48395 2005 Review of the PRS Approach: Balancing Accountabilities and Scaling Up Results THE WORLD BANK AND THE INTERNATIONAL MONETARY FUND September 2005 This review sought to elicit and reflects on the views of wide range of stakeholders. Consultations with officials in PRS countries took place in the context of a range of international events, such as the Special Partnership for Africa Plenary Meeting held in January in South Africa, a PRSP Forum for the Western Balkans held in May in Greece, a regional Conference on Engaging Citizens for Enhanced Accountability held in May in Ghana, and a regional Seminar on Voice and Accountability held in June in the Dominican Republic. In addition, a focus group meeting with officials from PRS countries was conducted in Paris in March 2005. The review draws on the work of various initiatives and associated meetings, including on alignment and harmonization and the associated March 2005 High Level Forum in Paris; and development effectiveness in fragile states and the associated January 2005 Senior Level Forum in London. Bilateral and multilateral donors, as well as other partners, were invited to make contributions, and a host of discussions have been held with multilateral and bilateral agencies, including UN agencies based in Geneva, New York and Rome and the OECD-DAC. In addition to a focus group discussions with Civil Society Organizations (CSOs) in Brussels, the Global Civil Society Form brought together a broader group of stakeholders, including southern- and northern-based CSOs, officials from PRS countries and donors, to discussion issues related to the key themes of this review. An e-discussion was conducted following this event. An extensive literature review was undertaken. Direct links to country PRS documents, more than one hundred recent studies, numerous additional references though links to websites of other partners and on related topics, and specific contributions by CSOs and bilateral and multilateral agencies are available at the PRS 2005 Review webpage (www.worldbank.org/prspreview). 2005 Review of the Poverty Reduction Strategy Approach: Balancing Accountabilities and Scaling Up Results Prepared by the Staffs of the World Bank and International Monetary Fund September 2005 CONTENTS: PROLOGUE--PRSP A to Z .............................................................................................. v Albania: Linking the PRS to budget processes............................................................... v Zambia: Rebalancing the motivation for engaging in a PRS process..........................viii 1. INTRODUCTION .......................................................................................................... 1 A. Overview.................................................................................................................... 1 B. 2005 PRS review........................................................................................................ 2 C. Development challenges in low-income countries..................................................... 4 2. BALANCING ACCOUNTABILITIES........................................................................ 10 A. Overview.................................................................................................................. 10 B. Domestic accountability........................................................................................... 11 Elements supportive of evidence-based decision making ................................... 12 An enabling environment for strong accountability ............................................ 26 Some special considerations in conflict-affected and fragile states..................... 35 C. External aspects of accountability............................................................................ 36 Donor accountability and aid quality................................................................... 37 D. Balancing the accountability framework ................................................................. 42 Factors that tilt the accountability balance........................................................... 43 How the PRS architecture can support a balance ................................................ 46 3. SCALING UP RESULTS............................................................................................. 48 A. Overview.................................................................................................................. 48 B. Analytical foundations for scaling up results........................................................... 49 Poverty analysis ................................................................................................... 50 Growth analysis ................................................................................................... 55 C. Supporting ambitious development plans ................................................................ 59 PRSs as a framework for scaling up .................................................................... 59 Dealing with macroeconomic impacts of scaling up aid ..................................... 61 Addressing capacity constraints........................................................................... 63 4. SUMMARY OF FINDINGS AND GOOD PRACTICES ........................................... 66 A. Overview.................................................................................................................. 66 B. Five themes .............................................................................................................. 67 Strengthening the medium-term orientation of the PRS...................................... 67 Utilizing the PRS as a mutual accountability framework.................................... 69 Enhancing linkages between the PRS and the MTEF and budget....................... 70 Sustaining meaningful participation .................................................................... 71 Tailoring the approach to conflict-affected and fragile states ............................. 71 C. Sustaining support for country-based development efforts...................................... 72 ANNEX 1: PRS STATUS BY COUNTRY ..................................................................... 73 ANNEX 2: SUMMARY OF PRS IMPLEMENTATION PROGRESS REPORTS........ 74 BIBLIOGRAPHY............................................................................................................. 77 References..................................................................................................................... 77 Other background reading............................................................................................. 85 BOXES: BOX 1: PRS CORE PRINCIPLES............................................................................................. 3 BOX 2: EXPERIENCES IN CUSTOMIZING HEALTH TARGETS TO COUNTRY CIRCUMSTANCES . 13 BOX 3: LINKING OBJECTIVES, STRATEGIES AND INDICATORS--PAKISTAN ENVIRONMENT. 15 BOX 4: LINKING OBJECTIVES, STRATEGIES AND INDICATORS--KENYA ROADS.................. 16 BOX 5: CLARIFYING AND AGREEING ON PRS PRIORITIES--ARMENIA................................ 17 BOX 6: AN EMERGING EDUCATION SECTOR MTEF--VIETNAM......................................... 20 BOX 7: A GRADUAL APPROACH TO MTEF REFORM--RWANDA......................................... 20 BOX 8: STRATEGIES FOR STRENGTHENING DEMAND FOR MONITORING SYSTEMS ............... 25 BOX 9: INSTITUTIONALIZING PARTICIPATION IN PRS FORMULATION--TANZANIA ............ 27 BOX 10: INSTITUTIONALIZING PARTICIPATION IN MONITORING SYSTEMS .......................... 28 BOX 11: REACHING VULNERABLE GROUPS--MONTENEGRO.............................................. 34 BOX 12: MAINSTREAMING GENDER IN THE PRS--BANGLADESH ...................................... 34 BOX 13: CONFLICT-SENSITIVE PARTICIPATORY PROCESS--BOSNIA AND HERZEGOVINA... 36 BOX 14: THE INTERNATIONAL AID HARMONIZATION AGENDA ........................................... 37 BOX 15: THE 2004 SPA AND DAC SURVEYS OF HARMONIZATION AND ALIGNMENT......... 39 BOX 16: BANK/FUND EXPERIENCE WITH HARMONIZATION AND ALIGNMENT ................... 40 BOX 17: INNOVATIVE SUPPORT FOR POVERTY ANALYSIS ................................................... 51 BOX 18: LINKING PSIA WITH PRSPS: EXAMPLES FROM SELECTED COUNTRIES ................ 53 BOX 19: ENHANCING THE PRO-POOR FOCUS OF POLICIES/IDENTIFYING REFORM OPTIONS.. 53 BOX 20: BINDING CONSTRAINTS TO ECONOMIC GROWTH................................................... 56 BOX 21: CONNECTING POOR PEOPLE TO ECONOMIC GROWTH ............................................ 57 BOX 22: INCORPORATING DTIS RECOMMENDATIONS--CAMBODIA.................................. 59 BOX 23: INDICATIVE CLASSIFICATION OF ABSORPTIVE CONSTRAINTS................................ 61 BOX 24: PUBLIC EXPENDITURE AND FINANCIAL MANAGEMENT CAPACITY BUILDING ........ 65 BOX 25: REACHING THE MDGS: FINDINGS FROM MAMS SIMULATIONS--ETHIOPIA........ 65 FIGURES: FIGURE 1: GDP GROWTH RATES 1990-2003 ........................................................................ 5 FIGURE 2: GDP PER CAPITA GROWTH RATES 1990-2003 ..................................................... 5 FIGURE 3: CPIA SCORES (1999)........................................................................................... 7 ii FIGURE 4: CHANGES IN CPIA SCORES (1999 TO 2004) ........................................................ 7 FIGURE 5: CPIA SCORES (2004)........................................................................................... 8 FIGURE 6: CHANGES TO KEY PUBLIC SECTOR MANAGEMENT CPIA SCORES (1999-2004).... 8 FIGURE 7: KEY PUBLIC SECTOR MANAGEMENT CPIA SCORES (1999 AND 2004).................. 8 FIGURE 8: GOVERNANCE INDICATORS (2004) BY COUNTRY GROUP ..................................... 9 FIGURE 9: VARIATIONS IN GOVERNANCE INDICATORS FOR IDA COUNTRIES (2004) ............ 9 FIGURE 10: ARE PRS DEVELOPMENT GOALS LINKED TO SPECIFIC TARGETS?.................... 14 FIGURE 11: RESULT-FOCUSED MONITORING SYSTEMS ....................................................... 22 FIGURE 12: STATISTICAL CAPACITY SCORES IN IDA COUNTRIES, 2005.............................. 23 FIGURE 13: STAKEHOLDER ACCESS TO DEVELOPMENT INFORMATION ............................... 23 FIGURE 14: DEVELOPING COORDINATED COUNTRY-LEVEL MONITORING SYSTEMS............ 25 FIGURE 15: USE OF COUNTRY SYSTEMS ............................................................................. 45 FIGURE 16: AVAILABILITY OF HOUSEHOLD SURVEYS TO MEASURE POVERTY .................... 51 FIGURE 17: WORLD BANK ECONOMIC AND SECTOR REPORTS (FY 2000-2005)................ 58 TABLES: TABLE 1: CHANGES TO SELECTED HUMAN DEVELOPMENT INDICATORS............................... 6 TABLE 2: POVERTY-REDUCING EXPENDITURE...................................................................... 6 TABLE 3: PROGRESS IN ENHANCING SPECIFICITY--TANZANIA .......................................... 18 ACRONYMS: APR Annual Progress Report CAS Country Assistance Strategy CFAA Country Financial Accountability Assessment CPIA Country Policy and Institutional Assessment CSO Civil Society Organization DTIS Diagnostic Trade Integration Studies ECA Europe and Central Asia HIC High-Income Country HIPC Heavily-Indebted Poor Country IDA International Development Association IEO Independent Evaluation Office (IMF) I-PRSP Interim Poverty Reduction Strategy Paper IMF International Monetary Fund JSA Joint Staff Assessment JSAN Joint Staff Advisory Note LIC Low-Income Country LICUS Low-Income Country Under Stress MDGs Millennium Development Goals MIC Middle-Income Country MTEF Medium-term Expenditure Framework NGO Non-governmental Organization ODA Overseas Development Assistance OECD-DAC Organization for Economic Cooperation and Development-Development Assistance Committee OED Operations Evaluation Department (World Bank) PAF Performance Assessment Framework PEFA Public Expenditure and Financial Accountability PEM Public Expenditure Management iii PER Public Expenditure Review PETS Public Expenditure Tracking Surveys PFM Public Financial Management PNoWB Parliamentary Network of the World Bank PRGF Poverty Reduction and Growth Facility PRS Poverty Reduction Strategy PRSC Poverty Reduction Support Credit PRSP Poverty Reduction Strategy Paper PSIA Poverty and Social Impact Analysis SPA Strategic Partnership with Africa UNDP United Nations Development Program WTO World Trade Organization iv PROLOGUE--PRSP A TO Z The Poverty Reduction Strategy (PRS) process in every country is unique--what has worked and what has not; how the process has unfolded over time; and what the priorities are going forward. This prologue looks at particular aspects of the stories in Albania and Zambia, first and last in alphabet of countries that are implementing poverty reduction strategies. In Albania, enthusiasm for an approach that allowed for country- determined priorities resulted in strong support by the Ministry of Finance and directed early attention to the need for a close link between the PRS and budget processes. Considerable progress has been made in strengthening planning and budget processes, although challenges remain in integrating various initiatives and in better coordinating donor assistance. In Zambia, the PRS process started off driven largely by the external compact--produce an adequate Poverty Reduction Strategy Paper (PRSP) to become eligible for debt relief. However, over time, a series of factors helped reorient the process to local constituents. There are well over fifty of these stories--each unique and multifaceted, and each deserving to be heard. The diversity of country experience is rich; this prologue aims to provide a small taste of that. ALBANIA: LINKING THE PRS TO BUDGET PROCESSES Albania's experience with the PRS process is one of strong government ownership, wide consultations with sustained involvement of civil society and parliament, and substantial emphasis on the monitoring and evaluation of results. The PRS became Albania's first strategy with a strong focus on poverty reduction and significantly contributed to the alignment of policy planning and budgeting. Struggling with a poor track record of governance and service delivery, Albania was able to make progress in building capacity in the civil service and addressing incentives by improving its salary structure. But Albania's experience also illustrates the challenges of coordinating donor assistance and designing sequenced reforms that are tailored to local conditions. The Albanian Government which came into office in 2002 welcomed the PRS approach "as a process which empowered the government to prepare its own strategy rather than simply signing off on strategies prepared by various donors."1 At the same time, the authorities recognized that shortcomings in the budget management process would impede effective implementation of its Growth and Poverty Reduction Strategy (NSSED).2 A medium-term expenditure framework--called the medium-term budget program (MTBP)--was needed to translate policy priorities into budgetary policy and actions, while also providing a realistic macro-fiscal framework within which to develop 1Prior to the PRSP, in 1997 a joint document of the World Bank, EC, EBRD, and the Albanian government was conceived as a basic strategic document for the Albanian government and for donor interventions. In 1998, an EU report identified the prerequisites for Albania participation in the Stabilization and Association Process (SAP) and provided a perspective for opening negotiations with Albania on an Agreement. There was no reference in either of the documents to `poverty'. 2 Shortcomings included that: (i) budget preparation was based on outdated budget norms with little prioritization; (ii) policy, programming, and budgeting functions in line ministries were not fully integrated; (iii) there was little policy debate on spending decisions; and (iv) budget monitoring paid little attention to the implementation of budget programs. v NSSED proposals. Attention was focused on the need for strengthening public expenditures systems to more effectively implement the NSSED. The MTBP and budget reform process led to a number of important achievements. In 2002 the budget was presented by program for the first time allowing identification of objectives and program costs to be tracked over the medium term. To clarify the role of the MTBP within the budget cycle, the roles of the different agencies involved in preparation of the MTBP were specified. Several line ministries established policy units responsible for developing sectoral policies. A revised budget calendar was developed and circulated. And budget reforms were grounded in a new organic budget law and implementation manual. Reforms were based on international good practice and emphasized the link between NSSED objectives and budgeting. For the 2005-07 MTBP, four ministries (education, health, social protection, and agriculture) were asked to undertake a detailed process to identify policy goals for each program, ensuring that they corresponded to the priorities outlined in the NSSED, while also providing detailed estimates on program costs and prioritization of resources between programs. Priority coefficients were introduced to achieve better prioritization of sectoral allocations in line with the NSSED. However, challenges remain. Over the years, a multitude of planning processes had evolved around the long-term goals of EU integration, NATO membership, poverty reduction, and achievement of the Millennium Development Goals (MDGs), often using different policy planning and budgeting tools. Integrating this multiplicity of processes proved institutionally complex, complicated in part by unclear responsibilities between the Ministries of Finance and Economy and a fragmented approach to external assistance.3 Even within the Ministry of Finance, there are two separate departments-- one charged with NSSED formulation, the other with the MTBP--without clarity on their respective roles and responsibilities, and modalities for cooperation and coordination. For example, the NSSED has assumed operational planning and monitoring roles (e.g., compiling ministerial action plans that covered one to four years) that may have been more appropriate to incorporate with the MTBP. Furthermore, leadership for budget reform was undermined by the relatively weak role and status of the budget department in the Ministry of Finance and the difficulties it faced in attracting qualified staff. As a result of multiple planning processes, line ministries have been overburdened with requests for (similar) information from central government institutions, undermining rather than reinforcing their planning and budgeting capacity. Design issues also complicated matters. For example, the budget calendar--while crafted to facilitate integration of the MTBP and budget--left most key decisions until too late for the Council of Ministers to provide strategic direction during budget preparation. Overly optimistic macro-economic forecasts and revenue projections often resulted in downward revisions of the MTBP and budget preparation ceilings after line ministries had already started to work of their submissions, creating credibility problems 3Albania counted more than 300 operations in 1997 supported by 27 donors. vi for the reform process. These regular within-year expenditure reductions are a serious disincentive to investing time in MTBP and budget preparation. Weak capacity in turn has contributed in part to the lack of clear prioritization within sector expenditure programs and the poor costing of programs. Similarly, the monitoring burden at the ministerial level is perceived as overwhelming and expanding. Although the government receives numerous monitoring reports during the year, they are not adequately tied to the planning cycle, and the identification of meaningful outputs continues to be a challenge. To address these challenges, the Prime Minister established a Technical Secretariat in 2004 consisting of key government departments, supported by the World Bank, the European Commission, UNDP, the Organization for Security and Co-operation in Europe, and bilateral donors. It diagnosed that Albania's budgeting problems did not result from "policy inaction but from excessive policy activity across a series of disconnected planning systems and frameworks."4 In response, the Technical Secretariat is developing a proposal for an integrated planning system to coordinate and combine existing processes and external assistance, clarifying the roles and responsibilities of various actors. Albania's experience illustrates that even with strong domestic ownership and commitment to reforms, institutional integration and donor coordination can still pose major challenges. Current efforts to establish an integrated planning system with coordinated donor support could provide important lessons for other PRS countries. In addition to improving institutional and donor coordination, the Albanian authorities intend to strengthen the capacity of their ministries to prepare realistic macroeconomic frameworks as a basis for the MTEF and the annual budgets; improve the costing of programs and projects; and strengthen prioritization in sector expenditure programs. Albania At a Glance POVERTY and SOCIAL Most recent estimate (latest year available, 1997-03) Lower- Albania ECA Middle-income Poverty (% of population below national poverty line) 25 .. .. Urban population (% of total population) 44 63 50 Life expectancy at birth (years) 74 69 69 Infant mortality (per 1,000 live births) 22 31 32 Child malnutrition (% of children under 5) 14 .. 11 Access to an improved water source (% of population) 97 91 81 Illiteracy (% of population age 15+) 1 3 10 Gross primary enrollment (% of school-age population) 107 103 112 Male 107 104 113 Female 107 102 111 LONG-TERM TRENDS (average annual growth) 1984-94 1994-04 2002 2003 2004 GDP -3.9 6.1 3.4 6.0 5.9 GDP per capita -5.2 6.5 2.8 5.4 5.3 Exports of goods and services .. 20.5 14.0 18.8 16.5 Source: World Bank Development Economics central database. 2004 Preliminary. 4Albania Technical Secretariat (2004). vii ZAMBIA: REBALANCING THE MOTIVATION FOR ENGAGING IN A PRS PROCESS Many, many people have contributed to the PRSP. It is a national document that belongs to all Zambians. Civil Society for Poverty Reduction, Zambia5 The Zambian experience with the PRS process has been influenced by the secular decline in the country's national income and the contentious elections of 2001. Soon after the PRS initiative was launched and linked by the Bank and the Fund to their concessional assistance, the Zambian authorities accepted the process as a way to secure debt relief. However, upcoming national elections quickly put the demands of increasingly well-organized constituent groups at the fore. Over time, the dynamics of the PRS process in Zambia has shifted from one that was initially externally focused towards one much more integrated with domestic policy-making processes. Decades of decline in living standards, coupled with repeated, but half-realized, adjustment measures supported by the Bank and the Fund, had resulted in pessimism in Zambia about its prospects. A brief look at the numbers explains why: from being a low- middle-income country with a per capita GDP of US$613 in 19656 (about the same as Egypt or Morocco), Zambia today is one of the poorest Sub-Saharan African countries with a per capita GDP in 2003 of only US$354. As a result, civil society's first reaction to the PRS process was that it was merely "[o]ld wine in new bottles."7 Many within civil society were reluctant to engage in the process. Given these concerns, the authorities focused early efforts on initiating a PRS process in order to secure debt relief, perhaps building up domestic support for reforms along the way. However, the exercise was viewed largely as an externally imposed burden. Timeline July, 2000 Interim Poverty Reduction Strategy Paper (I-PRSP) August, 2000 Start of PRSP Consultations December, 2000 HIPC Decision Point December, 2001 National Elections March, 2002 Poverty Reduction Strategy Paper (PRSP) March, 2004 First Progress Report December, 2004 Second Progress Report April, 2005 HIPC Completion Point The mobilization for the 2001 elections helped change these perceptions. In the months leading up to the elections, the incumbent President sought to change the 5Civil Society for Poverty Reduction (2004: 1). 6Both figures in constant 2000 USDs. 7Seshamani (2005). Also see Eurodad (2005a). viii constitution in order to run for a third term. This move was unpopular, and he backed down after months of campaigning by a civil society coalition and other constituent groups. This political controversy led many observers to believe that the PRSP--initiated under the previous government--would not be `owned' by the new one, especially since many members of Parliament were not involved in the PRS process. However, a network of civil society groups, formed in 2000 for coordinating civil society input into the PRS process, continued to work with the government in broadening the participatory framework. This group, which later came to be known as the Civil Society for Poverty Reduction (CSPR), contains over sixty non-governmental organizations (NGOs), trade unions, and church groups, many of which also campaigned on the "third term issue," from which they gained added legitimacy and voice. As Besinati Mpepo, a coordinator for the network, explains: "If civil society is effectively organized and mobilized on national issues, government can take it seriously. This has been the case with the debt campaign, the campaigns against the third term of Zambia's second republican president...and civil society input to the PRSP..."8 Since the completion of the PRS consultation process, CSPR has also played an active role in monitoring implementation--particularly in conducting participatory poverty monitoring and expenditure tracking--although capacity remains a constraint. The robustness of civil society participation is a promising aspect of Zambia's PRS process. As a next step, to solidify domestic ownership and keep poverty reduction firmly on the national agenda, there is a need for civil society to build bridges to civil servants, political parties, and members of parliament. While individual members of parliament might have been involved in the PRS process on an ad hoc basis, it is important now to formalize the institutional role that parliament will play. In an otherwise critical review, Seshamani notes that "[t]he constant dialogue and interaction of a dynamic civil society with government and donors is [nevertheless] producing some positive impacts. In the last year of the just-ended PRSP, the government released 100% of the amounts allocated to poverty reduction programs, as opposed to only 24% and 50% in 2002 and 2003. In his 2005 budget address, the Minister of Finance has affirmed `the commitment of the New Deal Government to refocus public expenditures to poverty reduction programs.' In sum, despite a donor-driven and donor- dependent agenda, there is room for optimism yet."9 This is a case where a process that was viewed initially as externally imposed has gained domestic currency. It shows how effective participation can help rebalance an external focus towards a domestic one. It also shows the importance of understanding the political economy factors that determine how a PRS process unfolds. 8Mpepo (2003). 9Seshamani (2005). ix Zambia At a Glance POVERTY and SOCIAL Most recent estimate (latest year available, 1997-03) Zambia SSA LIC Poverty (% of population below national poverty line) 73 .. .. Urban population (% of total population) 36 36 30 Life expectancy at birth (years) 37 46 58 Infant mortality (per 1,000 live births) 102 103 82 Child malnutrition (% of children under 5) 28 .. 44 Access to an improved water source (% of population) 64 58 75 Illiteracy (% of population age 15+) 20 35 39 Gross primary enrollment (% of school-age population) 79 87 92 Male 81 94 99 Female 76 80 85 LONG-TERM TRENDS (average annual growth) 1983-93 1993-03 2002 2003 2004 GDP 1.3 2.0 3.3 5.1 5.0 GDP per capita -1.6 -0.1 1.6 3.5 3.6 Exports of goods and services -2.1 5.5 6.8 10.1 12.6 Source: World Bank Development Economics central database. 2004 Preliminary x 1. INTRODUCTION If there is one thing we, in rich and poor countries, have learnt over the last five decades of development cooperation, it is that poverty is a hydra that is not so easy to pin down and eradicate. Cutting off one or two of its multiple heads will not do. We need to do more, comprehensively....[W]e have also learnt to come to terms with the reality that development is a process, and that it can only be sustained if it is owned and led by the targeted population. Development cannot be imposed, it can only be facilitated; it requires ownership, participation and empowerment, not harangues and dictates. African countries ... have to assume full ownership and responsibility for their development. [B]ilateral donors, multilateral agencies and externally-funded NGOs ...are entitled to demand transparency and evidence of results. But they must be ready to genuinely concede enough space for the African countries to innovate, develop and pilot their policy frameworks and processes. Tanzanian President Benjamin William Mkapa10 A. OVERVIEW 1. In December 1999, the Boards of the World Bank (the Bank) and the International Monetary Fund (the Fund) introduced a new approach to their relations with low-income countries. The approach--centered around the development and implementation of poverty reduction strategies (PRSs)--was in many ways novel. It made the successful preparation of a nationally owned poverty reduction strategy a precondition for access to debt relief and concessional financing from both institutions. These strategies were expected to be poverty-focused, country-driven, results-oriented, and comprehensive. They were also expected to serve as a framework for better coordination of development assistance among other development partners. The approach called for a fundamental change in the nature of the relationship between developing countries and donors. It sought to empower governments to set their development priorities and pushed donors to align their assistance around a country's priorities rather than their own. 2. The PRS approach has become widespread. Currently 49 countries have prepared national poverty reduction strategies.11 Just over half of these countries are in sub-Saharan Africa; a similar proportion are heavily-indebted poor countries (HIPCs). Twenty-nine of these countries have also produced at least one annual progress report (APR). Countries have been implementing their strategies, on average, for just over two and a half years. Several countries are in the process of revising their original strategies, and Burkina Faso, Tanzania, and Uganda have already done so (see Annex 1). Eleven 10Keynote speech. Africa Regional Workshop on Harmonization and Alignment, Dar es Salaam, November 2004. 11The numbers in the section reflect the PRS documents, of which staffs are aware, that were competed by June 30, 2005. 1 more countries have produced interim strategies, and ten have initiated processes that could result in a PRS. 3. After five full years of implementation, there is evidence in many countries that the PRS approach has helped put poverty reduction at the center of the public policy debate, highlighted the need to identify and address country-specific constraints to more effective development, heightened awareness of the need for sound analysis to underpin policy choices, and reinforced the incentives to monitor the results of public actions. The PRS approach provides the operational framework for governments to set their development priorities and to specify policies, programs, and resources needed to meet their goals. This process helps to crystallize political commitment and accountability--both for countries themselves and for their development partners--for accelerating progress towards the Millennium Development Goals (MDGs). 4. Yet, as this paper will discuss, the agenda ahead requires a sustained, long- term commitment. As noted in the 2002 PRS Review, the PRS approach is a major challenge for low-income countries, both in terms of analysis and organization. Besides managing a complex policy dialogue with a wide range of stakeholders, low-income country governments have to put together an integrated medium-term development and poverty reduction strategy, complete with short- and long-term goals and monitoring systems. These are a set of tasks that few industrial countries could systematically do well. Moreover, these tasks must be managed with limited technical and institutional capacity and in ways that reinforce--rather than undermine--existing national institutions, processes, and governance systems. Expectations, therefore, need to be realistic. 5. This review concludes that the core principles that underpin the PRS approach should be maintained, and that they provide the foundation on which results at the country level are achieved (box 1). But, there are no `magic bullets.' The range of institutions, systems, and underlying analysis that needs to be strengthened at the country-specific level is large. Progress is being made in some dimensions; in other areas, adjustments are needed to more fully realize the potential benefits. However, actions will need to be sequenced and tailored to country conditions. Now is the time to step back from unrealistic expectations, and to focus on consolidating gains and making adjustments as needed. B. 2005 PRS REVIEW 6. This review, conducted at the request of the Boards of the Bank and the Fund, has sought to elicit and reflect on the views of a wide range of stakeholders. As an output from this review, staffs have prepared two papers. This paper provides a more extensive discussion of implementation experience to date. However, it remains necessarily selective and cannot fully capture the wide range of experiences in about one- third of the world's countries over the past five years. A companion paper provides a synthesis of the main findings. The real value added of this review, however, will be structuring necessary follow-up actions, by countries themselves, the Bank and the Fund, 2 and other development partners, and distilling and disseminating lessons and the range of good practices across a diverse spectrum of PRS-related issues. Box 1: PRS Core Principles Recognizing that there may be, at times, limitations to their immediate applicability in particular country contexts, poverty reduction strategies should be: Country-driven (with broad based participation). A country-driven approach improves the prospects of designing reforms that are tailored to country circumstances and feasible to implement, by appropriately prioritizing and sequencing public actions and programs and by focusing on strengthening country processes. Participation can contribute to higher quality strategies, and strengthen the environment for governance and accountability. Medium- to long-term in perspective. The PRS approach encourages defining development objectives in the context of a long-term development vision, and sustained efforts over the medium to long term to address capacity, institutional, and governance constraints to faster growth and better development results. Comprehensive and results-oriented. The PRS approach recognizes that sustainable development will not be possible without rapid economic growth that benefits poor people, for which macroeconomic stability, structural reforms, and social stability are required. This implies a multidimensional view of poverty and of the approach to reducing it, with better understanding of the causal links between policies and intended outcomes, and attention to setting appropriate targets for measuring progress. Effective monitoring can help ensure that policies are well designed and effectively implemented. Partnership-oriented. A national poverty reduction strategy can provide the context for improved partnerships between government, external partners, and other domestic stakeholders. The PRS approach thus reinforces the shared responsibility for achieving improved results. 7. This review did not set out to attribute specific poverty reduction results to the PRS approach. This is because the PRS approach comprises a set of principles and actions that affect the environment in which policy is formulated, implemented and monitored. This prevents building a counterfactual--which is necessary to establish causality to poverty outcomes and which is possible for a project or specific policy measure--for two reasons: There are a wide range of factors at play. The approach is implemented in an ever- changing context, where countries simultaneously react to exogenous factors, political processes, and other country-specific changes that are outside of the PRS process. Some of these factors are influenced by the PRS process, but it is not possible to disentangle the contribution of each. The PRS approach is not limited to discrete activities. It promotes changes in a myriad of areas rather than constituting a clearly defined set of activities that would lend themselves to specific impact evaluation. 8. However, one can assess the effect that the PRS approach has on the quality of the various decision-making and implementing processes. By focusing on the environment within which interventions take place, one can discern the impact of the PRS approach on the processes and systems that lead to improved interventions and better outcomes. For example, one can consider the extent to which policies are informed by analysis, whether or not monitoring systems are in place to track progress, and if the 3 approach has fostered more accountable, efficient, and pro-poor budget processes. Given that this is the approach taken in this review, it avoids narrowly reviewing PRS documents against uniform benchmarks, and instead focuses on examining how selected processes have evolved over time. 9. This paper is structured as follows. The remainder of this introductory chapter provides a stylized overview of the development challenges in low-income countries. Chapter 2 focuses on balancing accountabilities. It discusses: (i) how the PRS approach can support the strengthening of domestic accountability; (ii) the accountabilities of donors in terms of providing better aid; and (iii) various factors that help external requirements to reinforce rather than undermine domestic accountability. In the context of scaling up results, chapter 3 discusses: (i) the analytic foundations which can support the strengthening of strategies and their results orientation; and (ii) how the PRS process can support ambitious development plans by providing a framework for scaling up assistance and addressing absorptive constraints so that additional aid is used well. The paper ends by summarizing selected findings and good practices within the context of the five themes that were the focus of this review.12 C. DEVELOPMENT CHALLENGES IN LOW-INCOME COUNTRIES 10. The development challenges facing low-income countries are enormous. Various recent reports have provided rich depictions of these challenges. For example, the 2005 Global Monitoring Report argues that without faster progress, the MDGs will be seriously jeopardized. At stake are the prospects for hundreds of millions of people to escape poverty, disease, and illiteracy. The UN Millennium Project's report, Investing in Development: A Practical Plan to Achieve the Millennium Development Goals, notes that while the world has made significant progress in achieving many of the MDGs, progress has been uneven across regions and across goals. It also notes huge disparities across and within countries, and finds that sub-Saharan Africa continues to fall further behind other regions with continuing food insecurity, a rise of extreme poverty, high child and maternal mortality, and large numbers of people living in slums, all of which is compounded by the HIV/AIDS pandemic. The report of the United Kingdom's Commission For Africa, Our Common Interest, provides a moving portrait of the development challenges facing Africa.13 11. Five years have passed since the PRS approach was introduced. Although it is not possible to disentangle the effects of the PRS approach on development outcomes from those of other factors, the remainder of this section briefly summarizes contextual evidence about what has happened in low-income countries over the past five years, in 12 The concept paper for this review identified five themes as central to the effectiveness of the PRS approach. These themes were: (i) strengthening the medium-term orientation of the PRS; (ii) utilizing the PRS as a mutual accountability framework between countries and donors; (iii) broadening and deepening meaningful participation; (iv) enhancing linkages between the PRS, the MTEF, and budgets; and (v) tailoring the approach to conflict-affected and fragile states. This report integrates these themes within the context of balancing accountabilities for and scaling up development results. 13World Bank/IMF (2005); UN Millennium Project (2005); UK Commission for Africa (2005). 4 terms of development outcomes and intermediate expenditure indicators. It also examines the policy, institutional, and governance environment that characterizes low-income countries. 12. When the PRS approach was launched, evidence suggested that, outside of East Asia, progress toward poverty reduction was disappointing, with the number of poor rising in many part of the world.14 Moreover, progress toward universal primary education had stalled, or even reversed, in many African countries. Countries severely affected by AIDS had seen all the gains in life expectancy since the Second World War wiped out. 13. Growth performance. On the whole, there has been relatively solid growth performance in low-income countries over the past ten years, although high population growth rates in low-income countries temper the per capita gains (figures 1 and 2). While these aggregate figures disguise substantial differences across countries and volatility over time, nonetheless, this performance provides an encouraging basis on which to build. Figure 1: GDP growth rates 1990-2003 (period average) 6 5 4 High income 3 Middle income 2 Low income 1 0 1990-95 1996-99 2000-03 Figure 2: GDP per capita growth rates 1990-2003 (period average) 6 5 4 High income 3 Middle income 2 Low income 1 0 1990-95 1996-99 2000-03 Source for figures 1-2: 2005 World Development Indicators, and Bank staff calculations. 14IMF/World Bank (1999a: 4). 5 14. Socio-economic indicators. While there is some evidence that over the past five years low-income countries have made some gains on various socio-economic indicators, in light of very low baselines, there is urgency to accelerate progress further.15 Between 1998 and 2002, low-income countries in sub-Saharan Africa have made considerable progress in improving some human development indicators, outstripping all other country groupings (table 1). However, starting from such a low base, low-income countries in sub-Saharan Africa continue to face enormous challenges. Table 1: Changes to selected human development indicators (in percent) Primary school Measles immunization Access to improved completion rate rate (12-23 months) water source 1998 2002 change 1998 2003 change 1990 2002 change HIC 98.6 100.0 1.4 89.7 90.5 0.8 99.8 100.0 0.2 MIC 93.5 90.8 -2.7 87.9 89.5 1.6 85.7 88.0 2.2 LIC non AFR 81.2 83.6 2.3 79.6 81.1 1.5 74.5 75.4 0.9 LIC AFR 42.4 51.0 8.6 60.6 68.3 7.7 49.8 61.1 11.3 Source: 2005 World Development Indicators, and Bank staff calculations. 15. Poverty reducing expenditures. While increased public spending does not necessarily imply better results, shifts in budget outlays can be early indicators of PRS implementation. Over the past five years, there is some evidence of increases in poverty-reducing expenditures. Data for a sample of 27 countries that reached the Decision Point under the HIPC Initiative suggest that poverty-reducing spending increased by close to two percentage points to 8.3 percent of GDP between 1999 and 2004 (table 2).16 Table 2: Poverty-reducing expenditure 27 countries that have reached HIPC decision points (in percent) 1999 2000 2001 2002 2003 2004 Ratio of poverty reducing expenditure to government revenue African Countries 38.6 39.2 41.3 44.9 47.8 48.7 Latin American Countries 47.6 46.5 49.4 50.8 51.2 51.4 Total 40.9 41.3 43.5 46.4 48.5 49.3 Ratio of poverty reducing expenditure to GDP African Countries 5.5 5.4 5.8 6.6 7.3 7.8 Latin American Countries 10.8 9.9 10.3 10.6 10.7 11.2 Total 6.4 6.3 6.7 7.4 7.9 8.3 Source: IMF/World Bank (2005). 15Very few indicators have good coverage across countries for two data points. Consequently, only a few are discussed in this section. 16It is important to note that the definition of poverty-reducing spending is not always based on a good understanding of the appropriate public interventions necessary to reduce poverty, and identified programs often fail to reach front line service providers or are skewed towards better-off households. Moreover, the spending classification may have improved, with better coverage of pro-poor spending. 6 16. Policy and institutional environment. In 1999 the policy and institutional environment in low-income countries was, on average, weaker than in middle- income countries.17 Not surprisingly, as measured by the World Bank's Country Policy and Institutional Assessment (CPIA) scores, policy performance in all groups of low- income countries--those that are not heavily indebted (LIC non-HIPC), those that are heavily indebted and had reached decision or completion point by June 2005 (HIPC-1) and those that had not yet reached decision point by that time (HIPC-0)--was below that of middle-income countries (figure 3). 17. Over the past five years, gains in policy performance by those low-income countries that had reached the decision or completion point under the HIPC Initiative substantially outpaced that in middle-income and other low-income countries. Even HIPC-0 countries witnessed robust improvements, at a rate on par with middle-income countries (figure 4). By 2004, the institutional and policy environment in heavily-indebted poor countries that had reached decision or completion points matched (or was slightly stronger than) that in low-income countries that were not heavily- indebted (figure 5). The focused attention which the Heavily Indebted Poor Countries (HIPC) Initiative placed on improved institutions and policies may be responsible in part for these developments. Figure 3: CPIA scores (1999) 6 5 MIC 4 LIC nonHIPC 3 HIPC-1 2 HIPC-0 1 0 Economic Structural Policies Policies for Social Public Sector Total management Inclusion Mgt/Institutions Figure 4: Changes in CPIA scores (1999 to 2004) 0.6 0.5 MIC 0.4 LIC nonHIPC 0.3 HIPC-1 0.2 HIPC-0 0.1 0 Economic Structural Policies Policies for Social Public Sector Total management Inclusion Mgt/Institutions 17In this section, LIC includes all IDA-only countries, those most likely engaged in a PRS processes. MIC includes all IDA-blend and IBRD countries. A few IDA-blend countries have also engaged in a PRS process. 7 Figure 5: CPIA scores (2004) 6 5 MIC 4 LIC nonHIPC 3 HIPC-1 2 HIPC-0 1 0 Economic Structural Policies Policies for Social Public Sector Total management Inclusion Mgt/Institutions Source for Figures 3-5: World Bank staff calculations. 18. Of particular note are the strong performance gains by HIPCs on two key indicators of public sector management: (i) the quality of budget and financial management; and (ii) transparency, accountability, and corruption in the public sector (figure 6). By 2004, performance on these variables by HIPCs that had reached decision or completion points had surpassed that of low-income countries that were not heavily- indebted (figure 7). Figure 6: Changes to key public sector management CPIA scores (1999-2004) 0.8 0.6 MIC 0.4 LIC nonHIPC 0.2 HIPC-1 0 HIPC-0 -0.2 Quality of budget and financial Transparency, accountability & management corruption in public sector Figure 7: Key public sector management CPIA scores (1999 and 2004) 1999 6 2004 1999 2004 5 MIC 4 LIC nonHIPC 3 HIPC-1 2 HIPC-0 1 0 Quality of budget and financial Transparency, accountability & management corruption in public sector Source for figures 6-7: World Bank staff calculations. 8 19. Governance quality. Notwithstanding improvements in policy and institutional performance over the past five years, the overall governance environment in most low-income countries remains difficult. The recent Governance Matters indicators measure six dimensions of governance on a scale from -2.5 to 2.5.18 In 2004, for developed countries, all six dimensions have, on average, strongly positive scores. Middle-income countries hover just under zero--the score which represents the average for the world as a whole. Low-income countries score less than -0.5 in all categories, with low-income African countries performing worse than other regions in all but one dimension (political stability and absence from violence) (figure 8). While low- income countries are not homogenous, few low-income countries score above zero (the worldwide average) on any dimension (figure 9). Figure 8: Governance indicators (2004) by country group 2.50 2.00 1.50 HIC 1.00 MIC 0.50 LIC not africa 0.00 LIC Africa -0.50 -1.00 -1.50 Voice and Political Stability Government Regulatory Rule of Law Control of Accountability Effectiveness Quality Corruption Figure 9: Variations in governance indicators for IDA countries (2004) 1 0 -1 -2 Voice and accountability Government Effectiveness Control of Corruption 19Source for figures 8 and 9: Bank staff calculations; 2004 Governance Research Indicators Dataset. 18 The indicators are (i) voice and accountability; (ii) political stability and absence of violence; (iii) government effectiveness; (iv) regulatory quality; (v) rule of law; and (vi) control of corruption. For definitions and caveats on data use, see Kaufmann, Kraay, and Mastruzzi (2005). 19 Median is indicated by the center line; average is the center dot; first and third quartiles are the edges of the box. Extreme values are the ends of the lines. Points at a greater distance from the median than 1.5 times the first or third quartile are plotted individually as open dots. 9 20. Given this broad sketch of the institutional, policy and governance environment in low-income countries, it should not be surprising that implementation of the PRS approach--which needs to be built on solid accountability mechanisms--will require sustained efforts. This requires understanding better the political economy context of policy formulation and implementation, and the implications these have for policy choice (e.g., when "second-" or even "third-best" solutions may produce better results than the theoretical first-best but non-implementable choice). 2. BALANCING ACCOUNTABILITIES A cautionary note: [A]ccountability is a complex and chameleon-like term [that] now crops up everywhere performing all manner of analytical and rhetorical tasks. Mulgan (2000: 555) A. OVERVIEW 21. This chapter discusses PRS implementation within a framework of balancing accountabilities--including that of governments to both domestic constituents and to donors, as well as that of donors to developing countries. While the PRS approach was intended to put countries more squarely in control of policy formulation, the link to the HIPC Initiative, combined with longstanding relationships between low-income countries and the Bank and the Fund based on conditionality, often emphasized external accountabilities rather than domestic ones. In countries where the approach shows the most promise, domestic stakeholders are at center stage. To encourage this across the board, it is helpful to review PRS implementation through an accountability lens. 22. Country ownership and participation are central to the PRS approach. The first core principle of the PRS approach was that it be country-driven and owned, based on broad-based participatory processes.20 Yet these concepts have not always been clearly defined. While there are rich definitions and a robust literature underpinning these concepts,21 their use as short hand notation at times obscures the complexity of the concepts and can lead to a range of problems, including different interpretations and, at 20IMF/World Bank (1999b). 21For example, on country ownership, the Bank's OED proposes four criteria to assess country ownership: (i) the locus of the initiative must be in the government; (ii) key policymakers responsible for implementation must be intellectually convinced that the goals to be pursued are the right ones; (iii) there is evidence of public support from top political and civic leadership; and (iv) government is building support among affected stakeholders and can rely on their cooperation. OED (1995: 64) based on Johnson and Wasty (1993). Killick (1998) proposes adding a fifth dimension--measures are institutionalized within a country's policy systems. Morrisey (2001: 6-8) takes issue with the first dimension, dismissing the notion that the intellectual origin of policies outside of government necessarily compromises its prospects. He argues that attention should be focused on commitment (encompassing both preferences and the political capacity to articulate them). Fund staff have defined ownership as a willing assumption of responsibility for an agreed program of policies by officials in a borrowing country who have the responsibility to formulate and carry out those policies, based on the understanding that the program is achievable and is in the country's own interests. IMF (2001: 6). 10 times, unrealistic expectations by different stakeholders.22 It has also meant that stakeholders, including the Bank and the Fund, have not always been clear about the facilitating conditions and types of support that are needed to foster country ownership and strengthen participatory processes.23 23. While the Bank and the Fund have spoken largely in the language of country ownership and participation, some early observers were calling attention to the fact that strengthening ownership and participation are intimately linked to supporting domestic accountability.24 Others have highlighted the intrinsically political context of the PRS approach. For example, Norton (2002) remarks that success of the PRS approach rests on several crucial `gambles' including: (i) that it will be easier to hold governments to account if they are obliged to discuss poverty and what they are going to do about it with their citizens; and (ii) that if donors and countries take the PRS process seriously, this will change the emphasis towards (political) accountability to citizens over (technocratic) accountability to donors. BMZ (2002: 2) observes that "PRSPs have a profound effect on political processes." Booth (2003: 137) notes that PRS approach "will work through the political systems...of countries concerned, or it will not work at all." 24. The intention of this chapter is to review implementation experience more explicitly from an accountability perspective. Section B considers several dimensions that are needed to support a strengthened framework for domestic accountability. Section C focuses on the external aspects, with a particular emphasis on donors' accountability to align and harmonize their assistance. Section D identifies some factors that can tilt the focus of governments towards external requirements rather than domestic ones, and how the PRS architecture can help redress this. B. DOMESTIC ACCOUNTABILITY 25. This section considers central aspects of the PRS approach in the context of their role in supporting domestic accountability. The first sub-section discusses how 22OED (2004) emphasized the disparate perceptions of different stakeholders. Some have argued that the Bank, and Fund need to establish standards for a participatory process in a country, for example Herz and Embrahim (2005: 43), although the Boards have provided explicit instructions to the contrary. Others indicate that a participatory process can only be successful if CSO views are reflected in the final document, CIDSE-CARITAS (2004: 45) or if a rights-based approach is used, World Vision (2005: 10). 23Recent work by Entwistle and Cavassini (2005) has attempted to unpack the notion of country-ownership and develop an operational approach to assessing country ownership. It includes a list of 17 indicators of country ownership, the importance of each of which is likely to vary across countries, ranging from the extent to which sector action plans and PRS priorities are systematically aligned; to the role of parliamentary committees in PRS formulation and monitoring; to the degree to which interruptions due to political and economic events, including electoral cycles, are anticipated and factored in. 24See, for example, Whaites (2000). Links between participation and domestic accountability have also been raised in various recent World Development Reports including the 1997 WDR, The State in a Changing World and the WDR 2003, Sustainable Development in a Dynamic World. The 2004 WDR, Making Services Work for Poor People, developed an analytical framework built around the notion that improving service delivery requires strengthening the relationships of accountability between policymakers. However, within the context of reporting on progress in PRS implementation, the link to accountability has been largely implicit rather than explicit. 11 the PRS approach supports evidence-based decision making--by focusing on the need to frame clear development goals and targets that are coupled with programs of action; to link these to budgets and effective public financial management systems; and to monitor implementation in order to facilitate adjustments in policies and programs. The second sub-section considers elements that can enable the demand for, and enhance the quality of, evidence-based decisions--including institutionalizing participation; the vital role of information; the need for greater space for policy dialogue; and the involvement of key stakeholders (such as parliaments and poor people themselves). The final sub-section discusses some special considerations in conflict-affected and fragile states. Elements supportive of evidence-based decision making 26. Evidence-based policy making means that, wherever possible, public policy decisions should be informed by careful analysis using sound and transparent data. This entails the systematic and rigorous use of information to inform program design and policy choice, monitor policy implementation, and evaluate policy impact.25 By enhancing the link between policy choices and intended outcomes, it sets the stage for defining accountabilities. This sub-section starts by looking at progress and challenges in framing development goals and setting targets. It then considers the linkages between goals and targets to specific policies, programs, and public actions, and between the PRS to the budget and medium-term expenditure framework. It concludes by considering the monitoring systems that are needed to track progress.26 Framing development goals and setting targets 27. Setting clear development goals and targets helps to clarify the priorities of government and increase accountability. The PRS approach provides the operational framework for governments to set their development priorities and to specify policies, programs and resources needed to achieve results. This process helps to crystallize political commitment and accountability--both for countries themselves and for their development partners--for accelerating progress towards the MDGs.27 28. In this regard, it is also important for long-term visions to be adequate linked with medium-term strategies. Increasingly, national development visions are providing a guiding framework for policy actions detailed in PRSs (e.g, Bhutan, Uganda). However, in many cases, there is still a disconnect between long-term development 25Scott (2005: i). 26In addition, evidence-based decision making needs to draw on sound analysis. This includes detailed poverty diagnostics; distributional analysis; analysis of policy and institutional levers for growth and the transmission mechanisms for connecting poor people to growth; as well as well-designed sector strategies built on solid understanding of how well public programs and actions meet the needs of poor people. Section 3B discusses the analytic underpinnings of PRSs in more detail. 27While the MDGs have been agreed at the global level, the development agenda to reach these goals needs to be anchored in coherent, country-specific development strategies that respond to local conditions and priorities and are nationally owned and led. 12 visions and medium-term strategies.28 Strengthening the PRS process to tie it more closely to ongoing long-term vision processes would help countries to develop common, country-owned frameworks.29 A number of countries (e.g., Cambodia, Yemen) are doing just that during PRS implementation. 29. Past implementation progress reports have encouraged countries to customize the MDGs to country priorities, while noting that this task can be technically challenging.30 Vietnam has effectively customized its development goals and targets. Other countries, such as Sri Lanka, are doing so. In this context, attention has also been called to the tension between realism (targets that seem overly ambitious relative to prior achievements and/or likely resource availability) and aspirations. Past implementation progress reports have noted that PRS indicators overlap significantly with the MDGs, although targets are at times more ambitious and at times less so.31 30. Stakeholder contributions to this review have also emphasized the need for MDGs to be tailored to country circumstances. Comments by the United Nations Office of the High Commissioner for Human Rights (2005: 1) caution against limiting the scope of the PRS approach to a narrow technocratic interpretation of the MDGs and its indicators framework, noting that countries must have sufficient flexibility in customizing their PRS targets. The Parliamentary Centre (2005: 2) urges that to ensure sound poverty reduction strategies, more attention is needed at the country level to examine the causal link between the MDGs and poverty reduction. The World Health Organization (2005: 2-3) cautions that narrow application of the MDGs could skew activities away from priority diseases and urgently needed institutional improvements. They encourage appropriate customization (box 2). WaterAid (2005: 3-4) notes that an undue focus on `quick wins' can detract from longer-term institutional reforms that are needed to improve the sustainability of service delivery. The views of CIDSE/Caritas Internationalis (2005b: 4) reinforce the above concerns. Box 2: Experiences in customizing health targets to country circumstances Adjusting MDG targets in PRSPs: Ethiopia has revised downward the targeted maternal mortality ratio in line with baseline indicators and country capacity. Uganda has made substantial progress toward the HIV/AIDS MDG target and has revised it to be more ambitions Focusing on country specific disease priorities in PRSPs: Yemen has identified tuberculosis as a priority while a preventative strategy is in place to stem the low spread of HIV/AIDS. The global focus on the latter needs to be adjusted to country needs. Vietnam places importance on addressing the use of tobacco, an area not included in the MDGs. Source: Adapted from WHO (2005: 2). 28World Bank (2005a: 3). This disconnect can emerge when donors support two separate processes that are not adequately coordinated across different government units. 29 See also section 3C for a discussion on the use of alternative scenarios within a PRS to support ambitious, longer-term development goals. 30IMF/World Bank (2004: 14-15). 31IMF/World Bank (2003a: 4-5); OED (2004: 16). 13 31. Over the past five years, the PRS process has been successful in encouraging countries to start setting concrete targets. The bulk of PRS countries have embarked on a process of defining development goals and associated targets (figure 10). For countries more advanced in this process, most have been implementing their PRSs for more than two years and long-term visions for even longer. Successive iterations have helped to sharpen these goals and targets.32 However, in about 20 percent of PRS countries--largely fragile states--defining country specific targets connected to broadly agreed goals remains elusive.33 There is also some evidence that governments have felt accountable for meeting the targets they set, leading to an iterative process of refining targets and interventions as additional information became available. Such processes should be encouraged.34 Figure 10: Are PRS development goals linked to specific targets? 100% 65% 80% 60% 40% 16% 22% 20% 0% Largely in place Actions being taken Some elements in place or being considered Source: World Bank (2005a). Sample includes 59 countries as of June 2005. 32. Specifying clear targets, for which data are available, and identifying intermediate indicators remains particularly challenging for countries. Past progress reports have noted that targets are often not clearly identified and lack good intermediate indicators that would help track implementation of public programs.35 Most PRSPs have focused on budgetary or expenditure analysis (input indicators) and survey-based measures of well-being (impact indicators). Intermediate outputs and outcomes have generally received less attention. For instance, a review of 21 PRSPs showed that in the health sector, many countries lacked the means to monitor significant components of the health strategy.36 Linking goals and targets to clear public actions 33. For a country to achieve its development goals, a medium-term development program needs to be linked to those goals. The PRS approach provides a framework 32World Bank (2005a: 16). 33World Bank (2005a: 14-15). 34For example, Uganda's poverty reduction strategy set ambitious targets to reduce infant mortality, underpinned by a fairly narrow set of health sector interventions. When results were disappointing, a more thorough analysis of the issues resulted in a shift to a broader multi-sector approach to reducing infant mortality. Targets were also revised downward, reflecting a more detailed assessment of the challenges and the pace with which actual improvements could be realized. 35IMF/World Bank (2002b: 35-36). 36WHO (2004). 14 for countries to elaborate such development programs. Its comprehensive nature also provides an opportunity to recognize that achieving most goals requires interventions across a range of sectors, and to address the tendency for many development interventions to be sector-oriented. In linking goals and targets to clear public actions, countries need to ascertain causal links, decide on the degree of specificity in their PRSPs, and make important choices with regard to prioritization and sequencing. This is a very challenging task, particularly in countries facing severe technical and institutional constraints. 34. Overall, progress on this front is mixed, and many PRSs would benefit from a more explicit link between goals and targets and the policies needed to achieve them. In some cases, development programs do not clearly support the achievement of specified goals. In other cases, cross-sector linkages are not well-articulated.37 However, there are good examples of strong mapping of objectives, strategies, and indicators, including the environment in Pakistan (box 3) and roads in Kenya (box 4). Box 3: Linking objectives, strategies and indicators--Pakistan environment Objective: To achieve environmental sustainability through better environmental management. Strategies: · Policy coordination and environmental governance. · Pollution control. · Eco-system management and natural resources conservation. · Dry land management. · Grass roots initiatives. Inputs indicator: Expenditure in the environment sector as a percent of GDP. Quarterly data from the Ministry of Finance will be disaggregated at the federal, provincial and district level. Output indicators: · Number of industries and municipalities complying with the environmental laws (% of total). · Number of development projects having an Environmental Impact Assessment (EIA) and Initial Environmental Examination (IEE). · Area brought under afforestation in hectares. · Number of new saplings planted. Data provided by the Ministry of Environment. Outcome indicators: · Percentage reduction in emissions of pollutants in air, water and land. · Percentage of arable land. · Percentage of change in forest cover. Data produced by the Ministry of Environment 37 Developing a coherent medium-term development agenda is a very complex task. It requires, for example, considering the cross-effects among goals. For example, health and nutritional status directly affects a child's probability of school enrollment, and access to safe water and sanitation is critical for child survival. The existence of such cross-effects means that improvements in one goal depend also on progress on other goals. Inter-linkages are also evident across sectors, meaning that isolated interventions may do little to achieve goals if bottlenecks remain in other sectors. 15 Box 4: Linking objectives, strategies and indicators--Kenya roads Objective: To achieve an expanded and well-maintained road network to increase access, affordability and reliability of the services, improve their competitiveness and quality, and improve road safety. Strategies/inputs (assumptions for success): · Build, rehabilitate, and upgrade the road network, key road links and rural roads and bridges. (roads program sustained and training of small-scale contractors continued). · Complete institutional reforms of road management (reform effectively managed and financing sustained). · Launch national road safety campaign (mobilization of the relevant stakeholders) · Enforce axle load limits · Improve the regulatory framework (privatization bill passed and private sector interested) · Have better trained staff in the existing regulatory agencies. · Develop investment road map. Implementing agencies: District Roads Committees, Kenya Road Board, Ministry of Roads and Public Works and Housing, and Ministry of Local Government. Outputs indicators and selected targets: · Gazettement of safety campaign notices. · Regulatory framework in place by 2006. · Number of regulatory agency staff trained by 2006. · Proportion of the road networks in bad or poor condition reduced from 43% to 20%. · The number of road links rehabilitated and updated. · Number of rural roads rehabilitated and the number of jobs created between 2003 ­ 2006. · Number of fatalities from road accidents. · Number of axle load limit cases enforced. Data sources: surveys by the Ministry of Roads and Public Works and Housing, Ministry of Local Government, Kenya Roads Board, Kenya Police and the Kenya Road Safety Council, and implementation reports. Outcome indicators: · Increased access, affordability and reliability of the infrastructure services ­ measured by the proportion of the poor household budget allocated to infrastructure services. · Improved competitiveness and quality of infrastructure services ­ measured by the proportion of infrastructure related production costs. · Improved roads network and road safety ­ measured by the cost of road transport. Data sources: Kenya Living Standard Survey, Ministry of Roads and Public Works and Housing, the Industrial survey, Central Bank of Kenya, Kenya Transport Association, and Kenya Association of Manufacturers. 35. Past progress reports have emphasized the challenge of formulating a comprehensive strategy that does justice to the range of competing demands, provides specificity, yet remains manageable in scope.38 Comprehensiveness is important in order to capture the complementary nature of public actions across sectors. Closely linking associated sector strategies to the PRS is one way that the scope of the PRSP itself could be contained, without sacrificing specificity in the process. It is also worth highlighting that a comprehensive strategy does not mean sacrificing priority setting. In fact, the more comprehensive the strategy, the more important it is to identify its main priorities.39 38IMF/World Bank (2003a: 40). 39Peretz and Wangwe (2004: ii, 13). 16 36. The fact that PRSs are often inadequately prioritized is also much discussed. OED notes that the general lack of prioritization in most PRSs is due in part to knowledge gaps about the impacts of various public actions but is also symptomatic of weakness in public expenditure management systems.40 Proposed solutions focus on systematic, long-term support to overcome capacity and technical constraints, continued support to strengthen sector strategies, and discipline by donors to respect country priorities and sequencing in their own advice and assistance strategies.41 Political economy issues, however, are also likely to be at play, with governments at times not wishing to make difficult decisions about prioritization. GTZ (2005: 14) suggests that the task going forward is for countries to boil down their priorities to a handful of core actions, as Armenia has done (box 5). The observation that countries that have close links between their PRS cycles and their budget processes make faster progress in prioritization, emphasizes the importance of more actively encouraging countries to embed the PRS process in their existing policy-setting processes.42 43 Box 5: Clarifying and agreeing on PRS priorities--Armenia While there are many processes by which countries can refine their priorities, the case of Armenia is an interesting example. In 2002, Armenia's draft PRSP proposed interventions in over 100 areas. Attempts to refine this list had come to a stalemate. To break the logjam, the Ministry of Finance and Economy held rounds of discussion, supported by a software tool that facilitates complex decision-making by visually presenting qualitative information in manner that makes more transparent underlying reasoning and interrelationships between different factors. A range of different scenarios were developed that participants could compare. This approach helped facilitate a consensus, concentrating on 12 areas that could be closely linked to the MTEF and well-specified policy actions. Source: Adapted from GTZ (2005: 14) 37. Linking goals and targets to specific policies and programs needs to be viewed as an ongoing process, with countries making progress in strengthening these linkages over time. Positively, as PRS implementation unfolds, there is a trend for countries to refine their development programs and better align them with specified goals. Tanzania provides a good example of a country that, over time, has considerably strengthened its PRS. Tanzania's PRSP, prepared in 2000, set out a relatively narrow strategy, focused on macroeconomic policy and key social investments. However, as implementation unfolded, the strategy was steadily broadened into a more comprehensive one for growth and development. Sector strategies were further developed and incorporated into the overall development agenda. By its third annual progress report (2004), Tanzania's medium-term development program was clearly articulated across a 40OED (2004: 16) emphasizes the importance of sound strategies to facilitate prioritization. For example, in Ethiopia, strong health and education strategies allowed government to present well-prioritized and costed PRS programs. Overall, they note that health and education strategies in PRSs tend to be the clearest in priorities and targets, while the weakest sector in almost all PRSs was private sector development. 41IMF/World Bank (2003a: 42). 42IMF/World Bank (2004: 29). 43Policy priorities are relatively better linked to specific targets in countries like Ethiopia, Mozambique, Rwanda, Tanzania, Uganda, and Vietnam. All of these countries also have been implementing their PRS for over two years. World Bank (2005a). 17 broad range of sectors, incorporated cross-cutting issues, and was linked to medium-term country goals. Table 3 highlights aspects of this progression. Table 3: Progress in enhancing specificity--Tanzania In Tanzania, progress was made both in deepening and broadening sectoral coverage. Education sector Agriculture sector PRSP Process of integrating education strategy Commitment to develop a national 10/2000 and PRS initiated. Need for more concrete agriculture strategy, including an intention details and costing. to rationalize taxes and remove impediments to efficient marketing. APR 1 Strategy linked to policy actions: school Strategy completed which specified an 8/2001 fees abolished at primary level; increased enabling role of government rather than budget allocations; capitation grants and direct intervention. Overall focus on investment fund to support schools at the increasing productivity. Growth targets set. local level; creation of fund to support children from very poor families. APR 2 Primary enrollment above target. Implementation plan for strategy 3/2003 Challenge is to implement specified articulated. Plan entails substantial measures to improve quality which budgetary outlays. Detailed costing and deteriorated due to rapid expansion. New assessment of impact on poor people are indicators include school completion rate needed. Performance targets and tracking (rather than just enrollment rates) to try to indicators lacking. capture quality dimension. Initiated development of secondary school strategy. APR 3 Focus on quality continues, with some Implementation plan made more 4/2004 improvements in primary sector noted. operational with specific measures. 95% of allocated funds used for intended Implementation of core programs initiated. purposes. Need to build capacity of MOE Taxes which inhibit trade repealed. to provide policy and technical assistance, as well as monitoring support, to district governments. Began to implement secondary school strategy. Source: Tanzania country documents 38. Moving forward, sustained support is needed to accelerate this process. When PRSs do not provide clarity on what is to be done, by which institutions, within what timeframes, or when they are insufficiently prioritized, they are not operationally useful--to governments in tracking their programs, or to citizens in monitoring their implementation. To facilitate this process, countries may need help to address analytic gaps, although attention is also needed to the institutional structures and incentives that encourage the use of that information. There is also a need to strengthen mechanisms for countries to evaluate the effectiveness of their development programs, and to make adjustments based on implementation experience. Donors need to recognize the magnitude of the task at hand, and avoid overloading the process by focusing only on specific issues that are of interest to them rather than supporting broader, sequenced efforts consistent with country priorities and capacity. 18 Linking the PRS to the MTEF/budget 39. Strengthening the link between the PRS and the budget process is essential to institutionalizing the PRS approach, ensuring that it is adapted to local circumstances, and helping countries better prioritize their strategies. More attention to costing the programs in PRSs, as well as alternative measures, also would facilitate this link. Medium-term expenditure frameworks (MTEFs) can also help link the PRS to the budget process through greater clarity of objectives, predictability in allocations, and more comprehensive coverage and transparency in the use of funds. They can also show the financial impact of new initiatives, both in the current and in future years. In doing so, the implications of delays or shortfalls in donor disbursements are made more clear, as is the financial burden of the new initiatives falling on the government budget in the years immediately after the donor support for the new initiatives has run out. 40. Although MTEF reforms are now common in many countries, they require political commitment and deep institutional reforms, and are constrained by weak institutional and technical capacity.44 They have thus proven more challenging to implement than initially envisaged.45 The more successful countries have benefited, although to varying degrees, from high-level political commitment to budget reform, active engagement of the cabinet in the PRS and MTEF process (Benin, Rwanda, Tanzania, Uganda),46 and a gradual opening of the budget review process to donors and civil society (Uganda, Tanzania, Rwanda).47 Even in the case of Uganda, however, where reforms of the public financial management (PFM) system began in 1992, a need for coordination and capacity building still exists, highlighting the long-term nature of the MTEF and PFM reform process. Continued efforts are also needed to ensure that the MTEF is linked to the budget process.48 41. Linking the PRS to the budget and MTEF has also been more successful where planning processes already existed and budget reforms were underway.49 44Holmes and Evans (2003); Andrews and Moon (2004). 45MTEFs are now common in countries where PRS implementation is more advanced. Of the 28 countries that have prepared MTEFs at various stages of implementation, eleven have been implementing PRSs for more than two years (Albania, Benin, Burkina Faso, Malawi, Mali, Mauritania, Mozambique, Rwanda, Tanzania, Uganda, Zambia), and an additional eight have at least one APR (Armenia, Azerbaijan, Cameroon, Ghana, Guinea, Kyrgyz Republic, Nepal, Tajikistan). Five are in the process of introducing their first MTEF to improve budget alignment with the PRSP (Ethiopia, Honduras, Nicaragua, Senegal, and Vietnam). 46Honduras, also, has approved an Organic Budgetary Law which sets the legal basis for developing this framework. 47This is confirmed by a number of studies that identify political support and engagement (through cabinet and parliament) and a strong role of the lead ministry (usually Finance) as central for the relevance and credibility of an MTEF. See Garnett and Plowden (2004); Holmes and Evans (2003); Andrews and Moon (2004); Piron and Evans (2004). 48This is an issue in many PRS countries in Eastern Europe and Central Asia, where dialogue between line ministries and the Minister of Finance remains weak and few of the MTEFs are located within an effective two-stage budget process where government sets the budget parameters for the preparation of the subsequent annual budget, see Andrews and Moon (2004). Continued attention is also warranted in other countries such as Cameroon, Gambia, Ghana, Guinea-Bissau, and Madagascar. 49GTZ (2005: 25). 19 Vietnam, for example, was able to rely on an existing planning system. Other countries, such as Tanzania, Uganda, and Burkina Faso have benefited from good pre-existing expenditure planning and information systems, inter-ministerial coordination mechanisms and, except in Burkina Faso, performance and results orientation at the outset. 42. MTEFs and sector strategies can be mutually reinforcing. Well-developed sector strategies (e.g., Uganda, Tanzania, Albania) have facilitated the adoption of MTEFs, while drawing up sectoral MTEFs has helped to build capacity and foster expenditure planning. Drawing up sectoral MTEFs can help establish a comprehensive view of sector expenditures while better grounding sector strategies in budget realities (box 6). Gradually strengthening the linkage between the PRS and the MTEF by developing the MTEF sector by sector has worked well in a number of countries, such as Ghana, Georgia, Kyrgyz Republic, and Rwanda (box 7). Box 6: An emerging education sector MTEF--Vietnam The government of Vietnam is beginning to pilot "bottom up" MTEFs in four sectors and four provinces, starting with the education sector. To date, work centers more on developing an analytical tool than a platform for policy dialogue between the Ministry of Education and Training, the Ministry of Finance and the Ministry of Planning and Investment. Nevertheless, useful insights are emerging from analysis of the existing cost structure of the sector and the interaction between existing policies and demographic trends. It is expected that the education MTEF work will become increasingly integrated with investment planning, state budget formulation and political dialogue and that similar approaches will be taken in other sectors. The sector MTEF can also provide an essential platform for sector budget support which is being considered by a number of donors. Source: OECD-DAC (2005c). Box 7: A gradual approach to MTEF reform--Rwanda In Rwanda the MTEF approach has been gradually mainstreamed in all ministries and districts, although the latter still require better integration with the national-level MTEF and budget process. Substantial progress has been made in costing and prioritization of strategies in some sectors (education, health, and more recently energy, water, infrastructure), although there is still room for improvements, especially at the district-level. The full integration of capital and current budgets is still outstanding as access to data on donor-financed development activities remains limited. Source: World Bank (2005h: 8). 43. A renewed focus on Public Expenditure Management (PEM). There have been some moderate improvements in PEM over the last few years, due to greater government attention and a high level of donor involvement.50 Measures to improve PEM are increasingly included as elements of poverty reduction strategies,51 and a shift has taken place from the focus on budget formulation issues in early PRSPs to more 50An average of seven donors per country provide PEM assistance in HIPC countries. 51Heavily-indebted poor countries are increasingly incorporating PEM reform measures from their HIPC Assessment and Action Plan into their PRSs, with the average number of PEM measures rising from eight in 2002 to over 16 in PRSPs produced in 2003, IMF/World Bank (2003b). However, while these measure help to track whether funds are put to the intended purposes, attention is also needed on whether funds are used efficiently (cost per activity) and whether they deliver the desired outcomes/benefits. 20 attention to budget execution and reporting. The extent of progress, however, is mixed across the 23 HIPC countries that were assessed in both 2002 and 2004. Total benchmarks met increased by about 10 percent from 2002, with budget reporting showing the largest improvement. Those PEM areas identified as the weakest in 2002 improved significantly by 2004, but substantial efforts are still needed. Only two countries need little upgrading to their PEM systems (Tanzania, Mali), while five need some (down from nine in 2002) and 16 still require substantial upgrading (15 in 2002).52 44. Experience has shown that a careful phasing of reform measures is important. In addition to ownership, reforms need to reflect a good understanding of local conditions, support government priorities, and capitalize and build on existing political support. While a number of basic interventions are critical for the formulation of an MTEF, others can be introduced more gradually. 45. Decentralization. Some countries with ongoing decentralization processes have begun to increase coordination of their PRS process with development planning at the regional and/or local level. In countries with decentralized service delivery, aligning local development plans with the PRS is seen as essential to rationalize planning and ensure accountability and transparency of local expenditures. Among others, Bolivia, Ghana, and Mauritania have made efforts to align national and local planning by using local development plans as inputs for the PRS and/or revising the plans to be consistent with the PRS. In the Kyrgyz Republic, strong local government involvement in PRS consultations has resulted in local development plans that reflect PRS priorities. In Pakistan and Bosnia and Herzegovina, the PRSP is used as a guiding document for MTEF and budget preparation at the regional level. 46. But capacity constraints and severe resource shortfalls at the local level make this process very challenging. During PRS discussions in Bolivia, local governments expressed the need to strengthen administrative and analytical capacity at the municipal and regional level, to be able to better coordinate local with national policies and implement priority projects. In Rwanda's Ubudehe approach,53 9000 communal cells prepared participatory poverty assessments, which fed into the development planning process and MTEF formulation of the district governments. Districts can then allocate block grants to communities, in addition to a cash transfer each receives, to realize their priorities.54 Consolidated district MTEFs inform the Ministry of Finance and the national MTEF process, although information overload can create significant management problems at the central level. 52 Caution should be exercised in comparing the findings of the re-assessment with the initial findings. More precise assessment criteria, coupled with stronger efforts at quality review, have improved the comparability of results across countries in 2004. This enhanced rigor, however, also resulted in somewhat stricter standards of assessment for some of the indicators. 53A grassroots approach used to identify and prioritize local development initiatives. 54This process is currently supported with financing from the EU. 21 Strengthening monitoring systems 47. The PRS approach and other initiatives have strengthened the focus on results. This has created a substantial incremental demand for data, underscoring the need for effective national monitoring systems. However, significant challenges remain in developing well coordinated monitoring systems with quality information that is accessible to various stakeholders. While only a handful of countries currently have such systems in place, nearly all countries are taking steps or at least considering actions to improve these dimensions (figure 11).55 Not surprisingly, developing coordinated monitoring systems--which requires complex institutional change--has proven more challenging than improving stakeholder access to information and the more technical aspect of improving data quality.56 In particular, most countries continue to have difficulties in tracking and measuring the effectiveness of pro-poor spending. Figure 11: Result-focused monitoring systems 100% 3% 3% 5% 80% 48% 46% 39% 60% Largely developed 40% Action being taken 49% 51% 53% Elements exist/being considered 20% Little or no action 3% 0% Quality of development Stakeholder access to Coordinated country information development level monitoring and information evaluation Source: World Bank (2005a). Sample includes 59 countries as of June 2005. 48. Quality of data. Although statistical capacity in IDA countries has improved over the past five years, substantial improvements are still needed. Based on a diagnostic framework that considers three dimensions--(i) statistical practice (the ability to adhere to internationally recommended standards and methods); (ii) data collection (frequency of censuses/surveys and completeness of vital registration); and (iii) indicator availability (availability and frequency of key socioeconomic indicators)--some progress has been in made between 1999 and 2005 along all three dimensions (figure 12).57 Of some concern, though, is the limited improvement in data collection in all IDA countries, and in African countries in particular. The indicators reveal differences in the strengths 55World Bank (2005a). 56 Coordinated monitoring systems include a range of functions: coordinating among data producers; organizing information flows among different stakeholders; compiling data from different sources; organizing analysis and evaluation; reporting on progress; providing the basis for the development of government activities and donor support; and disseminating information to other stakeholders. 57The diagnostic framework used and the resulting Country Statistical Information Database are available at: http://www.worldbank.org/data/countrydata/aboutcsidb.html. 22 and weaknesses of country systems, confirming that country-based programs for capacity building will be needed to produce sustainable results. Figure 12: Statistical capacity scores in IDA countries, 2005 Practice Practice Non African 100 African 100 IDA countries IDA countries 80 80 1999 60 60 2005 40 40 20 20 - - Availability Collection Availability Collection Source: World Bank, Development Data Group http://www.worldbank.org/data/countrydata/csid.html 49. Access. An important component for effective accountability is that existing data be accessible to various stakeholders. Although a range of factors are likely at play, there is some evidence that the PRS process has encouraged greater access to data, but more effort is needed. For countries that have had a PRS in place for at least two years, 70 percent provide access to development information, or are taking actions towards that end. The remainder are at least considering the necessary reforms. Progress, however, remains disappointing in low-income countries under stress (LICUS), where only a quarter of the countries have taken steps to improve data accessibility (figure 13).58 Figure 13: Stakeholder access to development information 100% 3% 9% 25% 80% 46% 50% 60% 61% Largely developed 40% Action being taken 51% 75% Elements exist/being considered 50% 20% 30% Little or no action 0% All PRS 2 years or PRS less than 2 LICUS more years Source: World Bank (2005a). Sample includes 59 countries as of June 2005. 50. Coordinated systems. Progress in building monitoring systems that coordinate the collection of data, its analysis, and its use for policy making has been limited in many countries. In fact, this is the area that joint staff advisory notes (JSANs), and 58World Bank (2005a). 23 previously joint staff assessments (JSAs), most frequently mentioned as a significant constraint to PRS implementation. The challenge is on both the supply and demand sides: (i) organizing a coordinated system; and (ii) encouraging the actual use of information in the policy process. Accelerating improvements will require treating the development of PRS monitoring systems as central to overall PRS design and implementation, rather than as a secondary process motivated by an obligation to donors. Since the participation of key actors depends on their interests and incentives, the formal and informal rules that govern PRS monitoring systems are therefore a key dimension of their success. 51. Countries usually have a range of existing monitoring mechanisms, often resulting from donor programs, but which lack a common framework, causing among other problems, duplication, excessive burdens, lack of compatibility of data, and poor information flows. Instead, efforts are need to strengthen coordinated, national systems. Adding new arrangements, even where technically superior, is unlikely to be successful, unless redundancies are eliminated and the overall burden minimized. While there does seem to be increased donor support for monitoring, it tends to continue to focus on each donor's own reporting requirements and priorities, thus impeding the development of national ones. To address this, there is a need for donor harmonization in this area, and for donors to clearly articulate their monitoring needs during the design process to ensure that their monitoring requirements are served through the national system. 52. While the challenges at hand are large, there is some evidence that the PRS process has contributed to improvements over time. For instance, in countries that have been implementing their PRS for at least two years, three-quarters had systems that were largely developed or were taking actions. The rest had elements in place or were considering actions. For countries that have been implementing their PRS for less than two years, only about a quarter were in the process of taking actions (figure 14).59 Good practice has pointed to some promising strategies for strengthening demand for monitoring, which is key to overall success (box 8).60 61However, the effectiveness of the PRS monitoring system will ultimately depend on the quality of the PRS itself. Only if the PRS is clearly prioritized and articulated can the PRS monitoring system be an effective way to improve accountability. Recognizing the centrality of effective monitoring systems, the Bank's Africa Action Plan indicates the Bank's willingness to assist all PRS countries in the region to develop and implement monitoring and evaluation plans and to build integrated monitoring and evaluation systems. This is intended to help put in place monitoring systems to track implementation and to encourage evidence-based decisionmaking. 59World Bank (2005a). 60Based on the analysis of 12 country studies (Albania, Bolivia, Guyana, Honduras, Kyrgyz Republic, Malawi, Mali, Mauritania, Nicaragua, Niger, Tanzania and Uganda) in Coudouel, Cox and Thornton (2005). 61Participation in the General Data Dissemination Standard (GDDS) has enabled some PRS countries to establish comprehensive statistical strategies and development frameworks. For example, Sierra Leone has integrated major components of the GDDS into its PRS, incorporating also the financing for the requisite reforms. 24 Figure 14: Developing coordinated country-level monitoring systems 100% 5% 13% 25% 25% 80% 39% 60% 61% Largely developed 40% 69% 70% Action being taken 53% Elements exist/being considered 20% 26% Little or no action 3% 0% 6% 5% All PRS 2 years or PRS less than 2 LICUS more years Source: World Bank (2005a). Sample includes 59 countries as of June 2005. Box 8: Strategies for strengthening demand for monitoring systems Strengthening the practice of analysis and evaluation: For PRS monitoring to influence policy, the practice of analysis and evaluation needs to be institutionalized. This is a striking deficit in most systems to date. Some countries have created central analytical units in the presidency, ministry of finance or national statistics institute, and these units have worked best where they remained small, close to government and focused purely on analysis. Another useful technique has been joint analytical exercises with donors, including Public Expenditure Reviews and Poverty and Social Impact Analysis. Tailoring and disseminating outputs: Monitoring information and analysis must be compiled into outputs and disseminated across government and to the public. This is another major weakness in existing systems. Most PRS monitoring systems are focused exclusively on the production of reports for donors, which are often inappropriate for domestic use. A PRS monitoring system must develop outputs which are tailored to the needs of the national policy cycle. Making the information accessible to the general public is also a key objective. Presenting monitoring data in a non-technical way is a new skill for governments, and is an area where cooperation with civil society partners would be valuable. Linking PRS monitoring to the budget process: The various occasions when government agencies bid for public funds--for example, during the elaboration of the annual budget, public investment plans, or a Medium-Term Expenditure Framework--present opportunities to require a justification of the bids by reference to PRS objectives and evidence on performance. This has been done in Uganda and Tanzania, with a notable (if uneven) boost to results-oriented policy making. To be more effective, it is helpful to have an agency close to the center of government with the capacity and authority to engage with the sectors on their policy choices. When linking the monitoring system to the budget, care needs to be taken to avoid creating perverse incentives which may distort the monitoring process. Involve parliaments: Parliaments should be a key user of monitoring information, but in practice have been involved very little in monitoring systems. This is partly explained by the low capacity of parliaments in many PRS countries, which are unable to engage effectively with the executive on policy issues without a strong committee system supported by analytical and research staff. Institutionalizing a role for parliamentary committees within the PRS monitoring system, supporting parliaments in this area, and forging alliances between parliaments and civil society could allow them to broaden their inputs into the policy process. 25 An enabling environment for strong accountability 53. While the previous sub-section focused on the technical needs for evidence- based decision making, this sub-section considers selected factors that can enable strong accountability mechanisms. It starts by reviewing progress in using participatory processes during PRS formulation and monitoring, noting the need to better ground those processes within a national agenda. Second, it considers the vital role of information in creating an enabling environment for accountability. It then considers the space for open dialogue around key policy choices including the use of poverty and social impact analysis and discussions on the macroeconomic policy choices that underpin poverty reduction strategies. Finally, it examines the criticism that neither parliaments nor poor people (and other marginalized groups) are sufficiently integrated in the PRS process.62 Institutionalizing participation 54. Participation can be a key factor to enable stronger accountability mechanisms.63 The original emphasis that the PRS approach placed on broad participation of civil society, other national stakeholder groups, and elected institutions was based on an expectation that such participation would improve the design and implementation of poverty reduction strategies. In addition, it was expected that civil society could play an important role in monitoring implementation and strengthening accountability.64 55. Relative to their starting points, in most countries the PRS approach has opened space for stakeholders to engage in a national dialogue on economic policy and poverty reduction. "In response to the question whether the PRSP represented the most open policy dialogue in their country to date, the uniform response from Oxfam offices and civil-society partners in virtually every country was a clear yes." (Oxfam, 2004: 6). However, participation in PRS formulation has tended to be broad rather than deep, with a wide range of stakeholders engaged, but to only a limited extent. Since the PRS process cannot immediately offset the shortcomings of weak institutions and political structures, there is substantial scope to better institutionalize participation in PRS formulation and monitoring. 56. The nature of stakeholder participation in PRS formulation has varied depending on countries' political structures, stability, traditions, and institutional capacity. Where there is little political openness, it provides an opening; where there is more openness and tradition of participation, it can promote greater and more tangible success. Some countries construed participation to mean one-time involvement of non- governmental organizations and interested members of parliament. Others, including 62See, for example, Venro (2005: 1) and O'Malley (2004: 1). 63In World Bank (2004: 89) participation is defined as the process through which stakeholders (those affected by the outcome of reform or capable affecting the reform) influence or share control over setting priorities, making policy, allocating resources, and ensuring access to public goods and services. The IDB (2005: 2) notes that participation is not a technical process but rather one that must be set in political context, and whose variables may have a dramatic impact on the institutional trajectories of countries. 64IMF/World Bank (1999b: 12). 26 Niger and Uganda, built their participatory processes on poverty reduction initiatives which pre-dated the PRS approach. In other cases, the PRS presented an opportunity to institutionalize dialogue with civil society as part of a broader nation-building effort (e.g., Bosnia and Herzegovina and Rwanda). 57. In many cases, different stakeholders had very different expectations regarding the participatory process. This points to the importance of communicating the goals of the participation process upfront in order to avoid the credibility and legitimacy of the process being undermined by expectations which may prove impossible to fulfill. For example, participation does not imply final consensus or that views garnered through participatory processes will necessarily be reflected in final programs. Instead, participation should enable policy choices to be better informed.65 Other lessons include the need for adequate time and information, and attention to representativeness. In response to lessons learned in formulating earlier PRSs, Uganda and Tanzania both developed consultation guidelines when they were revising their strategies to clarify expectations with regard to the participatory process (box 9 on Tanzania).66 Box 9: Institutionalizing participation in PRS formulation--Tanzania In its third annual progress report, Tanzania established "Consultation Guidelines for the PRSP Review" which stipulate that ongoing adjustment of PRS documents should provide for: Ample timeline for consultations Transparent planning of participation processes Expanded consultations across government agencies Leadership of consultations to be shared between government and domestic stakeholders Stakeholder participation to explicitly include business associations and the labor movement Expanded methods for engagement with civil society, including townhall meetings, radio programs, popular drama presentations, and television broadcasts Systematic analysis of civil society feedback and integration of key issues into the next PRS 58. While the initial emphasis was on participation in PRS formulation, efforts to institutionalize participatory processes in PRS monitoring are also essential. Early on, some countries were already developing and applying participatory monitoring tools. A recent acceleration of cross-country learning and knowledge sharing seems to have encouraged the spread of these tools, and some good practice is emerging. For example, public expenditure tracking surveys, report cards, and scorecards, which have been developed to promote the institutionalization of citizen feedback into periodic PRS assessments, show promise. Examples of institutionalized participation in formal monitoring systems are also emerging (box 10). 65World Bank (2004: 91). 66Morazan and Knoke (2004: 9) also cite Tanzania as an encouraging example of how ownership can be fostered throughout the implementation process. See Canagarajah and van Diesen (2005) for a discussion of the Ugandan experience. 27 Box 10: Institutionalizing participation in monitoring systems Armenia's PRSP Social Monitoring and Analysis System contains measures for collaboration in PRS monitoring, which explicitly involve central and local government, as well as community councils, NGOs, unions, business associations, religious organizations, vulnerable groups, television and written media. Cameroon's Integrated Monitoring and Evaluation of PRS Implementation, which established a special mechanism for managing and monitoring the use of HIPC funds by CSOs, has resulted in an unprecedented level of access to information by civil society. On-the-job capacity building has had a direct impact on the quality of CSO feedback on the process; and their ability to use independent analysis tools for monitoring national budget execution. CSOs are also empowered to conduct dialogue with government officials and donors on programming, budgeting and monitoring of poverty oriented developments. Malawi's Qualitative Impact Monitoring system, established in partnership between the government and other stakeholders to keep policymakers informed about progress in PRS implementing and impact, uses participatory methods. Uganda's emphasis on monitoring and evaluation has resulted in the mobilization of government departments and civil society stakeholders and the arrangements have produced tangible results. The Budget Directorate of the Ministry of Finance provides regular information on budget disbursements to CSOs for review. CSOs and parent associations collaborate at the district level, using public expenditure tracking surveys (PETS), to track budget implementation. CSOs provide independent reports which validate the work of local government. Value for money audits are conducted and CSO verify the quality of work. 59. During PRS implementation and monitoring, there is also scope to broaden the engagement of stakeholders that may have been less involved in the formulation stage. For example, while the involvement of the business community in PRS formulation has been uneven, there is potential for greater private sector engagement. In Burkina Faso, the private sector is pressing government for greater attention in the PRS to economic growth and business development. In Honduras, there is a focus on providing integrated services to micro-, small-, and medium-sized enterprises. In Malawi, a coalition of private sector and civil society organizations is conducting budget analysis to ensure government compliance with budgetary allocations. These efforts should be encouraged, as failure to include the business community can limit the potential synergies between public and private efforts to stimulate growth and poverty reduction. 60. There are also examples of local community participation in monitoring and updating PRSs. Honduras has formal mechanisms in place for considering inputs by local consortia, NGOs and community organization in the design and implementation of Strategic Plans for Municipal Development. In Nicaragua, space has been opened for municipalities, development councils, NGOs, private entrepreneurs, and civil society groups to identify and design locally-driven programs and projects, as well as for social accountability audits by civil society. Rwanda's Ubudehe process informs district level planning decisions and is reflected in aggregate plans at the national level. The role of communications 61. While stakeholder access to information is important, an `information-rich environment' alone does not ensure meaningful participation or, ultimately, country 28 ownership.67 Communication needs to be managed strategically in order to create demand for accountability and open space for more effective dialogue between governments and civil society. The quality and timeliness of communication and information has generally mirrored PRS participatory processes themselves.68 Where the process has worked well, governments have forged a more trusting relationship with civil society, and it has resulted in greater impact on policy and PRS review. Less successful experiences reflect communications being used largely to disseminate information, with little evidence that stakeholder feedback is taken into account at the policy level. Communication processes in the Cambodia PRS, which were problematic during the formulation phase, have experienced a marked improvement during implementation and revisions to the PRS. Government explicitly requested direct feedback from CSOs on the implementation phase, and consultations have resulted in regular dialogue with the general population. 62. Strategic communication can be usefully integrated in the PRS process, and some countries have applied innovative tools to engage the broader population. In Tanzania, simplified booklets in Kiswahili and English entitled "Tanzania without Poverty" were serialized in local newspapers and generated popular interest in PRS goals and their relevance for the lives of individual citizens. Simplified or local language versions of the PRSP were produced in Ghana, Vietnam, and Yemen for stakeholders with limited knowledge of the country's poverty reduction plans, and efforts are ongoing to sustain feedback from these groups. 63. Mass media also play an important intermediation role in this process through analysis and transmission of key policy information to the public, and provision of targeted feedback to authorities on the public's evaluation of progress. For example, in Rwanda, soap operas are used to engage the general public in the latest developments in PRS implementation. While accessible to only about a third of the population, the soap operas have been reinforced by additional coverage in local newspapers and monthly PRS update sheets published by NGOs. Opening the policy space for dialogue 64. Opening the space for policy dialogue is an integral part of domestic accountability. Broadening the national policy dialogue on poverty reduction strategies, with fuller involvement of all stakeholders, is seen as an opportunity to ensure more flexibility in carrying out national development policies, making them more inclusive and tailored to country circumstances. Calls for an expansion of this policy space, particularly with regard to macroeconomic policy, as well as the use of analysis to help clarify assumptions and to identify the potential effects of policy choices on poor people, has become an important element in the policy dialogue among governments, their development partners, including the Bank and the Fund, and civil society organizations. 67DfID/World Bank (2005: 12). 68Herz and Ebrahim (2005:39) note that access to information has been inadequate. Often key documents are either not provided or not sufficiently in advance. 29 65. Distributional analysis. Policy debate can be informed by distributional analysis (often referred to as poverty and social impact analysis). In reforms that are subject to substantial debate and disagreement, such analysis can create an opportunity for exploring conflicting hypotheses and being transparent about assumptions. While poverty and social impact analysis (PSIA) is not a panacea, and significant technical and capacity issues remain, there are cases where the PSIA process has helped to create a forum where stakeholders can raise their concerns. For example, in Ghana, an energy sector PSIA helped to dispel common misperceptions about utility reforms by examining the impact on poor people and taking those impacts and stakeholder views into account. In Congo (DRC), a mining sector PSIA raised awareness about the likely negative impacts of mine closure due to reductions in social services that the mines supplied. The PSIA identified vulnerable groups who would suffer most from reform that would entail shutting down schools and health stations. It resulted in 40,000 children being kept in school by putting in place a short-term assistance program to keep schools open providing more time to develop a sustainable, long-term solution. A fuller discussion of the use of distributional analysis is provided in section 3B. 66. Macroeconomic policies. The need to expand the openness of macroeconomic policy discussions figures prominently in the views of external stakeholders. Many observers believe that an open debate on what macroeconomic framework is most suited to reduce poverty is still lacking as a systematic part of the PRS process.69 In the view of some civil society groups, the opportunity to participate in their government's economic policy making is rendered meaningless because the policy space enjoyed by governments is so limited as to prevent any effective policy debate among different groups.70 Past implementation progress reports by Bank and Fund staff, and the 2004 evaluation of the PRS approach by the Fund' Independent Evaluation Office (IEO), have also concluded that the PRS process has had a limited impact in generating discussions of alternative policy options with respect to the macroeconomic framework and related structural reforms.71 67. There may be several reasons why macroeconomic policy options have not been discussed more broadly in the PRS process. In some instances, macroeconomic policy decisions (e.g., on the level of exchange rates or interest rates) are by their nature sensitive and cannot be made in a process of broad public debate. This is the case even in OECD countries. In other cases, governments are reluctant to discuss macroeconomic policy options in a broader public forum, as they consider this may slow decisionmaking 69Cordaid, Novib and Wemos (2005: 2). 70CIDSE-Caritas Internationalis (2005a: 6, 38). Some recognize, however, that this opportunity itself depends on the PRS process pursued by the government, which may be deficient and is often exacerbated by lack of transparency, institutional fragmentation, and widespread off-budget funding. See Water Aid (2005: 10). 71IMF/World Bank (2004: 19), and IEO (2004: 72, 111). The IEO report expressed doubts that PRGF program design had been guided by the PRSPs--in fact, the opposite influence has been more common. By contrast, Timor Leste provides an example where government encouraged broad-based consultations before adopting its petroleum savings policy which sets aside the bulk of petroleum revenues for future generations. Since effective implementation depends critically on its continued implementation by successive governments, the government considered it important to build popular consensus for the policy. 30 and complicate relations with donors. The institutional framework may also not be conducive to such discussions, 72 and the necessary information may not be available on a timely basis and in accessible form. Many potential participants in such discussions also lack the technical and institutional capacity to evaluate macroeconomic trade-offs and elaborate feasible alternative policy options.73 68. While it is true that the scope of public debate on the macroeconomic policy framework has been limited in early country experience with the PRS approach, there have been a number of recent examples of more flexible discussions between government officials and Fund staff on key macroeconomics issues during the development and implementation of the PRS. In some countries with more advanced PRS processes (e.g., Mozambique, Nicaragua), there have been discussions around setting alternative targets for growth, inflation, current account balances, and revenues and expenditures that are consistent with the objectives of macroeconomic stability and poverty reduction. There have also been discussions of the trade-offs between the possibly adverse macroeconomic effects of higher-than-projected externally financed spending and its poverty-reducing impact in Burkina Faso, Mozambique, Tanzania, and Uganda. However, these debates on policy options continue to be largely limited to a narrow circle of officials directly involved in macroeconomic programming and the negotiations with development partners, with relatively little direct involvement of the public. 69. More effective ways must be found to design, consider, and incorporate economic alternatives into PRSs.74 The country authorities, for their part, should establish a framework for debate on macroeconomic policies, ensure better information sharing with all stakeholders in the PRS process, and together with other partners, endeavor to build up the capacity, within government and among non-government stakeholders, to analyze macroeconomic information and to make informed alternative proposals. It would also be the responsibility of the government to set clear rules governing such public discussions, and to manage expectations.75 An improved macroeconomic policy debate will also require the willingness of Fund staff to consider 72The Fund's outreach and public activities in-country must take place within this framework. Fund staff can, however, encourage a government to open up the debate on macroeconomic policy issues, as this will promote better understanding and acceptance of the policies ultimately decided, and enhance the chances for their successful implementation. However, the Fund cannot organize a public debate of policies against the government's wishes. See IMF (2003a). 73There is some debate on whether the IFI staffs should themselves propose alternative policy options, or merely be willing to discuss the proposals of others on their merits. Alternative proposals by the staffs of the Fund and the Bank could run the risk of filling, at least partially, the policy space that should be left to the country authorities. 74 A variety of proposals have been made for achieving this, including: public oversight of the government's macroeconomic policy; better justification by the IFIs of their policy analysis, assumptions and recommendations in the light of government's PRS; or incorporating alternative proposals for stabilization and alternative scenarios for financing and expenditure, including in case of shocks, into Fund- supported programs. Some civil society organizations consider that they, together with parliaments and private sector groups, should participate in program discussions between the government and the Fund. 75For example, it should be made clear that some macroeconomic policy decisions cannot and should not be made in open fora, and that PRGF program negotiations will not be conducted in public. 31 all feasible policy options that are consistent with the objectives of poverty reduction and macroeconomic stability,76 and to inform the public debate on these issues through appropriate outreach activities and a clear presentation of the staff's views on macroeconomic policies, constraints and perspectives.77 The Bank can assist by analyzing the consistency of growth paths with medium-term development goals or the MDGs, and by examining the implications of different combinations of government spending for progress towards those goals. Neglected stakeholders 70. Many observers have noted concerns about the lack of involvement in the PRS process by critical political decision-making and representative bodies such as parliaments. They note that early focus on civil society engagement in PRS formulation may have drawn attention away from the involvement of representative bodies.78 There is concern that bypassing existing processes of deliberation and dialogue and creating parallel structures may undermine formal channels of representation such as local governments and parliaments. However, it is important to note that opportunities for legislative involvement in the PRSP process are country specific. Furthermore, legislatures in many PRS countries may lack the human and institutional capacities needed for effective engagement. 71. Despite the mixed record from 1999 to 2005, there has been an upward trend in parliamentary involvement in the PRS process. PRSPs have been formally presented to parliament in about one third of the countries with full PRSPs. Several PRSPs highlight the role of parliament in PRS oversight, either through a PRS standing committee (e.g., Azerbaijan, Ghana), other standing committees (e.g., budget committees), or through membership of PRS Steering Committees managed by the executive branch (e.g., Chad and Georgia). The parliaments of Bosnia and Herzegovina and Ethiopia are implementing changes to facilitate legislative oversight of PRS implementation. Many governments report to parliaments on budget and PRS progress (e.g., Benin, Kyrgyz Republic, Mozambique, and Tajikistan). 72. Parliaments are also increasing their ability to contribute to PRS prioritization, costing, policy coordination, and monitoring of poverty outcomes.79 Capacity building efforts that support PRS monitoring, including the link to the MTEF and budget, have resulted in more targeted policy recommendations on the PRS to the executive branch (e.g., Burkina Faso and Cameroon). Regular engagement with constituents through outreach activities and public hearings can also improve the quality of feedback to policymakers (e.g., Rwanda and Tanzania). Other observers have noted that strengthening links between active civil society groups and parliaments could help 76 Obviously, there is considerably more scope for this flexibility in countries where macroeconomic stabilization has already been achieved. See IMF (2005b), forthcoming. 77 However, for the Fund, at least, the scope for this type of activity is constrained by its limited representation in-country, as well as the relative infrequency and short duration of missions from headquarters. 78See, for example, Oxfam (2004: 7) and GTZ (2003: 27). 79Draman and Langdon (2005: 16). 32 create a common platform to support pro-poor priorities.80 With support from parliamentary associations worldwide, including the Parliamentary Network of the World Bank (PNoWB), informal networks have been formed in East Africa, the Middle East, and Latin American to strengthen parliaments' abilities to engage in country-level PRS processes. 73. There is also concern that the views of poor people and other marginalized groups have not been adequately reflected in poverty reduction strategies. This is in part because direct engagement of poor people takes more time than existing planning cycles allow, and empowerment of the most vulnerable members of society is fundamentally difficult to do. Vulnerable, marginalized, and disempowered populations generally have less voice, fewer assets, weaker networks, and suffer more from the effects of non-income poverty than the average poor person.81 This form of powerlessness results in a failure to include poor people systematically in the generation of information and adjustment of policies and actions which affect them.82 This, in turn, prevents policymakers' accountability to citizens through the "long route" of accountability.83 The use of participatory poverty assessments for raising the visibility of poor people's views, such as those in The Gambia, Rwanda, and Uganda, go some way in addressing this issue. 74. The relative importance of different groups will vary depending on country circumstances, and efforts should be made to identify these groups and put in place processes to capture their views and concerns.84 For example, in some countries, it may be important to include indigenous groups as an explicit stakeholder of the PRS process Examples of inclusion of indigenous people's concerns in the PRS are found in Bolivia, where indigenous and peasant men and women are elected as Presidents of the Social Watch Committees which are responsible for control and monitoring of public expenditures. In countries affected by conflict, the views of internally displaced persons, returnees, and refugees may be of particular concern. However, ensuring balanced representation of poor people and marginalized groups remains a problem, as their concerns may not be directly addressed by CSOs focused on specific issues. Box 11 provides an example of how Montenegro has included vulnerable groups in the PRS process.85 80Parliamentary Centre (2005: 2, 6). This type of approach is being to emerge, for example, in Malawi and Ghana. 81"Civil society must increasingly invest in grassroots-based and national-level policy monitoring and analysis, in order to ensure that evidence-based advocacy can be carried out and to enable the marginalized to fulfill their right to participate in the decisions that affect their lives." CIDSE-Caritas International (2004: 3). 82Venro (2005: 1) points out that significant improvements in the PRS approach can only occur if there is empowerment amongst the poorest sections of the population. 83The World Development Report 2004 contrasts short-route accountability through consumer monitoring of service delivery (for example through score cards) with long-route accountability through citizen monitoring of policy delivery. 84Kindernothilfe (2005: 4) and others stress that the views of children and youth need to be taken into account. 85In other cases, the PRS process included the preparation of a legal framework to guarantee rights and design targeted programs for disabled, elderly, refugees, and displaced populations (Cameroon, Kenya, 33 Box 11: Reaching vulnerable groups--Montenegro In Montenegro, analytical work conducted by the Center for Enterprise and Economic Development provides a good example of how poverty analysis and tracking indicators addressed the concerns of the Roma community. The work also focused on other vulnerable groups: internally-displaced populations (IDPs) and refugees from the wars in Bosnia and Kosovo. As a result, several important decisions were made. First, a national action plan for Roma inclusion was developed and supported by the parliament in early 2005. With the support of UNDP, a strong advocacy network was mobilized that was instrumental in securing broad support. Second, the authorities established a commission headed by the Minister of Labor and Social Welfare to develop and implement a national strategy for resolving the problems of refugees and internally displaced persons. This strategy, which includes concrete steps (such as amendments to the labor law and changes to social programs) was adopted in early 2005, and its implementation has started with active involvement of local NGOs. Both strategies are integrated in the PRS framework and their implementation is monitored as part of the PRS implementation. 75. While gender is addressed in an uneven way in PRS documents, it is generally recognized that poverty outcomes are likely to be more positive if gender dimensions are integrated fully into PRS planning and implementation. More progress is needed in mainstreaming the treatment of gender, which is still frequently considered under a social sector rubric. One challenge is that government ministries responsible for women's affairs and women's advocacy groups tend to have limited expertise in macroeconomic issues, and are therefore disadvantaged when it comes to negotiating gender-equitable policies. They also often lack authority and/or budget allocations for follow-up action (box 12).86 Box 12: Mainstreaming gender in the PRS--Bangladesh Bangladesh. The PRSP Steering Committee includes a Gender and PRSP Group (GPG), chaired by the Ministry of Women and Children Affairs. The GPG is composed of representatives from government and civil society. With support from donors, the GPG conducted gender analysis for the I-PRSP and identified critical gaps and a roadmap for mainstreaming gender in the full PRSP. GPG held extensive consultations with poor women and men, and developed a gender matrix and indicators for monitoring outcomes. Stakeholder consultations at national and regional levels resulted in the inclusion in the PRSP of a section on women's advancement which addresses gender-based dimensions of poverty. Seven strategic objectives were also identified to address gender vulnerabilities: (i) ensuring women's full participation in mainstream market-oriented economic activities; (ii) improving women's efficiency (reduce time use) in performance of household and reproductive roles and responsibilities and ensuring that infrastructure services reach women; (iii) building women's capacity in health and education in order to increase productive capacity and fulfillment of personal achievements; (iv) ensuring social protection for women against vulnerability and risk; (v) enhancing women's political empowerment and participation in decision-making; (vi) ensuring that women's voices are articulated in international fora; and (vii) strengthening gender-related institutions and capacity for monitoring and evaluation. Gender indicators are mainstreamed throughout. Mauritania and Senegal). The Senegalese monitoring system places special emphasis on assessing the ability of community organizations to target poverty reduction measures on vulnerable groups. 86UNIFEM (2005:2) notes that links between the Ministry of Finance and other line ministries, including the Ministry of Women's Affairs, are not strong. Capacity building is needed to strengthen these links. 34 76. There are also preliminary indications that gender budgeting can contribute to greater transparency and accountability.87 Such initiatives can illustrate the gender impact of budget decisions, e.g., the introduction of fees in the health sector may raise poor women's workloads because they must allocate time to caring for the ill when their families cannot afford the fees. Several countries are at various stages of integrating gender budget initiatives into their budget reforms (e.g., Bangladesh, Kenya, Mozambique, Rwanda, Tanzania, and Uganda). Some special considerations in conflict-affected and fragile states 77. The principles underpinning the PRS approach apply equally in conflict- affected and fragile states. And, while many issues related to developing and implementing a PRS are similar to other countries, constraints are often more binding. As such, it is even more imperative that flexible approaches be adopted and that the tools not be used mechanistically. The approach in each country needs to be realistic and tailored to country circumstances. This sub-section highlights a few special considerations. Issues related to aid are addressed further at the end of section 2C. 78. Tailoring PRS content in conflict-affected countries requires sensitivity to several dimensions. For example, in countries recently emerging from conflict, there is often a greater need to integrate security, political, humanitarian, and development concerns. With a focus on reconstruction and recovery, short-term priority setting and planning is likely to have greater urgency than long-term planning. Policy choices can also be strengthened by considering the sources of conflict and its legacies, even in cases where it may be politically difficult or counterproductive to peace and reconciliation to include detailed references in written documents. While severely weakened capacity, incomplete and outdated data, and poor access to certain regions often limit the ability to undertake poverty diagnostics, some countries have usefully examined the interrelationship between conflict and poverty, including through regional disparities, internal security, unequal access to services, and ethnic relations, in the analytic work that underpins their PRSs (e.g., Burundi, Rwanda, Sierra Leone).88 79. In fragile states, governance reforms and institutional strengthening often need to be at the heart of a poverty reduction strategy. However, governance aspects have generally been poorly covered in the PRSs that these countries have formulated. Countries should be encouraged to identify a few governance reforms that are central to PRS formulation and implementation, and to focus initial monitoring around those elements. This also requires a more proactive approach by countries in identifying their 87Gender budgeting analyzes the impacts of both public expenditure and revenue policies on women, men, girls and boys at national and local levels, and targets the re-prioritization of expenditures and revenues in favor of the needs and interests of poor women. The approach can be used to monitor whether allocations to a sector or program contribute to gender equality, and to uncover discrepancies between stated priorities and actual allocations. 88Transitional results frameworks, introduced in countries such as Liberia, Timor Leste, and Sudan, offer a simple application of the most basic PRS principles in the early stages of post-conflict recovery, and can build the basis for more comprehensive analysis and participatory planning. 35 technical assistance needs.89 Simple documentation formats and action matrices should be encouraged to facilitate implementation of the strategy. 80. Ownership of the process often needs to be built up over time, although the conditions for country ownership will vary, depending on whether a country is recovering from conflict, political crisis, or poor governance, or is facing a stagnant or declining governance environment. In all cases, there is some evidence that encouraging participatory processes can be beneficial, but with realistic expectations and a clear understanding of the political context in order to assess and manage risks.90 A special feature is to ensure that participatory processes are conflict sensitive. Several examples point to a strong rationale for addressing conflict through broad-based participation and engagement with war-affected groups during PRS formulation (e.g., Bosnia and Herzegovina, Sierra Leone, and Rwanda) (box 13). Maintaining a balance between the central government's steering of PRS design and implementation and greater local-level decisionmaking and choice in formerly unstable or underserved regions also presents a special challenge in conflict-affected countries. Box 13: Conflict-sensitive participatory process--Bosnia and Herzegovina The PRS participatory process in Bosnia and Herzegovina included all regions and ethnicities, and the strategy was reviewed and approved by the parliament. Genuine efforts were made to incorporate conflict-affected voices early in the process, including refugees and internally displaced persons, women, and single-headed households. Thematic consultations considered the rights of war veterans, the families of fallen soldiers, missing persons and military personnel disabled in the war, refugee return, corruption, human rights, unemployment, and rural poverty. Consequently, NGOs rated the process as highly participatory, and one which they committed to remain engaged with over time through monitoring. The PRSP Coordinator's office stated: "The Government was able to learn more about the problems faced by the most vulnerable categories of the population, while on the other hand, civil society representatives were able to mobilize and establish dialogue with government representatives." C. EXTERNAL ASPECTS OF ACCOUNTABILITY 81. This section considers the external aspect of accountability in the PRS approach--what do external partners expect from developing countries; and what should developing countries expect from donors. While the PRS approach encourages donors to align their assistance around country priorities rather than their own, governments receiving development assistance also have accountabilities for results to those who provide it. Donors can naturally require that development assistance be used in a transparent and efficient manner to produce tangible benefits. These requirements broadly coincide with the interests of domestic stakeholders. In addition, donors have also committed to provide coordinated, predictable, and harmonized aid that is aligned to country priorities. 89Thornton and Cox (2005: 11-13). 90Thornton and Cox (2005: 10) and World Bank (2005e). 36 Donor accountability and aid quality 82. In response to improvements by developing countries to their policies and governance, donor countries are expected to provide more and better aid. As part of the mutual accountability framework set out in the Monterrey Consensus91 and in the follow-up to the Rome High-Level Forum on Harmonization in February 2003, the Paris Declaration on Aid Effectiveness (March 2005) reaffirmed donor commitments to align development support to country-defined strategies and priorities; to use a country's own institutions and systems wherever possible (and when not possible, to ensure that safeguard measures introduced by donors help strengthen rather than undermine country systems and procedures); to provide reliable indicators of aid commitments over a multi- year framework and to disburse aid in a timely and predictable manner; and to implement common arrangements and simplified procedures. All of these measures are supportive of a balance in accountabilities. This sub-section considers aid quality in the context of the PRS approach.92 83. Aid quality. The PRS approach was intended to serve as a framework for coordinating donor activities, harmonizing donor support, and aligning it with country priorities so as to achieve development results.93 Consistency between donor supported programs and the development priorities of recipient countries is essential for recipient countries to assume full ownership of their development strategies and to enhance the development effectiveness of external assistance. Coordination and harmonization of donor procedures could also increase the reliance on domestic systems, helping enhance their quality and also reinforcing domestic accountability (box 14). Box 14: The international aid harmonization agenda In keeping with the commitments under the Monterrey Consensus, since 2002, the donor community, led by the OECD-DAC and the multilateral development banks, has developed good practice principles for simplifying and harmonizing donor procedures and changing aid delivery and management practices to enhance the effectiveness of aid. These include: using country systems to the extent possible for reporting and monitoring the use of aid resources; choosing aid modalities that minimize transaction costs to recipient countries and strengthen country systems; relying on joint diagnostic and analytical work; delegating responsibility for administering aid, including disbursement decisions, from donor headquarters to their representatives in-country; and coordinating donor support programs within a partnership framework led by the respective country authorities. In March 2005, the international community agreed on the Paris Declaration, which set out a series of commitments by donors and recipient countries for more rapid progress in the areas of ownership, harmonization, alignment, managing for development results, and mutual accountability. Progress in implementing these commitments will be measured through 12 indicators with specific time-bound targets for 2010 which are expected to be agreed by the time of the UN General Assembly meeting in September 2005. See OECD-DAC (2003a) and Paris Declaration (2005). 91In addition to providing more and better aid, the Monterrey Consensus also emphasizes the responsibility of developed countries to open their markets. 92Issues related to aid volume are addressed in section 3C. 93 The original PRS policy documents called on donors and multilaterals to "...make medium-term commitments in support of the strategy agreed in the PRSP", and "to increase aid inflows--particularly in grant form--in support of such policies with improved harmonization of aid procedures." Higher aid 37 84. A key ingredient is for country authorities to take a leadership role in coordinating external assistance. Donors can support such efforts by encouraging country authorities to chair sector and thematic groups, as well as Roundtables and Consultative Group meetings. Holding such events in-country should also be encouraged so that local institutions can play a stronger role. However, there is also much that donors can do on their own to improve the quality of their aid. 85. Overall progress on improving aid through alignment, harmonization, and better coordination has been mixed. While donors often cite problems with the PRS or weaknesses of country systems as constraints to alignment and harmonization, there is much donors can do to improve their practices. Donors are making progress in focusing their support on the priorities areas specified by countries; starting to use simplified procedures and practices; and collaborating in joint analytic work and sector- wide approaches. A limited start has also been made on the use of country systems. While this emerging good practice in a wide variety of country contexts has not yet become general practice,94 there is growing evidence that the PRS approach is bringing about an improvement in the coordination of donor support and a re-alignment of individual donor support programs with the country's priorities. Studies by the Strategic Partnership with Africa and the OECD-DAC of the alignment of budget support and sector programs with the PRS indicate some progress, but note that much more remains to be done (box 15).95 86. Donors have stepped up their efforts to improve the coordination and alignment of their support. This is reflected in the growing number of examples of good practice that have emerged in recent years, including an increasing willingness to provide direct support for budgets and sector programs; broader use of delegated cooperation and silent partnerships; and greater collaboration in diagnostic and analytical work.96 Two significant tendencies have emerged recently--the devolution of wider responsibilities to field offices, and the formation of joint government-donor groups, particularly among budget support donors. Box 16 discusses the experiences of the Bank and Fund with respect to alignment and harmonization. volumes and effectiveness were also a centerpiece of the 2002 Monterrey Consensus. See IMF/World Bank (1999b: 23-24). 94Past progress reports on PRS implementation by Bank and Fund staff, and the recent reports on the PRS approach by the independent evaluation units of the Bank and the Fund found only modest progress. See IEO (2004: 9), OED (2004: 8), and annex 2 which summarizes past progress reports. 95For example, Eberelei (2005: 4-5) in a study of the Zambian experience finds efforts on the part of some donors (though not, notably, by the IMF) to link their support strategies explicitly to the PRS, but little evidence of willingness to replace their individual strategies by a joint donor assistance strategy. AFRODAD (2003: 10) calls on donors to "respect local initiatives and resist the temptation of giving blueprints", and recommends that "Conditionality that negates ownership and true participation should be avoided". See also World Vision (2005: 47). 96For example, 38 percent of CPARs conducted by the Bank in FY04 were prepared in conjunction with other donors, with similar collaborative efforts for CFAAs. There is also broad support for the joint Public Expenditure Financial Accountability Initiative and its approach to a common assessment of public financial management systems and joint support for associated reforms. 38 Box 15: The 2004 SPA and DAC surveys of harmonization and alignment For the past two years, the Budget Support Working Group (BSWG) of the Strategic Partnership with Africa (SPA) has conducted a comprehensive survey of the progress made in aligning donor budget and balance of payments support with country PRSs. The 2004 survey in 15 African countries found some improvement in the coordination of capacity-building support, but noted a need for further progress on strengthening the alignment and predictability of budget support, and coordinating missions, conditionality, and reporting requirements: Less than a third of donors considered the content of the annual PRSP progress report (APR) sufficient for their disbursement decisions. The alignment of donor processes and cycles with national processes like the budget was generally partial at best. Despite progress, there continued to be problems in enhancing the predictability of disbursements. And finally, only in a few country cases was conditionality directly derived from the country's PRS. The SPA's sector work found some progress in aligning sector programs strategically with the PRS. However, there were still problems in establishing links to the budgets and MTEFs, and scope for greater donor collaboration in implementing sector support, including through delegated cooperation and silent partnerships. Building on the SPA methodology, the OECD-DAC conducted a survey of the alignment experience in 14 countries in 2004. This survey indicates that while donors have clearly internalized the principle of aligning their programming on the needs and priorities, issues remain in turning this commitment into practice. One issue was donors' view that the PRSs currently in place were often of qualified operational value. There were also still major problems of cumbersome and costly donor practices, with few examples of streamlined conditionality, delegated cooperation, joint missions, shared diagnostic reviews or other good practices. While the survey reasonably concludes that much needs to be done in terms of adapting donor programs to the priorities of the countries, it also found interesting examples in Ethiopia, Kyrgyz Republic, Tanzania, and Vietnam, where the PRS has clear action-oriented priorities supported by quantitative and qualitative targets that provide a strong basis for donor alignment and for objective performance assessment. See SPA (2005a); SPA (2005b); OECD-DAC (2004). 87. An increased field presence and greater willingness to decentralize operations and devolve more policy and operational decision-making responsibility to field offices allows greater awareness of the specificities of the country circumstances; shorter delays in making decisions and disbursements; and more effective participation by donors in the PRS process itself. This increases the likelihood that donor support programs will be articulated around the country's priorities as set out in its PRS.97 88. Reflecting this decentralization of responsibility, joint donor groups have become a more common feature over the past few years, ranging from informal groupings of "like-minded donors"(e.g., in Vietnam, Malawi), to formal arrangements under government leadership with negotiated memoranda of understanding or partnership framework agreements (e.g., in Tanzania, Mozambique, Cambodia, Ghana, and Burkina Faso). Such groups offer a range of advantages--joint performance monitoring and evaluation; coordinated conditionality; designation of "lead donor" roles; more intensive 97 There could also be a risk that decentralization leads to deeper donor penetration in domestic policy making with a risk to country ownership. 39 alignment of donor programs with PRSs; and, eventually, the establishment of mutual accountability frameworks.98 Box 16 : Bank/Fund experience with Harmonization and Alignment The World Bank has been a driving force in the harmonization and alignment movement, including as a leading participant in the work of the OECD-DAC, and in conjunction with other multilateral development banks. It has made considerable progress in the areas of joint analytical work, evaluations, and operations, and is coordinating its country assistance strategies with other donors in a number of countries (Bangladesh, Cambodia, Honduras, Nigeria, Tanzania, and Uganda). In fragile states, where governance quality and capacity are seriously impaired (e.g., Comoros, Somalia, Sudan, and Togo), Bank teams are collaborating closely with other donors on transitional support matrices. Twenty-three PRS countries are receiving assistance from the Bank in the form of poverty reduction strategy credits, many of which involve coordination of support and conditionality as reflected in cofinancing arrangements with other donors. Since 2002, when Bank policy changed to allow pooling arrangements, the Bank's lending using sectorwide approaches with pooled financing and similar arrangements has risen: Four such projects were approved in FY03, 13 in FY04, and nine in FY05. The Bank's greatly expanded local presence, particularly through its resident country directors, is seen as a positive factor in its contribution to the PRS process in-country. The Bank has been undertaking a substantial set of internal reforms to facilitate harmonization and alignment--these include requiring that CASs for IDA countries be based on the PRSP and country priorities and take into account other donor programs; policy reforms governing procurement, development policy lending, eligible expenditures for financing, audit requirements; and revised guidelines to encourage joint work with other donors in the preparation of CPARs and CFAAs. The experience of the Fund in the area of harmonization and alignment is more mixed. The IEO evaluation report and country cases studies noted signs of gradual alignment of the content of PRGF- supported programs with the PRSP--some evidence of flexibility in program design to protect and accommodate more pro-poor spending and adjust program targets to higher available aid flows; and some willingness to provide more policy space for government-led initiatives and flexibility. But content alignment is often complicated by the perceived tension between the need for realistic macroeconomic frameworks that maintain overall stability, and more ambitious frameworks for reducing poverty; and by the general absence of clear links between the PRSP and annual budgets. By contrast, there has been relatively little progress in aligning the IMF's work to the PRSP in terms of process. In some cases, the timing of missions and the coordination with other donors has improved, but despite increased outreach efforts in some areas, there has been relatively little open public debate on macroeconomic policies and options, or on institutions and transparency. 89. Realizing the potential of the PRS approach as a vehicle for donor coordination, alignment, and harmonization requires actions by partner countries 98Extending the formal performance assessment frameworks that are a feature of most of these groups to one where both the country authorities and their donor partners mutually assess progress toward jointly agreed commitments has been proposed by many commentators and in a variety of fora. See, for example, comments and contributions to the 2005 PRS review from Oxfam, CIDA, CIDSE/CI, Cordaid, Novib and Wemos, DfID, and the EC. Some of these proposals include the suggestion of an independent assessment, or an independent international structure to monitor and enforce donor compliance. Mozambique's joint donor group is instructive in this context. A Performance Assessment Framework (PAF) for Program Aid Partners was formulated in 2004 to monitor implementation of their specific commitments, as well as performance relative to broader aid effectiveness objectives. The PAF matrix contains 10 indicators covering five areas--predictability; alignment and harmonization; administrative burden; transparency; and capacity building. See Mozambique's Programme Aid Partnership website at http://www.pap.org/mz. 40 and donors alike. 99 There is scope for improvements in many aspects of donor procedures and practices--better coordination of their analytical, diagnostic, and sectoral work; more closely coordinated or joint missions; and coordinated support for capacity building, particularly for helping countries to strengthen their own monitoring and evaluation systems, on which donors can then rely. Of greatest impact, however, would be steps by individual donors to derive their conditionality to the extent possible from the country's PRS and to coordinate it with other donors to minimize overlap;100 to provide early and timely commitments of their annual support and indications of likely envelopes of medium-term assistance for inclusion in annual budgets; and to align with country PRS cycles rather than following their own internal timetables. 90. These suggestions apply equally to the Fund and the Bank in their operations in PRS countries. For the Fund in particular, its role in signaling its assessment of the macroeconomic situation of countries assumes rising importance in the context of the gradual shift toward more programmatic and budget support. This will require closer alignment of the processes with the PRS/budget cycle. From the harmonization and alignment context, too, as aid flows rise and are better coordinated and more predictably disbursed, the Fund will have to assist countries in assessing the macroeconomic implications, if any, of the higher flows, and in helping to design more ambitious macroeconomic frameworks to reflect the scaled-up efforts to achieve the MDGs (see section 3C below). 91. Conflict-affected and fragile states. In fragile states, donors face a dilemma. On the one hand, increasing aid substantially is often ineffective in improving outcomes because of lack of capacity or unwillingness to use money efficiently in reducing poverty. On the other hand, neglecting these countries perpetuates the poverty and deprivation of their people, and may lead to the collapse of the state, with possible adverse regional consequences. The challenge in these countries is, therefore, to find ways of balancing the countries' limited absorptive capacity and high risk with the need to remain engaged so that prospects for progress are not compromised. 92. Responding to this challenge calls for a pragmatic mix and sequencing of aid instruments, tailored to country circumstances and based on a sound understanding of local social and political dynamics. As noted in the draft principles for good international engagement in fragile states, "fragile states require a mix of aid instruments, including in particular for countries in promising but high-risk transitions, support to recurrent financing."101 102 Even in countries with deteriorating governance, long-term 99 The 2005 Comprehensive Development Framework Progress Report notes the importance of strong government leadership of the coordination process (see World Bank 2005a). This could include designating "lead" donors in specific areas; and discussing with donors how their conditions can be adapted to ensure their consistency with the objectives of the PRSP. It could also include integrating PRS documents with national documents, such as those associated with the budget, or using such national documents as PRS documents as well, where these are adequate to the requirements of a PRS. 100Coordinated, but not harmonized conditionality, could also reduce the risk of a total suspension of aid in the case of relatively minor slippages in program implementation. 101OECD-DAC (2005a: 3). 41 support to health, education, and other basic services is critical but must be carefully designed to avoid long-term dependence on parallel, unsustainable structures while at the same time providing sufficient resources to meet urgent basic and humanitarian needs. An engaged civil society can also play an important transitional role in providing services.103 The agreed principles of engagement in fragile states will be piloted in six countries, and the Bank will co-lead with DFID in Somalia. 93. In post-conflict countries, it is also important for international partners to take a coherent "whole of government approach" to their support, involving agencies responsible for security, political affairs, humanitarian assistance, and development aid.104 In terms of aid volumes, research suggests that there may be further room for improving the time profile of aid provision.105 Specifically, the rebuilding of an institutional environment capable of absorbing large aid flows effectively typically occurs after about three to four years following the end of a conflict, while actual aid flows generally peak in the first or second year after the conflict ends, and then taper off dramatically. Consideration of mechanisms for more targeted aid in the initial post- conflict period, while stretching the allocation of exceptional aid levels over time, offers the potential for improving development results in these countries.106 94. Efforts to align and harmonize support should remain a cornerstone of donor efforts in conflict-affected and fragile states. Where governments demonstrate the political will but lack capacity, international actors should fully align assistance behind government strategies. Timor Leste demonstrates a case of close donor coordination in the provision of budget support as the main implementation mechanism for their national development plan. Where alignment behind government-led strategies is not possible due to particularly weak governance, opportunities for partial alignment should be sought at the sectoral or regional level. In all cases, exceptionally strong aid coordination is essential. The fragility of reform programs implies that donors must avoid overwhelming the limited political and technical capacity that exists. Technical assistance and financial support should be focused on a small number of realistic reforms. D. BALANCING THE ACCOUNTABILITY FRAMEWORK 95. While there is no inherent tension between domestic and external demands for better financial management and results-oriented performance, various factors can tilt the balance toward externally-focused requirements rather than domestic ones. 107 This section first considers some of the factors that can tilt this balance. Many of 102See also World Bank (2005i) which provide guidance to Bank teams in applying development policy lending in the context of fragile states. 103OECD-DAC (2005a: 4). 104OECD-DAC (2005b: 3). 105See for example Collier and Hoeffler (2002). 106ADB (2005: 3) notes that, in these countries, the need for pooling information on support strategies and coordinating assistance is critical. 107While interests in better financial management and results-oriented performance align, tensions can arise, particularly around political issues, human rights, and deep-rooted governance problems, most of which are generally not addressed in PRSs nor in PAFs. Also, the trade arena is fraught with tensions 42 these factors stem from the absence in many low-income countries of effective domestic oversight of government programs and budgets. This gap is often filled by donor processes and conditionality in order to address short-term donor concerns about transparency and immediate results. However, in so doing, care must be taken not to tilt the accountability towards donors at the expense of domestic stakeholders. In this context, the remainder of the section discusses how the PRS architecture can support an environment where domestic and external interests are mutually reinforcing. Factors that tilt the accountability balance 96. This sub-section identifies selected reasons why in some countries the PRS process has in practice led to an unduly heavy focus on external accountability. While the areas identified are not exhaustive, the intention is to encourage all stakeholders to recognize more explicitly the issues, and to consider more systematically how various ways of doing business may have unintended consequences. 97. The external legacy. Because of its links to Bank and Fund concessional lending, many countries perceived the PRS approach as an externally imposed requirement, and the focus was on a narrow aspect of the external compact (produce an `adequate' PRSP, receive concessional assistance). Where this perception of a narrow (externally focused) compact has persisted, the approach is at its weakest. A rush to the `finish line'--as was at times encouraged by the link with the HIPC Initiative--may have also encouraged a `box-ticking' approach to processes like participation, undermining their meaningfulness as tools to enhance domestic accountability. Overemphasis on form (how the paper looks) has also at times shifted the focus away from what really counts (the underlying policies, systems, and outcomes). 98. PRSs that lack prioritization and specificity. Many PRSPs and APRs continue to lack prioritization, which can be result of a lack of analysis and capacity needed to do so effectively, unwillingness to make tough policy choices across competing demands, and/or because donors continue to encourage the inclusion of particular projects and priorities. It can also result from the lack of an adequate medium-term growth framework, let alone a long-term vision of development, around which to prioritize. Whatever the reason, the end result is that when PRSs lack operational detail, with identified priorities and sequencing, the notion of donor alignment is ephemeral. Lack of specificity has also created the need for a host of `derivative products'--such as performance assessment frameworks (PAFs)--which seek to make the PRS operational and around which donors can align.108 While in good examples, derivative products are carefully drawn from the PRS, in other cases the link is more tenuous. between low-income countries and donor countries. In addition, domestic constituents and advocacy groups in donor countries provide the authorizing environment for aid, and the positions which are advocated may not always meet the needs of the recipient government. 108GTZ (2005: 15). 43 99. Parallel processes. In many instances, PRSPs were produced with processes that ran parallel to existing planning processes.109 Such a situation is clearly undesirable, confusing and unsustainable.110 In some cases, this has contributed to weak links between the PRS and MTEFs and annual budgets. At other times, performance monitoring systems were set up in parallel to existing (albeit often weak) domestic ones. These practices have at times fragmented scarce human capacity and drawn attention away from strengthening existing processes. Disconnecting the PRS cycle from domestic political cycles can also detract from domestic accountability. 100. Volume and modalities of donor financing. When the level of development assistance is high, there are risks that it can shift the focus of attention to providers of aid. In addition, when assistance is "off-budget," line ministries (and CSOs) have incentives to align their interests to donors in a bid for financial resources, and to focus on donor reporting requirements. This can distort or short-circuit broader based efforts to enhance domestic accountability. It makes monitoring problematic and governments cannot plan for recurrent cost obligations or balance overall public expenditures across sectors since it is not clear what donors are actually disbursing. Vertical aid programs, such as global funds aimed at specific sectors or problems, can create the same distortions and are unlikely to produce sustainable country-level results unless they are aligned to and linked with country priorities, budgets, and systems. Furthermore, donor aid modalities at times entail setting up parallel systems for project implementation. 111 Unless carefully designed, these too can undermine rather than strengthen country systems. A survey of 14 countries (including 12 PRS countries) showed that, on average, only about 30 percent of the portfolio of projects were managed according to national procedures (figure 15).112 101. The weight of donor analysis. Efforts by external stakeholders to fill analytic gaps can at times outpace ownership of that analysis or capacity for it to be absorbed. Analysis that multilateral and bilateral partners undertake can significantly influence the content of PRSs, at times with the effect of absorbing policy space.113 109The EC (2005: 2) cites the risk that in some countries, PRS processes have been set up in parallel to domestic ones, becoming "donor oriented" rather than embedded in government's internal processes. 110In the first instance, existing planning processes may not have elicited stakeholder views; had an orientation towards poverty reduction; had a results orientation; nor been linked to budget allocations and policy decisions. Under these circumstances, there may have been some benefit in an initial parallel process that focused attention and momentum behind these issues. However, while in some cases this may have made sense in the first instance, perpetuating a parallel process going forward is likely to result in increasingly divergent processes. The focus going forward should be on consolidating country-specific lessons from formulating and implementing a poverty reduction strategy, and integrating those into existing domestic processes. 111Such parallel project implementation units (PIUs) often reflect underlying weaknesses in countries' project implementation and monitoring capacities. However, PIUs and off-budget funding further undermine country systems. 112OECD (2004: 18). Countries include Bangladesh, Bolivia, Cambodia, Ethiopia, Fiji, Kyrgyz Republic, Morocco, Mozambique, Nicaragua, Niger, Senegal, Tanzania, Vietnam, and Zambia. 113Playing off the popular analogy of "putting countries in the driver seat," Wilks and Lefraçois (2002: 10) note that "even if governments are allowed to get hold of the steering wheel, it is vital to see who is in charge of producing maps and deciding on the navigation." 44 Tensions have also arisen over the need for longer-term efforts to build in-country capacity to undertake crucial policy analysis and shorter-term donor requirements for analysis to inform their decision making. Figure 15: Use of country systems (percent of sample) Audit 28 Monitoring and Evaluation 28 Reporting 30 Disbursement 32 Procurement 34 0 10 20 30 40 Source: OECD (2004: 19). 102. Drive for the "wrong" results. While a focus on results is positive, and a key feature of the PRS approach, this drive can produce distortions. It can lead to `cherry-picking' of interventions where it is easier to show results (either because data are available, causality is easier to map, or because policy, institutional or capacity constraints are not as pronounced). These interventions may not actually have the highest returns. Similarly, international attention focused on specific indicators can also direct interventions to those areas--and influence the way interventions are designed--even if that particular intervention would not be a country-specific priority.114 The need to show `fast' results can also create perverse incentives in how donor-financed projects and programs are implemented. As noted above, shortcuts (such as stand-alone project implementation units) often short-circuit building sustainable systems, focusing the accountability of implementing agencies on donors rather than domestically. The understandable need for donor agencies to demonstrate results to their taxpayers can, at times, further reinforce these tendencies. 103. Use of conditionality. The perception that conditionality limits a country's policy space and constrains ownership has led some commentators to consider conditionality in its traditional forms to be fundamentally incompatible with the PRS approach, and to demand its complete elimination, or at least to call for substantial revision and streamlining of conditionality.115 However, conditionality can be useful for both donors and countries and there need not be a debilitating tension between country 114One consequence of this is that donor interventions are often concentrated in the social sectors, at the expense of support for investment in the productive sectors or in economic infrastructure. To some extent, however, this may be a legacy of the HIPC Initiative with its emphasis on using debt relief for pro-poor spending, which was often viewed as social sector spending. 115Most such calls pertain to any donor-driven conditions that are not specifically derived from the declared objectives of the PRSP. For example Oxfam (2005: 1) calls for the elimination of all "structural adjustment-type conditionality." Oxfam also recommends that remaining conditionality should be open to public debate among all domestic stakeholders, and restricted to the requirements for financial accountability measures and broadly agreed poverty reduction and gender equality goals. 45 ownership and conditionality. Ownership does not mean that donors have to accept whatever policies are proposed, nor does donor disagreement with some policies necessarily involve rejection of a government's entire program. Similarly, conditionality need not be separated from the government's intentions and policy priorities, and its role in monitoring progress toward defined objectives for the purposes of the donors can also coincide fully with the same requirements on the part of the government. If properly defined and applied, conditionality can reinforce the government's program, and donors' efforts to align their support with government priorities. Seen this way, conditionality can be a driver of mutual accountability, rather than an instrument of external accountability of government to their donors. 104. It is clear also that conditionality cannot substitute for country ownership--a key lesson on development effectiveness which has been recognized in the reviews of conditionality by the Bank116 and the Fund.117 To ensure country ownership of Bank- supported programs, the Bank is systematically aligning its Country Assistance Strategies with countries' own development strategies. Both institutions are also stressing the importance of focusing on a limited number of conditions that are critical to the success of country programs.118 The Fund's 2005 review of the implementation of its 2002 conditionality guidelines observed a reduction in the overall number of structural conditions in PRS countries and a refocusing of remaining conditions on the Fund's core areas of competence. The Bank's 2005 conditionality review shows that the aggregate conditionality covered by the Bank and the Fund has also declined. How the PRS architecture can support a balance 105. As originally envisioned, PRSs would be updated every three years with annual progress reports (APRs) in intervening years. It was expected that APRs would highlight whether development targets were attained and indicate the reasons for any deviations between actual and targeted outcomes. Modifications to the strategy to strengthen implementation in light of experience or to deal with shocks--both of which could entail changes to the macroeconomic framework and other policies--could be presented. It was also expected that the APR could serve as the basis for reporting to all external partners, thereby potentially alleviating excessive reporting requirements. Furthermore, it was recognized that the approach would need to be country-specific.119 However, as noted in past reviews, original expectations were overly ambitious with regard to how quickly countries could prepare comprehensive, prioritized, and costed strategies, with clear monitoring frameworks.120 Others have noted the potential tensions in using the APR for both domestic and external audiences.121 116For information related to the Bank's review of conditionality see: www.worldbank.org/conditionality. 117 See IMF (2005d and 2005e) and IEO (2004). 118 The framework for Bank-Fund collaboration introduced in 2001 encourages staffs of the two institutions to provide more coherent support to countries through early and systematic coordination on programs and conditionality. 119IMF/World Bank (1999b). 120 For example, the SPA 2004 survey of budget support found that less than half of the APRs examined contained a thorough review of performance in terms of indicators, or a revision of targets, while just over half reviewed key policy measures planned for the next year. However, several of the surveyed countries 46 106. For the PRS architecture to be supportive of a balance in accountabilities, it needs to encourage better customization at the country level. A key underlying imperative is that, where not already the case, PRS processes be (re)integrated with domestic ones. To facilitate this, the March 2002 PRS Review recognized that a three- year PRS cycle may be incompatible with existing planning or political cycles. It recommended that countries determine the appropriate periodicity for updating their strategies, generally within a three- to five-year timeframe. In the same vein, as was emphasized in last year's review of PRS implementation, when they do not already do so, low-income countries may wish to consider linking their APRs more closely to the budget planning and policy review cycles and domestic monitoring and evaluation process such as reporting to parliaments. It would be helpful to time sector reviews to feed into that process. Donors can further support these efforts by scheduling roundtables or consultative group meetings in a manner that matches country cycles. 107. Low-income countries also need to continue to focus on strengthening the prioritization, sequencing, monitoring framework, and costing of their PRSs, which will also facilitate donor alignment. However, stakeholders need to appreciate the magnitude of this endeavor, and recognize that progress in this direction will be gradual. Efforts should support movement toward this goal rather than bypassing the complex agenda at hand in search for `easier' options that undermine efforts to create sustainable development impact. To facilitate alignment and coordination, over time PRS matrices-- or the operational strategies closely linked to them--need to become more action- oriented and prioritized over time. To be useful for countries (and for the notion of alignment to provide a more binding constraint to donor activities), APRs will need to link PRS priorities to the annual budget more realistically, and signal any major adjustments to targets or priorities over the coming year. 108. In addition, using the process to prioritize analytic and capacity building needs would also help to better realize the potential benefits of the PRS approach. This process would help support greater donor coordination and alignment in of analytic support and capacity building efforts. With regard to externally supported analysis, efforts are needed to ensure that the results are disseminated in an understandable manner. 109. Donors will also need to turn high-level commitments and good practice into normal practice. For example, when PRSs are not specific, donors need to be particularly careful to ensure that the policy measures and programs that they support do reflect national priorities. Donors should also use country processes and institutions whenever possible. In cases where additional safeguards are needed, these should be put in place in a way that helps to strengthen domestic systems rather than undermine them. They should also demonstrate flexibility with regard to reporting requirements, allowing that had not reviewed policy measures taken during the year in 2003, did so in 2004; and many have included updates on planned policy measures, with revisions of targets in some cases. See SPA (2005a). 121See, for example, Booth, Christiansen, and Driscoll (2005). 47 more space for the governments to develop country-specific APRs that may serve multiple needs rather than requiring a separate, dedicated report.122 110. In sum, there can be balance only if countries find that the approach has inherent benefits for them. So, while the PRS process unfolds at the country level, multilateral and bilateral agencies need to be aware of how their behaviors can either provide space for countries to realize the benefits from the PRS approach or reinforce an externally driven process. As this section has noted, the latter outcome is not desirable if the PRS approach is to be sustained in the medium term. Given the role of the Bank and the Fund in introducing the PRS approach, they have a particular responsibility to be aware of how their actions influence the process. 3. SCALING UP RESULTS [A]s long as poverty, injustice and gross inequity persist in our world, none of us can truly rest. ... Massive poverty and obscene inequity are...terrible scourges of our times--times in which the world boasts breathtaking advances in science, technology, industry and wealth accumulation.... While poverty persists, there is no true freedom. ... [D]o not look the other way; do not hesitate. Recognize that the world is hungry for action, not words. Act with courage and with vision. ... Of course the task will not be easy. But not to do this would be a crime against humanity, against which I ask all humanity to now rise up. Nelson Mandela123 A. OVERVIEW 111. Commitments to rapidly scale up aid call for attention to ensuring that it is used effectively.124 In April 2005, the OECD-DAC estimated that on the basis of commitments made by DAC members, annual overseas development assistance (ODA) could increase by US$20 billion in real terms between 2003-04 and 2010. Since then, the European Union (EU) has announced sharply higher targets to increase ODA to 0.56 percent of Gross National Income by 2010, reaching 0.7 percent by 2015. Japan has announced a US$10 billion increase in its aggregate ODA over the next five years. Based 122A good example of this is the case of Mozambique, where the government's report to Parliament on implementing the national PRS (the PARPA) now serves as the annual progress report. The PRS process gave impetus to efforts to improve the analytic context of the national reporting instrument to parliament to the point where it fulfills the requirements of an APR. 123Speech in Trafalgar Square for the campaign to end poverty in the developing world, February 3, 2005. 124The Marrakech Roundtable (February 2004) and commitments made on managing for results in the Paris High Level Forum on Harmonization, Alignment, and Results (March 2005) call for ensuring effective use of aid. The Marrakech Action Plan endorsed actions in three areas: (i) strengthening the demand for and capacity to develop and use information on results at the country level; (ii) identifying and promoting interventions in development institutions' internal procedures and priorities that encourage a results orientation across all elements of their work; and (iii) participating in global level partnerships which aim at promoting harmonization, alignment, and managing for results at donor agency and country levels. 48 on new announcements, total ODA is now projected to increase by around US$50 billion per year in real terms by 2010. Much of this increase will go to low-income countries, through the framework of their poverty reduction strategies. The G8 has also made specific commitments to double aid to sub-Saharan Africa to US$50 billion by 2010.125 112. As noted in the last chapter, there has been progress in strengthening many of the foundations for scaled-up development results. Growth has resumed in many low-income countries. Countries are setting targets to track performance based on development goals. Availability and access to data are improving, and attention is focused on the need for coordinated monitoring systems. Medium-term expenditure frameworks are being established which can help link PRS objectives to the budget. Public expenditure management systems are improving. And partnerships are broadening and deepening. While all of this is encouraging, the agenda ahead remains large. Political will, civic engagement, and donor behavior will all influence how far and how fast progress is made. 113. This chapter considers several factors that can ground efforts to scale up results. It begins by selectively reviewing analytic foundations that can help improve the overall quality of poverty reduction strategies by providing information that can be used to assist in prioritizing policy actions and programs and linking them to specific goals and targets. It then considers how the PRS approach can be used as a framework for scaling up results and addressing absorptive constraints--whether they be macroeconomic, human or physical capital, or institutional in nature. These factors, when coupled with systems and processes that support evidence-based decision making and strong domestic accountability--the topics discussed in the previous chapter--can foster more effective use of domestic resources and aid flows and assist in efforts to scale up results. B. ANALYTICAL FOUNDATIONS FOR SCALING UP RESULTS 114. The PRS process has drawn attention to the strong linkages between policies and programs and poverty-related outcomes, including growth. It has underscored the need for more country-specific analysis on these issues.126 This section highlights selected issues related to: (i) poverty analysis; and (ii) growth diagnostics and the productive sectors. This brief description is not intended to underplay the need for continued strengthening of PRS analysis and content across a wide range of sectoral and cross-sectoral topics. Rather, any detailed assessment of PRS analysis and content is most useful at the country level, particularly since the definition of analytic priorities needs to be country-specific. 115. A well-articulated analytic framework can help to think through country- specific issues and to arrive at sequenced and prioritized public interventions. This calls for analysis that is operationally relevant and integrated into the PRS process, including better understanding of the linkages between public policies and expenditures 125World Bank (2005b). 126OED (2004: 48). 49 with growth, poverty reduction, and accelerated progress towards the MDGs. Development partners, including the Bank and the Fund, should continue to help countries address analytic gaps, but in a manner that recognizes that priorities will vary by country, and that gaps need to be addressed in a sequenced manner. Furthermore, attention should be paid to building capacity of local and regional research groups. Poverty analysis 116. Poverty diagnostics. Past implementation progress reports have noted that the PRS process has been an impetus for improved poverty information and knowledge base. Technical and capacity challenges remain in linking diagnostic results about who is poor and why they are poor with the determination of priority public actions.127 However, even where there are sound diagnostics, this analysis is not always clearly linked to a country's PRS.128 This may be due in part to lack of adequate coordination among those who undertake the analysis and those who formulate the strategy or difficulties in translating analysis into policy choices. However, political economy considerations and the fact that creating a culture of evidence-based decision takes time are also likely at play. 117. Most PRS countries have enhanced their poverty diagnostics, often with support from development partners. For example, in West and Central Africa, the Bank has supported poverty assessments in 20 out of 26 countries. In many of these cases, these reports were prepared jointly with national statistical offices or PRS units. Most reports include a poverty profile and an estimation of poverty trends, an analysis of the livelihood of households, and an analysis of the education and health dimensions of poverty. Some reports also contain innovative work on a range of topics, such as rural productivity, cash crops, time use, gender analysis, as well as the perceptions of poverty and priorities of poor people. As another example, in Europe and Central Asia, the Bank has adopted a programmatic approach to poverty assessments, in order to enhance the policy relevance of the Bank's poverty analysis and monitoring.129 Box 17 provides two examples of innovative mechanisms for building capacity for poverty analysis. 127IMF/World Bank (2002b: 29-33). 128GTZ (2005: 16). 129 This approach, being applied in Armenia, Azerbaijan, Bosnia and Herzegovina, Georgia, Kyrgyz Republic, Serbia and Montenegro, Tajikistan, and Uzbekistan, involves tailoring the poverty work to country-specific needs, developing a three-year program of poverty work, integrating capacity building and use of local consultants especially through mobilizing donor assistance (primarily DfID), providing a series of analytical inputs timed with the PRSP and budget cycle, and having a continuous dissemination process. This approach recognizes that effective poverty monitoring and analysis needs to be anchored in local institutions, and that this is best facilitated through an approach that builds local capacity and feeds into the policymaking process in the country. 50 Box 17: Innovative support for poverty analysis In Sri Lanka, the PRS process suffered from an initial lack of poverty- related data and analysis. In an effort to increase the availability of poverty-related research and analysis from sources that could be accepted across the domestic political spectrum, as well as by donors and civil society, the Center for Poverty Analysis (CEPA) was established as an independent, commercially-oriented provider of poverty-related research and training. With support by GTZ, CEPA, which was converted from the former Poverty Impact and Monitoring Unit linked to the Ministry of Finance, is a non-profit which works on commercial terms by offering fee-based research and training. CEPA has gained broad-based acceptance for the quality of its services, which are demanded not only by the IFIs and bilateral donors, but also from national government departments and NGOs. CEPA is currently carrying out a number of PSIAs on critical issues within the PRSP. This is a case where donor behavior has supported building national ownership and mechanisms that can enhance domestic accountability. By supporting CEPA through the commissioning of analytic and training services , the donor community has helped CEPA gain standing in the national process as a reliable domestic knowledge base on poverty issues. See GTZ (2005: 124-126). The Belgian Poverty Reduction Partnership Trust Fund, managed by the World Bank, supports PRS analysis in Sub-Saharan Africa, with a focus on Burkina Faso, Burundi, the Democratic Republic of Congo, Mali, Niger, Mozambique, and Rwanda, as well as sub-regional initiatives. To support the scaling up of poverty diagnostic work, PSIAs of major reforms and policies, and the links between PRSs and the budget and MTEF, it finances poverty economists located in World Bank field offices. An important priority is to establish a closer collaboration on the ground between the World Bank and PRS units to facilitate capacity-building efforts. The presence in the field of the poverty economists working closely with PRS units has been received favorably by country authorities. 118. However, continued attention is warranted to ensure that adequate data are available to underpin poverty analysis. While there have been some improvements in the availability of current household surveys which can be used to measure poverty, more progress is needed. In 1999, only 19 percent of low-income countries had a household survey that was less than three years old. By 2004, 29 percent did. This is still far below the 72 percent of middle-income countries that had current surveys. Also disturbing is the very large portion of low-income countries (42 percent) that continue to have either no survey or only very dated ones, a figure practically unchanged over the past five years (figure 16). Figure 16: Availability of household surveys to measure poverty 100% 19% 29% 70% 72% 80% 60% 37% survey less than 3 years old 29% 40% survey 3 to 5 years old 44% 42% 20% 15% 13% older or no survey 15% 15% 0% 1999 2004 1999 2004 LIC MIC Source: World Bank, Development Data Group http://www.worldbank.org/data/countrydata/csid.html 51 119. Distributional analysis. The limited use of distributional analysis (often termed PSIA) to inform PRS design was noted in the 2002 PRS Review.130 There was a recognition that such analysis can help inform and foster national policy debate on reform options and design, and support evidence-based, transparent policy making. Since then, the use of PSIA has become more common, although data limitations, methodological issues, and national capacity constraints necessitate gradual progress.131 120. Various civil society organizations and bilateral donors have noted areas where they feel PSIA work could be improved.132 For example, the need for better integration of PSIA into the PRS process, greater stakeholder involvement in the analysis, disclosure of work carried out by the Bank or the Fund, and more systematic use of PSIA in informing Bank- and Fund-supported operations, are frequently raised. Taking stock of the experience to date, there have been improvements in these areas, although more can still be done. However, misconceptions continue to persist as to the nature of PSIA, and unrealistic expectations are still placed on its applicability and effectiveness. 121. PSIA is most effective when applied to specific and well-defined reforms rather than on broad strategies where distributional outcomes cannot be meaningfully assessed. As most PRSPs still tend to lack specificity in terms of policy reforms and public actions, it is often more realistic for PRSPs to outline the priority areas for this work and to develop a PSIA agenda as part of the PRS implementation and monitoring plan (box 18). Selectivity remains essential. Since PSIA requires significant time and resources, it is important for low-income countries to focus analyses on key reforms which are likely to have significant distributional impact, and to prioritize the need for PSIA with other analytic gaps. 122. Most often, PSIA is conducted ex ante, but even when reforms are already under implementation, PSIA can still inform the pace/sequencing and identify or strengthen mitigation measures. Some ex-post PSIAs have assessed the actual distributional impacts of a completed reform in order to understand the likely impacts of future actions. For example, in Ethiopia, a PSIA is planned to analyze whether land reform undertaken in 2002 had any impact on access to land, investments in land, or perceived security of land tenure for poor people. The objective is to help inform the discussion of next steps for liberalizing the land market. 123. PSIAs are being used to inform the design and sequencing of policies and to develop policy options. There are also good examples where PSIAs has helped to enhance the pro-poor focus of policies. For example, the Mongolia Utility PSIA, results of which were presented on television, directly influenced the design of tariff reforms. In Zambia, a proposal by the Ministry of Land to convert the country's 94 percent of land under customary tenure into state-owned land was analyzed, leading to a decision to cancel the proposed reform. In Rwanda, the team working on the PSIA in the tea sector 130IMF/World Bank (2002b: 32). 131IMF/World Bank (2003a:12-13), IMF/World Bank (2004: 25-28). 132See for example CIDSIE-Caritas Internationalis (2005b); EURODAD (2005b); Oxfam (2005); GTZ (2005). 52 advised the Ministry of Agriculture to introduce a differentiated tea pricing policy, which would increase productivity and returns to laborers. Once the proposal was adopted, they also advised on its design. Some additional examples are provided in box 19. Box 18: Linking PSIA with PRSPs: Examples from selected countries Cambodia. The Cambodian PRSP recommended that "social land concessions" be used to distribute vacant state lands to poor households. A comprehensive land reform package was proposed which was intended to alleviate rural poverty and increase economic productivity. Rural households would be able to use idle land productively and as collateral for capital investment. The PSIA assessed the conditions and options for implementing such a land reform program. Results revealed that the main reason for losing land was health-related expenses. A sustainable land reform program would need to address issues of indebtedness, as well as affordability and accessibility of social and physical infrastructure. The analysis also revealed that finding suitable land for distribution, and releasing it, would be more difficult than expected. Strong political commitment from concerned groups is a prerequisite for the proposed reform. Tanzania. Tanzania's PRSP highlighted rural development and agricultural export growth as key to reducing income poverty. The first annual progress report identified the review of the roles and funding arrangements of the Crop Boards as a priority. The PSIA supported this review and contributed to the ongoing debate within Tanzania on the role of the Crop Boards in improving the growth performance of agricultural exports. The study identified crop-specific reform options on appropriate regulatory and service functions for the cotton, coffee, cashew and tea boards, and facilitated policy dialogue on reform alternatives among various stakeholders. Lesotho. Poor performance of state-owned utility companies--particularly electricity and telecommunications--was identified in Lesotho's I-PRSP as a key factor slowing growth and regional integration. The PSIA looked at the potential impact of tariff increases on various economic and social groups. The analysis helped inform decision making related to utility privatization, tariff re-balancing, and capacity development. The analysis also provided information about potential demand for connections; how to mitigate potential inequities; and ways to improve billing and collection services. In addition, the PSIA helped the newly created regulatory agency take into account consumer views. Box 19: Enhancing the pro-poor focus of policies/identifying reform options Pro-poor focus Burkina Faso's macroeconomic PSIA revealed the role that inequality was playing in curtailing poverty reduction efforts and enabled policymakers to take that into account. In particular, the PSIA emphasized the need to expand social services to reduce poverty and inequality. It recommended a school construction program, as well as an increase in the capacity of existing schools through provision of multi-grade and double-shift schools. The Mongolia cashmere PSIA led to the inclusion of pro-poor policy alternatives--including the suspension of moves to impose an export ban on raw cashmere and focusing on improving infrastructure maintenance for water wells. Following the PSIA work, the government and private sector set up a committee to work on improving the regulatory environment. Identification of reform options In Ghana, PSIA of the proposed reform of the petroleum pricing formula identified various approaches to mitigating the adverse impact on the real incomes of poor people. In line with these findings, the February 2005 increase in petroleum prices resulting from the price formula was accompanied by a number of budgetary measures to mitigate the impact on the poorest households, (e.g., removing primary and junior secondary school fees, investing in public transport, and extending electricity access). In Azerbaijan, PSIA of electricity tariff increases, to finance badly needed network maintenance and to balance supply and demand, found that the short-term impact on residential consumers was quite different in the capital city compared to elsewhere in the country. Consequently, different sets of complementary reform options were identified to mitigate the impact on poor people. 53 124. For its part, the Bank has encouraged staff to increase distributional analysis by providing dedicated funding for such analyses and requiring that poverty and social impacts be considered in Bank operations. The dedicated funding was intended to encourage methodological innovations and to incubate the use of PSIA while the approach was being mainstreamed in work programs.133 With the adoption of the Bank's Operational Policy 8.60 in September 2004, staff are required to determine "whether specific country policies supported by the operation are likely to have significant poverty and social consequences, especially on poor people and vulnerable groups." Guidance on PSIA is provided in Good Practice Notes for Development Policy Lending (October 2004). Additionally, the Bank's internal quality assurance review process developed benchmarks and standards for development policy lending that reflect the importance of analyzing and documenting distributional impacts. Less than three years after the first pilot studies analyzing distributional impacts of policy reforms, the approach has been firmly established in the World Bank's work and within several other development agencies. 125. The Fund has stepped up its efforts to integrate the results of PSIA in the design of its programs. The Fund has established a group within the Fiscal Affairs Department to pursue this objective and, in the process, improve the quality of program design by better understanding the likely impact of key reform measures that are integral to Fund-supported programs on different population groups, in particular poor people; assessing alternative measures; and where appropriate, integrating compensatory and complementary measures to mitigate any negative effects of reform policies into Fund- supported programs. The group pursues its objectives in several ways. First and foremost, it works with area department mission teams to specifically address program-related issues for which distributional effects are potentially important. Toward this end, the group identifies existing studies that could shed light on the potential distributional implications and, where necessary, performs supplemental PSIA itself.134 The group is also developing guidance notes on the types of issues that would benefit from PSIA and the methods with which PSIA could be performed with the goal of expanding the amount of PSIA performed by mission teams. Finally, the group liaises with development partners, other institutions, and scholars working on PSIA to keep them informed of the group's activities and help set priorities for future work. 126. While there has been substantial progress on the analytical side of PSIA, there is scope for more timely dissemination of PSIA information and results. However, there are instances where at the Government's request public disclosure is 133In the last quarter of FY03, US$400,000 in Bank budget funds were made available. An additional US$2.4 million was disbursed in FY04, followed by a total of US$1.8 million in FY05. This funding window has recently been extended through FY06, with an additional one million dollars available to task teams for conducting PSIAs. 134In its first eight months of operation, the PSIA group fielded seven missions, the majority of which focused on the reform of energy subsidies, particularly in the wake of the rise in world oil prices. The analyses found subsidies to be both costly and poorly targeted, but that their elimination would raise the prices of many goods consumed by poor people. The PSIAs have thus proposed measures to offset the increased costs for poorer households, and these have been included by the country authorities as part of their reform packages. 54 restricted. Sometimes, due diligence considerations call for adequate quality control before results are disclosed. Furthermore, as PSIA is progressively mainstreamed into Bank operations, dissemination of PSIA work will be increasing subsumed within existing practices for development policy lending and economic and sector work. 127. While it is not feasible for all stakeholders to be involved in all analytical work, PSIA needs to identify the characteristics, interests and impacts of key stakeholders. Experience, however, shows that consultations with stakeholders during the policy dialogue and disseminating PSIA results to stakeholders helps to create better understanding of reform imperatives. This can contribute to broadening the support for reforms. 128. The effort to encourage countries to use PSIA to inform PRS formulation and implementation while also using, when necessary, distributional analysis to underpin Bank- and Fund-supported operations, poses some challenges. Encouraging PSIA to be fully embedded in country-led processes often requires sustained capacity building and attention to the demand side. Undertaking analysis to inform the design of lending operations has a shorter time horizon and narrower focus. At a recent PSIA Network Meeting,135 members of bilateral and multilateral donor agencies and representatives of CSOs recognized the challenge of working on these two tracks, as well as the fact that some governments may not be interested in PSIA. It is also important to recognize that, as with other analytic work and decision-making and participatory processes, there are risks that capture by specific interest groups can undermine results. Growth analysis 129. In line with increased attention on productive sectors, countries are increasingly making growth a central element of their poverty reduction strategies. While only about 60 percent of the earliest PRSPs (14 of 25) noted a prominent role for growth, this had increased to over 80 percent (13 of 16) for the period July 2003 to June 2004. Over the last year, all six new PRSPs have incorporated growth as a central pillar in their strategies. This is consistent with evidence that underscores the importance of promoting strong and sustained growth as part of any pro-poor growth strategy.136 130. However, there has been considerable variation in the quality of PRS growth strategies, and improved growth analytics are needed at the country level to delve into the host of factors that influence the environment for growth and the mechanisms by which poor people are connected to it.137 138As noted in past progress reports, much of the discussion related to growth pillars has remained qualitative, with less specificity on policy and institutional reforms. For example, while countries with 135The Network is open to all parties interested in PSIA activities. See http://www.dgroups.org/groups/PSIAtools/index.cfm?op=act_login. 136See, for example, World Bank (2005c). JBIC (2005: 1) and SIDA (2005: 7) also emphasizes the importance of growth to poverty reduction. 137Diop, Gust and Khandelwal (2005, forthcoming) in a review focused on trade issues in PRSPs. 138World Bank (2005c) and World Bank (2005f) discuss connecting poor people to growth. 55 open trade regimes grow, on average, significantly faster than more closed economies, offering better opportunities for poverty reduction, trade is only now being mainstreamed in PRSs.139 Together with higher official development assistance, an improved trade performance has the potential to generate an important share of the resources low-income countries need to reach the MDGs.140 131. A key step going forward will be to understand better the country-specific binding constraints to economic activity so that this information can be used to help derive policy priorities (box 20). Understanding how poor people are connected to economic growth is also essential in this regard (box 21). Box 20: Binding constraints to economic growth There is broad consensus on the necessary conditions required for countries to accelerate and sustain growth--for example maintaining macroeconomic stability, ensuring that returns on private investment are predictable and appropriable, and promoting market-oriented incentives. As Rodrik (2004) and others have argued, however, necessary conditions do not translate readily into generalized policy prescriptions that apply across countries and successful growth strategies are more than meeting necessary conditions. This conclusion is confirmed by a recent review of growth experiences in the 1990s, which shows that the same policies can produce vastly different results in varying country circumstances, and that successful countries have followed varied policy and institutional approaches, World Bank (2005g). The same work also highlighted the importance for growth strategies to be country specific rather than generic, eschew "best practices," concentrate on identifying and addressing binding constraints to growth, rather than on any or all constraints, and for growth strategies to be based on sound growth diagnostics. Growth diagnostics are more an art than science, however, and refining this art will take practice and experience. A useful framework for the conducting diagnostics has been recently formulated by three academics, see Hausmann, Rodrik, and Velasco (2005), and is being piloted in twelve countries by Bank staff. The framework essentially consists of a decision tree to guide an empirical diagnostic of what is the most binding constraint to growth in a given country, and using that diagnostic to derive the policy options to address this constraint. Although piloting this framework is at an early stage, initial results are promising. 139Furthermore, "openness has important positive spillovers on other aspects of reform, so the correlation of trade with other pro-reform policies speaks to the advantage of making openness a primary part of the reform package." See Berg and Krueger (2003: 39). 140The Bank and the Fund have consistently stressed the responsibility of developed countries to meet their commitments to increase aid, to improve access to their markets for developing country exports, and to dismantle trade-distorting subsidies. Both rich and poor countries carry responsibilities in promoting the fuller integration of developing countries into the global trading systems. See, for example, de Rato (2004) and World Bank (2003a). 56 Box 21: Connecting poor people to economic growth A recent study entitled "Pro-Poor Growth in the 1990s: Lessons and Insights from 14 Countries," confirms that policymakers who seek to reduce poverty should implement policies that enable their countries to achieve a higher rate of growth. But growth is more effective in reducing poverty in some countries than in others, depending on the capacity of poor people to participate in and benefit from growth. The study highlights several broad policy options to help poor households take advantage of nonagricultural and urban employment opportunities. They include improving the investment climate; expanding access to secondary education; enhancing access of girls to all levels of education; designing labor market regulations to create more formal employment for poor workers; and improving access to infrastructure, particularly roads and electricity, to better link rural areas to small towns and urban centers. Similarly, the study identifies several policy interventions that were important in raising agricultural incomes of poor households in the 1990s. These include lowering transactions costs to access markets through investments in roads and encouraging contract farming and producers' organizations, as well as strengthening property rights to improve land access and investment incentives for smaller farmers. The study also recognizes the importance of expanding technology available to smallholder producers in arid climates and helping poorer and smaller producers deal with price and climate risk. Finally, it stresses the need to create an incentive framework that benefits all farmers and takes into account the differential impact of price and trade policy reforms on poor households. As with growth strategies, the binding constraints that need to be addressed to enhance the ability of poor people to participate in growth will vary depending on country conditions. For example, population density and the degree of urbanization affect transactions costs and the ability of poor households to participate in agricultural and nonagricultural growth. The extent of initial asset and income inequality not only influences the sensitivity of poverty to growth, but may also point to gender or ethnic discrimination or other inequality traps that keep certain groups of poor households from participating in growth. Source: World Bank (2005c) 132. There is also evidence, though, that while many PRSs have intended to incorporate productive sectors, most have found it challenging to do so effectively. Treatment has tended to be less detailed than for the social sectors. One challenge is that a different approach is often called for in addressing issues in the productive sector, where public actions are more likely to relate to the enabling environment rather than to the allocation and effective use of public expenditures.141 Lack of good analysis is also at times a constraint. For example, most PRSs have included rural priority actions, but often the analytic support to ground those priorities has been absent.142 133. While the PRS approach provides an opportunity to build coherence across productive sectors, it is not a substitute for clear strategies for each of the productive sectors. The development of such strategies remains an area for continued attention. For its part, over the past five years, while a significant portion of Bank's economic and sector reports continued to focus on social sectors, considerable attention has also been paid to rural issues and infrastructure (figure 17). Additional support has been provided through policy notes and technical assistance. 141Shepard, et al. (2005). 142As part of a review of analytical gaps on rural development in PRSs and of potential improvements in the participatory processes to discuss these issues in countries preparing or revising PRSs, IFAD recently launched a review of the experience of a number of Sub-Saharan Africa countries. Case studies are being prepared in partnership with others, including a team from the Africa region at the World Bank. 57 Figure 17: World Bank Economic and Sector Reports (FY 2000-2005) IDA-only countries, number of reports 50 40 39 42 45 30 35 31 20 24 10 0 Social Education Health Governance Law Environment Infrastructure Justice Agriculture Rural Urban Source: World Bank staff calculations. 134. Various factors can impede the effective integration of a cross-cutting issue, such as growth, into poverty reduction strategies. These factors range from lack of analysis, to difficulties in deriving operational implications from available analysis, to lack of engagement of important stakeholders in the PRS process. In the case of trade, which is an important underpinning for growth, a recent review highlights some of these issues.143 Selected findings from that trade review are also likely to be germane to other topics, such as labor markets and rural development. 135. Sustained efforts are needed to make trade strategies more prioritized and operational. While a majority of countries discussed at least one trade issue in their PRSs, a substantial number of the priorities mentioned did not contain operational measures. The exception was in the case of issues related to WTO accession, where specific reforms related to tariff and non-tariff barriers, as well as complementary fiscal reforms, were generally mentioned. 136. In many instances, prioritization and operational detail is hindered by the lack of background analysis or analysis that is compiled in a form that is not readily absorbed.144 While analysis is often available, too often PRSs do not consider trade- related opportunities and vulnerabilities. They are also weak in analyzing the macro- linkages of trade reforms. And at times, trade policies set forth in a PRS lack internal consistency. Positively, though, there are cases where sound diagnostics have influenced PRSs (box 22). Instruments like the Diagnostic Trade Integration Studies (DTIS) as well as distributional analysis can support a more objective debate and help identify suitable complementary or mitigating measures to trade reforms. 143Diop, Gust, and Khandelwal (2005). 144More generally, inadequacies in data availability, particularly macroeconomic data, often constrain analytical work, e.g., on PSIA or the sources of growth. 58 Box 22: Incorporating DTIS recommendations--Cambodia Cambodia's DTIS identified deficiencies in customs procedures as one of the factors affecting trade competitiveness. The PRSP incorporated this information and identified measures designed to reduce the timing and costs of customs clearing. Notably, the authorities committed themselves to (i) simplifying the tariff regime, (ii) harmonizing customs and trade facilitation procedures with neighboring countries; and (iii) rationalizing administrative responsibilities for border inspections. The incentives and accountability of customs officials are to be addressed in the context of the wage and employment reforms, and enforced with anti-corruption legislation and the codes of ethics. Source: Diop, Gust, and Khandelwal (2005). 137. Engaging key ministries and agencies in the PRS process is also important. In terms of linking trade issues to the PRS, this means enhancing the link between the Ministry of Trade and the agency charged with PRS design. Frequently trade diagnostics, and formal mechanisms of consultation among policymakers, business sector, civil society representatives, and development partners, have taken place under the purview of the Ministry of Trade. While these interactions have enabled a more fluid dialogue on trade and growth among those stakeholders, this frequently is not well connected to the PRS process. Strengthening this link, by more fully involving the Ministry of Finance and agencies charged with PRS design in the trade dialogue, would help foster greater integration of the trade agenda within the PRS. 138. Ex-post analysis of a particular policy area seems to enhance the likelihood of its coverage, and the quality of that coverage, in a poverty reduction strategy. The trade review found that discussing past trade reforms helps draw lessons, prioritize options for future reforms, and identify necessary complementary measures. 139. In sum, growth diagnostics will need to focus on identifying the binding constraints at the country level. But, in addition, attention is needed to a range of factors that help make the analysis operational and facilitate linking it into the PRS process. C. SUPPORTING AMBITIOUS DEVELOPMENT PLANS 140. This section first discusses how the PRS approach can be used as a framework for scaling up aid, noting the tension that needs to be resolved between a PRS that provides a realistic framework for day-to-day policy implementation and one that maps out efforts and financing required to achieve more ambitious longer-term goals. It then looks at addressing absorptive constraints, including managing the macroeconomic impacts of scaling up aid and addressing absorptive capacity constraints. PRSs as a framework for scaling up 141. Development assistance has increased since 2001, and the major donors have all made substantial new pledges, especially for the period after 2010. As noted at the start of this chapter, there are more external resources available for development than 59 ever before, with prospects for significant further increases over the medium term. Nonetheless, many low-income countries remain pessimistic about both increased aid volumes and improved aid modalities.145 Others, such as Uganda, have become increasingly cautious about taking on additional donor financing in terms of its implications for fiscal deficits, aid dependency, and the sustainability of poverty reduction efforts. To overcome this aid pessimism and to maximize effectiveness, it will be essential that pledged increases materialize in ways that are flexible and predictable. 142. Well-articulated PRSs should provide the framework for scaling up country- level efforts to reach the MDGs. A country's PRS can help elaborate its plans for utilizing higher aid levels while providing donors with the information necessary to enable them to make firm commitments to increase aid and improve its delivery. However, one of the major criticisms of PRSs is that they are generally not formulated as an operational framework for reaching the MDGs--the underlying strategies are seen as realistic, but not ambitious enough.146 143. There is a tension between the need for realism in PRSs, if they are to provide an operational framework for poverty reduction, and the need to aspire to more ambitious results related to longer-term development goals, such as the MDGs. 147 Surrendering the link to a realistic assessment of what a country can achieve today, based on its own capacities and available domestic and foreign resources, would undermine the credibility of the PRS, both as a framework for coordinating donor support, and as an instrument to enhance accountability. On the other hand, the PRS approach also needs to accommodate ambitious goals and targets consistent with a long- term vision of development and poverty reduction. 144. One way to marry the need for a realistic framework, as the basis for day-to- day policy implementation, with an ambitious framework that maps out how a country will ramp up its efforts to attain more ambitious development goals, would be to use alternative scenarios.148 In addition to the assumptions on higher external assistance, the alternative scenario would map out efforts to mobilize domestic resource 145As a result, several countries presently use "discounted" values of donor indications of aid as the basis for their budget formulation. However, success in establishing joint budget support donor groups in some countries has improved the situation--for example, in Tanzania, some 70 percent of budget support is now disbursed in the first quarter of the fiscal year, considerably improving the government's ability to plan the use of the resources effectively. 146Under this view, "[v]ery few PRSPs are ambitious or comprehensive enough to achieve the goals, largely because they have been prepared in a context of insufficient donor assistance." See UN Millennium Project (2005:25). However, a review of PRSs found that countries were setting targets in most areas which implied a significant improvement over their performance in the preceding decade, and which were often times at least as ambitious as the MDGs. Harrison, Klugman, and Swanson (2004). 147Past progress reports on PRS implementation and the IEO/OED evaluations of the PRS approach have raised this point. 148The use of alternative scenarios has been suggested in past progress in implementation reports. It was discussed by the IMF's Executive Board in 2003 in the context of a paper on aligning the PRGF and the PRSP approach in 2003. See IMF (2003b). 60 mobilization149 and the improvement in results that can be expected from additional public spending programs. This approach would also identify the consequences, at the country level, of failing to mobilize the additional resources.150 145. Building alternative scenarios also requires a country-specific understanding of constraints to absorptive capacity, so that alternative scenarios can take into account sequenced interventions to alleviate them. Depending on the country and the sector, certain constraints are likely to be more binding than others, both in the short term and in the long term. Box 23 provides an indicative classification of some of these constraints. Clearly the specific nature of the constraint, and the timeframe for addressing it, will be country-specific. Managing potential macroeconomic constraints, as well as capacity building to address human capital and institutional constraints, and infrastructure investments to address physical ones, are discussed in the following two sub-sections.151 Box 23: Indicative classification of absorptive constraints Macroeconomic Debt sustainability Dutch disease effects Physical and human Technical and managerial skills of public officials (doctors, teachers, accountants) Lack of adequate infrastructure and equipment Institutional and policy Perverse incentives in due to compensation and recruitments systems Inadequate public expenditure management systems Social/cultural factors determining demand for services Donor behavior Aid volatility Uncoordinated donor behavior Difficulties in full donor shift to improved behavior Source: adapted from ODI (2005b: 3). Dealing with macroeconomic impacts of scaling up aid 146. Although the cases where prospective increases in aid flows could seriously threaten macroeconomic stability are expected to be few, there are a number of potential macroeconomic impacts that need to be assessed and managed at the country level. Aid allows more domestic resources--workers, natural resources, and physical capital--to be devoted to satisfying investment and consumption needs in low- income countries, because fewer domestic resources are needed for producing exports or 149While external aid is important, the bulk of investment needed to scale up to MDGs will need to come from domestic sources, which are more reliable than foreign aid and debt relief. See Vandemoortele and Roy (2004). 150To date, few countries have made use of alternative scenarios, and none of these mapped out clearly a path to achieving the MDGs. Moreover, even in countries which included a higher-case scenario as part of the PRS, it did not seem to play a significant catalytic role in mobilizing additional donor support, nor did it address the binding capacity constraints. 151Constraints generated by donor behavior (e.g., fragmented interventions that impose heavy transaction costs and detract from, rather than build, domestic capacity; and lack of predictability of aid flows which can hamper medium- and long-term planning) were discussed in sections 2c and 2d. 61 import substitutes. Insofar as aid-financed projects carry a reasonable rate of return, this reallocation of resources toward domestic uses will help increase growth. A real exchange rate appreciation may be part of the adjustment process. But if the traded goods sector is an important source of productivity improvement and positive spillovers to the rest of the economy, growth may be lowered by diverting resources away from the sector. (This is referred to as Dutch disease). 147. Managing the macroeconomic impacts of aid inflows can involve difficult, country-specific trade-offs.152 Policymakers will have to balance the benefits of higher aid-related spending against the possibility of real exchange rate appreciation. In many cases, the government spends the aid and the central bank resists the real exchange rate appreciation--but this can raise domestic interest rates, thus choking off private sector activity and increasing interest costs to the budget. 148. The most effective way out of these difficulties is ensure that aid improves productivity by enough to offset the otherwise dampening effects of reduced competitiveness. This requires focusing spending on more productive projects and ensuring sound expenditure management. A carefully designed and scaled-up investment program may result in a traded goods sector even larger than it would have been without the aid if, for example, productivity gains from better roads, education or health outweigh the effects of the real exchange rate appreciation. Similarly, using aid to import factors of production, such as skilled labor, would tend to create fewer pressures on the real exchange rate. However, making such adjustments so as to mitigate the potential negative effects of a large increase in aid may be a challenge. It may be difficult to re-direct aid- related spending toward imported inputs, for example, as some projects may require substantial use of scarce domestic inputs, such as skilled labor. 149. In managing the transition to significantly scaled up aid flows, countries will need to assess existing bottlenecks (e.g., skilled personnel, physical infrastructure), in light of available policy instruments, and on this basis formulate plans for addressing these capacity constraints in a timely manner (see section below). Policymakers will also need to coordinate fiscal and monetary policy closely, so as to balance fiscal, exchange rate, and reserve management objectives. In some cases, the spending of incremental aid flows may also need to be more gradual. 150. The impact of significantly scaled-up aid on medium-term debt and fiscal sustainability will also need to be considered. First, if additional aid is in the form of loans, even if concessional, it will have a bearing on a country's medium-term fiscal outlook. Concerns about debt sustainability are, of course, mitigated to the extent the additional aid inflows are in the form of grants. Second, aid can set in motion government projects and programs that will need funding well into the future if they are to be effective. If it is likely that future aid will not fully cover the recurrent costs associated 152IMF (2005a) reviews recent experience with the macroeconomic management of large increases in aid inflows. Of the five cases reviewed (Ethiopia, Ghana, Mozambique, Tanzania, Uganda), none managed fully to absorb the aid surge and ensure a real resource transfer to the domestic economy. 62 with scaled-up programs, then the government will have to find other sources of revenues--such as increased tax revenues--to fund such activities. 151. The Bank and the Fund can assist countries in assessing potential macroeconomic implications of increased aid, and developing plans to address them. The Fund could play an important role in helping countries to assess the likely country- specific macroeconomic impact of the additional aid inflows, and the adjustments to the macroeconomic policy framework needed, immediately and over time, to accommodate these. This includes helping to clarify the potential economy-wide implications of increased aid-financed spending for inflation and the real exchange rate, and ways to manage any country-specific risks. The Bank can help by examining how the planned composition of spending and use of aid contributes to growth, exports, and additional public revenue generation. Both the Bank and the Fund have stepped up work at the country level to assess the implications for fiscal and external debt sustainability of the assumed volume of external financing and the likely mix of loans and grants. Addressing capacity constraints 152. In order for significantly scaled-up aid to contribute effectively to sustained growth and poverty reduction, capacity constraints also need to be addressed. In addition to the macroeconomic considerations discussed above, countries will need to alleviate a combination of human and physical capital constraints, and institutional weaknesses, the relative importance of which will vary across countries. An understanding of these constraints, and the ways in which they are binding, is critical to developing sequenced interventions to address them. Efforts to address these constraints must also be long term, systemic, and effectively deal with both the demand and supply for improved public sector performance. While external assistance can provide support, it cannot easily influence the cultural norms and political economy underpinning demand for public sector performance. Consequently, for capacity-building efforts to succeed, they need to take adequate account of local politics and institutions, and be country- owned rather than donor-driven.153 153. The PRS approach has the potential to help countries better prioritize capacity building activities and to guide support from donors, but this dimension deserves considerably more attention. Some countries (e.g., Mauritania) do discuss in their PRSPs the adverse impact of capacity constraints on PRS implementation, but the vast majority of PRSs have thus far failed to address this issue in a comprehensive manner.154 There are, however, some encouraging examples. Ethiopia, for instance, has developed a comprehensive capacity-building strategy, integrated into its PRS, around which most development agencies are coalescing their support. To capture cross-sector 153OED (2005: 7-8). 154The IEO, in its report on the IMF's technical assistance, pointed to the absence of comprehensive capacity-building strategies as one of the weaknesses of most PRSPs, complicating efforts to put together coordinated, well-sequenced, and effectively implemented capacity-building support programs that are consistent with the objectives of the PRSP. See IEO (2005). 63 synergies and lessons, a specific ministry is charged with formulating, implementing, and monitoring capacity-building projects. 154. To support the development of effective capacity-building programs, country diagnostics activities should be country-owned. Fostering full government ownership of the change agenda remains an area for continued attention. For example, in Ethiopia, while there was limited interest and involvement by the authorities in past public expenditure reviews (PERs), and consequently little capacity building, strong government demand for, and involvement in, the CFAA resulted in a positive transfer of skills.155 Another positive example is in Tanzania, where government and donors have redefined the traditional PER from a study that primarily fulfils fiduciary requirements to one that is part of the government's work plan and informs the annual budgetary decision-making cycle.156 155. Where diagnostics are donor-supported, they also need to be better coordinated and result in actionable plans that are customized to local context. For example, a recent survey of 13 African countries found that, on average, there were four separate public financial management reviews in each country.157 The Public Expenditure and Financial Accountability (PEFA) program is an example of good practice for all countries, and could be usefully extended to other sectors (box 24). The emphasis on country leadership and donor coordination could help address problems with many sector capacity-building efforts that are still often fragmented and donor-driven. Other good practice includes ensuring that international practices are adapted to the local context. 156. Using additional aid productively also requires complementary investments in physical capacity, as well as improvements in policy and governance. A boost in aid-financed infrastructure, for example, could contribute to increased sustainability of long-term development efforts.158 However, private sector capacity for both construction and maintenance, and the ability of government to regulate and manage contracts effectively are important factors. All in all, better country-level analysis of the composition of expenditures and how they should evolve to make faster progress towards the MDGs is required. Incorporating an analysis of these relationships at the country level into PRSs is critical to ensuring that incremental aid is used well. In Ethiopia, where considerable work has been done to assess needs, limitations, and capacity, it has been possible to use the Bank's Maquette (model) for MDG Simulations (MAMS) to examine various constraints and their implications for aid absorption. Some insights from that analysis are summarized in box 25. 155OED (2005: 15). 156OECD-DAC (2003b: 51). 157SPA (2005a: 15-18). 158See, for example, the Bank's Infrastructure Action Plan, World Bank (2003c) and OED (2005: 4). 64 Box 24: Public expenditure and financial management capacity building The goals of the PEFA Program are to strengthen recipient and donor ability to: (i) diagnose the condition of country public expenditure, procurement and financial accountability systems; and (ii) develop a practical sequence of reform and capacity-building actions, in a manner that: Encourages country ownership Reduces the transaction costs to countries Enhances donor harmonization Better addresses developmental and fiduciary concerns Leads to improved impact of reforms A strengthened approach to supporting PFM reforms has been identified, building on three components: A country-led PFM reform strategy and action plan A coordinated program of multi-year support aligned with the government's PFM strategy A shared information pool with a framework for measuring results that provides consistent information on country PFM performance, including progress over time This PEFA PFM Performance Measurement Framework has been developed as a contribution to the collective efforts of many stakeholders to assess and develop essential PFM systems, by providing a common pool of information for measurement and monitoring of PFM performance progress, and a common platform for dialogue. Source: www.pefa.org. Box 25: Reaching the MDGs: findings from MAMS simulations--Ethiopia Reducing poverty by half will require higher economic growth. Key to raising growth is substantial investment in basic infrastructure, particularly roads, energy and water. Reaching the human development MDGs will require spending to boost the quality and quantity of public services. Initial estimates suggest that foreign grant financing requirements might rise from $16 per capita at present to around $60 per capita in 2015. This is roughly twice the average level of ODA per capita in sub- Saharan Africa in the early 1990s. Careful sequencing of public investment is important to minimize the cost of reaching the MDGs. In Ethiopia, priority investment is needed in basic infrastructure to generate the basis for higher productivity growth and improved linkages within and across sectors. Among the MDG services, accelerating education spending is a priority since skilled labor can only be produced with a lag and is a binding constraint on absorptive capacity. The macroeconomic impact of significantly higher aid flows on the tradable sector (Dutch disease) is a serious concern. This underscores the importance of a liberal trade regime, market access in OECD countries, and addressing behind-the-border barriers to trade. Large-scale frontloading of aid disbursement (other than in infrastructure) is costly as it pushes against capacity constraints, intensifies the premium on skilled wages, bids labor away from the private sector (depressing growth), and could incur more serious Dutch disease effects. On the other hand, frontloading also results in earlier success to social outcomes. The relative costs and benefits need to be weighed. Improvements in underlying governance and institutional structures are important for securing broad productivity improvements in the public sector. Source: Sundberg and Lofgren (2005). 65 4. SUMMARY OF FINDINGS AND GOOD PRACTICES The superior man is modest in his speech, but exceeds in his actions. Confucius159 A. OVERVIEW 157. This review has considered implementation of the PRS approach over the past five years. It concludes that the approach has focused attention more squarely on poverty reduction--including the analysis of the causes and different dimensions of poverty--and the need to address country-specific constraints to development. While implementation experience has been diverse and varied, what is common is the need to address a challenging development agenda in an environment with generally weak institutions and capacity. In many countries, the PRS approach has helped catalyze the strengthening of policy processes and systems that are central to meeting their development challenges. In particular, there is increased focus on: setting clear goals and targets that are linked to public actions; improving budget and monitoring systems; opening the space for discussing national priorities and policies for poverty reduction and growth; filling country-specific analytic gaps; and aligning and harmonizing donor assistance with national priorities. It is also clear, however, that the PRS approach is not a panacea; it cannot compensate for generalized policy failures, poor governance, or broad institutional lapses. 158. The findings of this review do not point to the need for a radical shift in approach, but rather they speak to the need to consolidate progress, address outstanding issues, and focus on implementation. Past implementation progress reports have pointed to a range of issues that need continued attention, by country authorities, country stakeholders, and other development partners, including by the Bank and the Fund, to strengthen the PRS approach. The conclusions of this review are consistent with those messages. Effective states, engaged societies, and more and better aid are all needed. 159. With this country focus in mind, and realizing that after five years of annual reviews a body of good practice is now available, the issue then becomes one of tailoring these practice to individual country circumstances. Countries are at different stages in the PRS process, and, for sustainability, PRS processes need to be increasingly embedded in domestic policy formulation and implementation processes. Therefore, it would be beneficial to focus any future reviews on particular aspects of the approach that are important for effective implementation. Countries and donors alike are more apt to benefit from this more selective and targeted approach to experiences on the ground than from additional general reviews. 159The Confucian Analects. 66 B. FIVE THEMES 160. At the start of the review process, Bank and Fund staffs identified five themes that are central to the effectiveness of the PRS approach. This review has confirmed the importance of these themes all of which are central to a country-based development model. These themes consist of: (i) strengthening the medium-term orientation of the PRS; (ii) utilizing the PRS as a mutual accountability framework between countries and donors; (iii) enhancing linkages between PRSs, MTEFs and budget processes; (iv) sustaining meaningful participation; and (v) tailoring the approach to conflict-affected and fragile states. This section highlights selected findings and good practice for each of these themes. Going forward, the key is to translate good practice into normal practice, while maintaining momentum and support behind a process that requires commitment over the long haul.160 Strengthening the medium-term orientation of the PRS 161. Strengthening the medium-term orientation of the PRS requires progress along several dimensions. The review points to the need for countries to set clear goals and targets (paragraphs 27 to 32)161, which in turn supports greater transparency and accountability. To provide a framework for supporting ambitious development plans (paragraphs 141 to 145), alternative scenarios need to be mapped out, including measures to address bottlenecks to absorptive capacity (paragraphs 146 to 156). Effective monitoring systems (paragraphs 47 to 52) are needed to track progress and adjust policies and programs during implementation. 162. Medium-term goals and targets. Setting clear development goals and specific targets in the PRS helps to provide clarity as to the priorities of government. Efforts to frame goals and targets against the backdrop of the MDGs, taking into account local conditions and priorities, can help to crystallize deeper commitment and accountability. The PRS process has encouraged countries to set concrete targets, and there is some evidence that governments have felt increasingly accountable for delivering results. However, many PRSs would benefit from more explicit links between goals/targets and the policies, programs, and public actions needed to achieve them. 163. For countries, good practices include setting goals and targets, framed against the backdrop of the MDGs, but customized to country conditions. Also needed are appropriate intermediate indicators that enable monitoring and feedback and complement measures of poverty outcomes. Gaining a better understanding of causalities and complementarities can help strengthen the links between goals and targets and public actions. When goals, targets, and indicators are not clearly defined, or when links to public actions need to be strengthened, efforts should be made to improve them over time 160As these five themes are quite interrelated, various findings and good practice apply across more than one theme. For presentational purposes, it was necessary to present such cross-cutting issues under one heading. It is important to note that there is often broad applicability. 161Paragraph numbers refer to relevant paragraphs in this report. 67 through an iterative process. Development partners can support the above practices through coordinated analytical work and capacity building. 164. Framework for supporting ambitious development plans. PRSs should provide the framework for scaling up country-level efforts to reach the MDGs. A country's PRS can elaborate plans (and present them in alternative scenarios) for utilizing higher aid levels, including measures to deepen absorptive capacity, and deal with the macroeconomic implications of absorbing higher aid flows. This is now impeded by a combination of technical challenges, and skepticism that such efforts will yield more and better aid and other priorities. 165. Absorptive capacity. PRSs should address the combination of constraints-- human and physical capital, institutional, and macroeconomic--that can reduce the contribution of aid to sustained growth and poverty reduction. Efforts to address human and institutional constraints need to be long term, country owned rather than donor driven, and effectively address both the supply and demand for improved public sector performance. To date, while the PRS approach has the potential to help countries develop prioritized plans for coordinated capacity building, this is not yet the norm. In terms of potential macroeconomic constraints, the use of scaled-up macroeconomic scenarios has been suggested as a way to assess and manage the likely macroeconomic impact of additional aid inflows. However, the experience in using such scenarios has been limited. 166. For countries, in addition to developing alternative macroeconomic scenarios, good practices include using the PRS process to identify the most binding constraints to absorptive capacity, and to develop holistic plans for alleviating them. Development partners need to ensure that their support for capacity development is demand-driven, coordinated, and adequately takes into account local conditions. Involving country stakeholders in diagnostic work can also help build capacity and ownership of a reform program. Support is also needed to help countries identify and manage the likely macroeconomic impacts of additional aid flows. 167. Monitoring systems. For PRSs to support evidence-based decisions and policy choices, mechanisms for evaluating the effectiveness of development programs, and for making adjustments based on implementation experience are needed. This requires attention not just to the more technical issues of the supply of information, but also the more complex and politically sensitive factors that create demand for its use. Developing effective, coordinated monitoring systems--which require complex institutional change--has proven particularly challenging. This process is also at times impeded when PRS monitoring systems are not sufficiently integrated with country systems, focusing more on reports for donors that may not be appropriate for domestic use. 168. For countries, good practices include paying sufficient attention to the incentives for building and using PRS monitoring systems. Often, placing the institutional lead for PRS monitoring systems close to the center of government or budget process can give the 68 system greater authority. It is also useful to disseminate monitoring information and analysis across government and to the public. Cooperation with civil society partners can be helpful in presenting monitoring information to the general public in a non-technical way. Development partners need to provide sufficient coordinated support to overcome capacity and technical constraints. Setting up parallel monitoring systems should be avoided. Also, development partners need to be flexible with regard to their reporting requirements, allowing more space for the governments to develop country-specific APRs which are closely tied to (or integrated with) domestic reporting arrangements. Bearing that in mind, it is also important that partners articulate any essential monitoring needs during the design process to ensure that they can be served through the system. Utilizing the PRS as a mutual accountability framework 169. The PRS approach is intended as an instrument to support mutual accountability. To strengthen it in this regard, the review findings point in two directions. Countries need to develop PRSs that include sufficiently prioritized and specific public actions (paragraphs 33 to 38) around which donors can align. At times this process requires strengthening analytic foundations (paragraphs 114 to 139) to help inform choices, particularly those focused on sustaining higher growth; it also requires the political will to make difficult choices. Donors need to improve aid quality (paragraphs 83 to 90) by making it more predictable, aligned, and harmonized. These efforts by countries themselves and donors are both central to maintaining a balance of accountabilities (paragraphs 95-110). 170. Prioritization and specificity of public actions. Sufficiently specific operational plans are needed to support PRS implementation and donor alignment. Many PRSPs and APRs continue to lack prioritization, which can be a result of the level of analysis and capacity needed to do so effectively, unwillingness to make tough policy choices across competing demands, and/or because donors continue to encourage inclusion of "their" projects and priorities. There is some evidence that improving sector strategies and ensuring their close link to the PRS has a positive impact in enhancing specificity and prioritization--these efforts need to be sustained and strengthened. 171. For countries, good practice includes continuing efforts to make PRSs--or the (operational) plans closely linked to them--more action-oriented and prioritized. For specific actions that are highlighted, it is useful to specify the agency responsible and timeframe. Development partners should provide timely and constructive feedback, but resist making comments that are really intended to push their own priorities. Country priorities and sequencing should be respected in designing assistance strategies. When a PRS is not yet sufficiently specific to derive operational priorities, care should be taken to ensure that the policies and programs supported by donors benefit from real political commitment and a sense of national priority. 172. Analytic foundations. The results-orientation of PRSs can be strengthened over time with better analytic foundations. Such analysis includes poverty diagnostics that are regionally and demographically disaggregated; distributional analysis of the effects of specific policies on different groups; analysis of policy and institutional levers 69 for growth and the transmission mechanisms for connecting poor people to growth; and sector strategies built on solid understanding of how well public programs and actions meet the needs of poor people. When high-quality sector strategies exist, the PRS process has proven useful in drawing linkages across them, and countries are increasingly balancing the much-needed emphasis on human development issues with strategies in the productive sectors and attention to the enabling environment for private sector development. However, continued attention is needed to strengthen weak or poorly- defined sector strategies. 173. For countries, good practice includes regularly updating poverty profiles and refining them when needed. Given the wide range of areas where analysis may be needed, it is important to identify high-priority gaps, and to focus capacity on filling those first. For example, distributional analysis should focus on the key reforms that are likely to have significant distributional impact. Development partners should help countries to address analytic gaps, recognizing that priorities and sequencing will vary by country. Attention should also be paid to building capacity of government counterparts, as well as local and regional research groups. 174. Aid quality. Aid that is provided in an effective manner in support of the priorities set out in the PRSP can reinforce, rather than detract from, domestic accountability. While some progress has been made in improving aid quality and in aligning donor support around the PRS, it falls short of what was initially expected. High- level commitments (and good practice) must now be translated into the normal way of "doing business." 175. For countries, good practice includes actively leading coordination efforts, and alleviating the obstacles to harmonization, including by making their PRSPs more operationally relevant. Donors need to deliver aid in ways that build rather than undermine local capacity, using country systems whenever feasible, and designing necessary safeguards in a way that helps strengthen rather than undermine existing systems. They need to respect and align their assistance with national cycles, particularly the budget, and ensure the consistency of their interventions with the development priorities set out by the country in its PRSP. Enhancing linkages between the PRS and the MTEF and budget 176. Strengthening the institutional link between the PRS and MTEF and budget processes is key to institutionalizing the PRS approach. Close links between a country's PRS, MTEF, and budget processes (paragraphs 39 to 44) helps foster prioritization and is a requisite for linking scaled-up resources to PRS priorities. Strengthening these links requires a sequenced approach that also addresses weak institutional and technical capacity. Moreover, it depends critically on political will. The PRS process has helped raise the profile of the need for improving public expenditure management and for developing country-specific reform programs. While early PRSs focused mostly on budget formulation, more attention is now being paid to budget execution and reporting. Evidence suggests some increase in the transparency, openness and pro-poor character of budgeting processes. 70 177. For countries, good practices include assessing the current state of PEM systems and developing realistic plans for improving them, along with sequenced plans for full integration of the PRSs, MTEF, and budget process. Local development plans and PRSs also need to be aligned, recognizing that capacity constraints and severe resource shortfalls at the local level make this process very challenging. Development partners need to be aware of local conditions and avoid undermining existing capacity with multiple, disconnected reform initiatives. Coordinated, demand-driven capacity-building efforts are called for. The changes to the PRS architecture which were introduced last year were intended to encourage stronger links to domestic processes. Dissemination of information about these changes and how they are being taken up at the country level should be a priority for the Bank and the Fund. Sustaining meaningful participation 178. Participation can be a key factor to enable strong accountability mechanisms. (paragraphs 53 to 80). Relative to starting points, the PRS approach has opened up space for stakeholders to engage in a national dialogue on economic policy and poverty reduction. However, it is clear that a PRS process cannot immediately offset the shortcomings of weak institutions and political structures. Consequently, continued efforts are needed to improve these and to better institutionalize participation in PRS formulation, implementation, and monitoring. This often calls for closer integration with existing participatory processes and representative bodies. An informed policy debate can be supported by exploring conflicting hypotheses and making assumptions transparent, although building in-country capacity to conduct such debates is a process that requires time and more attention. Similarly, not all decisions can or should be made in public fora, but there is still considerable scope to open up space for a public discussion on some of the policy choices that underpin the poverty reduction strategy. 179. For countries, good practices include communicating the goals of participation process upfront and providing relevant information in accessible forms. Efforts should be made to embed participation within existing mechanisms, with due consideration to representative bodies, strengthening these where necessary. Participatory monitoring tools can be helpful in institutionalizing citizen feedback into periodic PRSP assessments. During implementation, there is also an opportunity to broaden the engagement of stakeholders, such as the business community, that may have been less involved in the formulation stage. Development partners need to be aware of (and not bypass) participatory and representative processes and institutions that already exist. When such processes are weak, consideration needs to be given as to whether and how external assistance can contribute to improving them. Development partners should also encourage the country authorities to establish a conducive framework for open public debate on policy choices, including on macroeconomic policies. Tailoring the approach to conflict-affected and fragile states 180. The broad principles of the PRS approach apply to conflict-affected and fragile states. Tailoring the PRS approach to country circumstances is important 71 everywhere, but is particularly critical in these countries. While the broad principles of the PRS approach apply to conflict-affected and fragile states, extra care is needed to tailor the approach (paragraphs 77 to 80 and paragraphs 91 to 94) to ensure that it is appropriately customized and that expectations are realistic. 181. For countries, good practice includes to integrate security issues into their strategies when this is an important element for effective poverty reduction. The causes and legacies of conflict, and their strategic implications should also be considered, even in cases where it may be politically difficult or counterproductive to the peace and reconciliation process to include detailed references in written documents. It can also be helpful for PRS processes to draw upon, as well as contribute to, parallel peace processes that may be underway. The PRS process can promote the inclusion of poverty-related issues in peacebuilding processes, and conversely reconciliation issues can be integrated into the PRS. Identifying a few governance reforms that are essential to formulating and implementing a PRS, and focusing early attention on those, can be a useful way to initiate a PRS process. Development partners should tailor their aid instruments based on a sound understanding of local circumstances and avoid mechanisms that create long-term dependence on parallel, unsustainable structures. Participatory processes, which take into account local conditions and are conflict-sensitive, should be encouraged, but expectations also need to be realistic. In post-conflict countries, it is also important that aid volumes not diminish just at the time when they can be most effective. C. SUSTAINING SUPPORT FOR COUNTRY-BASED DEVELOPMENT EFFORTS 182. The focus of country efforts, and of donor support, looking forward should be on building and implementing country-driven strategies that are consistent with the core PRS principles. Specifically, these strategies need to: lay out results and the policies to achieve them; rely on processes and systems that are fully integrated into existing domestic decision-making processes; focus on monitoring of implementation and of intermediate progress so as to make adjustments as needed; and emphasize building the analytic, human and institutional capacity to strengthen formulation and implementation. Such strategies will contribute to maximizing the long-term sustainable impact of aid. Donors must consistently ensure that their international commitments to align their assistance, harmonize their procedures, and increase the volume of their aid translate into actual practice at the country level. Above all, they must be careful to provide their assistance in ways that help strengthen domestic systems and accountability. 183. The priority in this next "generation" of PRSs should, therefore, be to renew the focus on the core PRS principles and ensure their even application across countries. This will enable the PRS approach to realize its potential as a framework for effectively coordinating public action and donor support, as well as for balancing accountabilities so that countries and their development partners can meet their shared responsibility for achieving the desired development results. 72 ANNEX 1: PRS STATUS BY COUNTRY March 2000-June 2002 July 2002 - June 2003 July 2003-June 2004 July 2004-June 2005 Albania ** Azerbaijan * Armenia * Bhutan Bolivia (CP) Benin * (CP) Bosnia & Herzegovina Burkina Faso Burkina Faso **** (CP) Cambodia * Djibouti Cape Verde Gambia, The * (DP) Cameroon * (DP) Kenya Lesotho Guyana * (CP) Chad * (DP) Lao, P.D.R. Nigeria Honduras ** (CP) Ethiopia * (CP) Madagascar * (CP) Sao Tome and Principe (DP) Malawi ** (DP) Georgia * Moldova Sierra Leone (DP) Mali * (CP) Ghana * (CP) Mongolia Tanzania Mauritania ** (CP) Guinea * (CP) Pakistan Timor Leste Mozambique *** (CP) Kyrgyz Republic * Serbia and Montenegro Uganda Nicaragua ** (CP) Nepal Niger ** (CP) Sri Lanka Rwanda ** (CP) Senegal * (CP) Tajikistan * Tanzania *** (CP) Uganda *** (CP) Vietnam ** Yemen Zambia ** (CP) Dates correspond to those on country documents. This list reflects PRS documents that staffs were aware of as of June 30, 2005. *, **, ***, and **** indicates one, two, three, or four APRs, respectively. Countries in italics have completed an APR during the past year. Countries in bold have completed a fully revised PRSP. (DP) and (CP) indicates countries that have reached the decision point and completion point, respectively, under the HIPC Initiative. 73 ANNEX 2: SUMMARY OF PRS IMPLEMENTATION PROGRESS REPORTS This annex highlights key messages from joint Bank/Fund progress reports on PRS implementation. Initially progress reports were prepared twice a year. In March 2002, the progress report reflected the findings of a larger Review. Since the September 2002 annual meetings they have been prepared annually. Progress Report on Poverty Reduction Strategy Papers (April 2000). This report primarily discusses consultations around the PRS Initiative and actions by the Bank and Fund to gear up to support countries in the preparation of their PRSPs. The report emphasizes the need for countries to tailor the PRSP to reflect individual country circumstances. Poverty Reduction Strategy Papers--Progress in Implementation (September 2000). Drawing mainly from the country experience in preparing I-PRSPs, this report highlights likely capacity constraints due to the institutional and technical demands and administrative costs of preparing PRSPs; rising expectations for coverage; and uncertainty of development partners about their specific roles. The report also signals the need for greater country-specific analysis on a range of issues, including better understanding the linkages between expenditures and results and the determinants of pro- poor growth. Several tensions in the PRS approach are identified including: (i) the need for countries to prepare their PRSP quickly in order to obtain concessional assistance and debt relief versus country ownership secured through broad participation; and (ii) country ownership versus the prerogative of the Bank and Fund Boards to determine if the PRSP forms a sound basis for concessional assistance. Poverty Reduction Strategy Papers--Progress in Implementation (April 2001). At the time of this report, four countries had prepared PRSPs and 32 countries had prepared I- PRSPs. This report describes steps taken by the Bank and Fund to facilitate the PRS process, including developing guidelines for the JSAs of full PRSPs, expanding learning programs, and improving information available to countries and their development partners through the PRSP sourcebook and external websites. The report reflects the intention of the Fund to streamline conditionality under PRGF programs and notes the Bank's creation of the PRSC instrument to support implementation of PRSPs. Finally, the report highlights a range of outreach efforts with the UN, EU, bilateral donors, multilateral development banks and NGOs. Poverty Reduction Strategy Papers--Progress in Implementation (September 2001). At the time of this report, five countries had prepared PRSPs and 36 countries had prepared an I-PRSPs. The report highlights that initial country timetables for developing full PRSPs have been overly optimistic, and that countries and development partners have underestimated the time needed to develop an inclusive participatory process and to undertake the necessary analytical work. The report discusses how countries could use Poverty and Social Impact Analysis to help understand the growth, poverty and 74 distributional impact of policy actions. It notes, however, that countries are likely to face significant methodological and analytic challenges in conducting PSIA, that it will be a long-term endeavor and that it is important to be realistic about how quickly progress can be made, that countries will need to be selective in the reforms that are analyzed. The report also briefly discusses the need to strengthen public expenditure management systems to track poverty-reducing spending, Bank-Fund coordination in program design and conditionality, and outreach to other partners. The report notes that the PRSP process has been accepted as the basis for country-level monitoring of progress towards achieving medium-term development goals. Review of the Poverty Reduction Strategy Paper (PRSP) Approach Main Findings (March 2002). At the time of this report, ten countries had prepared PRSPs, of which three had produced annual progress reports. The central message from the review is that there is broad agreement among low-income countries, civil society organizations and their development partners that the objectives of the PRSP approach remain valid. The review found broad agreement on four key achievements of the PSRP approach to date: (i) a growing sense of ownership among governments of their PRSs; (ii) a more open dialogue within governments, and with at least some parts of civil society than had previously existed; (iii) a more prominent place for poverty reduction in policy debates, extending beyond social sector interventions to focus on reducing income poverty through higher and more broadly shared growth; and (iv) more systematic data collection, analysis, and monitoring of outcomes. The key challenges identified include: (i) alignment by partners, including the Bank and the Fund, to support PRS implementation; (ii) shifting beyond process, to content and implementation, and greater understanding of the linkages between policies and poverty outcomes; and (iii) realism in setting goals and targets, as well as in managing expectations, both within countries and among their development partners. While there have been improvements over time in both process and content, substantial scope for further enhancement remains. Based on country experience, high priorities include: (i) improving public expenditure management systems; (ii) placing greater emphasis on, and buildings capacity for, monitoring and evaluation; and (iii) strengthening and institutionalizing participatory process. The report emphasizes that the PRS approach requires flexibility so that both the process and content of poverty reduction strategies can vary across countries in light of national circumstances. It also notes that lack of capacity, and the inability to use existing capacity effectively, remain important constraints to preparation, monitoring and implementation of PRSPs in many countries. The report highlights a range of good practices for countries and development partners, and notes actions to be taken by the Bank and Fund, on a range of topics (participatory processes; conflict-affected countries; poverty diagnostics, targets and indicators, monitoring systems; priority pubic actions; public expenditure management; integration of the PRS into other decision-making processes; and improving donor alignment). Poverty Reduction Strategy Papers--Progress in Implementation (September 2002). At the time of this report, 18 countries had prepared PRSPs, of which five had produced 75 annual progress reports. The report concludes that the approach remains relevant at the country level, but highlights challenges which need to be addressed over time. These include capacity building, opening up of policy dialogue, aligning external assistance, and integrating PRSs into budgets. The report concludes by noting that need for more systemic examination of progress in implementation of the results achieved, and note that future progress reports would feed into the next joint review of the PRSP approach scheduled for Spring 2005. Poverty Reduction Strategy Papers--Progress in Implementation and Detailed Analysis of Progress in Implementation (September 2003). At the time of this report, 32 countries had prepared PRSPs, of which seven had produced at least one annual progress report. The report finds that while there is evidence of improvement and progress, as recent PRSPs build on the efforts of earlier PRSPs, and countries more advanced in the process are successfully adapting and implementing their strategies, the PRS instrument is charged with multiple objectives, may of which result in tensions. This report concludes that this inevitably means the PRS will reflect compromises and that attaining some ideal level of performance along every line is impossible. The report points to these tensions being particularly manifest in the following respects: (i) concerns about the breadth of government's commitment beyond the team responsible for preparing the PRS; (ii) countries continue to find it difficult to strike an appropriate balance between ambition and realism in setting PRS targets; (iii) weak PEM and difficulties in linking the PRS to the budget strain countries' administrative capacity; and (iv) there is an urgent need to improve donor alignment and harmonization around national strategies, in order to achieve successful PRS implementation. Poverty Reduction Strategy Papers--Progress in Implementation (September 2004). At the time of this report, 42 countries had prepared PRSPs, of which 23 had produced at least one annual progress report. The report notes that given the country-specific nature of PRS process, country experience has varied with regard to both process and content. However, in general, it has helped: (i) countries focus more squarely on poverty reduction in formulating and implementing their development strategies; (ii) open up the participatory process in many countries; (iii) focus more attention on monitoring poverty- related outcomes; and (iv) draw attention to the importance of understanding and addressing the country-specific constraints to more effective development. The key finding was that while countries have made good progress in addressing the more straightforward challenges inherent in the approach, the challenges that remain are technically complex and institutionally challenging. As implementation proceeds, continuing attention on several key issues is warranted, including: (i) integrating the PRSP process with existing decision-making processes, particularly the budget, and expanding the involvement of sectoral ministries and parliaments; (ii) deepening the links to the MDGs, and identifying the financial, policy and institutional constraints that need to be addressed to accelerate progress towards these goals; (iii) continuing to strengthen the results focus of country strategies and the complementary monitoring and evaluation systems; and (iv) speeding the pace of progress in aligning donor support with country strategies; harmonizing donor processes and procedures; and increasing aid flows. 76 BIBLIOGRAPHY REFERENCES AFRODAD (African Forum and Network on Debt and Development), 2003. "Africa's Experience with the PRSP: Content and Process--A Synthesis Report of 10 African Countries." AFRODAD PRSP Technical Paper, Zimbabwe. Albania Technical Secretariat, Council of Ministers, 2004. "Establishing an Integrated Planning System in Albania," Tirana, Albania. Andrews, Matthew and Allister Moon, 2004. "A Study of Public Expenditure Management Capacity and Reforms in ECA PRSP Countries." Washington. D.C. ADB (Asian Development Bank), 2005. "Comments on Poverty Reduction Strategy Initiative and Country Experiences." Manila, Philippines. Berg, Andrew and Anne O. Krueger, 2003. "Trade, Growth, and Poverty: A Selective Survey." IMF, Working Paper WP/03/30, Washington, D.C. Booth, David, Karin Christiansen, and Ruth Driscoll, 2005. "Progress Reviews and Performance Assessments in Poverty-Reduction Strategies and Budget Support: A Survey of Current Thinking and Practice." Report submitted to the Japan International Cooperation Agency, Overseas Development Institute, London, UK. Booth, David (ed.), 2003. "Fighting Poverty in Africa: Are PRSPs Making a Difference?" Overseas Development Institute, London. BMZ (Bundesministerium fur Wirtschaftliche Zusammenarbeit und Entwicklung), 2002. "German Contribution to the World Bank/IMF PRSP/PRGF Review." Germany. Canagarajah, Sudharshan and Arthur van Diesen, 2005. "Uganda's Experience with the Poverty Reduction Strategy: Approach an Overview of Lessons Learned," draft mimeo. CIDA (Canadian International Development Agency), 2005. "CIDA Comments: World Bank's 2005 Review of Progress in Implementing the PRS Approach." Ottawa, Canada. CIDSE-Caritas Internationalis, 2004. "PRSP as Theatre: Backstage Policy Making and the Future of the PRSP Approach." CIDSE-Caritas Internationalis Background Paper, Brussels and Vatican City. ----, 2005a. "More than a Numbers Game? Ensuring that the Millennium Development Goals Address Structural Injustice." Brussels and Vatican City. ----, 2005b. "Submission to World Bank/IMF 2005 PRS Review." Brussels and Vatican City. Civil Society for Poverty Reduction, 2004. "An Easy Look at Zambia's Poverty Reduction Strategy Paper 2002-2004." Zambia. Cordaid, Novib and Wemos, the Netherlands, 2005. "Input to the PRSP Review." the Netherlands. 77 Collier Paul, and Anke Hoeffler, 2005. "Resource Rents, Governance, and Conflict." in Journal of Conflict Resolution. 2005; 49: 625-633 Coudouel, Aline, Marcus Cox, and Nigel Thornton, 2005 (forthcoming). "Monitoring Systems for Poverty Reduction Strategies-an Analysis of Institutional Arrangements." World Bank, Washington, D.C. DFID (Department for International Development), 2005. "DFID's Response to the 2005 PRS Review Concept Note." Staff Comments, London, UK. DFID and World Bank, 2005. With the Support of Multitudes: Using Strategic Communication to Fight Poverty through PRSPs, Masud Mozammel and Sina Odugbemi, (eds.). jointly published by the Information and Communication for Development, DFID, and the Development Communication Division, External Affairs, The World Bank. Diop, Ndiami, Charleen Gust, and Padamja Khandelwal, 2005. "Critical Review of Trade and Growth in PRSPs." IMF and World Bank, Washington, D.C. Draman, Rasheed and Steven Langdon, 2005. "PRSPs in Africa: Parliaments and Economic Policy Performance." State and Democracy Division, Project on Democracy and the Rule of Law, GmbH Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ), Eschborn, Germany. Dutch Federations of Trade Unions: FNV (Federatie Nederlandse Vakbeweging) and CNV (Christelijk National Vakverbond), 2005. "Letter on the PRS approach." The Netherlands. Eberlei, Walter, 2005. "Donor Politics in Zambia: Promoting Poverty Reduction or Fuelling Neopatrimonialism?" Presentation at a conference on "The Political Dimensions of Poverty Reduction--the Case of Zambia", organized by the University of Zambia, Lusaka, March 9-11, 2005. Entwistle, Janet and Filippo Cavassini, 2005. "An Operational Approach for Assessing Country Ownership of Poverty Reduction Strategies, Volume 1: Analysis and Implications." Operations Policy and Country Services, World Bank, Washington, D.C. EURODAD, 2005a. "Eurodad Country Update: Zambia's Poverty Reduction Strategies in 2005." Brussels, Belgium. EURODAD, 2005b. "EURODAD Submission to the World Bank / IMF 2005 PRS Review." Brussels, Belgium. EPEP (Eurodad/Economic Policy Empowerment Program), 2005. "PRSP­Politics, Power, and Poverty ­ A Civil Society Perspective." Brussels, Belgium. EC (European Commission), 2005, "2005 PRS Review Concept Note: Comments." Brussels, Belgium. Garnett, Harry, and William Plowden, 2004. "Cabinets, Budgets, and Poverty: Political Commitment to Poverty Reduction." in Brian Levy and Sahr Kpundeh (eds) Building State Capacity in Africa, World Bank WBI Development Studies, Washington, D.C. 78 GTZ (Gesellschaft fur Technische Zusammenarbeit), 2004. "National Monitoring of Strategies for Sustainable Poverty Reduction / PRSPs." Study conducted on behalf of the German Federal Ministry for Economic Cooperation and Development (BMZ) by GFA ­ Management GmbH, Eschborn, Germany. ----, 2005. "Making Poverty Reduction Strategies Work­Good Practices, Issues, and Stakeholder Views." Eschborn, Germany. Harrison, Makiko, Jeni Klugman and Eric Swanson, 2004. "Are Poverty Reduction Strategies Undercutting the Millennium Development Goals? An Empirical Review," (preliminary draft). World Bank, Washington, D.C. Hausmann, Ricardo, Dani Rodrik, and Andres Velasco, 2005. "Growth Diagnostics," mimeo, Harvard University, Cambridge, MA. Herz, Steve and Alnoor Embrahim, 2005. "A Call for Participatory Decision Making: Discussion Paper on World Bank-Civil Society Engagement." Commissioned and presented by the Civil Society Members of the World Bank-Civil Society Joint Facilitation Committee (JFC), Washington, D.C. Holmes, Malcolm and Alison Evans, 2003. "A Review of Experience in Implementing Medium Term Expenditure Frameworks in a PRSP Context: A Synthesis of Eight Country Studies." Study commissioned by the Africa Policy Department of DFID in collaboration with the European Commission DG Development, Overseas Development Institute, London, UK. IEO (Independent Evaluation Office), 2004. "Evaluation of the IMF's Role in Poverty Reduction Strategy Papers and the Poverty Reduction and Growth Facility." IMF, Washington, D.C. ----, 2005. "Evaluation of the Technical Assistance Provided by the International Monetary Fund." IMF, Washington, D.C. IMF (International Monetary Fund), 2001. "Strengthening Country Ownership of Fund- Supported Programs." Washington, D.C. ----, 2003a. "Guide for Staff Relations with Civil Society Organizations." External Relations Dept., IMF, Washington, D.C. ----, 2003b. "Aligning the PRGF and PRSP Approach--Issues and Options." Washington, D.C. ----, 2005a. "The Macroeconomics of Managing Increased Aid Inflows: Experiences of Low-Income Countries and Policy Implications." Draft, Washington, D.C. ----, 2005b. "Monetary and Fiscal Policy Design Issues in Low-Income Countries." Draft, Washington, D.C. ----, 2005c. "The Design of Fund Supported Programs-Overview." SM/04/403, Washington, D.C. ----, 2005d, "Review of the 2002 Conditionality Guidelines." March 3, 2005, Washington, D.C. 79 ----, 2005e. "Review of the 2002 Conditionality Guidelines--Selected Issues." Washington, D.C. IMF and World Bank, 1999a. "Building Poverty Reduction Strategies in Developing Countries." DC99-29, Washington, D.C. ----, 1999b. "Poverty Reduction Strategy Papers: Operational Issues." SM/99/290, 12/10/99 and IDA/SecM1999-99/789, Washington, D.C. ----, 2002a. "Review of the Poverty Reduction Strategy Paper Approach--Main Findings and Issues for Discussion." SM/02/53 and IDA/SecM2002-0086, 2002b, Washington, D.C. ----, 2002b. "Review of the Poverty Reduction Strategy Paper Approach--Early Experience with Interim PRSPs and Full PRSPs." SM/02/54 and IDA/SecM2002-0174, 2002c, Washington, D.C. ----, 2003a. "Poverty Reduction Strategy Papers: Progress in Implementation." IDA/SecM2003-0468, IMF SM/03/279, Washington, D.C. ----, 2003b. "Heavily-Indebted Poor Countries Initiative-Status of Implementation." Washington, D.C. ----, 2004. "Poverty Reduction Strategy Papers: Progress in Implementation." IDA/SecM2004-0598/1, IMF SM/04/403, Washington, D.C. ----, 2005. "Heavily-Indebted Poor Countries (HIPC) Initiative-Statistical Update." IDA/SecM2005-0135, Washington, D.C. Inter-American Development Bank, 2005. "Comments on the "Concept Note Joint World Bank and IMF Report on Poverty Reduction Strategy Papers­Progress in Implementation 2005 PRS Review." Washington, D.C. Japan Bank for International Cooperation, 2005. "JBIC's inputs for 2005 PRS Review." Tokyo, Japan. Johnson, J.H. and Wasty, S., 1993. "Borrower Ownership of Adjustment Programs and the Political Economy of Reform." World Bank Discussion Paper No. 199, Washington, D.C. Kaufman, D., A. Kraay, and M. Mastruzzi, 2005. "Governance Matters IV: Governance Indicators for 1996-2004." Draft, May 9, 2005, World Bank, Washington, D.C. Killick, Tony, 1998. "Aid and the Political Economy of Policy Change." Routledge and ODI, London, U.K. Kindernothilfe, 2005. "Recommendations from a Child Rights Perspective with regard to joint World Bank and IMF 2005 PRS Review." Germany. Lampietti, Julian and Nils Junge, 2005. "PSIA of power sector reforms in various ECA countries," mimeo. World Bank, Washington, D.C.. Morazan, Pedro and Irene Knoke, for Institute SÜDWIND in consultation with Church Development Service (EED), Bread for the World and Misereor, 2005. "PRSP Is About People--Experiences and Reflections with Partner Organisations in Five 80 African Countries: Cameroon, DR Congo, Ethiopia, Ghana and Tanzania." Germany Morrissey, 2001. "Pro-Poor Conditionality for Aid and Debt in East Africa." Presented at WIDER Development Conference, August, 2001, Helsinki Finland. Mpepo, Besinati, 2003. "Civil Society Engagement in Zambia's Poverty Reduction Strategy Paper Process: Bringing on the Voices of the Poor." Presented in Nairobia Kenya. Mulgan, Richard, 2000. "Accountability': An Ever-expanding Concept?" Public Administration, Vol. 78 No. 3, 2000 (555-573), Oxford, UK. Netherlands Development Cooperation, 2005. "Input to the WB/IMF 2005 PRS Review." Coordinated by BUZA/DGIS, The Hague, Netherlands. Norton, Andrew, 2002. "Approach Paper: Politics and the PRSP Approach," PRSP Monitoring and Synthesis Project, Overseas Development Institute, London, UK. O'Malley, Kate, 2004. "Children and Young People Participating in PRSP Process: Lessons from Save the Children's Experiences." Save the Children, London, UK. OECD-DAC, 2003a. DAC Guidelines and Reference Series: Harmonising Donor Practices for Effective Aid Delivery, Volume 1: Framework for Donor Co- operation; Country Analytic Work and Preparation of Projects and Programmes; Measuring Performance in Public Financial Management; Reporting and Monitoring; Financial Reporting and Auditing; Delegated Co-operation, Paris, France. ----, 2003b. "Harmonising Donor Practices for Effective Aid Delivery." Good Practice Papers, A DAC Reference Document, DAC Guidelines and Reference Series, Paris, France. ----, 2004. "Survey on Harmonisation and Alignment­Measuring Aid Harmonisation and Alignment in 14 Partner Countries." Paris, France. ----, 2005a. "Principles for Good International Engagement in Fragile States." Draft, Paris, France. ----, 2005b. "Chair's Summary: Senior Level Forum on Development Effectiveness in Fragile States." Paris, France. ----, 2005c. "Harmonising Donor Practices for Effective Aid Delivery Volume II." Paris, France. ----, 2005d. "Scaling up versus Absorptive Capacity: Challenges and Opportunities for Reaching the MDGs in Africa." Briefing Paper, May, London, UK. OED, 2004, "OED Review of the Poverty Reduction Strategy (PRS) Process." CODE2004-0052, Operations Evaluations Department, World Bank, Washington, D.C. ----, 2005. "World Bank Support for Public Sector Capacity Building in Sub-Saharan Africa: An OED Evaluation." Approach Paper, Operations Evaluations Department, World Bank, Washington, D.C. 81 ODI (Overseas Development Institute), 2005a. "Harmonization and Alignment in Fragile States." (DRAFT), Background paper for the Senior Level Forum on Development Effectiveness in Fragile States, 13-14 January, London, commissioned by the DAC Learning and Advisory Process on Difficult Partnerships. ----, 2005b. "Briefing Paper: Scaling Up Versus Absorptive Capacity: Challenges and Opportunities for Reaching the MDGs in Africa," London, UK. ----, 2005c. "Toward Sector Program Alignment­SPA­2004 Sector Program Tracking Report." January 12, 2005. Oxfam, 2004. "From 'Donorship' to Ownership? Moving Towards PRSP Round Two." Briefing paper 51, London, UK. Oxfam International, 2005. "Oxfam International Submission to the World Bank/IMF 2005 PRS Review." London, UK. Paris Declaration on Aid Effectiveness: Ownership, Harmonisation, Alignment, Results and Mutual Accountability, High Level Forum, 2005. Parliamentary Centre, 2005. "Parliamentary Centre Contribution to World Bank/IMF PRS Review." Ottawa, Canada. Peretz, David and Samuel Wangwe, 2004. "Monitoring Donor and IFI Support Behind Country-Owned Poverty Reduction Strategies in the United Republic of Tanzania." Report for the Commonwealth Secretariat, London, UK. Piron, Laure-Hélène and Alison Evans, 2004. "Politics and the PRSP Approach: Synthesis Paper." Working Paper 237, Overseas Development Institute, London, UK. Rato, Rodrigo de, 2004. "Africa Needs International Help to Turn Promises Into Progress." in Business Day, South Africa, November 12, 2004. Rodrik (2004). "Growth Strategies," mimeo, Harvard University, Cambridge, MA. Scott, Christopher, 2005. "Measuring up to the Measurement Problem: The Role of Statistics in Evidence-Based Policy-Making." Partnership in Statistics for Development in the 21st Century, London School of Economics, London, UK. Seshamani, Venkatesh, 2005. "The Same Old Bottle? Content, Process and Donor Conditionalities of the PRSP." Zambia. Shepard, Andrew and Verena Fritz with inputs from Caroline Ashley, Neil Bird, David Booth, Tom Slaymaker, Tim Williamson, and Yasemin Yenersoy, 2005. "Key Issues in Sharpening the Rural Focus of Poverty Reduction Strategy Process: Draft Literature Review for IFAD Issues Paper." ODI, London, UK. SPA (Strategic Partnership with Africa), 2004. "SPA Budget Support Working Group, Report of the Kigali Workshop." June 16-18, 2004. ----, 2005a. "Survey of the Alignment of Budget Support and Balance of Payments Support with National PRS Processes." Budget Support Working Group. 82 ----, 2005b. "Toward Sector Program Alignment--SPA 2004 Sector Program Tracking Report," January 12, 2005. Sundberg, Mark and Hans Lofgren, 2005. "Absorptive Capacity and Achieving the MDGs: The Case of Ethiopia." Presented at the Seminar on Foreign Aid and Macroeconomic Management organized by the IMF Institute, March 14-15, 2005, Maputo, Mozambique. Sweden­SIDA (Swedish International Development Agency), 2005. "Swedish comments to Joint World Bank/IMF 2005 PRS Review." Stockholm, Sweden. (Note: This paper reflects work in progress and, as a consequence, the views expressed in the paper do not necessarily represent official Swedish policy.) Thornton, Nigel and Marcus Cox, 2005. "Developing Poverty Reduction Strategies in Low Income Countries Under Stress (LICUS)." Agulhas Development Consultants Ltd., London,UK. Trocaire, 2004. "The Other Side of the Coin: An Alternative Perspective on the Role of the International Monetary Fund in Low-Income Countries." Ireland. UNIFEM (United Nations Development Fund for Women), 2005. "UNIFEM comments: Joint World Bank and IMF 2005 Review of Poverty Reduction Strategy." New York. UK Commission for Africa, 2005. "Our Common Interest: Report of the Commission for Africa." UK. UNDP (United Nations Development Programme), 2005. "Comments on "The Concept Note: Joint World Bank and IMF Report on Poverty Reduction Strategy Papers- Progress in Implementation, 2005 PRS Review." New York. United Nations Millennium Project, 2005. "Investing in Development: A Practical Plan to Achieve the Millennium Development Goals." New York. United Nations Office of the High Commissioner for Human Rights, 2005. "Comments on PRS 2005 Review Concept Note." New York. Vandemoortele, Jan and Raythin Roy, 2004, "Making Sense of MDG Costing." UNDP Poverty Group, New York. VENRO (Verband Entwicklungspolitik Deutscher Nichtregierungs Organisationen), 2005. "Fighting Poverty without Empowering the Poor?" Germany Water Aid, 2005. "Broadening and Deepening Meaningful Participation: WaterAid's Experience of Involvement in PRSPs." London UK. Whaites, Alan, 2000. "PRSPs: Good News for the Poor? Social Conditionality, Participation and Poverty Reduction." World Vision International, Milton Keynes, U.K. Wilks, Alex and Fabien Lefraçois, 2002. "Blinding with Science or Encouraging Debate? How World Bank Analysis Determines PRSP Policies." Bretton Woods Project and World Vision International, UK and USA. 83 World Bank, 1996. World Development Report 1997: The State in a Changing World. Washington, D.C. ----, 2001. World Development Report 2000/2001: Attacking Poverty. Washington, D.C. ----, 2003a. Global Economic Prospects 2004: Realizing the Development Promise of the Doha Agenda. Washington, D.C. ----, 2003b. World Development Report 2004: Making Services Work for Poor People. Washington, D.C. ----, 2003c. "Infrastructure Action Plan." Washington, D.C. ----, 2004. "Good Practice Notes for Development Policy Lending." Washington, D.C ----, 2005a. "Enabling Capacity to Achieve Results-2005 Comprehensive Development Framework (CDF) Progress Report." Operations Policy and Country Services, World Bank, Washington, D.C. ----, 2005b. "Aid Financing and Aid Effectiveness," draft. World Bank, Washington, D.C. ----, 2005c. Pro-Poor Growth in the 1990s: Lessons and Insights from 14 Countries. World Bank, Washington, D.C. ----, 2005d. World Development Indicators. Washington, D.C. ----, 2005e. "Towards a Conflict-Sensitive Poverty Reduction Strategy: Lessons from a Retrospective Analysis." Washington, D.C. ----, 2005f. World Development Report 2006. Washington, D.C. ----, 2005g. "Economic Growth in the 1990s: Learning from a Decade of Reform." Washington, D.C. ----, 2005h. "Rwanda ICR: Institutional Reform Credit." Report# 32483, Washington, D.C. ----, 2005i. "Good Practice Note for Development Policy Lending: Development Policy Operations and Program Conditionality in Fragile States." Washington, D.C World Bank and IMF, 2005. Global Monitoring Report 2005 ­ Millennium Development Goals: From Consensus to Momentum. Washington, D.C. WHO (World Health Organization), 2005. "PRSPs: Their Significance for Health: Second Synthesis Report." WHO/HDP/PRSP/04.1, MDGs, Health and Development Policy, Geneva, Switzerland. ----, 2004. "PRS(P) Implementation: A Health Sector Perspective." Geneva Switzerland. World Vision UK, Fletcher Tembo (ed.), 2005. "Poverty Reduction: Are the Strategies Working?" UK. 84 OTHER BACKGROUND READING Adablah, Cornelius, 2003. "The Role of Civil Society in Decentralized Governance for Poverty Reduction: The Experience in West Africa." Paper presented at the UNDESA Workshop on Poverty Alleviation and Social Inclusion, Dakar, Senegal. ActionAid International USA and ActionAid International Uganda, 2004. "Rethinking Participation: Questions for Civil Society about the Limits of Participation in PRSPs." Washington, D.C. Agir ici, 2005. "Contribution d'Agir ici au Rapport conjoint Banque mondiale et du FMI sur l'avancement de la mise en oeuvre des Documents de stratégie pour la réduction de la pauvreté (DSRP)." Alonso, Rosa, Lindsay Judge, and Jeni Klugman, 2005. "PRSPs and Budgets a Synthesis of Five Case Studies." Paper presented at the Practitioners' Forum on Budget Support, Cape Town, South Africa. Bird, Graham, 2004. "The IMF and Poor Countries: Towards a More Fulfilling Relationship," Paper prepared for a conference on Developing Countries, Global Finance, and the Role of the IMF: Towards a New Relationship." The Hague, Netherlands. Bojö, Jan and Rama Chandra Reddy, 2003. "Status and Evolution of Environmental Priorities in the Poverty Reduction Strategies: An Assessment of Fifty Poverty Reduction Strategy Papers." Environmental Economics Series, Paper no. 93, World Bank, Washington DC. Bojö, Jan, et al., 2004. "Environment in Poverty Reduction Strategies and Poverty Reduction Support Credits." Paper no. 102, Environment Department, World Bank, Washington DC. Bonnel, René, Miriam Temin, and Faith Tempest, 2004. "Poverty Reduction Strategy Papers: Do They Matter for Children and Young People Made Vulnerable by HIV/AIDS? Results of a Joint UNICEF and World Bank Review." Africa Region Working Paper Series No. 78, World Bank, Washington, D.C. Booth, David, Karin Christiansen, and Paolo de Renzio, 2005. "Reconciling Alignment and Performance in Budget-Support Programmes: What Next?" Paper presented at the Practitioners' Forum on Budget Support, Cape Town, South Africa. Booth, David, 2005. "Poverty Monitoring Systems: An Analysis of Institutional Arrangements in Tanzania." Working Paper 247, Overseas Development Institute, London., UK. Booth, David and Xavier Nsabagasani, 2005. "Poverty Monitoring Systems: An Analysis of Institutional Arrangements in Uganda." Working Paper 246, Overseas Development Institute, London. Boughton, James M., 2003. "Who's in Charge? Ownership and Conditionality in IMF- Supported Programs." IMF Working Paper WP/03/191, September 2003, Washington, D.C. 85 Bulir, Ales; A. Javier Hamann, 2001. "How Volatile and Unpredictable Are Aid Flows, and What Are the Policy Implications?" Policy Development and Review Department, IMF Working Paper No. 01/167, October 1, 2001, Washington D.C. Bwalya, Edgar, et al., 2004. "Poverty Reduction Strategy Processes in Malawi and Zambia." R 2004: 8, Chr. Michelsen Institute (CMI), Norway. Caillods, F. and J. Hallak, 2004. "Education and PRSPs: A Review of Experiences." Review undertaken with the support of DFID, UNESCO: IIEP, Paris, France. Christian Aid, et al., 2005. "Kept in the Dark: A Briefing on Parliamentary Scrutiny of the IMF and World Bank." CIDSE-Caritas Internationalis, 2004. "PRSP: Are the World Bank and IMF delivering on Promises?" CIDSE-Caritas Internationalis Position Paper, Brussels and Vatican City. CUTS International, 2004. "Economic Development Priorities: What Next After HIPC Completion Point and PRSP?" Roundtable discussion organized by Consumer Unity & Trust Society-Africa Resource Centre (CUTS-ARC), in partnership with Southern Africa Regional Poverty Network (SARPN), Lusaka, Zambia. Dante, Idrissa, 2002. "Aid Coordination and Donor Reform." in Africa Report: Assessing the New Partnership. (pp.37-53), The North-South Institute, Ottawa, Canada. De Renzio, Paolo, et al., 2004. "Incentives for Harmonisation and Alignment in Aid Agencies." Report to the DAC Working Party on Aid Effectiveness, Overseas Development Institute, London, UK. Department for International Development (DfID), 2005. "Partnerships for Poverty Reduction: Rethinking Conditionality." a UK Policy Paper, March 2005, London, UK. Dorotinsky, William and Rob Floyd, 2004. "Public Expenditure Accountability in Africa: Progress, Lessons, and Challenges." in Brian Levy, Sahr Kpundeh eds Building State Capacity in Africa. World Bank WBI Devleopment Studies, Washington, D.C. Dorotinsky, William, forthcoming. "Performance Budgeting and Management: Comparative Study: Lessons Learnt from 16 Developing and Developed Countries." World Bank, Washington, D.C. Dorman, Sara Rich, 2004. "Democrats and Donors: Studying Democratization in Africa." in Tim Kelsall and Jim Igoe, (eds.), Donors, NGOs and the Liberal Agenda in Africa. Driscoll, Ruth and Alison Evans, 2004. "Second Generation Poverty Reduction Strategies­Paper prepared from the PRSP Monitoring and Synthesis Project." Overseas Development Institute (ODI), London, UK. Eberlei, Walter, 2004. "Fighting Poverty and Corruption: Integrating the Fight Against Corruption into the PRS Process-Analysis and Recommendations for Development Cooperation." a paper commissioned by the German Federal 86 Ministry for Economic Cooperation and Development (BMZ), Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) GmbH, Eschborn, Germany. Egulu, Lawrence, 2004. "Trade Union Participation in the PRSP Process." Social Protection Discussion Paper Series, No. 0417, Social Protection Unit, World Bank, Washington, DC. Entwistle, Janet, Natasha Bajuk, Filippo Cavassini, and Federico Steinberg, 2005, "An Operational Approach for Assessing Country Ownership of Poverty Reduction Strategies, Volume 2: Country Case Studies." Operations Policy and Country Services, World Bank, Washington, DC. Eurodad, 2005. "EURODAD Submission to the World Bank / IMF 2005 PSIA Review." Brussels, Belgium. European Commission, 2005. "EC Budget Support: An Innovative Approach to Conditionality." DG Development, Brussels, Belgium. Fox, James W., 2004. "The Treatment of the Private Sector in African PRSPs and APRs." Report to the Bureau for Africa, USAID, Washington, D.C. G-24 Secretariat Briefing Paper on the Poverty Reduction Strategy Paper (PRSP) Approach. 2003. Gould, Jeremy and Julia Ojanen, 2003. "'Merging in the Circle': The Politics of Tanzania's Poverty Reduction Strategy." Policy Papers 2/2003, Institute of Development Studies, University of Helsinki. GTZ (Gesellschaft fur Technische Zusammenarbeit), 2003. "Parliaments in Sub-Saharan Africa: Actors in Poverty Reduction?" Eschborrn, Germany. Gunning, Jan Willem, 2005. "Budget Support, Conditionality, and Impact Evaluation." Paper presented at the Practitioners' Forum on Budget Support organized by the World Bank, May 5-6, 2005, Cape Town, South Africa. Hakikazi Catalyst, 2004. "Participatory Monitoring of PRS and Pro-Poor Expenditure in Selected Districts and Areas of Arusha Region, Tanzania." Programme funded by Oxfam Ireland, Arusha, Tanzania. Heller, Peter, 2005. "Back to Basics: Fiscal Space: What It Is and How to Get It." Finance and Development, June 2005: 42: 2, IMF, Washington, D.C. Holden, Pat and Dagmar Walter, 2004. "Poverty Initiatives in the ILO: A Review of Past and Present Approaches." Working Paper no. 8, Policy Integration Department, National Policy Group, International Labour Office, Geneva, Switzerland. Hubli, K. Scott and Alicia P. Mandaville, 2004. "Parliaments and the PRSP Process." Series on Contemporary Issues in Parliamentary Development, WBI Working Papers, World Bank, Washington, D.C. Hunt, Sarah, 2004. "Honduras Update." Trocaire, Ireland. Institute of Social Studies, 2003. "Can Poverty be Reduced? Experience with Poverty Reduction Strategies in Latin America." 2003 Regional Report, 5 year review 87 project on Evaluation and Monitoring of PRS in Latin America, commissioned by Swedish International Development Cooperation Agency (Sida), The Hague. International Labour Organization, 2005. "Decent Work and Poverty Reduction Strategies: A Reference Manual for ILO Staff and Constituents." Geneva, Switzerland. IMF, 2005. "Strengthening the Fund's Ability to Assist Low-Income Countries Meet Balance of Payments Needs Arising from Sudden and Exogenous Shocks." EBS/05/88 Washington, D.C. IMF and World Bank, 2005. "Moving Forward: Financing Modalities towards the MDGs." Development Committee background document, DC2005-0008/Add. 1, Washington, D.C. ----, 2005. "Update on the Assessment and Implementation of Action Plans to Strengthen Capacity of HIPCs to Track Poverty-Reducing Public Spending." April 12, 2005, Washington, D.C. Japan Bank for International Cooperation, 2005. "The Poverty Reduction Impact of Economic Infrastructure Project (2)-Key Issues and Lessons Learned from the Vietnam's CPRGS Process." Tokyo, Japan. Johnson, Alison, Matthew Martin,and Hannah Bargawi, 2004. "The Effectiveness of Aid to Africa Since the HIPC Initiative: Issues, Evidence and Possible Areas for Action." (DRAFT), Background paper prepared for Commission for Africa (Secretariat), Development Finance International, London, UK. Kubalasa, Dalitso Kingsley, 2004. "PRSPs. a Positive Development from the Debt Cancellation Campaign? Malawi's Perspective `Under the Microscope'." Presented at the Khanya College Winter School, Johannesburg, 4-11 July, Malawi Economic Justice Network. Ladd, Paul, 2003. "Too Hot to Handle? The Absence of Trade Policy from PRSPs." Christian Aid, UK. Le Houerou, Philippe and Robert Taliercio,2002. "Medium Term Expenditure Frameworks: From Concept to Practice. Preliminary Lessons from Africa." World Bank Africa Region Working Paper 28, Washington, D.C. Levinsohn, Jim, 2002. "The World Bank's Poverty Reduction Strategy Paper Approach: Good Marketing or Good Policy?" Paper prepared for the Group of 24 research program, Gerald R. Ford School of Public Policy, Department of Economics, University of Michigan, Ann Arbor, Michigan. Lucas, Henry, David Evans, and Kath Pasteur, 2004. "Research on the Current State of PRS Monitoring Systems." commissioned by the Department for International Development (DFID), The Institute of Development Studies (IDS), Sussex, UK. Martin, Matthew and Hannah Bargawi, 2004a. "The Role of the IMF in Low-Income Countries." Study for the Swedish Ministry of Finance and Foreign Affairs, Debt Relief International and Development Finance International, London, UK. 88 Martin, Matthew and Hannah Bargawi, 2004b. "Protecting Africa Against Shocks." Africa Commission Background Paper, Development Finance International, London, UK. Maxwell, Simon, 2005. "The Washington Consensus is Dead! Long Live the Meta- Narrative." Working Paper 243, Overseas Development Institute, London, UK. McKinley, Terry, 2004. "Economic Policies for Growth and Poverty Reduction: PRSPs, Neoliberal Conditionalities and `Post-Consensus' Alternatives." Paper presented at the IDEAs International Conference on The Economics of the New Imperialism, Jawaharlal Nehru University, New Delhi, India. McKinley, Terry, 2005. "MDG-Based PRSPs Need More Ambitious Economic Policies." United Nations Development Program Policy Discussion Paper, New York. McKinnon, Don, 2004. "Transforming the Business of Development: By Process, Not Paper Alone." Statement by the Secretary-General of the Commonwealth Secretariat, London, UK. McLeod, Helena, 2004. "Trade and Poverty Mainstreaming: Lessons from Southern Africa." (DRAFT), Department for International Development (DFID), London, UK. Marcus, Rachel and Jenni Marshall, 2004. "Tackling Childhood Poverty in Central and East Asia: Donor Aapproaches in Kyrgyzstan and Mongolia." CHIP Report No. 13, Childhood Poverty Research and Policy Centre, London. Marshall, Jenni and Esther Ofei-Aboagye, 2004. "Donors and Childhood Poverty in Sub- Saharan Africa: Approaches and Aid Mechanisms in Ghana and Tanzania." CHIP Report No. 12, Childhood Poverty Research and Policy Centre, London, UK. Martin, Matthew and Hannah Bargawi, 2004. "The Role of the IMF in Low-Income Countries. Study for Swedish Ministries of Finance and Foreign Affairs by Debt Relief International/Development Finance International." September 27, 2004, London, UK. McGillivray, Mark, 2005. "Aid Allocation and Fragile States." Background paper for the Senior Level Forum on Development Effectiveness in Fragile States, 13-14 January, London, commissioned by DFID for the Learning and Advisory Process on Difficult Partnerships of the OECD/DAC. National Democratic Institute and UNDP, 2004. "Legislative-Executive Communication on Poverty Reduction Strategies: Strengthening Parliamentary Involvement in the Millennium Development Goals and the Poverty Reduction Strategy Process." Parliament and Poverty Series, Toolkit no. 1, Washington, D.C. Nørlund, Irene, Tran Ngoc Ca, and Nguyen Dinh Tuyen, 2003. "Dealing with the Donors: The Politics of Vietnam's Comprehensive Poverty Reduction and Growth Strategy." Policy Papers 4/2003, Institute of Development Studies, University of Helsinki, Helsinki, Finland. Odereo, Kenneth K, 2004. "PRSPs in Decentralized Contexts: Comparative Lessons on Local Planning and Fiscal Dimensions, Uganda Study." 89 OED, 2002. "OED Review of the HIPC Initiative." Operations Evaluations Department, World Bank, Washington, D.C. OECD-DAC, 2004. Organisation for Economic Co-operation and Development­ Development Assistance Committee (OECD-DAC). "Results of OECD-DAC Survey on Harmonisation and Alignment." Paris, France. ----, 2005. DAC Guidelines and Reference Series: Harmonising Donor Practices for Effective Aid Delivery, Volume 2: Budget Support, Sector-Wide Approaches and Capacity Development in Public Financial Management. Paris, France. ---- , 2005. "Incentives for Harmonisation in Aid Agencies." Study prepared by the Overseas Development Institute for the OECD-DAC Working Party on Aid Effectiveness, Paris, France. ---- , 2005. "Harmonisation, Alignment, Results: Report on Progress, Challenges and Opportunties." Report prepared by the OECD-DAC Working Party on Aid Effectiveness, Paris, France. ---- , 2005. "Survey on Harmonisation and Alignment: Measuring Aid Harmonisation and Alignment in 14 Partner Countries." Paris, France. Overseas Development Institute, 2004. "PRSP Annual Progress Reports and Joint Staff Assessments-A Review of Progress." Briefing Note 9, London, UK. ----, 2004. "The PRS Process and DFID Engagement: Survey of Progress 2003." London, UK. ----, 2004. "Second Generation Poverty Reduction Strategies." London, UK. ----, 2005. "Incentives for Harmonization in Aid Agencies." Study prepared for the DAC Working Party on Aid Effectiveness, London, UK. Piccioto, Robert, Charles Alao, Eka Ikpe, Martin Kimani, and Roger Slade, 2005. "Striking a New Balance: Donor Policy Coherence and Development Cooperation in Difficult Environments." Background paper for the Senior Level Forum on Development Effectiveness in Fragile States, 13-14 January, London, commissioned by the Learning and Advisory Process on Difficult Partnerships of the Development Assistance Committee of the OECD. Renard, Robrecht and Nadia Molenaers, 2003. "Civil Society Participation in Rwanda's Poverty Reduction Strategy." IDPM­UA discussion paper 2003-5, Institute of Development Policy and Management, University of Antwerp, Belgium. Robb, Caroline M., 2003. "Poverty and Social Impact Analysis­Linking Macroeconomic Policies to Poverty Outcomes: Summary of Early Experiences." IMF Working Paper Wp/03/43, Washington D.C. Robinson, Shirley, 2003. "Children First in the Poverty Battle! A Review of the Poverty Reduction Strategy Papers in the Southern African Region--From a Child Rights Perspective." Save the Children, Sweden. Rogerson, Andrew, 2005. "What if Aid Harmonisation and Alignment Occurred Exactly as Intended­A Reality Check on the Paris Forum on Aid Effectiveness." Draft 90 Paper for an informal workshop co-hosted by the OECD Development Center and the Overseas Development Institute (ODI) in collaboration with the OECD-DAC, January 28, 2005. Runge, Peter and Klaus Wardenbach (eds), 2003. "Poverty Reduction and Crisis Prevention: How can Poverty Reduction be Shaped Conflict-Sensitively?" VENRO Project: Prospects for 2015­Combating Poverty Requires Involvement, VENRO, Germany. Sanchez, Diana and Katherine Cash, 2003. "Reducing Poverty or Repeating Mistakes?" A Civil Society Critique of Poverty Reduction Strategy Papers, Church of Sweden Aid, Diakonia, Save the Children Sweden and The Swedish Jubilee Network. Seppänen, Maaria, 2003. "Transforming the Concessional State? The Politics of Honduras' Poverty Reduction Strategy." Policy Paper 3/2003, Institute of Development Studies, University of Helsinki, Helsinki, Finland. Stewart, Frances and Michael Wang, 2003. "Do PRSPs Empower Poor Countries and Disempower the World Bank, or Is It the Other Way Round?" QEH Working Paper Series--QEHWPS108, Oxford, UK. Strategic Partnership with Africa, SPA-6 Budget Support Working Group (BSWG), 2004. "Mission to Mozambique, March-April." Stevens, Mike, 2004. "Institutional and Incentive Issues in Public Financial Management Reform in Poor Countries." World Bank, Washington, D.C. Southern African Regional Poverty Network (SARPN), 2005. "The Political Dimensions of Poverty Reduction­the Case of Zambia." selected papers from the Conference on Zambia's PRSP in Lusaka, 9-11 March, hosted by the Department of Development Studies at the University of Zambia and University of Duisberg- Essen. Swiss Coalition of Development Organizations, 2003. "Civil Society and Monitoring the PRSP in Burkina Faso: Achievements, Lessons Learned and Perspectives." Study Commissioned by State Secretariat for Economic Affairs (seco) and Swiss Agency for Development and Cooperation (SDC), Debt-for-Development Unit, Berne, Switzerland. Trocaire, 2004. "The Other Side of the Coin...: An Alternative Perspective on the Role of the International Monetary Fund in Low-Income Countries." Policy Briefing, Ireland. ----, 2004. "Are PRSPs Combating Rural Poverty in Honduras and Nicaragua? Lessons for a New Generation of PRSPs," Policy Briefing, Ireland. ----, 2005. "More Than a Numbers Game? Ensuring that the Millennium Development Goals Address Structural Injustice." UNDG (United Nations Development Group), 2004. "Guidance Note on UN Country Team Engagement in PRSPs." New York. ----, 2005. "Using Results-Based Frameworks in Fragile States." New York. 91 UNDP, 2004. "Overview of UNDP's Support to Poverty Reduction Strategies." New York. UNHCR, 2004, "Poverty Reduction Strategy Papers: A Displacement Perspective." Geneva, Switzerland. Wardenbach, Klaus (ed.), 2004. "How Can the Poor Gain Their Rights? Poverty Reduction and Human Rights." VENRO Project: Prospects for 2015 ­ Combating Poverty Requires Involvement, VENRO, Germany. Wardenbach, Klaus (ed.), 2003. "PRSP -- Prospects and Limits of Civil Society Involvement." VENRO Project: Projspects for 2015­Combating Poverty Requires Involvement, VENRO, Germany. Water and Sanitation Program, 2003. "Water Supply and Sanitation in Poverty Reduction Strategy Papers in Sub-Saharan Africa: Developing a Benchmarking Review and Exploring the Way Forward." World Bank, Washington D.C. Ware, Helen, 2003. "The World Bank, PRSPs and African Poverty." African Studies Association of Australasia and the Pacific, 2003 Conference Proceedings­African on a Global Stage. Weaver, Richard, 2004. "Poor Relations? PRSPs and the Response to HIV/AIDS and Children." HIV/AIDS Briefing Paper no. 2, Tearfund, UK. Wood, Angela, 2004. "One Step Forward, Two Steps Back: Ownership, PRSPs and IFI Conditionalities." A special report in the 60th year of the IMF/World Bank, World Vision International, UK. World Bank, 2004. "Poverty Reduction Strategies in Decentralized Contexts: Comparative Lessons in Local Planning and Fiscal Dimensions." A Draft Source Book: A Synthesis of Four Reports commissioned under a research program funded by a World Bank-Netherlands Partnership Program (BNPP) Trust Fund, Washington, D.C. ----, 2004. "A Review of Population, Reproductive Health, and Adolescent Health & Development in Poverty Reduction Strategies."Health, Nutrition and Population Central Unit, Washington, D.C. ----, 2004. "The Poverty Reduction Strategy Initiative: An Independent Evaluation of the World Bank's Support Through 2003." Operations Evaluation Department, Washington, D.C. ----, 2004. "A Review of Rural Development Aspects of PRSPs and PRSCs, 2000- 2004," Agriculture and Rural Development Internal Report, World Bank, Washington, D.C. World Development Movement, 2005, "Denying Democracy: How the IMF and World Bank Take Power from the People." London, UK. World Vision, 2005. "Poverty Reduction: Are the Strategies Working?" U.K. 92 Youash, Michael, 2003. "The PRSP Process and Parliament: Rethinking the Boundaries of Parliamentary Oversight." Gauteng Provincial Legislature, Johannesburg, South Africa. 93