territorial based finance, planning, and community-based accountability, thinking that infrastructure should be provided by sectoral ministries outside of the territorial system. On the other hand, the budget guidelines for 2008 partially clarified the situation, making division between the 7 billion (that were kept for productive investments--until today) and the FIL that is still used for public investments. Over time the FIL has become more important (in terms budgetary allocations) than the 7 billion. 6.9 Another moderate shortcoming is the lack of clarity on criteria for selection of private investment projects for funding. The inability to produce a decentralization policy as had been agreed is disappointing. Agreement on such a policy is challenging because of the legacy of a recent civil war with a territory-based insurgency and a strong desire by the leadership to preserve the hard fought unity of the nation. There are also continuing concerns among some in the leadership over giving recognition to traditional community leaders (Makgetia 2010). 6.10 By the time of the Mid-Term Review in 2006, the government had only approved US$205,500 out of a budgeted amount up to that time of about US$480,000. In addition, there were problems with the timely availability of counterpart funds, leading to payment delays to providers of goods and services, and delays in project implementation. To address this, the legal agreement was amended in 2008 to allow 100% Bank financing of all project activities. By project close, 82 percent of the original total counterpart contribution agreed at appraisal had been paid, and no payments were outstanding at the end of the project due to the shortfall and delayed availability of counterpart funds. 22 6.11 Implementing Agency Performance. Satisfactory. The Ministry of Planning and Development was an appropriate choice as implementing agency because it had the power and influence to change procedures and processes. It also had the ability to work across sectors, and the power to protect the project and to scale up the decentralization process as conditions allowed it. The Ministry constituted a DPFP National Steering Committee to oversee the project. The Ministry motivated and guided provincial directors of planning in achieving project objectives. The Ministry also worked to get important procedural changes from other ministries, including the first budget transfers made to districts, the first auditing of district financial statements, and a significant increase in budget transfers. A moderate shortcoming is that provincial governments didn't take the lead in three out of the four provinces for designing subprojects as had been anticipated, despite the stationing of staff from the project implementation unit in the provinces. In response, some investment cycle operations needed to be pulled back to the project implementation unit in Maputo, making the operation in this respect more like a micro-projects program rather than a governance one. A minor shortcoming was weak coordination with other ministries in some cases. There was also lack of full compliance with environmental and social safeguard procedures at some points in project implementation; however, appropriate action was taken following Bank recommendations. Taking the above into account, overall Government performance is rated Moderately Satisfactory. Monitoring and Evaluation 6.12 M&E Design. The monitoring framework was written up during preparation in an M& E manual that provided clear instructions for use by provincial and district administrations, tailored to Mozambique's needs, capacity and institutions. M&E implementation was supported by project financed M&E specialists to compensate for lack of qualified staff in local administrations. Following project closure, project-financed specialists were absorbed by the district administrations. 6.13 One of the six indicators directly measured achievement of the project objective to improve the institutional performance of district administrations to plan and manage small infrastructure investments in response to community demands, by tracking those districts rated good or excellent on annual local investment performance based on a performance management framework supported by the project. The other five indicators all measured outputs that contributed to achievement of the objective. All had baselines starting from zero which was appropriate. 6.14 A minor shortcoming was the differing statements of objectives in the DGA and PAD. The statement in the PAD was much broader, and more difficult to monitor. The statement in the DGA was specific and monitorable based on the selected indicators. Another shortcoming is the lack of impact assessments linking the training provided to improve job performance. 6.15 M&E Implementation. There is full reporting on status of the six key project development objective (PDO) indicators starting with the third Implementation Status Report of April, 2005. The project's third PDO indicator aimed at 60 percent of districts participating in the Local Investment Grants, in which participants in planning processes 23 express satisfaction with responsiveness and accountability of district administration, but could not be monitored during the operation. A study to measure this took place in 2009, carried out by DPFP financed consultants. However, it is not known what the sample size was, how the sample was selected, and what specific questions were asked. The operation also supported pilot citizen report cards in two districts, but the effort was discontinued, awaiting a comprehensive effort for use of citizen report cards as part of the national statistical system. Another shortcoming was the limited sharing of M&E information with stakeholders between districts and across different levels of administration. Another shortcoming is the lack of monitoring on the cost effectiveness of small scale infrastructure supported by the Bank. Although productive investments supported by the Government's fiscal transfer system are not supported by the project, some district consultative councils are reportedly monitoring repayment practices, trying to ensure that beneficiaries make payments as specified in their contracts. 6.16 Use of M&E Data. M&E information was used to address shortcomings in the quality of district plans and underlying participatory processes. For example, M&E findings on the need for structural reform contributed to the establishment in 2008 of a unit within the district governmental structure dedicated to planning and finance. It also led to production of training manuals, guidelines, and support for private contractors, and greater supervision at the provincial level of the planning process. For example, reports on works implementation raised serious capacity weaknesses of contractors. The response was the development of training on preparing bidding proposals, the introduction of standard timeframes to verify and monitor implementation progress, and rollout of a prices database for reference purposes, and a template for preparing supervision reports. 6.17 Overall M&E Quality Rating. Taking all these factors into account, IEG rates Monitoring and Evaluation as Substantial. 7. Lessons 7.1 There are three key lessons coming out of this review. First, for countries at an early stage in the decentralization process, a learning-by-doing approach can be effective using iterative adaptation, where the focus is on solving locally nominated and defined problems, encouraging experimentation and experiential learning, and engaging a broad set of agents so that development efforts are viable, legitimate, relevant and supportable. A well-designed M&E framework can help to inform the work of planning and finance units, and to inform revisions in operational programs and training manuals. For district administrations, the process of working each year with consultative councils, sectoral directorates, and other stakeholders to plan and implement small-scale infrastructure projects will bring up a variety of issues and challenges. Responding to these helps build capacity for solving future challenges, while enhancing legitimacy to rural populations. While well designed training courses can contribute to building up essential skills, there is no substitute for on-the-job learning by managing resources according to a basic set of rules, with participation from local citizens. 7.2 Second, in political settings without democratically elected subnational authorities, a process of controlled/limited participation within a deconcentrated framework can be 24 politically feasible, and help to ensure accountability and efficient use of resources. The political impetus may come from close national elections, pressure from donors, and a desire to manage ethnic and regional issues. The low initial cost of the approach helps to mitigate opposition from other claimants to government resources. The process of improving greater accountability may be uneven, with setbacks along the way as political interests are accommodated. Yet with a combination of strong technical direction and sustained interest from civil society, an overall positive trend can be maintained. 7.3 Third, while managing small scale, public infrastructure investments is useful for building subnational administrative capacity, managing lines of credit for private investments that are vulnerable to manipulation, and not well monitored or regulated, can weaken legitimacy and the good will built up through participatory processes. 25 References African Elections Database. undated. "Elections in Mozambique". http://africanelections.tripod.com/mz.html Ahmad, Ehtisham and Giorgio Brosio. 2009. "What do we know? Evidence on decentralization and local service provision". In Ehtisham Ahmad and Giorgio Brosio, eds. Does Decentralization Enhance Service Delivery and Poverty Reduction? Cheltenham, UK: Edward Elgar. Andrews, Matt, Lant Protchett, and Michael Woolcock. 2012. "Escaping Capability Traps through problem driven iterative adaption. Cambridge, Mass: Harvard University. Accessed 7 September, 2012 at http://www.hks.harvard.edu/var/ezp site/storage/fckeditor/file/pdfs/centers- programs/centers/cid/publications/faculty/wp/239.pdf Anokbonggo W.W., J.W. Ogwal-Okeng, C. Obua, 0. Aupont and D. Ross-Degnan. 2004. "The Impact of Decentralization on Health Services in Uganda: A Look at Facility Utilization, Prescribing and Availability of Essential Drugs." East African Medical Journal 81 (12): S2-S7. Boex, Jamie and Eduardo Nguenha, 2007. Intergovernmental Fiscal Relations In Mozambique: Towards A Coherent (Fiscal) Decentralization Policy. Accessed 15 May, 2013 at http://www.undp.orz.mz/en/content/download/1585/7122/file/%E2%80%8E Concerned Citizens of Abra for Good Government (CCAGG), undated. Accessed May 15, 2012 at http://www.ccagg.com/index.php?option=com content&view=article&id=62:based-project- monitors-of-abra-philippines&catid=1:latest-news De Renzio, Paolo and Joseph Hanlon. 2007. Contested Sovereignty in Mozambique: The Dilemmas of Aid Dependence. GEG Working Paper 2007/25. Oxford: Oxford University. Accessed 15 October, 2012 at http://www.globaleconomicgovemance.org/wp- content/uploads/Derenzio%20and%2OHanlon Mozambique%20paper%20rev%20120107.pdf Handley, Geoff. undated. "Mutual Accountability at the Country Level: Mozambique Country Case Study". London: Overseas Development Institute. Accessed 7 September, 2012 at http://www.oecd.org/dataoecd/52/35/41178552.pdf Independent Evaluation Group. 2010. Mozambique Country Program Evaluation. Washington: D.C.: World Bank. Lawson, Andrew, Paolo de Renzio and Mariam Umarji. 2006. Assessment ofPublic Finance Management in Mozambique 2004/05 Based On PEFA Methodology - Final Report Current status of PFM systems & processes, overview of reforms and perspectives for 2006. Maputu and London: SAL Consultoria e Investimentos Lda and Overseas Development Institute. Accessed 11 January, 2013 at http://www.pefa.org/en/assessment results?country value%5B%5D=MZ Lawson, Andrew, Mariam Umarji, Jodo Guilherme, and C61ia Chachine. 2008. Assessment ofPublic Finance Management in Mozambique 2006 - Based On PEFA Methodology - Public Expenditure & Financial Accountability. Foma; Report, English Version Maputu and London: Fiscus Public Finance Consultants, Ltd. And SAL Consultoria e Investimentos Lda. Accessed 11 January, 2013 at http://www.pefa.org/en/assessment results?country value%5B%5D=MZ Makgetia, Tumi. 2010. Innovations for Successful Societies: Embracing the Power of Tradition: Decentralization in Mozambique, 1992-2000. Princeton. http://www.princeton.edu/successfulsocieties/content/data/policy note/PN idl41/Policy Note ID 141.pdf Mansuri, G. and V. Rao. 2004. "Community-based and -driven Development: A Critical Review", World Bank Research Observer 19 (1): 1-39. Pyndt, Holger and Jesper Steffensen. 2005. Effective states and engaged societies: capacity development for growth, service delivery, empowerment and security in Africa: World Bank Review of selected experiences with donor support to decentralization in East Africa. Copenhagen: Local Government Denmark, International Consultancy Division. Republic of Mozambique. 2001. Action Plan for the Reduction of Absolute Poverty (2001-2005) (PARPA). Maputo: Government of Mozambique. Accessed 7 September, 2012 at http://siteresources.worldbank.org/INTPRS 1/Resources/Country-Papers-and- JSAs/Mozambique PRSP.pdf .2004. Constitution of the Republic ofMozambique. Maputo: Mozlegal, Lda. Accessed 7 September, 2012 at 26 http://confmder.richmond.edu/admin/docs/Constitution (in force 21 01 05)(English)- Mozlegal.pdf 2010. Poverty and Wellbeing In Mozambique: Third National Poverty Assessment. Maputo: Ministry of Planning and Development National Directorate of Studies and Policy Analysis. 2012a. Relat6rio De Execuglio Do Orgamento Do Estado. Maputo: Ministry of Finance. 2012b. Cenario Fiscal De M dio Prazo 2013-2015. Maputo: Minist6rio Da Planificaqao E Desenvolvimento, Minist6rio Das Finanqas Republic of Mozambique, Intemational Food Policy Research Institute (IFPRI), and Purdue University. 2004.Poverty and Well-Being in Mozambique: The Second National Assessment (2002-2003), Maputo. Directorate of Planning and Budget, Ministry of Planning and Finance Economic Research Bureau, Ministry of Planning and Finance, Intemational Food Policy Research Institute, Purdue University. http://www.sarpn.org/documents/d0000777/P880- Mozambique_032004.pdf Serrano, Rodrigo. 2002. Participation, Transparency, and Downward Accountability in District Planning in Mozambique. Washington, D.C.: World Bank Water and Urban Group 1, Eastem and Southem Africa. Serrano-Berthet, Rodrigo, Louis Helling, Julie Van Domelen, Warren Van Wicklin, Dan Owen, Maria Poli, and Ravindra Cherukupalli, 2010. Lessons from Africa. In Hans P. Binswanger-Mkhize, Jacomina P. de Regt, Stephen Spector, eds. Local and Community Driven Development - New Frontiers Of Social Policy - Moving to Scale in Theory and Practice. Washington, DC: World Bank: 121-151. http://www- wds.worldbank.org/extemal/default/WDSContentServer/WDSP/IB/2010/03/05/000333037 20100 305000306/Rendered/PDF/533000PUBOcommlB1Official0Use0Onlyl.pdf Thomsen, Thomas and Alda Saide. 2011. Development Of Case Studies And Preparation Of A Paper On How To Improve Aid Effectiveness In The Field Of Decentralisation And Local Govemance- Final Report - Case Study Mozambique. Maputo: IBF Intemational Consulting and European Union. World Bank. 2000. World Development Report, 2000-1. Washingon, D.C.: World Bank. 2003a. "Project Appraisal Document on a proposed Grant from the Intemational Development Association in the amount of SDR 29.9 Million (US$42.0 Million Equivalent) to the Republic of Mozambique for Decentralized Planning And Financing Project". Report No: 26365 - MOZ. Washington, DC: World Bank. 2003b. "Development Grant Agreement, Decentralized Planning And Financing Project". Washington, DC: World Bank. 2006. "Decentralized Planning And Financing Project - Joint Review Mid Term Review Mission Final Draft AideM6moire Post Wrap up". Washington, D.C.: World Bank. 2009. "Implementation Completion and Results Report on a Grant in the Amount of SDR 29.9 million (US$42.0 million equivalent) to the Republic of Mozambique for the Decentralized Planning and Finance Project". Washington, D.C.: World Bank. . 2011. Implementation Status & Results Mozambique Strengthening M&E for PARPA III (P116352). Washington, DC: World Bank. 27 ANNEX A Annex A. Basic Data Sheet The Republic of Mozambique Decentralized Planning and finance Project - P001807 (LOAN IDA-HO670) Key Project Data (amounts in US$ million) Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 46.29 48.68 105 Loan amount 42.00 45.17 108 Cofinancing Cancellation Cumulative Estimated and Actual Disbursements FY04 FY05 FY06 FY07 FY08 FY09 FY10 Appraisal estimate (US$M) 7.45 13.95 22.15 32.55 42.00 42.00 42.00 Actual (US$M) 2.67 7.68 14.20 25.53 40.13 45.19 45.17 Actual as % of appraisal 35.84 55.05 64.11 78.43 95.55 107.60 107.55 Date of final disbursement: July 2009 Project Dates Original Actual Initiating memorandum 10/15/1999 10/21/2002 Negotiations 07/23/2003 07/23/2003 Board approval 0712/2001 11/20/2003 Signing 12/19/2003 12/19/2003 Effectiveness 02/16/2004 04/16/2004 Closing date 06/30/2008 03/31/2009 ANNEX A 28 Staff Inputs (staff weeks) Staff Time and Cost (Bank Budget Only) Stage of Project Cycle US$ thousands (including travel and consultant costs) LENDING FY00 18.63 96.5 FYO1 18.24 124.8 FY02 40.34 186.6 FY03 41.28 316.6 Total: 118.49 724.5 SUPERVISION/ICR FY04 25.23 130.2 FY05 25.39 157.3 FY06 23.62 153.7 FY07 64.32 237.2 FY08 56.30 209.5 FY09 30.72 49.3 Total: 225.58 937.2 Task Team Members Names Title Unit 1esponsibilil Specialty LENDING Lance Morell Team Leader AFTU1 Local government Alan Corell Team Leader Decentralization Kate Kuper Urban Specialist TUDUR Urban and local government Aniceto Bila Operations Officer AFCO2 Project quality Joao Tinga Financial Management FTFM Financial management Analyst Marius Koen Senior Financial Management Financial management Specialist Subhash Dhingra Senior Procurement Specialist AFTPC Procurement Natalino Nascimento Consultant AFTU1 Civil engineering Lauren Wojtyla Financial Specialist AFTFM Local government finance Louis Helling Decentralization Specialist AFTU1 Decentralization Alberto Ninio Senior Counsel LEGAF Legal agreement John Hatton Environmental Specialist AFTEN Environmental safeguards Bridie Champion Disbursement Officer LOA Disbursements Lu Kang Program Assistant AFTU1 Program assistance Isabel Nhassengo Procurement Assistant AFTPC Procurement assistance Rildo Santos Team Assistant AFTU1 Team assistance Supervision/ICR Lance Morell Team Leader Local government Kate Kuper Senior Urban Specialist AFTUW Urban and local government 29 ANNEX A Ali Alwahti Urban Specialist AFTUW Urban and local government Boris Utria Sector Leader AFTEG Design and implementation Antonio Chamuco Procurement Specialist AFTPC Procurement Theresa Marissa J. Procurement Analyst AFTUW Procurement Gamulo Isabel Nhassengo- Procurement Assistant AFCS2 Procurement Assistance Massingue Cary Anne Cadman Forestry Specialist AFTEN Environment Jonathan Nyamukapa Senior Financial Management AFTFM Financial management Specialist Joao Tinga Financial Management FTFM Financial management Analyst Suzanne Morris Senior Finance Officer CTRFC Financial management Brighton Musungwa Senior Finance Officer CTRLP Financial management Teresa De Jesus S. Finance Analyst CTRDM Financial management McCue Uri Raich Urban Specialist AFTUW Urban and local government Caroline Winter Senior Social Development AFTCS Social development Specialist Arbi Ben-Achour Senior Social Development AFTCS Social development Specialist Serigne Omar Fye Senior Environmental AFTEN Environmental issues Specialist Victor Vergara Capacity Building Specialist WBI Capacity building Adelina Mucavele Team Assistant AFCS2 Team assistance Salma Chande Team Assistant AFCS2 Team assistance Rafael Saute Communications Officer AFREX Communications Lurdes Samuel Malate Executive Assistant Team assistance Amaral Anne Louise Grinsted Extended term Consultant AFCS2 Project quality Slaheddine Ben- Consultant AFTUW Project implementation Halima Natalino Nascimento Consultant AFTUW Civil engineering Jose Macamo Consultant AFTUW Public sector issues Mafalda Duarte Consultant AFTUW Capacity building Wendy Ayres Consultant AFTUW M&E and Implementation and Results Report ANNEX A 30 Other Project Data Borrower/Executing Agency: Follow-on Operations Operation Credit no. Amount Board date (US$ million) 31 Annex B. Mozambique Decentralization timeline Year National political, Institutional capacity building World Bank and other donors constitutional, legal events 1990 Constitution (provincial governments) 1992 Peace Accord concluding 16 year civil war signed by the leaders of Frente de Libertago de Moambique (FRELIMO) and the Resistencia Nacional de Moambique (RENAMO); Law no 5/92, of 6 May, which defines the role of the Tribunal Administrativo 1993 Rural Rehabilitation Project, approved 1993, closed 2000: Pilot experiences with participatory planning and decentralized small works management in Sofala and Zambezia Provinces; Capacity Building Human Development Project, approved 1993, closed 2001. Expanded quantity and improved quality of university graduates, and enhanced the quality of upper secondary education. 1994 first multiparty election for president and parliament 1995 First piloting of 5 year district strategic and development plans: Plano Estrategico de Desenvolvimento Distrital (PEDD). 1996 Local authority law approved ANNEX B 32 Year National political, Institutional capacity building World Bank and other donors constitutional, legal events (Law 9/96) 1997 Municipal Legislation (Pacote Autairquico) created 33 urban municipalities with full financial and political autonomy. (Law 2/97 and 11/97 ) The law sets out clear criteria for determining the distribution of the two annual transfers that municipalities receive from central government, namely the Fundo de Compensaqio Autarquica (FCA) to cover recurrent expenditure and the Fundo de Investimento de Iniciativa Local (FIIL) for investment expenditure. It also indicates that overall transfers should reach 1.5-3 percent of domestic revenue. Distribution of transfers should be based on four criteria: population, area, revenue collection, and degree of economic development. In this sense, the system is transparent and rules-based. Score for PEFA PI-8 on intergovernmental transfers is C+, referring only to practices concerning municipalities based on assessments in 2004 33 ANNEX B Year National political, Institutional capacity building World Bank and other donors constitutional, legal events and 2006 (Lawson and others, 2006: 23-4 and Lawson and others, 2008: 40-1); also, Law no 13/97, of 10 July, which establishes the regime for the audit and inspection of public expenditures by the Tribunal Administrativ 1998 Decentralization of rural water National Water Development Project, approved 1998, supply provision through closed 2005: reform and capacity building Community participation and beneficiary responsibility through demand driven approach to water supply provision 1999 multiparty election for president Education Sector Strategic Program Project, approved and parliament 1999, closed 2006: Decentralized management of the delivery process Greater role for communities in decision making process in the schools Increased community roles in planning and implementation of school construction with more sustainable construction technology 2000 Decree 15/2000 legally Coastal and Marine Biodiversity Management recognized Community Project, approved 2000, closed 2007: Adoption of Authorities as interlocutors strategic spatial development plans by national, between rural communities and provincial and district authorities in pilot areas district administration. ANNEX B 34 Year National political, Institutional capacity building World Bank and other donors constitutional, legal events 2001 Action Plan for the Reduction Municipal Development Project, approved 2001, of Absolute Poverty (PARPA or closed 2007: Pilot municipal grant system for local PRSP), that identified investment planning, financing and management; decentralized planning and capacity building for municipal officials; finance as a key element of its decentralization policy and regulatory development program in rural areas, adopted for municipalities. Decentralized maintenance of by Council of Ministers. tertiary and non-classified roads Increased district responsibility and community participation in non- classified road maintenance and spot improvements Roads and Bridges Management and Maintenance Program Project (Phase I) , approved 2001, closed 2007; Phase II approved 2077 and ongoing Rural Travel and Transport Program, (regional program) Mozambique technical workshop 2001 see http://www4.worldbank.org/afr/ssatp/Resources/SSATP- WorkingPapers/SSATPWP61.pdf 2002 Global Strategy for Public Sector Reform identified decentralization as a key area of action. Law no 9/2002, of 12 February, which approves the State Financial Administration System (Sistema de Administraqdo Financeira do Estado - SISTAFE) 2003 Local Administration Reform 2003 DPFP National Steering Public Sector Reform Project, approved 2003, closed (Lei das Org6as Locais do Committee constituted DPFP 2009; see other PPAR Estada - LOLE) legislation, National Steering Committee 35 ANNEX B Year National political, Institutional capacity building World Bank and other donors constitutional, legal events inter alia makes district a budget (Comit6 de Supervis5o or CS) DPFP, approved 2003; 2008 amendment to the entity, and emphasizes the role comprised of National Directors legal agreement to allow for 100 percent Bank of districts as development from MPF, the Ministry of State financing of all project activities, because poles, and the foundation for Administration (MAE), the government couldn't provide counterpart funds; Mozambique's planning system. Ministry o f Agriculture and Rural closed 2009. (law 08/03). development (MADER) and the Guidelines for community Ministry of Public Works and participation and consultation in Housing (MOPH).. district planning published in the official gazette. These provide for the composition and operation of consultative councils. Due to these guidelines, all consultative councils include at least 30 percent women, and representatives of vulnerable and disadvantaged groups and of geographically remote communities. 2004 Decree nt 23/2004, of 20 August, which approves the SISTAFE Regulations; linked to Law 9/2002. Constitution (provincial assemblies). Election. 2005 Regulations (decrees 3 networks for community 11/2005 and 13/2005) to participation established. implement the 2003 OLE Adoption of 13 standardized law. Procurement Law manuals for district (Contracts for Public Works, participatory planning covering Supply of Goods and issues such as district ANNEX B 36 Year National political, Institutional capacity building World Bank and other donors constitutional, legal events Services for the State - diagnostics, preparation of Decree 54/2005 of 13 development strategies and December) Inter alia, action plans, production of documents for all competitive district economic and social plan contracts must be submitted and budget, monitoring and to the Administrative Court evaluation, and organization and wtoi okn asfo functioning of local consultative within 5 working days fromGuidelines for the date of award, which can preparation of the district land-use either approve or request plan prepared. additional information. 2006 Procurement Regulation District Performance 2006 PEFA based on 2004 assessment (Lawson and adopted in July to implement Assessment System. Some 18 others, 2006) Decree 54/2005. land-use plans prepared. The work included diagnosis of the present situation, production of thematic maps of current land use, administrative divisions, population, infrastructure and facilities, physical conditions, and trends in settlement expansion. Communication strategy formulated. Audits of 48 districts' financial statements conducted, the first audits of district finances in the country's history. Audit manual elaborated. First direct budget transfers to the districts, replacing previous discretionary allocation of funds by provinces. Provincial budget allocations to districts were established for 37 ANNEX B Year National political, Institutional capacity building World Bank and other donors constitutional, legal events district infrastructure maintenance through budget instructions to the line ministries (education, health, water, and the like). 2008 Operational rules and 2008 PEFA based on 2006 assessment (Lawson and procedures for district finance others, 2008) were published 2008 Intergovernmental fiscal relations study (Boex and others). This describes the current situation and lays out options for transferring responsibilities from the central level and the provincial levels to the district level. The study has informed the development of the government's decentralization strategy. 2009 Election, including first election All 56 districts participating in 2009 Analysis of Experiences Relating to Community of provincial assemblies. DPFP had prepared Participation and Consultation in District Planning in participatory district Mozambique, completed. This provides an in-depth development plans and annual assessment of the mechanisms for community investment plans approved by participation established under the project in the district development councils. All 128 districts in Mozambique North, et, a t nigIn assa had adopted the DPFP approach by the end of the project. All the Gaza Provinces). districts are using the project- supported strategic planning 2009 Analysis of District Finances in Mozambique and annual planning and completed. This analyzes the current situation of budgeting approaches. district finances in Mozambique and proposes Government decided to increase improvements based on theoretical considerations the proportion of the state and international good budget allocated for financing ofi district plans, while keeping Strengthening M&E for PARPA c11. Approved 2009, ANNEX B 38 Year National political, Institutional capacity building World Bank and other donors constitutional, legal events funds for productive closed 2013. To strengthen the Government of investments flat, indicating a Mozambique's capacity to coordinate the overall M&E commitment to supporting framework for the upcoming third Poverty district capacity to plan and Reduction Strategy (referred to as PARPA) and, manage small infrastructure particularly, to strengthen the results orientation at sector investments. level. Source: World Bank documents. Bolded entries supported by project. 39 Annex C. Poverty in Mozambique The DPFP was designed to support the Government's PARPA, and its commitment to greater citizen participation in providing poverty reducing infrastructure and public services at the local level. Thus it is important to examine what happened to poverty in the provinces supported over the life of the project, and whether any change in poverty levels can be attributed to project interventions. At the time of appraisal, an estimated 54 percent of Mozambique's population lived in absolute poverty6, caused by: "(i) slow growth of the economy until the beginning of the 1990s; (ii) low levels of education of working age household members, particularly women; (iii) high dependency rates in households; (iv) low productivity in the family agriculture sector; (v) lack of employment opportunities within and outside of the agricultural sector; and (vi) poor infrastructure, especially in rural areas." The poor were also vulnerable to natural disasters (floods) and economic shocks (agricultural price declines) (Republic of Mozambique, 2001: 2). Table C.1. Poverty Headcount, 1996-2009 Diece. % points .h% 1996-97 to 2002-03 to 1996-97 2002-03 2008-09 2002-03 2008-09 Narnl 694 54.1 54i7 -153 06 Uban 62 515 496 -105 -1.9 Rural 71.3 $5.3 56.9 -16 1.6 North 663 553 46.5 -11 -8.8 Center 738 45.5 59.7 -28.3 14.2 S0tth 65.8 6.5 5.9 0.7 -9.6 Nassa 706 521 319 -185 -20. Cabo DeI_wdo 57.4 63.2 37.4 5.8 -25.8 Nanpula 68.9 52.6 54.7 -16.3 2.1 Zambez681 44.6 70 5 -23.5 25 9 Tere 82.3 59.8 410 -225 -17,8 Mamine 62.6 43.6 55.1 -19 11.5 solfah 87.9 36.1 5-8.0 -51L8 21 9 Inhlnkbanw 826 80.7 579 -1.9 -I7.8 Gam 64.6 60.1 62.5 -4.5 2.4 Mapito Prollnce 65.6 69.3 67.5 3.7 -1.8 Mapito Cry 478 53.6 36.2 5.8 -17.4 Disperioin. Provnres 117 12.6 13.2 12 18.3 Corretion coeftient.r: ConeitrivP strnwys -0.001 -0.006 -0.633 Iitul ket auxd chns -0 6834 -0 668 Destiarn kvelary ciuLge 0.731 0.726 Noteg. D ospe f the poverty rates and dages in ponty rates is rMasured by the standard deraion of the prouneI rates and chages in ates across provinces. The conelauion coeffiient for consecutne ureys shows tw corlations for eah coln with the pxeeding columi Initial Invl and cungt dhow he corrlataw stariag wth the t6i levels cott=n an the fist rates colmn Destinatio lcvd and change shows conrrelAo starting with the second levels colum and the firm rates colum. All correlaions are performed on province Source: Government of Mozambique, 2010: 26. 6 Using flexible bundle approach, which takes changes in relative prices since 1996-97 into consideration (Government of Mozambique and others 2004). ANNEX C 40 7.4 Looking at what happened to poverty in the target provinces prior to appraisal, Zambezia, Tete and Sofala were provinces showing the largest reduction in poverty between two household surveys in 1996-7 and 2002-3. Reasons included unusually high poverty levels at the starting point due to flooding and the impact of previous civil conflict, and favorable agricultural production and prices in 2002-3. 7.5 Poverty is affected by many factors, so the operation rightfully did not target poverty reduction as a PDO. Never the less, the logic of project design is that the outputs supported are: "...central to the growth strategy to reduce poverty." (World Bank, 2003a: 2) Thus, it is important to understand what the poverty trends were, and what effect the project might have had on them. Based on data from 2008-9, poverty decreased from levels in 2002-3 in only two of the five provinces targeted: in Tete (from 60 percent to 42 percent), and Maputo Province (from 70 to 68 percent), see Table D. 1. Poverty increased in the other three, with two provinces posting among the highest increases in poverty of all provinces: a 26 percentage point increase in Zambezia and 22 percentage point increase in Sofala, and Manica showing a 12 percent increase. The results in other provinces receiving decentralization support from other donors were also mixed, with a 2.1 percent poverty increase in Nampula province, and a 23% reduction in poverty in Inhambane. By comparison, there was a 1.6 percent average increase in poverty nationwide in rural areas over the same period. 7.6 These data suggest that changes in poverty result from other factors than the early stage decentralization support provided by the Bank and other donors. For example, while it is difficult to pinpoint the reasons for the steep increases in poverty in Zambezia and Sofala, one factor could be declines in agricultural production. Survey data indicates that per capita production of food crops was 32 percent lower in Sofala measured by calorie values, and 19 percent lower in Zambezia in 2008 compared to 2003. This is likely to be connected to more severe climate shocks (for example, flooding in late 2007), along with low rates of access to agricultural technologies such as fertilizers, more extreme agricultural price variability, and greater severity of the HIV/AIDS epidemic in the central region than in other parts of the country (Republic of Mozambique, 2010: 74-77). The sharp reduction in poverty in Tete was likely influenced by overall economic trends driven by rapid expansion of coal production during the period. 7.7 However, the ambitious targets of the Government's development strategy such as reducing poverty will not be achieved without much larger resource flows to support education, health, rural roads and other essential public goods. 41 Annex D: Elements of Intergovernmental design supported by DPFP 7.8 Table E. 1 summarizes the elements of intergovernmental design supported by the Bank, through this and other projects, and the value added over the project life. Many of these elements were also supported by the Bank and other donors in other ways, including investment projects, technical assistance, common funds, and advisory work. Table E.1. Elements of Intergovernmental Design Supported by the Bank and in Which Sequence Elements of Assessed at project initiation Value added over project life Intergovernmental Design Enabling legislation: 1997 municipal legislation Supported implementation of Clearly mandated spheres constitutionally limited to cities laws: LOLE (2003 and 2005), the of government and towns; PARPA that identified State Financial Administration Viable and sustained decentralized planning and System (SISTAFE 2002, 2004), jurisdictions finance as a key element of its and procurement (2005,2006) Expenditure and revenue program in rural areas; assignments 200 uiGlobal Strategy for Public Sector Reform identified decentralization as a key area of action. In 2000, "community authorities" legally recognized as interlocutors between rural communities and district administration. Laws on Local Organs of State (LOLE 2003) Intergovernmental fiscal The intergovernmental fiscal Starting in 2006, districts started design to pay for the (i) system assigned a portion of the receiving direct fiscal transfers one-off transitional costs of annual state for recurrent and capital creating local governments budget (OE) to provinces, expenditures, replacing previous and (ii) the recurrent costs Recurrent expenditures to districts discretionary allocation from of service delivery and econ were determined at provincial provinces. LOLE requires a five- dev't through: level as part of the provincial year strategic and development Optimized own revenues sector budgets. The district budget plan, called Plano Estrat6gico de Transfers had no legal standing. Thus, Desenvolvimento Distrital Access to credit and districts did not manage the (PEDD) for each district. Some borrowing capital budget nor the of the pilot districts have evolved civil service component of the beyond their first plan and are recurrent budget. on second or third generation strategic plans. These plans, together with provincial and central level sector and strategic plans provide the basis for annual planning and budgeting via an operational plan and budget known as the Plano Econ6mico SSocial e Oramento Distrital ANNEX D 42 Elements of termen D g Assessed at project initiation Value added over project life Intergovernmental Design (PESOD). Cross-cutting resource FM: high fiduciary risks: material FM: PEFA indicators show management systems: receipts and payments are improvement from 2006 to 2008 Public financial excluded from the budget and & 2010 as the PFM reforms led management (both revenue from Government accounting and to successful improvements. and expenditure) reporting system; accounting Mozambique is now among the Human resource systems and standards are highest performing developing management (see below) outmoded; internal and external countries of Africa in terms of auditing require substantial average PEFA score. However, support; and parliamentary there are gaps between policies, oversight requires strengthening, laws and regulations and their implementation, particularly at the subnational level with weak interministerial coordination, and resistance to change. In some critical areas re sectors, performance has gotten worse, for example, P-2 from B to D. Administrative structures: Very weak capacity in target All target districts have PEDD Local government districts, but government and PESOD. 2/3 rated good or structures and commitment to staff District excellent on local investment administrative cadres Accountant and District Works grant performance. All districts Reoriented line ministries, Technicians posts. Local in the country have adopted agencies, and departments contracting industry capacity also DPFP approaches for very weak (re infrastructure sub- participatory planning. Sofala projects) increased number of sr. staff from 242 to 292 (supported by PSRP). Ms Health, Education and Pub Works have reasonable capacity at prov level with support from SWAPs; Agriculture and other ministries do not. Accountabilities Weak procurement, audit and e-SISTAFE rolled out, increased Vertical (to finance/sector internal controls on FM coverage of internal and external ministries) vs. horizontal Link to community authorities and audit. However sectoral (communities) PEDD piloted since 1995: limitations in health controls and UNCDF and GTZ support procurement (cf Health PEFA), education (teacher absenteeism), agriculture and justice. Consultative councils rolled out to support planning, but limits on civil society monitoring of managerial weaknesses. Provincial assemblies elected from beginning 2009 mainly to discuss plans and budgets for previous and coming years. iConsultative committees at 43 ANNEX D Elements of Assessed at project initiation Value added over project life Intergovernmental Design village level (localidade), intermediate level (posto administrativo) and district level, but weak capacity. Source: Bank documents 44 ANNEX E Annex E. List of Persons Interviewed In Washington, DC (including teleconferences) Alwahti, Ali Y. DPFP TTL Carroll, Allan - DPFP TTL prior to appraisal Morrell, Lance -DPFP TTL at appraisal Kuper, Kate - DPFP TTL Helling, Louis - Consultant Boex, Jamie, Urban Institute Pomerance, Phyllis - former CD until 2000, Zambia Mozambique. Now at Duke van Holst Pellekaan, Jack - Consultant, member of Mozambique CAE team In Mozambique Government Alberto, Domingo's Calicos - technician Nhamatanda District, Sofala Province Alberto, Manuel Rodrigues - National Director, National Directorate of Municipal Development, Ministry of State Administrateion Alfane, Rocha Amborete - Deputy Director Finance and Planning, Sofala Province (was in Zambezia Province during DPFP) Antonio, Rui - Provincial Technical Team, Sofala Province Banze, Olegario dos Angos - National Deputy - Director, National Directorate for Promotion of Rural Development, Ministry of State Administration Belessone, Ana - District Administrator, Tsangano District, Tete Province Cardozo, Agostinho Manuel Maria - technician and manager of DPFP since 2006, Chiuta District, Tete Province Carlos, Tenday - Provincial Technical Team, Sofala Province Coanai, Miguel Taembera - Director of Maintenance, National Road Administration Costa, Ana Virgilio da silva - Permanent Secretary, Nhamatanda District, Sofala Province Diaz, Chimonio - Provincial Technical Team, Sofala Province Erskog, Hans - Senior Financial Analyst, Ministry of Environmental Coordination Fonseca, Maria de Lurdes de - Provincial Director, Provincial Directorate of Planning and Finance, Tete Province Francisco Saize Chiria, district technician for buildings, joined 2005, Chiuta District, Tete Province Gideon, Claudio - Nhamatanda District, Sofala Province Grachane, Cecilio - Director General, National Road Administration Head, Gabriel George. - Head, Administrative post of Thika, Nhamatanda District, Sofala Province Jutha, Momad Piaraly - National Director, National Directorate of Planning, Ministry of Planning and Development Lambo, Domingos - National Director, National Director of Social Welfare, Ministry of Finance Lemane, Alfredo - Provincial Technical Team, Sofala Province 45 ANNEX E Lubrino, Grilo - National Director of Studies and Administrative Procedures, Ministry of Public Service Mafunga, Adeline - Permanent Secretary, Chiuta District, Tete Province Majacunene, Paolo - District Administrator, Gorongosa District, Sofala Province Makeringi, Helder - Team Leader, NDPFP Tete Maleji, Rimundo- technician Nhamatanda District, Sofala Province Matavel, Antonio Augusto - provincial coordinator, Sofala decentralization Program GIZ Moiana, Sergio Sional - District Administrator, Nhamatanda District, Sofala Province Motsinhe - Civil Engineer, National Directorate of Buildings, Ministry of Public Works and Housing Mucudos, Custudio dos - Head NDPFP Project Implementation Unit, National Directorate of Planning, Ministry of Planning and Development Nhahingo, Antonio, Vice President of the Tete Provincial Assembly Nhmatande, District Administrator, Nhamatanda District, Sofala Province Nyangwni, Raphael M - Education Director Nhamatanda District, Sofala Province Pereira, Herminio - Trainer, Tete Provincial Directorate Sakut, Ibrahim Abdul - accounts specialist, Ministry of State Administration Soca, Brito Antonio - Civil Engineer, National Directorate of Buildings, Ministry of Public Works and Housing Somane, Elisa Arissone G. - Provincial Permanent Secretary, Provincial Secretariat, Province of Sofala Tui, Atalia - Director Finance and Planning, Sofala Province Vasquez, Ivan - Chief Technical Advisor, National Directorate for Promotion of Rural Development, Ministry of State Administration Victor, Armando - Technician, Gorongosa District, Sofala Province Vinfani-Tecuico, Smacio Jose- technician Nhamatanda District, Sofala Province Yemela, Chadreque - Nhamatanda District, Sofala Province Zambeze, Francisca Alberto - Permanent Secretary, Changara District, Tete Province Civil Society and Private Sector Aefimar, Antonio Augusto - Nhamatanda District Council, Sofala Province Allmeida, Fernando Jose Martins Ruas - Businessman and recipient of loan from "productive economic" grant, Changara District, Tete Province Antonio, Jose - Nhamatanda District Council, Sofala Province Antonio, Jusand - Nhamatanda District Council, Sofala Province Armando, Adeline - District Council member since 2003, Gorongosa District, Sofala Province Banze, Felix A.- Executive Director, APDCOMA Barry, Manuel Oaulo - Nhamatanda District Council, Sofala Province Dalusa, Joaozinho - Former Chair, Village Council, Gorongosa District, Sofala Province Dava, Lourino - Program Coordinator, Center for Public Integrity Faekm Baltazar Jorge - Researcher, Center for Public Integrity Jacob, Deolindo - Former member, District Council Gorongosa District, Sofala Province Jenmee, Antinio - Nhamatanda District Council, Sofala Province Jenmee, Antinio - Nhamatanda District Council, Sofala Province ANNEX E 46 Khanda, Celestine - District Council Gorongosa District, Sofala Province Mabunda, Antonio Albino - Managing Director, AM International Procurement Macuane, Jose - Political analyst, UNDP and Dfid consultant Manuel, Ibrahim - District Council Gorongosa District, Sofala Province Munhequete, Aida - Interpreter Nkamate, Salvador - Liga Moqambicana dos Direitos Humanos (Mozambican Human Rights League) http://www.ldh.org.mz Osman, Magid - Owner, Lex Mozambique (former Finance Minister) Quenece, Alfredo - Nhamatanda District Council, Sofala Province Seboural, Jeremia - Nhamatanda District Council, Sofala Province Solomon, Nhamatanda District Council, Sofala Province Development Partners Antonopoulou, Fotini - Attache, European Union Barnes, John - Technical Advisor, UNDP Bronselaer, Dirk - Sr. Procurement Specialist, World Bank Bruschi, Francesca - Policy Analyst, Italian Cooperation Chamuco, Antonio L. - Sr. Procurement Specialist, World Bank Figueiredo, Carlos Mauricio Cabral - Expert on Supreme Audit Institutions, Decentralization Program, GIZ Hawkins, Peter - Sr. Water and Sanitation Specialist, World Bank Imparato, Ivo - TTL, National Decentralized Planning and Finance Program Ising, Josef - Program Director, Decentralization Program, GIZ Karlsen, Anders B. - Head of Cooperation, Royal Danish Embassy Lefebvre, Anne-Lucie - Sr. Public Sector Specialist, World Bank Malate, Amos - Procurement Analyst, World Bank Mason, Jocelyn T. - Country Director, UNDP Raich, Uri - TTL Mozambique Urbanization and Municipal Development Project and NDPFP (successor to Imparato) Saleem, Furqan Ahmad - Sr. Financial Management Specialist, World Bank Ulens, Wim - Attache, European Union Verissimo, Patrick - Sr. Sector Economist, World Bank Vilissa, Datto - Analyst, Royal Danish Embassy